An application of FBGs sensors to measuring the strains of industrial boiler hanger rods Authors: Magdalena Palacz, Bolesław Bąk, Łukasz Felkowski, Marek Krawczuk Published: 2025 ABSTRACT Background of the research presented is the undoubted fact that industrial boiler maintenance is essential for efficient operation and preventing failures. Traditional stress measurement methods for hanger rods (with hydraulic pressure pump) can be unreliable due to operator subjectivity. Therefore, the objective is to introduce a more accurate method for measuring the stress state of hanger rods in industrial boilers using fibre Bragg grating (FBG) sensors. The study compares traditional hydraulic pressure pump measurements with FBG sensors, which allow continuous strain measurement without shutting down the boiler. The new approach follows from the fact, that FBG sensors have been used in structural health monitoring but not in this specific application. The experimental strain measurements have been conducted on an industrial boiler hanger rod in an operational facility. The FBG sensors provided more reliable and continuous measurements compared to the hydraulic pressure pump method. After the analysis of the results obtained it has been clear that the use of FBG sensors offers a promising new approach for continuous strain measurement in boiler hanger rods, contributing to the development of diagnostic methods for estimating stress levels in industrial boilers, what is a promising conclusion. Introduction For the last few decades, researchers in structural health monitoring have made significant attempts to address the challenges of monitoring structures in harsh environments, particularly when it comes to strain measurements. For example, the development of robust sensor technology has resulted in the elaboration of sensors that can withstand extreme conditions such as high temperatures, corrosive environments, and mechanical stress. Recent developments in smart and networked sensors have significantly impacted structural health monitoring systems. These sensors can collect and transmit data in real time, enabling researchers to monitor structures in harsh environments online more effectively1. In addition, novel results for understanding how temperature variations, humidity, and vibrations impact deformation measurements are crucial for accurate monitoring and maintenance of structures in harsh environments. Ongoing research and advances in sensor technology continue to improve the monitoring capabilities and reliability of structural health monitoring systems under challenging conditions. Therefore, several measurement solutions have also been available to determine the uniaxial force value of the hook. Systems are widely known on the market, and some of them are strictly related to the pipeline support rod process2, but certain ones can be successfully applied to the boiler hanger rod system. The importance of proper maintenance The boiler hanger rod suspension system is designed to maintain a balanced force distribution according to the boiler load distribution design. Overload of the individual hanger rod can lead to unexpected areas of stress concentration in a boiler pressure part3. One of the diagnostic methods commonly used so far in industry to test the stress state of the hanger rod is based on the mechanical rule using the design of the hydraulic jack impact boiler suspension system and has been described by4. However, a practical application of a hydraulic jack system requires time-consuming preparation, including heavy hydraulic unit goods transportation to the site location. The steps of the hydraulic pressure setup procedure must be repeated one by one on each hanger rod to calculate the individual load force of the hanger rod. Those operations do not allow for the collection of input data for continuous boiler operation, but only for a limited time when the measurement system is performed. Hanger rod load force data of the bearing rod gives a picture of the force distribution in a boiler hanger. In that sense, the force adjustment steps must be followed for each individual hanger rod, taking into account the nearby hanger rods. The specific force value is applied for each by the set pressure of the hydraulic system in the hook rod, and then the hook rod is locked with counter nuts under the required conditions. The impact of the force adjustment is observed when one hanger rod is taken into account at a time. The periodic return to the previously adjusted hanger rod is required to check the impact of the already adjusted force on the hanger rod. This method is simple and easy to perform, but the results need to be evaluated, and the impact of the physical condition needs to be taken into account. Techniques used for measurement of force in hanger rods Another type of measurement method used so far for tension analysis is the application of conventional electrical strain gauge measurements5. However, this method is also known for its measurement uncertainty that has been widely expressed in detail in research publications6–8. Additionally, the high number of hanger rods in the boiler suspension system might cause difficulties in implementing proper wire connections, as well as reduce the possibility of further correct connection systems due to the complex configuration of the measurement set. Some researchers have also tried to overcome the problems of electrical strain gauges by using measurement systems based on acceleration sensors that measure changes in the vibration frequency of the stressed hanger rod, which have been presented as a stress monitoring system by8. However, the main disadvantage of the measurement techniques described above is that the boiler usually has to be taken out of operation to carry out such tests. The approaches described above do not provide continuous feedback. On the other hand, there are methods that allow for continuous measurement of the rod’s condition. Among others, a measurement method using FBG sensors is such an approach. FBG for strain/stress monitoring FBG sensors have emerged as a prominent technology for strain measurement in demanding environments. These sensors consist of a short segment of optical fibre that features periodic variations in its refractive index, known as the Bragg grating. This grating acts as a wavelength-specific reflector, reflecting light of a specific wavelength, termed the Bragg wavelength. The Bragg wavelength is directly determined by the period of the refractive-index variations within the fibre. Any alteration in the fibre’s physical parameters, such as strain or temperature, induces a shift in the Bragg wavelength. This shift is precisely measurable and can be correlated with the corresponding physical parameter. For longitudinal strain measurement, the FBG sensor is securely attached to the structure or material under investigation. When the structure undergoes longitudinal deformation, the FBG sensor experiences a change in length, resulting in a corresponding shift in the Bragg wavelength. Monitoring this wavelength shift allows an accurate determination of the longitudinal strain. The applicability of FBG sensors for longitudinal strain measurement is extensive and is widely used in the monitoring of structural health of civil infrastructure, including bridges, buildings, and pipelines. According to the present state of the art9 the FBG sensors offer several advantages for longitudinal strain measurement, including:High sensitivity - FBG sensors can detect minute changes in strain with exceptional accuracy.Distributed sensing - Multiple FBG sensors can be multiplexed along a single optical fibre, enabling distributed strain measurement along the fibre’s length.Immunity to electromagnetic interference - FBG sensors are impervious to electromagnetic interference, making them suitable for harsh environments.Long-term stability - FBG sensors exhibit excellent long-term stability, ensuring reliable and accurate strain measurements over extended periods.Some limitations of the FBG strain sensor applied for tests shall be also mentioned, those listed below have been based on the sensor manufacturer data sheet. Sensor limitations are specific to the specific type applied for the measurement system [source]:Operation temperature [°C] min -40 max +100.Operation and storage humidity \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\le$$\end{document} 95%.Fatigue [number of cycles] \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\epsilon _w= \pm 1000\mu /m$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta \epsilon _m \le 30\mu /m>>10^7$$\end{document}.Sensor bend radius must be bigger than 25 [mm].These characteristics make FBG sensors a valuable tool for strain measurement in various applications, particularly in demanding environments. As severe because of dust pollution and elevated ambient temperatures reaching 60 °C, industrial conditions should be considered. It is a typical operating environment for industrial boiler hanger rods. Therefore, due to their mechanical parameters, which are substantial in size, a specific configuration of the measurement system10–13 is essential. It must contain elements for temperature compensation, which can be achieved through the integration of an additional temperature sensor within the measurement system. This inclusion allows for the assessment of temperature cross-sensitivity and the coefficient of thermal expansion effect on the host material, thereby enhancing the quality of the measurement. Using FBG sensors also mitigates the complexity induced by cable connections, ensuring a durable and simplified linkage. Incorporating the FBG temperature compensation sensor within the same optical fibre as the strain FBG sensor does not compound the complexity of the measurement system. The merits of using FBG sensors include sustained signal stability in signal accuracy, attributable to a high-fidelity interrogator system. Furthermore, optical FBG sensors have been noted to have minimal signal loss over long distances. The superiority of FBG strain sensors relative to electrical strain gauges has gained broad recognition14–16, prompting the development of several international standards and guidelines to define measurement uncertainty, such as ISO 61757-1-1 and VDI/VDE 266017,18. High sensitivity - FBG sensors can detect minute changes in strain with exceptional accuracy. High sensitivity - FBG sensors can detect minute changes in strain with exceptional accuracy. Distributed sensing - Multiple FBG sensors can be multiplexed along a single optical fibre, enabling distributed strain measurement along the fibre’s length. Distributed sensing - Multiple FBG sensors can be multiplexed along a single optical fibre, enabling distributed strain measurement along the fibre’s length. Immunity to electromagnetic interference - FBG sensors are impervious to electromagnetic interference, making them suitable for harsh environments. Immunity to electromagnetic interference - FBG sensors are impervious to electromagnetic interference, making them suitable for harsh environments. Long-term stability - FBG sensors exhibit excellent long-term stability, ensuring reliable and accurate strain measurements over extended periods. Long-term stability - FBG sensors exhibit excellent long-term stability, ensuring reliable and accurate strain measurements over extended periods. Operation temperature [°C] min -40 max +100. Operation temperature [°C] min -40 max +100. Operation and storage humidity \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\le$$\end{document} 95%. Operation and storage humidity \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\le$$\end{document} 95%. Fatigue [number of cycles] \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\epsilon _w= \pm 1000\mu /m$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta \epsilon _m \le 30\mu /m>>10^7$$\end{document}. Fatigue [number of cycles] \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\epsilon _w= \pm 1000\mu /m$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta \epsilon _m \le 30\mu /m>>10^7$$\end{document}. Sensor bend radius must be bigger than 25 [mm]. Sensor bend radius must be bigger than 25 [mm]. Measurement uncertainty considerations Measurement uncertainty and the definition of true value are delineated by several standards, including the ISO Guide to the Expression of Uncertainty in Measurement (GUM) and the International Bureau of Weights and Measures (BIPM) as described in ISO / IEC Guide 9819,20. The VIM3 (measurement standard) is referenced for its dual approaches to measurement21. These guides play a pivotal role in the advancement of expressing measurement uncertainties22–26. Therefore, also a low-incertitude FBG interrogator is essential for the measuring system, equipped with traceable wavelength reference components to ensure the accurate detection of received wavelength values. FBG interrogators, which incorporate reference components aligned with the National Institute of Standards and Technology (NIST), facilitate internal calibration of the interrogator with each wavelength measurement. Although periodic calibration of the FBG interrogator is not mandatory due to the built-in reference components, additional calibration can improve accuracy27. The stability and repeatability of the interrogator are critical factors that affect the performance of the measuring system, and the resolution of the FBG sensor [\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m] is directly contingent upon the interrogator’s resolution in wavelength measurement [pm]. The performance and accuracy of the FBG sensor and interrogator system have been widely acknowledged27,28. Consequently, the overall accuracy of the components of the strain measurement system must be considered in a comprehensive way. Purpose and scope of the article The manuscript under consideration details the initial quantification of strain within a hanger rod located in a power boiler, with the objective of evaluating the fatigue levels of the corresponding expansion joint. This evaluation is predetermined by the dynamic inner force fluctuations of a designated hanger rod in the boiler’s operational phase. Quantification has been performed using FBG sensors, thus marking their novel application in this particular context. The use of these sensors is optimal due to their dual function of strain measurement and ambient temperature compensation. A distinctive characteristic of this measurement system is its capacity for continuous and non-invasive monitoring of the subject structure. The following sections of this document will provide a detailed structural analysis, including an assessment of its industrial relevance, the identification of critical components, and a discussion of the limitations of conventional methodologies. The subsequent section will describe the experimental framework in detail and provide a comprehensive explanation of the methodology and the underlying assumptions. The results of the analysis will be presented, including the measurement timeline and a selection of exemplary data representations. Finally, the manuscript will conclude with a synthesis of the findings and recommendations for future analytical efforts. Structure under investigation Fig. 1Method of suspending the CFB power boiler (1 – drum, 2 – furnace, 3 – separator, 4 – convection part, 5 - tubular heater, 6 – structural grid, 7 - hanger rods for furnace, 8 - hanger rods for separators, 9 – hanger rods for convection part) prepared using Inkscape v.1.3.2. Method of suspending the CFB power boiler (1 – drum, 2 – furnace, 3 – separator, 4 – convection part, 5 - tubular heater, 6 – structural grid, 7 - hanger rods for furnace, 8 - hanger rods for separators, 9 – hanger rods for convection part) prepared using Inkscape v.1.3.2. A series of experimental measurements have been carried out in one of the energy plants located in Poland on the operational performance of the existing combined heat and power plant (CHP) during its annual maintenance period. The measurement system has been installed in the circulating fluidised bed (CFB) boiler, with a maximum continuous rate (MCR) of 134.2 [kg/s] of steam load capacity. The study focused on the part of the convection cage of the CFB boiler shown as position 4 in Fig. 1.Fig. 2(a) Hanger rod assembly, (b) Dimensions of hanger rod under investigation, prepared using Inkscape v.1.3.2. (a) Hanger rod assembly, (b) Dimensions of hanger rod under investigation, prepared using Inkscape v.1.3.2. The CFB boiler is located within a structure subjected to severe industrial environmental conditions, characterised by particle contamination.Vibrational forces from the auxiliary boiler apparatus in operation, as well as from the processes of combustion and steam generation, have been measured. Temperature gradients ranging from 150 °C in the lower extremity of the insulated hanger rod to an ambient temperature of 50 °C, where unrestricted access to the hanger rod is allowed, have been documented. Throughout the testing phase, the thermal readings have been systematically documented and the total weight of the boiler unit has been determined as 2391 kN. The distribution of forces across individual hanger rods has also been monitored, allowing the development of various methodologies to quantify the forces, displacements, and strains exhibited by the boiler hanger rods. This ensures the measurement of these physical parameters with adequate accuracy.Fig. 3Hydraulic unit setup (1 – control unit, 2 – pressure measurement, 3 – Hydraulic pump, 4 – hydraulic hoses, 5 – structural grid, 6 – hydraulic jack, 7 – hydraulic jack plunger, 8 – hanger rod hex nut, 9 – additional support, 10 – counter nut, 11 – CFB boiler hanger rod, 12 – CFB boiler), prepared using Inkscape v.1.3.2. Hydraulic unit setup (1 – control unit, 2 – pressure measurement, 3 – Hydraulic pump, 4 – hydraulic hoses, 5 – structural grid, 6 – hydraulic jack, 7 – hydraulic jack plunger, 8 – hanger rod hex nut, 9 – additional support, 10 – counter nut, 11 – CFB boiler hanger rod, 12 – CFB boiler), prepared using Inkscape v.1.3.2. As illustrated in Fig. 1, the rods located at position 9 are essential to maintain the structural integrity of the convection cage. The additional hanger rods, identified as position 7 (a schematic view of a typical hanger rod assembly is presented in Fig. 2), serve as the suspension mechanism for the combustion chamber of the boiler at position 2, while those at position 8 perform a similar function for the separator at position 3. It is important to note that an unequal distribution of the boiler’s mass can induce disproportionate stress and deformation within the pressure components. This, in turn, can result in the development of stress fractures. The design of the Circulating Fluidized Bed (CFB) boiler requires that the upper hanger rods bear the weight of the boiler chamber and its constituent elements, thereby accommodating the anticipated thermal expansion during the transition from a stable, ambient state to an operational, elevated temperature state, as well as during operational load fluctuations within the Maximum Continuous Rating (MCR).These hanger rods are attached directly to the boiler and structural grid at position 6. The distribution of force across the individual hanger rods is affected by variations in the boiler’s operational load, as well as by alterations in mass and thermal expansion, resulting in shifts in the centre of gravity. Disproportionate force distribution can lead to material fractures and damage, potentially resulting in unanticipated steam release and necessitating an emergency cessation of operations to facilitate the cooling and subsequent repair of the boiler. Recovery efforts that are required to rectify such issues can have a detrimental effect on the operational reliability of the boiler and result in significant financial expenditure, both in terms of the costs of the repairs and the interruption of steam generation. However, these issues can be mitigated by real-time monitoring of the force distribution across the hanger rods, ensuring adherence to the prescribed design parameters throughout the boiler’s active and maintenance phases. The hanger rod under examination has an external diameter of 0,060 m and an aggregate length of 3,110 m, as depicted in Fig. 2(b). One end of the hanger rod is equipped with a lug, which is designed to attach to the support elements of the boiler pressure part via pin insertion. The opposing end of the hanger rod is manufactured with threading, enabling it to be attached to nuts, which are to be applied after the rod has been inserted through the structural grid.The hanger rod is fabricated from carbon steel, specifically the 16Mo3 variant, which is known for its strength and resilience at high temperatures. The detailed assembly of the hanger rod to the pressure part on the bottom side and to the structural grid on the upper part is illustrated in Fig. 2(a). A method for assessing the hanger rods load force Non-invasive hydraulic system This part of the article discusses a non-invasive hydraulic system to determine the axial forces that are applied to the hanger rods. It is simple and commonly used but requires the transport of heavy equipment and sometimes gives results based only on the operator’s experience, which can be misleading. The proposed system uses a hydraulic pump, hoses and a hydraulic jack to measure the load on each hanger rod, as shown in Fig. 3. An additional support system equipped with a counternut (components 9 and 10) is required to enable this measurement. The hanger rod (component 11) is attached to the lower part of the CFB’s convection cage and transfers its weight to the underlying structural grid (component 5). The hydraulic pump (component 3), controlled by the regulating unit (component 1), generates the hydraulic pressure that is transmitted through the hoses (component 4) to the hydraulic jacks (component 6). These jacks are placed on the structural grid (component 5) under the additional mechanical support of the hanger rod (component 9). The support is secured to the hanger rod (component 11) by a threaded lock nut (component 10). To initiate measurement, the hex nut (component 8) supporting the hanger rod in the convection cage is loosened. Subsequent incremental increases in hydraulic pressure raise the hydraulic jack plunger (component 7) until it contacts the underside of the additional mechanical support. This process continues until the hanger rod nut (component 8) separates from the structural grid (component 5), causing the support and attached hanger rod to rise. This occurs when the upward hydraulic force exerted by the piston exceeds the downward load on the hanger rod from the convection cage. The hydraulic pressure applied is recorded by the pressure gauge (component 2). This data, combined with the dimensions of the hydraulic jack plunger, is used as input to calculate the axial force on the boiler hanger rod during the test. Accurate determination of the moment of release of the hanger rod nut from the structural grid is essential for correct recording of hydraulic pressure and subsequent calculation of the actual load on the hanger rod.Fig. 4Support deflection details for hanger rod lift up (1- CFB boiler hanger rod, 2- structural grid, 3- hanger rod hex nut, 4- additional support, 5- counter nut, 6- hydraulic jack, 7- hydraulic jack plunger), prepared using Inkscape v.1.3.2. Support deflection details for hanger rod lift up (1- CFB boiler hanger rod, 2- structural grid, 3- hanger rod hex nut, 4- additional support, 5- counter nut, 6- hydraulic jack, 7- hydraulic jack plunger), prepared using Inkscape v.1.3.2. The deformation of the support is illustrated in Fig. 4. First, pressure is applied to the hydraulic jack piston (position 6 in Fig. 4a) until it contacts the underside of the additional support (position 4 in Fig. 4a). Subsequently, the continued upward movement of the hydraulic jack piston (position 6 in Fig. 4b) induces deflection of the support (position 4 in Fig. 4b), which absorbs a significant portion of the applied force without vertical displacement. Further increases in hydraulic pressure initiate upward movement of the hanger rod, despite the already deflected counter nut support. When the hydraulic jack piston has reached its maximum displacement (position 6 in Fig. 4c) and the deflection of the support is at its maximum (position 4 in Fig. 4c), the applied force overcomes the resistance of the support, causing it to lift together with the hanger rod (position 1 in Fig. 4c). This process releases the hanger rod nut (position 3 in Fig. 4c) from the structural grid (position 2 in Fig. 4c).Fig. 5FGB measuring system configuration (1 - PC computer, 2 - Ethernet data transmission, 3 - FBG interrogator, 4 - fibre optic, 5 - FBG strain sensor, 6 - FBG temperature sensor, 7 - CFB boiler hanger rod), prepared using Inkscape v.1.3.2. FGB measuring system configuration (1 - PC computer, 2 - Ethernet data transmission, 3 - FBG interrogator, 4 - fibre optic, 5 - FBG strain sensor, 6 - FBG temperature sensor, 7 - CFB boiler hanger rod), prepared using Inkscape v.1.3.2. FBG-based measurement setup The FBG measurement system used in the analysis discussed and shown in Fig. 5 is composed of a PC computer (component 1) connected via Ethernet data transmission (component 2) to an FBG interrogator (component 3). The interrogator is connected to a single mode optical fibre (component 4), which is equipped with a strain FBG sensor (component 5) and a temperature FBG sensor (component 6). The fibre connection uses the FC/APC standard.Fig. 6Sensor installation at CFB industrial boiler convection cage hanger rod, (a) general view, (b) enlarged view, edited in Iknscape v.1.3.2. Sensor installation at CFB industrial boiler convection cage hanger rod, (a) general view, (b) enlarged view, edited in Iknscape v.1.3.2. The installation of the specific sensor is shown in Fig. 6. An FBG strain sensor has been attached to the surface of the target hanger rod to allow real-time monitoring of longitudinal force variations. Before sensor attachment, the surface had been subjected to rigorous preparation, including thorough cleaning to remove paint and rust. To enhance adhesion, a slightly roughened surface has been achieved through abrasive and fine cleaning processes. A fast setting cold-curing adhesive has been used to secure the sensor. A reference FBG temperature sensor to compensate for the influence of temperature variations on the strain measurements has been installed at the same time (glued in its central part only to avoid strain impact to its grating, as recommended in the instruction29). By correlating the temperature cross-sensitivity of the strain sensor with the thermal expansion coefficient of the hanger rod, thermal compensation calculations (using the data of the reference sensor) have been performed. The resulting strain-compensated values have then been converted into load forces for each individual hanger rod. The performance of the strain and temperature sensors, as shown in the table 1, is characterised by a sensitivity of 1.2 picometres per microstrain and a resolution of 0.5 microstrain30 for strain and a sensitivity of 30 picometres per °C and a resolution of 0.02 °C for temperature sensor respectively.Table 1FBG sensors characteristics.Sensor FS62PSS31k-factorn.a.0.79 ± 0.03Sensitivitypm/(\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m)1.2Resolution\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m0.5Measurement range\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m (%)±20000Sensor FS63LTS29Sensitivitypm/°C30Temperature compensation factor(\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m)/°C20Resolution °C0.02Max calibration error °C0.79 ± 0.03Measurement range °C-40 ... +100 Table 1FBG sensors characteristics.Sensor FS62PSS31k-factorn.a.0.79 ± 0.03Sensitivitypm/(\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m)1.2Resolution\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m0.5Measurement range\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m (%)±20000Sensor FS63LTS29Sensitivitypm/°C30Temperature compensation factor(\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m)/°C20Resolution °C0.02Max calibration error °C0.79 ± 0.03Measurement range °C-40 ... +100 FBG sensors characteristics. For the measurements reported in this study, a single-mode FBG sensor capable of measuring both strain and temperature has been selected. This sensor operates within the interrogator’s spectral range without overlap with other FBG systems, accurately translating wavelength shifts into the corresponding physical parameters. According to the manufacturer’s specifications, the FBG exhibited negligible strain-induced birefringence and was readily adaptable to the target structure. Furthermore, its low optical reflectance and compliance with the fibre bending guidelines ensured minimal signal attenuation. The sensor has been attached to the specimen using the method illustrated in Fig. 6. The passive nature of FBG technology, coupled with its inherent self-referencing, long-term stability, and compatibility with most interrogators, made it suitable for the harsh environment while delivering a wavelength measurement resolution of 0.5 [pm], translating to a strain sensor resolution of ±0.5 [\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m]. The interrogator’s stability and reproducibility, critical for accurate measurements, were determined to be 1 pm, aligned with the National Institute of Standards and Technology (NIST) Technical Note 1297 definitions of repeatability and reproducibility. Instrument calibration against a NIST-traceable gas cell ensured measurement accuracy. For further technical details, see the HBM FiberSensing technical notes32. Force calculation considerations A comprehensive analysis of the uncertainty of the FBG measurement system is essential, given the complexity of the system. The system consists of an FBG strain sensor, a temperature FBG sensor, and an interrogator, each element potentially contributing to the overall uncertainty. FBGs are typically used as strain sensors through direct application to deforming materials, and FBGs also function as temperature sensors when strain is isolated from the grating region33. The unique properties of FBGs enable the creation of transducers in which measurements are transferred to the deforming FBG. The operational principle of FBGs is based on the reflection of a narrow spectral band centred at the Bragg wavelength (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda _0$$\end{document}) when broadband light is incident on the grating. The remaining light propagates through the fibre. Variations in reflected wavelength (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta \lambda$$\end{document}) are induced by strain and temperature, making FBGs suitable for sensing applications33. The quantitative relationship between these phenomena is mathematically expressed in Eq.(1)33:1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \frac{\Delta \lambda }{\lambda _0} = k\Delta \epsilon +\left( \alpha + \zeta \right) \Delta T \end{aligned}$$\end{document}where:k is the dimensionless strain sensitivity of the sensor (also known as gauge factor or k factor). It is indicated on the characteristic sheet. For the FS62PSS Patch Strain Sensor \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$k=0,79\pm 0,03$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document} is the thermal expansion of silica,\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\zeta$$\end{document} is thermo-optic coeffcient of the fibre.When taking measurements with FBG-based sensors, care must be taken of the effects that temperature changes impose on that measurement. The measured strain on a mechanically loaded specimen that is also subjected to a temperature change is affected by the load (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon _{load}$$\end{document}) and the thermal expansion of the specimen material (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon _{CTE}$$\end{document}):2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \varepsilon _{RealE} = \varepsilon _{load} + \varepsilon _{CTE}. \end{aligned}$$\end{document}The thermally induced wavelength variation on an FBG that is attached to a certain material can be taken from Eq.(1) considering that there is no strain and that the thermal expansion of the silica is also constrained because the sensor is attached to the material and silica will not expand. item3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \frac{\Delta \lambda }{\lambda _0} = \zeta \Delta T \end{aligned}$$\end{document}The temperature compensation of strain sensors using temperature measurements should take into account the compensation of both thermal effects. Therefore, when considering encapsulated sensors, the temperature effect on the measurement is experimentally determined and provided in the sensor documentation, as it might differ from the theoretical silica behaviour. This influence is expressed by the Temperature Cross Sensitivity (TCS) as follows:4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \frac{\Delta \lambda }{\lambda _0} \cdot \frac{1}{k} \cdot 10^6= \varepsilon +CTE\cdot \Delta T + TCS\cdot \Delta T \end{aligned}$$\end{document}where:CTE is the coefficient of thermal expansion of the material onto which the strain sensor is installed (also known as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document} of the material). CTE is expressed in (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upmu$$\end{document}m/m)/°C,\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon$$\end{document} is the pure mechanical strain, in \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upmu$$\end{document}m/m, distinguished from the temperature-induced strain signal,TCS is the thermal cross-sensitivity of the strain sensor, indicated on the characteristic sheet. The TCS is expressed in (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upmu$$\end{document}m/m)/°C.From the previous equation, it is straightforward to calculate the temperature-compensated strain signal, denoted by \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon$$\end{document}. To induce temperature variations, an auxiliary temperature sensor can be employed in conjunction with the strain sensor, ensuring that both sensors are at the same temperature. Temperature can be measured using temperature sensors based on FBG or other technologies. k is the dimensionless strain sensitivity of the sensor (also known as gauge factor or k factor). It is indicated on the characteristic sheet. For the FS62PSS Patch Strain Sensor \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$k=0,79\pm 0,03$$\end{document}, k is the dimensionless strain sensitivity of the sensor (also known as gauge factor or k factor). It is indicated on the characteristic sheet. For the FS62PSS Patch Strain Sensor \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$k=0,79\pm 0,03$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document} is the thermal expansion of silica, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document} is the thermal expansion of silica, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\zeta$$\end{document} is thermo-optic coeffcient of the fibre. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\zeta$$\end{document} is thermo-optic coeffcient of the fibre. CTE is the coefficient of thermal expansion of the material onto which the strain sensor is installed (also known as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document} of the material). CTE is expressed in (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upmu$$\end{document}m/m)/°C, CTE is the coefficient of thermal expansion of the material onto which the strain sensor is installed (also known as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document} of the material). CTE is expressed in (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upmu$$\end{document}m/m)/°C, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon$$\end{document} is the pure mechanical strain, in \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upmu$$\end{document}m/m, distinguished from the temperature-induced strain signal, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon$$\end{document} is the pure mechanical strain, in \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upmu$$\end{document}m/m, distinguished from the temperature-induced strain signal, TCS is the thermal cross-sensitivity of the strain sensor, indicated on the characteristic sheet. The TCS is expressed in (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upmu$$\end{document}m/m)/°C. TCS is the thermal cross-sensitivity of the strain sensor, indicated on the characteristic sheet. The TCS is expressed in (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upmu$$\end{document}m/m)/°C. Care must be taken on the reference instant that should correspond simultaneously to the “zero strain” moment and the reference temperature:5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \varepsilon = \frac{\Delta \lambda }{\lambda _0} \cdot \frac{1}{k} \cdot 10^6 - (CTE + TCS)\cdot (T-T_0) \end{aligned}$$\end{document}where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda _0$$\end{document} is the reference wavelength at zero moment of the strain sensor, in nm, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_0$$\end{document} is the temperature, in °C, measured at the same instant. The configuration equation for the temperature FBG sensor has the following form:6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} T = (\lambda - \lambda _0)^2 \cdot S_2 + (\lambda - \lambda _0) \cdot S_1 + S_0 \end{aligned}$$\end{document}where:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda _0$$\end{document} is the reference wavelength at 30 °C [nm],\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda$$\end{document} is the measured wavelength [nm],\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_2$$\end{document} is the 2nd order calibration factor [ °C/nm2],\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_1$$\end{document} is the 1st order calibration factor [°C/nm],\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_0$$\end{document} is the 0 order calibration factor [°C].The traced uncertainty is defined by the manufacturer in the sensor calibration sheet for the full range of temperature measurement as equal to 0.4 [°C]. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda _0$$\end{document} is the reference wavelength at 30 °C [nm], \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda _0$$\end{document} is the reference wavelength at 30 °C [nm], \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda$$\end{document} is the measured wavelength [nm], \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda$$\end{document} is the measured wavelength [nm], \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_2$$\end{document} is the 2nd order calibration factor [ °C/nm2], \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_2$$\end{document} is the 2nd order calibration factor [ °C/nm2], \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_1$$\end{document} is the 1st order calibration factor [°C/nm], \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_1$$\end{document} is the 1st order calibration factor [°C/nm], \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_0$$\end{document} is the 0 order calibration factor [°C]. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_0$$\end{document} is the 0 order calibration factor [°C]. To determine the forces in the rods from the measured deformations, Hooke’s law has been applied. Hooke’s law is applicable to steel rods within the elastic region, where stress is linearly proportional to strain. This linear elastic behaviour persists until the material’s yield strength is reached. The relationship between tensile stress \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sigma$$\end{document} and strain \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon$$\end{document} is defined by the modulus of elasticity E of steel defined by the equation:7\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} {\sigma } = {E} \cdot {\varepsilon } \end{aligned}$$\end{document}where:E is hanger rod’s steel Young modulus,\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sigma$$\end{document} is tensile or compress stress,\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon$$\end{document} is strain as proportional deformation in the linear elastic range.The tensile stress of the steel hanger rod \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sigma$$\end{document} for its length L and cross-sectional area A equals:8\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \sigma = \frac{F}{A} \end{aligned}$$\end{document}where:F is the force applied [N],A is the initial cross section area of steel hanger rod [mm2].Based on the simple co-combination of Eq.(7) and Eq.(8), the linear correlation between strain values and the cross-section force can be obtained with the known material Young modulus. Changes in the area of the cross section A can be neglected as the measurement experiment is limited to the linear elastic range of the hanger rod. In such case the force is also linear to the strain and can be easily calculated as:9\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} {F} = {\varepsilon }\cdot {A} \cdot {E}. \end{aligned}$$\end{document}In order to confirm that experimental strain measurements have been limited to the elastic range of the hanger rod, the Finite Element Method (FEM)-based analysis has been performed. This analysis justifies the application of the linear Eq.(9) for force calculation. The force values have been subsequently determined using the following rod geometry values: length 3.110 [m], diameter 0.0060 [m], and material type: 16Mo3 steel. The material properties are enumerated in Table 2.Table 2Rod material 16Mo3 properties.ParameterValueUnitTemperature20°CYield point Rp0,2260MPaTensile strength Rm440MPaYoung’s modulus211.28GPa E is hanger rod’s steel Young modulus, E is hanger rod’s steel Young modulus, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sigma$$\end{document} is tensile or compress stress, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sigma$$\end{document} is tensile or compress stress, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon$$\end{document} is strain as proportional deformation in the linear elastic range. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon$$\end{document} is strain as proportional deformation in the linear elastic range. F is the force applied [N], F is the force applied [N], A is the initial cross section area of steel hanger rod [mm2]. A is the initial cross section area of steel hanger rod [mm2]. Table 2Rod material 16Mo3 properties.ParameterValueUnitTemperature20°CYield point Rp0,2260MPaTensile strength Rm440MPaYoung’s modulus211.28GPa Rod material 16Mo3 properties. Based on the FEM analysis results it has been possible to state that the values of the determined forces are within the linear range of strain limits. The corresponding numerical data have been gathered in Table 3.Table 3Measured strain values and calculated force values.FBG strain measurements [\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m]Force calculated FBG [kN]Force calculated FEM [kN]2716.1215.968651.3750.8414687.2286.31209124.86123.56272162.49160.80344205.50203.37 Table 3Measured strain values and calculated force values.FBG strain measurements [\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m]Force calculated FBG [kN]Force calculated FEM [kN]2716.1215.968651.3750.8414687.2286.31209124.86123.56272162.49160.80344205.50203.37 Measured strain values and calculated force values. Fig. 7Comparison of hanger rod force values calculated by FEM and obtained from FBG measurements, prepared using QGLE V. 4.3.0. Comparison of hanger rod force values calculated by FEM and obtained from FBG measurements, prepared using QGLE V. 4.3.0. Experimental results The measurement procedure performed within this study can be outlined as follows. A reference force of known magnitude has been applied to the analysed hanger rod using hydraulic actuators, and the resulting reaction of the rod has been measured using FBG sensors. The deformation and stress values in the rod have been then determined from the recorded data taking into account the FBG sensor parameters as k-factor, initial wavelength, CTE, TCS, T and T0 - rasulting in application of Eq.(9). Such approach constitutes an indirect method of determining the load state in an analysed hanger rod. Experiment validation through FEM analysis To check whether the determined stress level falls within the elastic range of the strain/stress characteristics a comparative analysis has been performed of the stress values determined on the basis of recorded wavelength changes (FBG sensor) and using the developed numerical FEM model incorporating the experimental hanger rod and hydraulic jack system (see Fig. 8). The numerical model includes the material data of the rod, i.e. Young’s modulus and geometry data. The Fig. 7 shows a comparison of stress-strain relationship obtained by the FBG measurements and following calculations as well as obtained by FEM analysis. The detailed values have been presented in Table 3. It may be observed that the values in both cases are within the elastic range. The linear force function presented in Eq.(9) is applicable only if the hanger rod behaves in a purely elastic mode, characterised by a linear stress-strain relationship. A finite element model incorporating the experimental hanger rod and hydraulic jack system (see Fig. 4)) has been developed to identify the stress-strain section of the curve. The result of this analysis can be seen in Fig. 7.Fig. 8Hanger rod 3D model for FEM analysis: (a) schematic overview: 1 – counter nut, 2 – additional support, 3 – hanger rod hex nut, 4 – CFB boiler hanger rod, 5 – support welded to structural grid, 6 – hydraulic jack; (b) initial conditions: F1 – hydraulic jack force, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon$$\end{document} – maximum strain measured, prepared using Ansys 2025 R2. Hanger rod 3D model for FEM analysis: (a) schematic overview: 1 – counter nut, 2 – additional support, 3 – hanger rod hex nut, 4 – CFB boiler hanger rod, 5 – support welded to structural grid, 6 – hydraulic jack; (b) initial conditions: F1 – hydraulic jack force, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon$$\end{document} – maximum strain measured, prepared using Ansys 2025 R2. In order to answer the question of how the hydraulic method affects the strength of the hanger rod, a numerical simulation of this measurement has been carried out. The experimental data have given a maximum longitudinal strain of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon$$\end{document} = 344 [\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}m/m]. Subsequently, this value has been utilised as the applied load for hanger rod position 4 in the FEM analysis, as illustrated in Fig. 8. The FEM analysis has then been carried out in two separate phases: first, the hanger rod has been subjected to the measured strain alone; then an additional hydraulic jack force has been applied to simulate system balancing, with the aim of achieving a loose position of the hanger rod hex nut (position 3) relative to the structural grid (position 5). The resulting hydraulic jack force is equivalent to the force exerted on the hanger rod. The results of the analysis have been presented in Fig. 9.Fig. 9Hanger rod normal stress: (a) before hydraulic jack load force applied, (b) after hydraulic jack load force applied, prepared using Ansys 2025 R2. Hanger rod normal stress: (a) before hydraulic jack load force applied, (b) after hydraulic jack load force applied, prepared using Ansys 2025 R2. As illustrated in Fig. 9, the normal stress value in the hanger rod at the location of the FBG strain sensor attachment is 73 MPa. This value is consistent with the material properties of the 16Mo3 steel used to manufacture the hanger rod, in particular it remains below the yield point (detailed material specifications have been presented in Table 2. The FEM analysis validates the equation Eq.(9) used to calculate the force of the hanger rod at the location of the FBG strain sensor. Furthermore, the results of the analysis indicate that the maximum calculated hanger rod stress do not exceed 170 MPa at any point along the length of the rod, thus remaining below the yield point of the 16Mo3 steel throughout the component. The linear force function derived from Eq.(9), which is based on the strain measurement data for the hanger rod operating within its linear elastic range, is shown in Fig. 10. As demonstrated in figure, the graphs provide a comparison of the two linear relationships between deformation and force in the tested rod. In both cases, the force value has been determined in six measurement steps – the force value has been established at the achieved deformation level. For the curve marked FBG, Eq.(9) was applied, and for the curve marked HydJack, the value indicated by the manometer included in the measuring assembly was recorded. The curves represent the mean value from four measurement series.Fig. 10Hydraulic jack force calculation with its maximum error based on pressure gauge 1.0 CL accuracy, prepared using prepared using QGLE V. 4.3.0.Table 4Standard uncertainties values of the measured hanger rod forces - type A evaluation, 4 independent observations, applied pressure values in the first column.ApplPress[bar]meanqmean [kN]variances2(qk)std deviations q(k)variance of the means2(qmean)std dev of the means(qmean)5017.623311.06353.32612.76581.663010050.331110.32003.21242.58001.606215089.610414.51353.80973.62841.9048200125.902629.35425.41807.33862.7090250165.181922.00824.69135.50212.3456300204.71000.55520.74510.13880.3725 Hydraulic jack force calculation with its maximum error based on pressure gauge 1.0 CL accuracy, prepared using prepared using QGLE V. 4.3.0. Table 4Standard uncertainties values of the measured hanger rod forces - type A evaluation, 4 independent observations, applied pressure values in the first column.ApplPress[bar]meanqmean [kN]variances2(qk)std deviations q(k)variance of the means2(qmean)std dev of the means(qmean)5017.623311.06353.32612.76581.663010050.331110.32003.21242.58001.606215089.610414.51353.80973.62841.9048200125.902629.35425.41807.33862.7090250165.181922.00824.69135.50212.3456300204.71000.55520.74510.13880.3725 Standard uncertainties values of the measured hanger rod forces - type A evaluation, 4 independent observations, applied pressure values in the first column. Fig. 11Comparative results of force/strain relation for analysed hanger rod obtained by hydraulic jack and by FBG sensor for 4 independent measurement series from (a) to (d), prepared using QGLE V. 4.3.0. Comparative results of force/strain relation for analysed hanger rod obtained by hydraulic jack and by FBG sensor for 4 independent measurement series from (a) to (d), prepared using QGLE V. 4.3.0. Experiment performance on a real object In order to control the level of possible measurement error, a hydraulic pressure gauge of the highest possible accuracy class, designated Class 1.0, has been selected for the hydraulic unit. The maximum permissible error is ±1.0\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\%$$\end{document} of its full-scale value, which corresponds to a calculated force error of ±7 [kN] for the specified hydraulic jack dimensions. A graphical representation of this error range is provided in Fig. 10. It is important to note that when the maximum error exceeds the specified accuracy of the gauge, the nonlinearity and hysteresis of the gauge must be considered as characteristic parameters. A rigorous uncertainty assessment has been performed according to the Guide for the Expression of Uncertainty in Measurement (GUM)20. This international standard provides a comprehensive framework for evaluating and expressing uncertainty in measurement systems, encompassing experimental design, data analysis, and complex components. In a manner consistent with the ISO/IEC Guide, a Type A evaluation of standard uncertainty has been performed based on the experimental data set. The resulting uncertainty estimates are graphically represented in Fig. 10 and detailed values of the standard uncertainties of the measured hanger rod force are reported in Table 4. The calculation of measurement uncertainty indicates that, despite the standard deviations observed in the measurements, the measured values can be considered to be at an acceptable level. However, efforts should be carried out to improve the accuracy of the measurements in order to avoid these discrepancies. As demonstrated in Fig. 11, a comparison is presented between the values recorded by a hydraulic jack and those determined by FBG sensors. The subsequent graphs illustrate four exemplary series of measurements (Fig. 11 a-d) for sequential pressure values (from 50 to 300 bar in increments of 50) applied by the cylinder. The difference between the recorded and calculated force/tension values can be observed clearly. The analysis of measurement errors indicates that the values calculated based on the FBG sensor measurements are closely aligned with the numerically calculated values (Table 3), thereby providing a more reliable description of the strain and stress field occurring in the rod during the measurements.Fig. 12Registered value: (a) of the measured force in the rod, (b) of the force in the rod with the influence of TCS and CTE of the rod material, (b) of external temperature, prepared using QGLE V. 4.3.0. Registered value: (a) of the measured force in the rod, (b) of the force in the rod with the influence of TCS and CTE of the rod material, (b) of external temperature, prepared using QGLE V. 4.3.0. As illustrated in Figure 12, the experimental research that has been performed on an industrial facility using an FBG sensor can be summarized. The figure contains graphs showing the changes in the forces acting on a selected hanger rod over the period of ten days of continuous measurements. For the purpose of readability, the graphs have been divided into specific single days. The force values have been calculated from the recorded FBG strain rate curves. The curves marked (a) represent the determined force value in the rod without consideration of the influence of temperature fluctuations, while the curves marked (b) demonstrate the actual force values, including thermal stresses. During the experiments, the boiler was in operation of a test run of the boiler from cold to operating condition. It is worth noting that within this time frame, the following activities are documented on the boiler: on the 6th of October, the burners have been activated, resulting in an increase in the rod load due to higher temperature of the boiler. On the 7th of October, the temperature at which the process of drying the boiler refractory commenced reached 350 °C. This was followed by the initiation of the second burner and the subsequent fuel feeding to the boiler. On the 8th of October, the process of drying the boiler refractory continued. The process of cooling the boiler was initiated on the the 9th of October and continued until the 16th of October. The temperature curves of the rod have been marked with (c) in all cases. The graph from the 7th of October demonstrates a marked increase in the force value of the analysed rod, resulting from the variability of the temperature field at that time. A comparison of the obtained values with the numerically obtained results allows us to conclude that the use of FBG is a good direction in this application. Conclusions and discussion In conclusion, the presented research shows that hydraulic systems are widely used in industrial applications due to their versatility and ability to generate significant forces. Although the technology is generally sufficiently precise for many engineering tasks, its inherent limitations limit its use in real-time, continuous monitoring applications. However, hydraulic systems remain a viable solution for industrial processes that are temporally constrained. The performance of these systems can be optimised through the implementation of rigorous calibration and maintenance protocols. The integration of reference force transmitters during hydraulic jack calibration offers a potential pathway to enhance overall system accuracy. Moreover, the hydraulic unit method presents several limitations. Primarily, it precludes the establishment of repeatable measurement conditions. Furthermore, the method necessitates the utilisation of substantial equipment, including a hydraulic pump and a lifting jack, to generate the requisite lifting forces. Furthermore, the restriction of measurements to static boiler states (either cold or hot) prevents continuous monitoring during operation. The testing process is time-consuming, involving multiple repeated measurements for each hanger rod, and this is another significant drawback. In addition, the inherent hysteresis in the hydraulic pump control unit, in conjunction with the necessity for regular system calibration, results in a reduction in both accuracy and efficiency. In contrast, FBG strain sensors offer several key advantages. Their response to applied strain is highly linear, ensuring accurate data acquisition. Additionally, FBG sensors demonstrate exceptional repeatability, enabling reliable and consistent measurements. The inherent long-term stability of the technology ensures the integrity of the data over extended periods. The utilisation of FBG sensors facilitates both short-term and long-term continuous monitoring applications, thereby providing real-time insights into the behaviour of structures. The multiplexing of multiple sensors onto a single optical fibre further simplifies system expansion and enhances measurement capabilities. Consequently, the utilisation of FBG sensors for the surveillance of the stress state of industrial boiler hanger rods emerges as a promising technique. Nevertheless, further experimental work is necessary to develop an effective algorithm for the processing of measurement signals and to establish a method for determining the pre-stress value in the rods under investigation. This approach holds great promise, but it is essential to meticulously plan the subsequent experiments on the actual object, taking into account the operational characteristics of the facilities, which require special attention to the continuity of the technological equipment.