Optimal sizing and placement of STATCOM, TCSC and UPFC using a novel hybrid genetic algorithm-improved particle swarm optimization Authors: Urbanus Mwanzia Ngei, Abraham Mutunga Nyete, Peter Musau Moses, Cyrus Wekesa Published: 2024 ABSTRACT The increase in global power demand has caused most of today's power networks to become overloaded especially in Sub-Saharan Africa. The increased load demand can be met through expansion of existing generation and transmission system. However, construction of new power infrastructure is limited by financing and technical constraints. Thus, power networks have been left to operate at overload conditions with high power losses and many power quality (PQ) problems. Flexible AC Transmission System (FACTS) devices can improve the power transfer capability of the existing transmission networks without the need of constructing new power infrastructure. In this paper, a multi-objective function comprising of minimization of power loss (PL), voltage deviation (VD) and operational cost (OC) was formulated and solved using a novel algorithm. A novel Genetic Algorithm-Improved Particle Swarm Optimization (GA-IPSO) technique is proposed in this paper for optimization of size and location of FACTS devices. Static Synchronous Compensator (STATCOM), Thyristor Controlled Series Capacitor (TCSC) and Unified Power Flow Controller (UPFC) are the three FACTS devices considered. The proposed technique was validated using IEEE-33 Bus Test System, which is a popular benchmark Radial Distribution System (RDS). The three FACTS devices were optimized separately and also in a combined manner. Under the separate optimization, the size and location of individual FACTS devices were optimized. For combined optimization, the sizes and locations of more than one device were optimized in the same test system. For separate optimization, UPFC produced the best results by reducing the active power losses by 38.44 % and OC from $1.59×105 to $ 1.15×105. Under the combined optimization, combination of TCSC, STATCOM and UPFC gave better results by achieving active power loss reduction of 56.09 % and reducing OC from $1.59×105 to $ 1.03×105. Comparison of GA-IPSO technique with other algorithms such as Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Improved Grey Wolf Optimization (IGWO) and Differential Evolution Algorithm (DEA) showed that the proposed hybrid technique was superior and more efficient in solving the FACTS optimization problem. Introduction The world has experienced a drastic increase of power demand in the recent years. The witnessed increase in electricity demand has been caused by several factors such as population growth, urbanization and heightened industrial activities [1]. One solution for meeting the increased load demand is through expansion of the existing generation and transmission networks. However, expansion of the power infrastructure is not a feasible solution because of the economic and technical constraints [2]. Furthermore, expansion of existing transmission network is an inappropriate remedy because it can cause congestion, undesirable environmental impacts and health hazards [3]. Another alternative is to increase the transfer capability of power networks by using conventional compensators such as capacitor banks. However, conventional compensators cause harmonics and lead to extreme switching transients. Thus, most of the power networks are left to operate at an overload condition with high power losses, which affects the stability and security of the power system [4,5]. The growth of the global load demand is expected to increase in the future with an approximated demand of 1134.6 GW by the year 2040 [1]. This will increase power quality issues and make future power networks unreliable. Hence, optimal utilization of the existing power networks is the best solution to maintain healthy power networks. FACTS allows for efficient utilization of the existing transmission and generation capacities with less investment cost [6]. FACTS devices efficiently mitigates voltage instability issue and losses among other power quality problems [7]. These devices use swift power electronics to improve power transfer capability and controllability of power networks [8,9]. Unlike the conventional electromechanical compensators, FACTS devices are compact, which makes them suitable for improving current highly interconnected power systems without adding bulkiness. Since FACTS devices use solid state technology to improve power system performance, these devices do not inject harmonics to the system if controlled by robust controllers and algorithms. FACTS devices can control reactive power under dynamic state; which is suitable for mitigating sudden unexpected changes occasioned by the intermittent Renewable Energy Sources (RES) in today's power networks [10]. Impedance, phase angle and voltage magnitude are fundamental attributes of electrical power systems. The ability of FACTS devices to alter these basic components of power systems allows for enhancement of electrical power networks [11]. The concept of FACTS devices was first introduced in 1999 by Hingorani and Gyugyi [3]. Since then, more advanced and useful FACTS devices have emerged. Different FACTS devices exist and the classification of these devices depend on the method applied to connect them to the power system network. In terms of evolution, FACTS devices can be categorized in two distinct generations. The first generation involve FACTS devices that employ the use of thyristor reactors [3]. Thyristor Controlled Series Capacitor (TCSC) is an example of a FACTS device under this initial generation. Voltage Source Converters (VSC) characterizes FACTS devices under the second generation and a common example is the Static Synchronous Compensator (STATCOM) [5]. Based on the topology used to connect FACTS devices to power network, these devices can be classified as series, shunt and shunt-series [12]. For series FACTS devices, the device is in series with the power network. For a system supplying an inductive load, the voltage drop is Vdrop=IRcosφ+IXLsinφ. Connecting a series FACTS device of reactance XC reduces the line reactance to XL−XC, which lowers the line's voltage drop. TCSC, Static Series Synchronous Compensators (SSSC) and Thyristor Switched Series Reactor (TSSR) are examples of series FACTS devices. FACTS devices that are connected to power system through parallel connection are referred to as shunt FACTS. This category of FACTS devices operate by consuming or injecting reactive power to the network [13]. Example of shunt FACTS include Static Var Compensator (SVC) and STATCOM. FACTS devices that utilize both series and shunt components to control power system parameters fall under series-shunt FACTS. Thyristor Controlled Phase Shifting Transformer (TCPST) and UPFC are examples of series-shunt FACTS. Optimization is important in engineering field and all other scientific fields because it allows for efficient use of the available resources. To reap the maximum benefits of using FACTS devices in power networks, optimization of their size and position is inevitable. Failure to optimize the rating of these devices and location where they should be positioned on a power network can increase the cost of running power networks because FACTS devices require a high initial investment cost [14]. Recently, many researchers have directed their attention in finding the best methods that can be applied to place FACTS devices in power systems. Analytic approaches and use of meta-heuristic algorithms are some of the techniques that have been employed to solve the FACTS optimization problem [1]. Meta-heuristic algorithms are robust and flexible and these desirable attributes have made many studies to apply meta-heuristic algorithms to solve the highly complex and highly constrained problem of FACTS devices optimization [5,12,15]. In Ref. [14], GA was applied to optimize TCSC and SVC with an objective to minimize voltage deviation without considering the cost of the FACTS. Adaptive Moth Flame Optimization Algorithm (AMFOA) was used in Ref. [16] to optimize TCSC to improve system power transfer. Again, the investment cost of the FACTS device was ignored in the formulation. Neglecting the cost of the FACTS devices makes it impossible to test their economic feasibility when applied to a particular network. In [16], STATCOM was optimized using Particle Swarm Optimization (PSO) and the authors did not include the cost in their work. Reference [17] used Particle Swarm Optimization (PSO) for STATCOM optimization on a power network. However, the authors used constant acceleration coefficients limiting the particles' ability to exploit the search space in its entirety. Use of constant acceleration coefficients causes the standard PSO to have reduced exploitation capabilities and can lead to premature convergence. A multi-objective optimization of UPFC was done in Ref. [18] using Lightning Algorithm (LA). The study fell short in testing the same approach on other FACTS devices rather than relying on a single device when testing the method. Different types of FACTS devices control different power system parameters. Thus, combining different FACTS devices can result in better control and performance of power networks. FACTS devices were optimized in Ref. [19] using Artificial Bee Colony Algorithm (ABCA) with two objectives of reducing losses and improving system stability. As other many studies in this research area, authors in Ref. [19] ignored FACTS cost in their design. In Ref. [20], Imperialist Competitive Algorithm (ICA) was used to minimize voltage deviation by optimal placement of TCSC and TCPST. The work did not incorporate the cost for the FACTS installation in the multi-objective function. In Ref. [21], Cuckoo Search Algorithm (CSA) was used to optimize STATCOM without consideration of the FACTS cost. Also, this work did not test the proposed approach in other FACTS devices. In Ref. [22], Firefly Algorithm (FA) was applied to optimize the position of UPFC while CSA was used to optimize it's size. Although this hybrid approach incorporated the cost of UPFC in the objective function, the study considered only one type of FACTS devices. Table 1 is a summary of some past works in the area of FACTS optimization.Table 1Past works on flexible AC transmission system (FACTS) optimization problem.Table 1ReferenceFACTS DevicesObjective(s)Optimization Approach[14]TCSC, SVCVoltage deviation, line violationGA[16]TCSCPower transfer capabilityAMFOA[17]STATCOMPower loss, costPSO[18]UPFCFuel cost, voltage deviationLA[19]SVC, TCSC, STATCOMPower losses, system stabilityABCA[20]TCSC, TCPSTVoltage deviationICA[21]STATCOMPower lossesCSA[22]UPFCPower losses, voltage stabilityFA-CSA[23]TCSCLine utilization factorPSO[24]TCSC, SVCPower losses, voltage deviationPSO[25]STATCOMPower lossesPSO Table 1Past works on flexible AC transmission system (FACTS) optimization problem.Table 1ReferenceFACTS DevicesObjective(s)Optimization Approach[14]TCSC, SVCVoltage deviation, line violationGA[16]TCSCPower transfer capabilityAMFOA[17]STATCOMPower loss, costPSO[18]UPFCFuel cost, voltage deviationLA[19]SVC, TCSC, STATCOMPower losses, system stabilityABCA[20]TCSC, TCPSTVoltage deviationICA[21]STATCOMPower lossesCSA[22]UPFCPower losses, voltage stabilityFA-CSA[23]TCSCLine utilization factorPSO[24]TCSC, SVCPower losses, voltage deviationPSO[25]STATCOMPower lossesPSO Past works on flexible AC transmission system (FACTS) optimization problem. A review of the existing literature in FACTS optimization as shown in Table 1 reveal that PSO is a commonly used method when optimizing the location and size of FACTS devices. Features such as less parameters and ease of implementation makes PSO suitable for solving the FACTS optimization problem that is non-linear and involves many constraints [12]. However, conventional PSO suffer from the problem of slow convergence and getting stuck at the local optimum [4]. This paper proposes Genetic Algorithm-Improved Particle Swarm Optimization (GA-IPSO) as a candidate algorithm for finding the optimal size and location of FACTS. The hybrid approach exploits the strengths of the two algorithms while overcoming the weaknesses of the individual optimization techniques. By utilizing the GA-IPSO, the exploitation of the search space will be improved and optimal sizes and positions of different types of FACTS devices will be obtained. A multi-objective function consisting of PL, VD and OC is used. This addresses a common gap where many studies do not consider the FACTS investment cost when optimizing the FACTS devices as indicated in Table 1. In addition, the work is a trailblazer in the field of FACTS optimization because it involves combined optimization of TCSC, STATCOM and UPFC; as opposed to many studies that only consider one type of FACTS device. A combination of different FACTS with unique features can offer better performance of power network and this paper optimizes different combinations of devices from each of the major categories of FACTS devices; shunt (STATCOM), series (TCSC) and series-shunt (UPFC). This paper's main contributions are.i)Use of a novel hybrid GA-IPSO algorithm to optimize the location and size of FACTS devices. Hybridization of GA with Improved PSO that uses constriction coefficient restricts the algorithm from getting trapped to local optima.ii)Combined optimization of different types of FACTS devices to improve performance of power network while ensuring economical operation of the entire power system.iii)Inclusion of the investment cost of FACTS devices in a multi-objective function of FACTS optimization to ensure economical operation. i)Use of a novel hybrid GA-IPSO algorithm to optimize the location and size of FACTS devices. Hybridization of GA with Improved PSO that uses constriction coefficient restricts the algorithm from getting trapped to local optima. Use of a novel hybrid GA-IPSO algorithm to optimize the location and size of FACTS devices. Hybridization of GA with Improved PSO that uses constriction coefficient restricts the algorithm from getting trapped to local optima. ii)Combined optimization of different types of FACTS devices to improve performance of power network while ensuring economical operation of the entire power system. Combined optimization of different types of FACTS devices to improve performance of power network while ensuring economical operation of the entire power system. iii)Inclusion of the investment cost of FACTS devices in a multi-objective function of FACTS optimization to ensure economical operation. Inclusion of the investment cost of FACTS devices in a multi-objective function of FACTS optimization to ensure economical operation. Modelling of FACTS FACTS devices utilize power electronics technology to improve controllability and the capability of a power grid to deliver power. Through FACTS devices, power transfer parameters of a network can be altered to a achieve a new level of efficiency from an existing network [26]. Power transfer through a transmission system is defined by equation (1).(1)P=VsVrX∗sin(θs−θr)Where Vs is the voltage at the sending side, Vr represents the voltage at the receiving side, x is the total reactance of the line, while θs and θr are phase angles at the sending and receiving side respectively. STATCOM improves power transfer by controlling the voltage, TCSC alters the line reactance to improve the power transmitted; while UPFC can control the voltage, power angle and reactance to adjust the power transmitted [27]. Fig. 1 represents the different parameters controlled by different types of FACTS devices.Fig. 1Parameters controlled by different flexible AC transmission system devices.Fig. 1 Parameters controlled by different flexible AC transmission system devices. The combination of different types of FACTS devices allows for control of all the three parameters, which improves controllability and power transmitted leading to improved performance of the overall electrical network [28]. In this paper, the major classes of FACTS devices are considered: series (TCSC), shunt (STATCOM) and series-shunt (UPFC). TCSC modelling TCSC is one of the mostly used series FACTS devices that adds either inductive or capacitive reactance to the transmission line. This device has a thyristor-controlled reactor shunted with a fixed capacitor [29]. By changing the firing angle α of the thyristor, the compensator's reactance can be varied, which alters the overall impedance of the transmission line to improve power transfer. The relationship between the reactance of a TCSC and the firing angle is represented in equation (2).(2)XTCSC=XCXL(α)XL(α)−XCWhere XC is the capacitive reactance of the fixed capacitor and XL(α) is inductive reactance of the thyristor-controlled reactor. The value of XL(α) is changed by changing the firing angle, and this alters the reactance of TCSC; which is abbreviated as XTCSC. Fig. 2 is a schematic diagram of a TCSC connected between Bus i and Bus j.Fig. 2Schematic diagram of thyristor controlled series capacitor (TCSC).Fig. 2 Schematic diagram of thyristor controlled series capacitor (TCSC). TCSC can be modelled as a variable reactance that is connected in series with the impedance of the transmission line as shown in Fig. 3.Fig. 3Equivalent Circuit of thyristor controlled series capacitor (TCSC).Fig. 3 Equivalent Circuit of thyristor controlled series capacitor (TCSC). The inclusion of TCSC changes the line's reactance as described in equations (3), (4).(3)ZTL=Rij+jXij(4)XN=Xij−XTCSC Further simplification of equation (4) yields equations (5), (6) where the ratio of the TCSC reactance to the line reactance before compensation determines the rate of compensation.(5)XN=(1−τ)Xij(6)τ=XTCSCXijWhere ZTL is the impedance of the transmission line, Rij and Xij are the line's resistance and reactance respectively between buses i and j, XTCSC is the reactance of TCSC, XN is the line's new reactance after introduction of the FACTS device and τ is the rate of compensation. Changing the lines reactance alters the conductance and the susceptance of the line as described in equations (7), (8) respectively.(7)Gij=RijRij2+XN2(8)Bij=RijRij2+XN2 The power flow between bus i and j after incorporating TCSC is describe by equations (9), (10), (11), (12) [30].(9)Pij=|Vi|2Gij−|Vi‖Vj|(Gijcosδij+Bijsinδij)(10)Qij=−|Vi|2Bij−|Vi‖Vj|(Gijsinδij−Bijcosδij)(11)Pji=|Vj|2Gij−|Vi‖Vj|(Gijcosδij−Bijsinδij)(12)Qji=−|Vj|2Bij−|Vi‖Vj|(Gijsinδij+Bijcosδij)Where δij is the voltage phase angle difference between bus i and bus j. STATCOM modelling STATCOM is a shunt-connected FACTS device that consists of a Voltage Source Converter (VSC) and a coupling transformer [31]. Fig. 4 represents a STATCOM connected in parallel with bus i.Fig. 4Schematic diagram of static Synchronous compensator (STATCOM).Fig. 4 Schematic diagram of static Synchronous compensator (STATCOM). STATCOM is modelled as a voltage source that is in series with the impedance of the coupling transformer. This model of the STATCOM is represented in Fig. 5.Fig. 5Equivalent Circuit of static Synchronous compensator (STATCOM).Fig. 5 Equivalent Circuit of static Synchronous compensator (STATCOM). Through the VSC, DC input voltage is converted to AC output voltage of controllable amplitude, phase and frequency. STATCOM works by delivering or absorbing reactive power from the system [32]. Under circumstances when the voltage amplitude of the system is smaller than that of the output of VSC (VSTATCOM>VBus), reactive power is supplied to the system by STATCOM as indicated in Fig. 5. Contrary, when the voltage amplitude of the system is bigger than that of the VSC output (VSTATCOM UPFC + STATCOM > UPFC > TCSC + UPFC > STATCOM > TCSC > STATCOM + TCSC. The operating cost for all the different FACTS configurations is illustrated in Fig. 15. More economic benefits can be achieved if the three FACTS devices are combined and used in the same power network. Since each FACTS device has unique attributes, optimal combination of the FACTS leads to a high performing power network. Use of a single UPFC is more economically feasible on IEEE-33 bus network as compared to a combination of TCSC + UPFC and STATCOM + UPFC.Fig. 15Operating cost for separate and combined installation of FACTS devices.Fig. 15 Operating cost for separate and combined installation of FACTS devices. Fig. 16 compares the voltage profiles of the system under the different FACTS configuration for combined optimization case. The voltage profiles obtained after using combined FACTS are better as compared to the voltage profile of the base system. However, the best voltage profile is obtained when STATCOM, TCSC and UPFC are jointly optimized together and installed in the test power network as shown in Fig. 16.Fig. 16Voltage profiles for combined optimization of FACTS.Fig. 16 Voltage profiles for combined optimization of FACTS. Comparison of GA-IPSO approach with GA and IPSO To show the superiority of GA-IPSO algorithm, convergence curves for all the seven cases considered in this paper were compared under identical number of fitness evaluations (NFE). The convergence curves for the three cases considered under separate installation of FACTS are shown in Fig. 17. Under combined installation of FACTS, the convergence curves for the four cases considered are illustrated in Fig. 18. The initial population for all the algorithms is set at the same value at the start to achieve a fair comparison. Under separate installation of FACTS, GA-IPSO found the global optimum faster than GA and IPSO for all the three cases as shown in Fig. 17. This trend repeated itself under the combined installation of FACTS devices with an exception of only case (STATCOM + TCSC + UPFC) as shown in Fig. 18. Thus, judging from the convergence curves, GA-IPSO converges faster than GA and IPSO when solving the problem of FACTS optimization. The superior convergence behavior of GA-IPSO can be attributed to hybridization that balances exploitation and exploration. Hybridization also reduces time complexity of GA-IPSO by simplification of the crossover process as explained in section 4.4. Hence, apart from improving the accuracy of the solution, GA-IPSO reduces the computation time when used in FACTS optimization problem.Fig. 17Convergence curves for separate installation of FACTS devices (Case 1).Fig. 17Fig. 18Convergence curves for combined installation of FACTS devices (Case 2).Fig. 18 Convergence curves for separate installation of FACTS devices (Case 1). Convergence curves for combined installation of FACTS devices (Case 2). The performance of GA-IPSO was compared with that of GA and IPSO as delineated in Table 6 and Table 7. In Table 6, GA-IPSO achieved the highest Power Loss Reduction (PLR) for all the three cases considered under separate installation of FACTS devices. Similarly, GA-IPSO reduced power losses by the highest percentage except for one case (STATCOM + TCSC) under the combined installation of FACTS as shown in Table 7. Overall, GA-IPSO out performed GA and IPSO for the six out of the seven cases considered in this paper. This shows prolific convergence traits of GA-IPSO algorithm when solving the problem of FACTS optimization. Again, Table 6, Table 7 shows that GA-IPSO converges faster than GA and IPSO in the cases considered except for one case (STATCOM + TCSC + UPFC). Fast convergence of GA-IPSO is attributed to the reduced computational burden of the proposed algorithms.Table 6Comparative results between GA, IPSO and GA-IPSO for separate installation of FACTS.Table 6FACTSMeasureGAIPSOGA-IPSO(Proposed)TCSCPLR (%)3.202.183.58Time (s)3.212.962.58STATCOMPLR (%)3.523.524.21Time (s)2.742.702.67UPFCPLR (%)23.1129.3538.44Time (s)4.854.173.54Key: PLR-Power Loss Reduction.Table 7Comparative results between GA, IPSO and GA-IPSO for combined installation of FACTS.Table 7FACTSMeasureGAIPSOGA-IPSO(Proposed)TCSC + UPFCPLR (%)30.5630.2431.17Time (s)10.236.173.22UPFC + STATCOMPLR (%)40.2342.1242.82Time (s)7.977.926.51STATCOM + TCSCPLR (%)3.124.503.07Time (s)3.843.112.54STATCOM + TCSC + UPFCPLR (%)51.1752.3256.09Time (s)8.896.777.44Key: PLR-Power Loss Reduction. Table 6Comparative results between GA, IPSO and GA-IPSO for separate installation of FACTS.Table 6FACTSMeasureGAIPSOGA-IPSO(Proposed)TCSCPLR (%)3.202.183.58Time (s)3.212.962.58STATCOMPLR (%)3.523.524.21Time (s)2.742.702.67UPFCPLR (%)23.1129.3538.44Time (s)4.854.173.54Key: PLR-Power Loss Reduction. Comparative results between GA, IPSO and GA-IPSO for separate installation of FACTS. Key: PLR-Power Loss Reduction. Table 7Comparative results between GA, IPSO and GA-IPSO for combined installation of FACTS.Table 7FACTSMeasureGAIPSOGA-IPSO(Proposed)TCSC + UPFCPLR (%)30.5630.2431.17Time (s)10.236.173.22UPFC + STATCOMPLR (%)40.2342.1242.82Time (s)7.977.926.51STATCOM + TCSCPLR (%)3.124.503.07Time (s)3.843.112.54STATCOM + TCSC + UPFCPLR (%)51.1752.3256.09Time (s)8.896.777.44Key: PLR-Power Loss Reduction. Comparative results between GA, IPSO and GA-IPSO for combined installation of FACTS. Key: PLR-Power Loss Reduction. Comparison of GA-IPSO approach for FACTS optimization with other works The use of GA-IPSO was effective in solving the FACTS optimization problem. In Ref. [2], PSO algorithm was used for combined optimization of TCSC and UPFC and active power losses were reduced by 20.60 %. In this work, optimal allocation of TCSC and UPFC by GA-IPSO algorithm reduced the system's active power losses by 31.17 %, which shows the effectiveness of the hybrid optimization approach. Use of GA in Ref. [47] to optimize UPFC reduced the system losses by 1.25 %. Optimization of UPFC in IEEE 33 bus system using Differential Evolution Algorithm (DEA) reduced the system losses by 0.015 % in Ref. [48]. In this paper, installation of single UPFC using the GA-IPSO technique reduced the losses of the test power system by 38.44 %. When compared to Refs. [2,5,17,48] where Whale Optimization Algorithm (WOA), Autonomous Group Particle Swarm Optimization-Grey Wolf Optimization (AGPSO-GWO), Adaptive Particle Swarm Optimization (APSO) and Improved Grey Wolf Optimization (IGWO) were used respectively to optimize FACTS size and location, the proposed GA-IPSO approach produced better results as shown in Table 8.Table 8Comparison of GA-IPSO with different optimization algorithms.Table 8FACTSAlgorithmPower Loss ReductionFACTSAlgorithmPower Loss ReductionTCSCIGWO [49]3.35 %STATCOMAPSO [17]2.87 %AGPSO-GWO [5]3.34GA-IPSO4.21 %GA-IPSO3.58 %UPFCGA [46]1.25 %TCSC + UPFCWOA [2]21.18 %DEA [47]0.015 %PSO [2]20.60 %IGWO [49]13.66 %GA-IPSO31.17 %GA-IPSO38.44 % Table 8Comparison of GA-IPSO with different optimization algorithms.Table 8FACTSAlgorithmPower Loss ReductionFACTSAlgorithmPower Loss ReductionTCSCIGWO [49]3.35 %STATCOMAPSO [17]2.87 %AGPSO-GWO [5]3.34GA-IPSO4.21 %GA-IPSO3.58 %UPFCGA [46]1.25 %TCSC + UPFCWOA [2]21.18 %DEA [47]0.015 %PSO [2]20.60 %IGWO [49]13.66 %GA-IPSO31.17 %GA-IPSO38.44 % Comparison of GA-IPSO with different optimization algorithms. Conclusion Different Meta-Heuristic Optimization (MHO) algorithms have been applied extensively to solve the FACTS optimization problem. This paper focused on optimization of size and location for Static Synchronous Compensator (STATCOM), Thyristor Controlled Series Capacitor (TCSC) and Unified Power Flow Controller (UPFC). Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) are popular algorithms used in optimization of FACTS devices. The standard forms of these algorithms suffer from the problem of local trapping. For improved convergence characteristics and efficiency, a hybrid approach (GA-IPSO) has been suggested in this paper. The GA-IPSO hybrid approach was applied to optimize FACTS devices on IEEE-33 Bus System. TCSC, STATCOM and UPFC were optimized both individually and in a combined manner. The FACTS devices were optimized to minimize power losses, voltage deviation and operating cost. The weights for the multi-objective function used in this paper were ω1= 0.8 for PL, ω2= 0.1 for VD and ω3= 0.1 for OC. These weight factors provided the best balance of the objective function with the highest significance given to power losses. Under separate allocation of FACTS devices, UPFC produced the best results. Under Combined optimization of FACTS devices, a combination of STATCOM, TCSC and UPFC emerged as the best combination that offers minimum power losses and minimum operating costs. When compared to other algorithms such as PSO, GA and DEA, the proposed hybrid technique produced better results at a good convergence rate. Since the model developed in this work allows for optimization of different FACTS configurations, the paper findings can assist power system planners in examining and selecting the most appropriate FACTS that fits their power networks. One challenge of the proposed GA-IPSO is that it has many tuning parameters because it combines the parameters of the individual GA and IPSO. Thus, applying the proposed method in a highly dimensional problem is worth of attention. In future, the authors intend to apply and evaluate the proposed GA-IPSO algorithm on practical and benchmark functions with different dimensions. CRediT authorship contribution statement Urbanus Mwanzia Ngei: Writing – original draft, Software, Methodology, Investigation, Data curation, Conceptualization. Abraham Mutunga Nyete: Supervision, Methodology, Conceptualization. Peter Musau Moses: Writing – review & editing, Supervision, Methodology, Conceptualization. Cyrus Wekesa: Writing – review & editing, Supervision, Methodology, Conceptualization. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.