Multi-port network based modeling and selection of capacitor for desired voltage regulation of a standalone six-phase short-shunt induction generator for application in remote areas

• Multi-port network analysis simplifies SPIG modeling, enhancing efficiency. • Binary search algorithm (BSA) streamlines excitation capacitor selection. • Focus on determining shunt and series capacitors for voltage regulation. • Validation of simulation results ensures method reliability. This paper gives a straightforward method to determine the values of excitation capacitors of a standalone short-shunt six-phase induction generator (SPIG) to maintain the voltage profile within predetermined percentage of voltage deviation (VD). In this envisioned study, the value of the capacitor is meticulously chosen to optimize the number of capacitor switching, ensuring minimal system cost and complexity. The theory of multi-port network analysis has been applied for modelling of the SPIG, thus, the complex mathematical derivation to obtain the model equations is avoided. The system is expressed as a multivariable nonlinear optimization problem. The resultant admittance of the SPIG is calculated from its per phase equivalent circuit and is used as an objective function, which is solved using Binary Search Algorithm (BSA). The main novelty of this work is the determination of the model equations of the SPIG system in an efficient and simple way using the multi-port network analysis approach. Along with this, the BSA is employed for optimal selection of excitation capacitors because of its simplicity and less computational time. The results, on a 3.7 kW induction machine, reveal that to maintain a 4 % VD, a fixed series capacitor of 140 µF and two switched shunt capacitors (34.4 µF, 91.8 µF) are required. For 2 % VD, four shunt capacitors (24.2µF, 36.2µF, 64.7µF, 91.2µF) are necessary. The performance of the machine is evaluated with the help of magnetic characteristics and other equations obtained from its per phase equivalent circuit. The experimentation has been carried out in a hardware prototype system developed in the laboratory. The experimental and the simulated results are compared and found that both are nearly same for different operating conditions, which indicates the efficacy of the proposed approach.