The Model Order Reduction For The Load Frequency Control Of Two Area Power System Network Using Genetic Algorithm And Pole Clustering

Abstract

The article presents the model order reduction of the load frequency control (LFC) of the multi-area or two area power system network using genetic algorithm (GA) and pole clustering method. Load disturbance causes persistent frequency deviations in the multi-area or two-area power system network which have an impact on other parameters. The complete system has been developed by considering various generating units of each area such as solar photovoltaic (SPV), wind turbine (WT), battery energy (BESS), and conventional source (CS) which are connected through AC grid. The system has been modelled in terms of transfer function and the order of the overall transfer function is too high. It is need to reduce the order of transfer function which will be assessed separately for both numerator and denominator. The coefficient of denominator of the transfer function has been determined by using pole clustering method and coefficient of numerator of the transfer function has been determined by using genetic algorithm. The performance of frequency deviation is analysed using performance measures like integral square error (ISE), integral absolute error (IAE), and integral time absolute error (ITAE). To analyse the frequency deviation with 1% disturbances, a network of two area power systems has also been established. For the 1% disturbances in the two-area model, GA achieves the lowest frequency deviation and least mechanical deviation in contrast to the traditional technique. In addition to this, ISE, IAE and ITAE are also found to be minimum with GA in contrast to higher order system. This all shows the superiority of GA for reducing the model order reduction.