ASSESSMENT OF THE VOLTAGE STABILITY LOAD INDEX OF ELECTRIC NETWORKS BASED ON THE ECHO STATES NETWORK USING THE PHASOR UNIT

Abstract

The issue of power system voltage stability, which is one of the key issues in the process of planning and operation of the power system, is considered in the paper. It is noted that neural networks have now been tested in a number of technological incidents in the form of power system voltage instability in various countries. The problem of possible risk of voltage failure due to the loss of its stability in the energy system during its operation is investigated. It is noted that in this regard, conducting a voltage stability analysis is an important procedure in order to timely identify buses with low reliability within the power system. Thus, the control personnel of the power system should take certain measures to avoid any cases of voltage drop within the system. The purpose of this paper is to present a new method for estimating the voltage stability load index (L-index) for a power system using data from optimally placed phasor units (PMUs). The voltage stability load index is evaluated in the work using a recurrent neural network known as the echo state network (ESN). The PMU optimal placement is made taking into account the island operating conditions. The results of the PMUs optimal placement for normal and boundary operating conditions, as well as the evaluation of the L-index using ESN in the IEEE 14 bus system, are presented. A technique for estimating the voltage stability load index in a power system based on a network of echo states using PMU measurements is presented. An estimate of the ESN performance for the L-index has been performed; the results have shown a high accuracy of its estimate under normal and disturbed conditions. It has been proven that ESN development is efficient and provides an accurate evaluation in computations.