Application of Meta-Heuristic Optimization Algorithms in Electric Power Systems

Abstract

Optimization of solutions on expansion of electric power systems (EPS) and their control plays a crucial part in ensuring efficiency of the power industry, reliability of electric power supply to consumers and power quality. Until recently, this goal was accomplished by applying classical and modern methods of linear and nonlinear programming. In some complicated cases, however, these methods turn out to be rather inefficient. Meta-heuristic optimization algorithms often make it possible to successfully cope with arising difficulties. State estimation (SE) is used to calculate current operating conditions of EPS using the SCADA measurements of state variables (voltages, currents etc.). To solve the SE problem, the Energy Systems Institute of Siberian Branch of Russian Academy of Sciences (ESI of SB RAS) has devised a method based on test equations (TE), i.e. on the steady state equations that contain only measured parameters. Here, a technique for EPS SE using genetic algorithms (GA) is suggested. SE is the main tool for EPS monitoring. The quality of SE results determines largely the EPS control efficiency. An algorithm for exclusion of wrong SE calculations is described. The algorithm using artificial neural networks (ANN) is based on the analysis of results of the calculation performed solving the SE problem with different combinations of constants. The proposed procedure is checked on real data.

Volterra Models in Load Leveling Problem

“…The detailed description of regularizing direct numerical algorithms is given in [29]. It is worthwhile to note that it is very difficult to apply these algorithms to solve equation (1) in the form of (3) because of the kernel discontinuities (2). The adaptive mesh should depend on the curves of the jump discontinuity for each number N of divisions of the considered interval and, therefore, this mesh cannot be linked to the errors in the source data.…”

“…(2) Here 0 ( ) ≡ 0, 0 ( ) < 1 ( ) < ⋯ < ( ) ≡ , (0) = 0. It is assumed that the kernels ( , ) and the right-hand side ( ) in the equation (1) are continuous and sufficiently smooth functions. The functions…”

“…A number of alternative methods have been examined to solve this challenging problem. For example, Lagrangian relaxation, mixed integer linear programming, particle swarm optimization, evolutionary methods, and genetic algorithms, [1]. These methods have many advantages and disadvantages compared to the traditional approach.…”