It is shown that managing electricity consumption by regulating the power of consumer-regulators is an important factor in reducing the loads of industrial enterprises. It is shown that during peak hours of the power system, it is advisable to reduce the capacity of enterprises to level the overall schedule of electrical loads. The procedure for reducing the total load can be achieved by regulating the power consumption of consumers - regulators or by disconnecting them. To rank consumers-regulators, it is necessary to perform an optimization procedure. For the optimization procedure, we propose the use of combinatorial algorithms based on heuristic schemes for finding solutions. The method of finding the optimal list of such consumers using a genetic algorithm, a heuristic search method used to solve optimization and modeling problems by randomly selecting a combination and variation of the noise parameters using mechanisms similar to natural selection, is investigated. To solve the problem, a software model of the evolutionary process - the development of a certain population of individuals - is created. An algorithm is described that can be used to obtain the most optimal solution for the choice of consumers at an enterprise. The genetic algorithm is considered to solve the problem of selecting the composition of consumers at the lower level of the enterprise's power grid. In the process of selection, the functions of loss and number of switching are analyzed, which tend to minimize when searching for a solution. The program model of the evolutionary process includes the following stages: creation of the initial population; crossing; mutation. A detailed description of the genetic algorithm, its essence, and the principle of its application for load regulation by selecting the optimal consumer option for regulating the electric load are considered. The criteria for selecting the most optimal list of consumer-regulators are the functions of loss and the number of switching operations.
Read full abstract