Optimization of a power supply system is one of the main directions in power engineering research. The reactive power compensation reduces active power losses in transmission lines. In general, researches devoted to allocation and control of the compensation units consider this issue as a static optimization problem. However, it is dynamic and stochastic optimization problem that requires a real-time solution. To solve the dynamic optimization NP-hard problem, it is advisable to use Swarm Intelligence. This research deals with the problem of the compensation units power control as a dynamic optimization problem, considering the possible stochastic failures of the compensation units. The Particle Swarm Optimization and the Bees Algorithm were applied to solve it to compare the effectiveness of these algorithms in the dynamic optimization of a power supply system.

Deep reactive power compensation allows for reduction of active power losses in transmission lines of power supply systems. The efficiency of the compensation depends on the allocation of reactive power compensation units (RPCUs) at the nodes of a network. In general, investigations devoted to the study of optimal allocation of the compensation units have revealed that it is a static and deterministic optimization problem that can be solved by heuristic methods. However, in real systems, it is reasonable to consider such optimization problems, taking into account the dynamic and stochastic properties of the problems. These properties are the result of equipment failures and operational changes in technical systems. In addition, optimizing the allocation of the compensation units is the NP-hard multifactor problem. Under these circumstances, it is advisable to use the swarm intelligence algorithms. Swarm intelligence is a relatively new approach to solving the optimization problem, which takes inspiration from the behaviour of ants, birds, and other animals. Advantages of swarm algorithms are most evident if problems involve the dynamic or stochastic nature of the objective function and constraints. Contrary to a number of similar studies, this research considers the problem of the optimal allocation of compensation units as a dynamic problem, taking into account the possible random failures of the compensation equipment. The optimization problem has been solved by two Swarm Intelligence algorithms (the Particle Swarm optimization and the Artificial Bee Colony optimization) and Genetic algorithms. It has been aimed at comparing the effectiveness of the algorithms for solving such problems. It was found that swarm algorithms could be successfully applied in the operation control of compensation units in real-time.

The issue of electric power grid mode of optimization is one of the basic directions in power engineering research. Currently, methods other than classical optimization methods based on various bio-heuristic algorithms are applied. The problems of reactive power optimization in a power grid using bio-heuristic algorithms are considered. These algorithms allow obtaining more efficient solutions as well as taking into account several criteria. The Firefly algorithm is adapted to optimize the placement of reactive power sources as well as to select their values. A key feature of the proposed modification of the Firefly algorithm is the solution for the multi-objective optimization problem. Algorithms based on a bio-heuristic process can find a neighborhood of global extreme, so a local gradient descent in the neighborhood is applied for a more accurate solution of the problem. Comparison of gradient descent, Firefly algorithm and Firefly algorithm with gradient descent is carried out.

<p><span lang="EN-US">The article dues to the arrangement of the reactive power sources in the power grid to reduce the active power losses in transmission lines and minimize cable cross-sections of the lines. The optimal arrangement is considered from two points of view. In the first case, it is possible to minimize the active power losses only. In the second case, it is possible to change the cross-sections of the supply lines to minimize both the active power losses and the volume of the cable lines. The sum of the financial cost of the active power losses, the capital investment to install the deep reactive power compensation, and cost of the cable volume is introduced as the single optimization criterion. To reduce the losses, the deep compensation of reactive power sources in nodes of the grid are proposed. This optimization problem was solved by the Genetic algorithm and the Particle Swarm optimization algorithm. It was found out that the deep compensation allows minimizing active power losses the cable cross-section. The cost-effectiveness of the suggested method is shown. It was found out that optimal allocation of the reactive power sources allows increasing from 9% to 20% the financial expenses for the enterprise considered.</span></p>

The load forecasting of a coal mining enterprise is a complicated problem due to the irregular technological process of mining. It is necessary to apply models that can distinguish both cyclic components and complex rules in the energy consumption data that reflect the highly volatile technological process. For such tasks, Artificial Neural Networks demonstrate advanced performance. In recent years, the effectiveness of Artificial Neural Networks has been significantly improved thanks to new state-of-the-art architectures, training methods and approaches to reduce overfitting. In this paper, the Recurrent Neural Network architecture with a small-size model was applied to the short-term load forecasting of a coal mining enterprise. A single recurrent model was developed and trained for the entire four-year operational period of the enterprise, with significant changes in the energy consumption pattern during the period. This task was challenging since it required high-level generalization performance from the model. It was shown that the accuracy and generalization properties of small-size recurrent models can be significantly improved by the proper selection of the hyper-parameters and training method. The effectiveness of the proposed approach was validated using a real-case dataset.

This work aims to analyze and manage the optimal power consumption of the autonomous power system within the Pamir region of Republic of Tajikistan, based on renewable energy sources. The task is solved through linear programming methods, production rules and mathematical modeling of power consumption modes by generating consumers. It is assumed that power consumers in the considered region have an opportunity to independently cover energy shortage by installing additional generating energy sources. The objective function is to minimize the financial expenses for own power consumption, and to maximize them from both the export and redistribution of power flows. In this study, the optimal ratio of power generation by alternative sources from daily power consumption for winter was established to be hydroelectric power plants (94.8%), wind power plant (3.8%), solar photovoltaic power plant (0.5%) and energy storage (0.8%); while it is not required in summer due to the ability to ensure the balance of energy by hydroelectric power plants. As a result, each generating consumer can independently minimize their power consumption and maximize profit from the energy exchange with other consumers, depending on the selected energy sources, thus becoming a good example of carbon-free energy usage at the micro- and mini-grid level.

Modern energy strategies aimed at the development of energy industry presuppose a significant change in the structure of process of formation, transmission, consumption of electrical energy and increasing energy efficiency by introducing modern technologies at all stages. The growth of capacities of industrial enterprises in the conditions of wholesale market of electrical energy and capacity in the modern energy system determines the need for development technologies of predictive control of power consumption process of these enterprises. The introduction of such technologies at the control rooms of the operational management of enterprises will allow to reduce the number of human errors, the number of emergency stops of technological process, increase the reliability of power system mode, rationally manage the process of power consumption of enterprises. In this regard, forecasting the load demand and consumption is an important stage in the functioning and planning of modern power systems. An accurate, correctly compiled forecast is the key to effective management of energy consumption process and reliable operation of the enterprise. Forecasting errors lead to imbalanced supply-demand, which negatively affects operating costs, reliability and efficiency.

SUMMARYThe article is devoted to the electrical network optimization by genetic algorithms-the optimization method based on the simulation of the biological evolution. This problem is a many-dimensional nonlinear discrete optimization problem, which requires the usage of heuristic or combinatorial algorithms. The mathematical model and algorithm of the given problem are developed. The presented methodology is applied for the IEEE 30-bus test system.