DEVELOPMENT OF A CONTROL ALGORITHM DEVICE LOAD BALANCING DEVICE

Abstract

The work is devoted to improving the control algorithm of the load balancing device for 0.4 kV electrical mains that are in pilot operation. The action of the device is aimed at balancing the load at the nodal point of the electrical network, where several 3-phase outgoing lines are connected, by changing the connection diagram of the phases of one of them. The device is capable of realizing 6 combinations obtained by circular permutation and changing the phase rotation. The classification of analogs of the device according to the type of connection to the network is proposed. The technique of collection and processing of experimental data on the operation of the device is presented. Analysis of primary data shows that among the consumers there are powerful energy receivers with automatic two-position control. As a criterion for the expediency of one or another method of connecting the phases of the outgoing line to the phases of the source, the coefficient of non-uniformity of conductivities is taken. Strict observance of the condition of the minimum coefficient of unevenness is impossible in practice with the existing nature of the load due to the need for too frequent operation of the device. This and other limitations given in the work are the reason for the systematization of requirements for the device control algorithm, one of which is the need to filter the initial data before executing the main algorithm. The paper provides a rationale for the use of a digital IIR low-pass filter. Based on the frequency analysis of the curves of the conductivity unevenness coefficients, the filter design was carried out and the digital transfer function coefficients were calculated. A difference equation is determined that is suitable for implementation in the program for the control microcontroller of the device. The developed filter is universal and suitable for use in the control algorithm of a load balancing device of any power and for any type of load. The results of modeling the proposed filter are presented using the example of experimental data. The prospect of the research is the collection of statistics on various types of consumers, analysis of practical situations, in order to further refine the algorithm.