BUILDING A MODEL AND EVALUATING ITS ROBUSTNESS IN THE TASK FORECASTING FOR CONSUMERS WITH ADDITIVE TECHNOLOGIES ELECTRICITY CONSUMPTION PROFILES

Abstract

The construction of a robust model and an assessment of its accuracy in problems of forecasting electrical loads with additive consumption profiles are considered. A study was conducted on the influence of neural network parameters (data packet size; number of neural network layers; neuron activation functions; optimizers) on the error in predicting power consumption. Graphs comparing the profiles of actual and projected consumption and the deviation of the forecast for electricity consumption above the average value for the period under review are presented. Optimal parameters of the predictive neural network model have been selected in manual mode. The result of the study of the varieties of genetic algorithms revealed the optimal hybrid algorithm for learning a neural network model based on the rapid convergence of the solution. A Python-based algorithm for selecting network hyperparameters based on power consumption data with different patterns of electricity consumption has been tested. The conducted training and testing of the genetic algorithm confirmed the possibility of obtaining forecasts of greater accuracy and the possibility of automating the selection of optimal hyperparameters. In the tasks of forecasting power consumption using a neural network model, regardless of the method of creating the structure, the optimal metric has been selected. It is revealed that for consumers with additive profiles of electricity consumption, it is advisable to use the robust Huber loss function, at the same time, for consumers with a unique or regular profile of electricity consumption, the use of a sliding window increases the error, unlike additive consumers. It is shown that the use of a genetic algorithm significantly increases the accuracy of forecasting due to the individual selection of optimal parameters for a specific consumer. A block diagram of an intelligent device for predicting energy consumption modes has been developed. A decisionmaking assistance system has been introduced that allows for the implementation of planned proactive management based on data taken from the electricity meter and obtained as a result of the neural network forecasting model. The decision–making assistance system calculates the deviation of the projected power consumption values from the actual ones and, as a result, issues recommendations to the dispatcher of the distribution power grids. Based on data from the decision-making assistance system, the distribution grid operator can make a decision on ordering the required amount of electricity, gets the opportunity to monitor possible spikes and decreases in consumer electricity consumption, abnormal equipment operation, and additionally monitor the adequacy of the neural network model