Intelligent algorithms for managing a hybrid solar panel-based power supply system to ensure uninterrupted power supply to educational laboratories

The expansion of the distributed energy segment and the active introduction of educational laboratories powered by renewable sources have significantly increased the requirements for the sustainability of local energy complexes [1]. Solar installations are characterized by significant variability of output, which creates sensitive fluctuations in power, especially in small laboratory systems with limited reserves [17]. In hybrid circuits combining photovoltaic modules, batteries, and network input, the accuracy of the short-term forecast plays a key role: it determines the charging strategy, load distribution, and overall energy efficiency [5]. More accurate estimates of future generation make it possible to reduce consumption peaks, reduce unnecessary switching of system elements, and maintain stable operating modes of educational complexes [18]. The purpose of the article is to determine which algorithms for predicting solar generation ensure the most stable functioning of a laboratory—type hybrid power supply system. To achieve the goal, the tasks are formulated: 1. to present modern approaches to forecasting solar generation and identify the conditions for their application in engineering energy systems; 2. describe the architecture of a hybrid laboratory power system and highlight the features of its operation; 3. evaluate how the accuracy of forecasts affects the management of the hybrid system and what measures can improve its efficiency. The object of consideration is a hybrid power supply scheme, and the subject is photovoltaic generation forecasting algorithms that affect control modes [3]. The material is aimed at identifying how a correct assessment of future capacity increases the stability and functional flexibility of local energy complexes.