Renewable energy systems, such as photovoltaic (PV) systems, have become increasingly significant in response to the pressing concerns of climate change and the imperative to mitigate carbon emissions. When static converters are used in solar power systems, they change the current, which uses reactive energy. A proportional-integral controller regulates active and reactive powers, whereas energy storage batteries enhance energy quality by storing c. Renewable energy systems, such as photovoltaic (PV) systems, have become increasingly significant in response to the pressing concerns of climate change and the imperative to mitigate carbon emissions. When static converters are used in solar power systems, they change the current, which uses reactive energy. A proportional-integral controller regulates active and reactive powers, whereas energy storage batteries enhance energy quality by storing current and voltage as they directly affect steady-state error. The utilization of artificial intelligence (AI) is crucial for improving the energy generation of PV systems under various climatic circumstances, as conventional controllers do not effectively optimize the energy output of solar systems. Nevertheless, the performance of PV systems can be influenced by fluctuations in meteorological conditions. This study presents a novel approach for integrating solar PV systems with high input performance through adaptive neuro-fuzzy inference systems (ANFIS). A fuzzy neural inference-based controller regarding energy generation and consumption aspects was designed and examined. This study examines the importance of artificial intelligence in facilitating continuous power supply to clients using a battery system, hence emphasizing its significance in energy management. Moreover, the findings demonstrated promising outcomes in energy regulation and management.••Management controlArtificial Intelligence (AI): Photovoltaic panelPV systemFuzzy-neural network controlThe world's interest in renewable energy is attributed to several reasons. First, many industrial countries suffer from high pollution levels, pushing people to stand firmly against building new conventional power plants. Therefore, the only solution to overcome this problem is to introduce renewable energy plants as replacements for conventional ones. Second, some countries depend on nuclear power plants for their electricity. After several atomic disasters in Japan and Russia, there was a solid public movement against this type of energy. The result was a gradual replacement of nuclear plants with renewable energy ones. Germany was one of the pioneering countries that went this way. Third, some countries do not have enough resources for fossil fuels, such as gas and coal, which are necessary for the energy industry and economic development. Therefore, non-oil countries started exploiting their renewable energy resources, including solar, wind, biomass, tidal, and other resources. Hybrid systems based on renewable energy sources (RES), including solar and wind energy sources, offer new solutions for remote areas outside the power grid and significantly reduce emissions.While these energy sources, in the appropriate combination, use the strengths of one to compensate for the weaknesses of the other, they can cause unreliable energy supplies due to unpredictable changes in weather and clima. The ability of systems to predict energy production and consumption allows for excellent optimization and efficiency. By using machine learning algorithms to analyze historical data and weather patterns, it is possible to forecast the energy output and adjust system settings accordingly. Additionally, artificial intelligence (AI)-controlled PV systems can monitor and react to changes in energy demand in real time, ensuring that energy is used efficiently and cost-effectively. Overall, applying AI to energy management is a promising solution for improving the sustainability and effectiveness of PV systems. This stability has encouraged several renewable energy companies to invest in this technology and consider it a center for promoting and marketing their products. This role was facilitated by a robust engineering system that could transfer this technology to the local community and neighboring countries. The reason for successfully applying renewable energy is the prevailing moderate climate. This makes it suitable for optimum renewable energy and PV system operation. The annual moderate average temperature and direct solid solar radiation in Jordan are essential conditions for the optimum operation of PV systems. These features have attracted all manufacturers to measure the performance of their PV products and test them to obtain accurate feedback on their new equipment and systems. The standalone PV system is widely applied in many fields where no electric power grid is used. The off-grid PV system includes PV panels. Integrating renewable energy sources into power grids and buildings is crucial for sustainable energy use. In this context, PV systems have become popular due to their proven effectiveness in generating electricity from solar radiation. However, the intermittent nature of solar energy must be addressed to ensure that the generated power can be utilized efficiently, which is where forecasting PV performance comes in. Through accurate predictions of energy generation, systems can be designed to handle fluctuations and have a more stable and reliable output.Regression models for solar output power and battery SOC have been built using MATLAB's ANN ToolBox, with the input values being measured daily. The real-time data characteristics, such as current as inputs, the solar array, and power, are taken every hour. The data is used for training to forecast the solar output power at intervals of 30 min on the same day. The following day's solar output power is anticipated using this 30-minute predicted data. In this study, the architecture of the feed-forward neural network is considered.Battery SOC is calculated using real-time system data of solar power and battery charging current as input signals as desired values. The measured data at one-hour intervals is combined with the estimated data at 30-minute intervals to anticipate the battery SOC for the following day. The load linked to the system is kept constant during this procedure.
