Integration of energy storage systems with adjustable curtailment of photovoltaic generation in power distribution networks for enhanced grid efficiency

Although renewable energy sources can produce electricity efficiently, there are certain operational challenges
when integrating them into the power grid. To mitigate these issues, this manuscript proposes a new approach
for integrating Energy Storage Systems (ESS) with adjustable curtailment of photovoltaic generation in power
distribution networks to develop grid efficiency, reduce energy costs, and improve system reliability. The proposed
technique involves the Logarithmic Differential Convolutional Neural Network (LDCNN) and Eel and
Grouper Optimizer (EGO) technique, which is commonly referred to as LDCNN-EGO. The objective is to improve
grid efficiency and reduce costs and enhance system reliability. The LDCNN method is used to predict the load
demand, and the EGO is used to optimize the locations and size of ESS in power distribution networks. The
proposed method is implemented using MATLAB to evaluate its performances and compared with the existing
methods like sizing algorithms (SA), Coot Bird Search Algorithm-Genetic Algorithm (CBSA) and Crayfish Optimization
Algorithm (COA). The proposed method attains the lowest operational cost of $27.2, outperforming SA
($28.3), CBSA ($29.3), and COA ($30.3). This approach effectively improves grid reliability while supporting a
smoother transition toward a sustainable energy system.