Multi-objective hybrid grasshopper optimization and differential evolution algorithm for wind-integrated power system expansion with energy storage solutions

• Hybrid GOA-DE algorithm for multi-objective generation expansion planning. • ANN-based load forecasting integrated into GEP optimization framework. • Four-level hierarchical model for wind integration with storage evaluation. • Multi-objective optimization: cost, capacity, EENS, LOLP and emissions. • Demonstrated benefits and decision complexity of wind integration with storage. The major challenges for electric energy utilities are ensuring optimal planning and resource management across diverse power production technologies. An artificial neural network is used in this study to forecast power consumption and propose a hybrid evolutionary algorithm called the grasshopper optimization algorithm with differential evolution to solve the multi-objective generation expansion planning. In Load forecasting, this paper utilizes Bayesian regularization and the Levenberg-Marquardt algorithm to predict the electricity consumption of a region. The model’s performance was evaluated using coefficient determination (R 2 ), Mean Absolute Error, Mean Absolute Percentage Error and Mean Squared Error. The forecasted indicator values for the BR model are 0.9962, 0.2553, 0.0239 and 0.1373 and the corresponding values for the LM model are 0.9941, 0.3191, 2.98 and 0.1677. In the GEP framework, we proposed a mathematical study model based on introducing a wind plant as a candidate plant with effective storage. The study involves a four-level hierarchy based on (i) investment policies that introduce WP as a replacement for existing high-emission plants, (ii) WP as an alternative candidate plant, (iii) WP capacity with and without energy storage and (iv) inclusion of treatment/penalty charges from HEP. The optimisation problem aims to minimise total cost, increased capacity, expected energy not served and loss of load probability, while mitigating the environmental footprints under varying forced outage rate per cent for a wind power plant with and with no storage across planning spans of 6 and 14 years.