Journal of Intelligent Computing Zhang Xiaoping 14 3 2023 https://doi.org/10.6025/jic/2023/14/3/69-77 https://www.dline.info/jic/fulltext/v14n3/jicv14n3_2.pdf With the rapid development of the new energy industry, how to accurately evaluate and predict customer value to meet customer needs better and provide high-quality services has become an important issue the industry faces. To address this issue, we propose an optimization model based on genetic algorithms to improve the value and satisfaction of new energy customers. The model first collects relevant customer data, including historical electricity consumption, electricity consumption behaviour, complaints, etc. Then, we used genetic algorithms to process and analyze these data to find the optimal customer value evaluation model. A genetic algorithm is an optimization algorithm that simulates the evolution process of nature and can automatically search for and optimize solutions to problems. In model optimization, we used simulation experiments to evaluate the effectiveness and performance of different models. A simulation experiment is based on real data, which can simulate actual operations and predict future development trends. Through simulation experiments, we can compare the advantages and disadvantages of different models and select the optimal model for promotion and application.