In smart grids, the amount of data is large and complex, and there is a lack of sufficient real-time data processing capabilities, which cannot timely reflect the dynamic changes of grid loads. This paper aimed to optimize load scheduling based on genetic algorithms (GAs), achieve carbon neutrality goals, and improve the flexibility and adaptability of smart grids. Through the collection and preprocessing of historical load data, combined with K-means clustering, ARIMA (Autoregressive Integrated Moving Average) model and random forest method, the accuracy of load forecasting is improved, and on this basis, a multivariate linear regression model is used for real-time adjustment. GAs are used for real number coding and fitness evaluation, combined with multi-objective optimization settings to reduce carbon emissions and minimize dispatch costs, to ensure the safety and stable operation of the power grid. The weight method is used to transform the multi-objective problem into a single-objective problem for comprehensive optimization. The results show that the smart grid load optimization scheduling based on GA has less carbon emissions, energy utilization efficiency of up to 70%, and high grid stability. GA can well complete the smart grid optimization scheduling and promote carbon neutrality.
Liang et al. (Thu,) studied this question.