The need for smart grid user-side management is becoming more critical as smart power distribution networks evolve. This study suggests a short-term power load forecasting model based on K-means and Clustering to increase the precision of short-term electric load forecasting for individual customers. Local comparable daily data are used as characteristics for users with poor correlation at adjacent times. Based on the -means cluster analysis technique, a clustering module was created. The clustering module was validated using home load profiles based on smart meters. Utilizing daily and segmented load profiles from individual and collective smart meter data, many case studies were put into practice. The smallest segmentation time window yields the lowest clustering ratio, according to simulation findings described in terms of the correlation between the clustering ratio and the time window. Results also indicate that for strongly coupled load profiles, a limited number of clusters is advised. Last but not least, the feather vectors are sent into the BP Neural Network, which forecasts the short-term load. According to experimental findings, the suggested clustering algorithm matches the features of customers' power consumption pattern. The accuracy of clustering-based load forecasting is higher than that of non-clustering load forecasting when using the same forecasting methodology. This model offers superior predicting accuracy when compared to conventional BP, RBF, and GRNN Neural Network.
