关键词: 机器学习, 用电预测, 偏最小二乘, 高相关性

Abstract:

With thedevelopment of power reform, the construction of electricity marketization has further accelerated. The analysis and prediction of electricity consumption have become the focus of attention for power supply enterprises and electricity consuming enterprises. Electricity consumption can also reflect the current operation status and development trend of enterprises, which is of great significance to production-oriented enterprises. However, the data of power consumption is less and the data accuracy is insufficient. The common machine learning methods are not suitable for this application. Therefore, the partial least squares (PLS) method is used to solve these problems. At the same time, the problem of high correlation between the acquired data can also be handled. Based on the PLS method, the regression model of enterprise electricity consumption prediction is established, and the validity and accuracy of the model are verified by the data simulation.

Key words: machine learning, electricity cousumption prediction, partial least squares(PLS), high correlation