Electric Power Consumption and Pollutant Emission: A Study Based on Big Data and Machine Learning Algorithm

Using electric power big data of industrial enterprises, a pollutant emission concentration prediction method based on Apriori-RF algorithm is proposed in this paper. Specifically, we use the Apriori algorithm to analyze the correlation between the electric power consumption and pollutant emission level, and apply the random forest algorithm to construct a model to predict the pollutant emission concentration. The example test shows that there is strong correlation between electric power consumption data and pollutant emission concentration data. Further, using the electric power consumption data in the forecasting model of pollutant emission concentration, we find that the prediction model using random forest algorithm can well fit the real value. Collectively, our study constructs a monitor model not relying heavily on the hardware devices of pollutant emission monitoring, filling the gap left by the extant monitor methods heavily relying on the hardware devices of pollutant emission monitoring. Our empirical results can provide both theoretical and practical implications for industrial enterprises to reduce environmental pollution and prevent pollutant emission from exceeding the standard.