Design practice of a power data-driven auxiliary decision-making platform for corporate pollution management based on electricity data

Currently, the traditional emission inventory constructed based on macro annual statistical data is an important tool for supporting enterprise pollution control in China. However, this method struggles to accurately reflect the near-real-time emission status of pollution sources, severely limiting precise pollution control and emergency management for heavily polluting industries. Therefore, this paper proposes a power data-driven auxiliary decision-making platform for enterprise pollution control, aiming to enhance the real-time monitoring and accuracy of pollution sources. The platform first utilizes the z-score algorithm and machine learning techniques to identify and correct anomalies in multiple datasets. Furthermore, a regression analysis model is employed to establish a quantitative relationship between enterprise electricity consumption and regional pollution levels, enabling accurate judgment of production restrictions and dynamic evaluation of air pollution. This effectively realizes auxiliary decision-making for enterprise pollution control. The application case analysis from a leading city in the northwest demonstrates the effectiveness and application prospects of the platform in practical environmental.