Operation Data Analysis and Performance Optimization of the Air-Cooled System in a Coal-Fired Power Plant Based on Machine Learning Algorithms

Abstract

Air-cooling technology has been widely used for its water-saving advantage, and the performance of air-cooled condensers (ACC) has an important impact on the operation status of the unit. In this paper, the performance of ACC in a typical coal-fired power plant is optimized by using machine learning (ML) algorithms. Based on the real operation data of the unit, this paper establishes a back pressure optimization model by using back propagation neural network (BPNN), random forest (RF), and genetic algorithm back propagation (GA-BP) methods, respectively, and conducts a comparative analysis of performance optimization and power-saving effect of the three algorithms. The results show that three algorithms offer significant power savings in the low-load section and smaller power savings in the high-load section. Moreover, when the ambient temperature is lower than 10 °C, the power-saving effect of the three algorithms after optimization is not much different; when the ambient temperature is greater than 10 °C, the power-saving effect of the performance optimization of BPNN and RF is significantly better than that of GA-BP. The optimization method has a good effect on improving the performance of ACC.

Keywords

• coal-fired power plant
• optimal back pressure
• machine learning algorithms
• operation data analysis
• performance optimization