Research On Optimal Operation Of Power Plant Based On Data Analysis

The state and society are increasingly concerned about energy conservation and emission reduction in the coal-fired power plants,since resource and environmental constraints have become more and more serious. With increasing automaticity of the power plants, a large amount of historical operating data has been stored in the database. This paper uses the strong data analysis ability of data mining to find the history optimal conditions from the database, so that the units can save energy and reduce emission with a practical optimal method. The research of this paper are expanded in following areas:1. The paper summarizes the advantages and disadvantages of Apriori algorithm, and improves its performance. The tests confirm that the improved algorithm has higher efficiency in data mining process. And the universal model of data analysis in power plants is built based on the improved algorithm;2. In the optimal operation of coal-pulverizing system for a 600 MW unit, the paper analyzes the comprehensive effects of the operating parameters on the boiler efficiency and power consumption, and finds the optimum ones of fully operating conditions,which meet the demand of boiler efficiency and power consumption.3. In the low NOx combustion optimization of a 300 MW unit, combined with NOx formation rules and data analysis methods, the paper builds the soft measurement model of secondary air and clustering model of oxygen content in flue gases. Meanwhile, the improved algorithm is adopted to optimize the conditions to meet the requirements of NOx emissions in the power plant from the database. According to the results of optimization, oxygen content in flue gases, combinations of coal mills, secondary air damper and SOFA gatages all conform to the conclusion of combustion adjustment test, and the NOx concentration is effectively reduced on the premise of ensuring the boiler thermal efficiency.