Research On Operation Optimization Of Thermal Power Unit Based On Operation Data

With the development of the times,the Distributed Control System(DCS)has been greatly applied and popularized in the power industry.The power station has mostly stored a huge amount of date after a period of operation and these data are closely related to the operation.We will obtain the information about how to optimize the operation of plant which provide data support for dealing with various problems in power station if we analyze these data deeply.Among all these optimization,combustion optimization is an important part of power station data analysis.In this paper,we firstly extract the operating data in steady state based on the historical operating data of a 300 MW coal-fired power plant for five months.Secondly,we preprocess these data by removing the error point to lay the foundation for the next forecast and optimization.According to the collection and collation of the formula for calculating the cost,the cost per kW·h calculation model was established through the combining with the actual operation of the plant prior to forecasting and optimizing.Then the cost per kW·h based on historical data was calculated and the proportion of each cost was analyzed.Finally,we select 20 auxiliary variables used in the prediction model and divide them into 6 groups according to the load and coal quality.The cost per kW·h prediction model based on Support Vector Machine(SVM)was established.The results show that the model can accurately predict the cost per kW·h and has good generalization performance.Based on the cost per kW·h prediction model established above,the Support Vector Machine was coupled with the Chaotic Particle Swarm Optimization(CPSO).Then the power cost optimization model was established.The Optimization results showed that the optimized cost per kW·h in each group has a certain degree of decline.Then a group of data was taken as an example to conduct comparative analysis of the date before and after optimization.A detailed comparison of diversification about the cost of change in the cost per kW·h was carried out and the reason why the cost per kW·h is reduced was analyzed.It is because that the improvement of boiler efficiency brought by the optimization of total coal,total air volume and throttle opening reduced the cost per kW·h.