Study On Oxygen Content Optimal Control And Economic Evaluation Of Coal-Fired Power Plant

With high-speed growth of China’s economy and population climbing,the problem of high energy consumption and low utilization rate has become increasingly prominent.To this end,China has put forward higher requirements of energy-saving tasks in its thirteenth five-year plan.As the main force of China’s power generation,the energy saving of coal-fired power plant is significant to the overall goal.Oxygen content in the flue gas is an economic indicator reflecting the efficiency of combustion,and its optimal control is one of the effective ways to improve the combustion efficiency and realize the energy-saving.Evaluating the economic performance of power plant’s operation could help optimize the operation setting values,improve the operation efficiency of the power plant and finally achieve the purpose of saving energy and reducing consumption.Monitoring key economic indicators of the power plant and real-time evaluating the whole unit’s energy consumption is meaningful to the optimizing operation.Evaluation results could provide the direction and basis for optimization.Firstly,the predictive model of oxygen content is studied.To solve problems of the nonlinear dynamic characteristics of combustion system and the wide range of load,a data-driven fuzzy multiple-model modeling strategy for oxygen content is proposed.Predictive model is presented as multiple-model network form.Based on the site data from the power plant,fuzzy C-means clustering algorithm is used to design the network structure parameters,and the subspace identification algorithm is used to identify the model parameters.The simulation results show that the model’s ability of fitting and predicting is good.The fuzzy multi-model network data-driven model established in this paper can be further used as the predictive model in advanced control.Secondly,the predictive control method of oxygen content is studied.The optimal oxygen content setting value varies with the load.Thus the control problem of the oxygen content is characterized as a real-time tracking problem of nonlinear system.Considering the problems of model mismatch,environment disturbance and production condition constraints in industry,the predictive control method is used for oxygen content tracking control.In the predictive control method,the established fuzzy multi-model network model is viewed as the predictive model,the real-time output error feedback is used to correct the predictive model output,and the control optimization sequence is obtained by solving the quadratic programming problem.The real-time control simulation results show that under the action of the predictive control method proposed in this paper,the oxygen content can quickly and accurately track the varying optimal set value.Finally,the operation economic evaluation of coal-fired power plant is studied.Coal-fired power plants include boilers,steam turbines,auxiliary systems and other equipment.The overall system is very complex.Due to the difficulty of evaluating the whole system economic performance,this paper establishes an economic comprehensive assessment framework,which combines module monitoring with overall evaluation strategy.According to the structure of the coal-fired power plant,the main economic part of the plant can be divided into the boiler module,the turbine module and the power supply module.In each module,the module evaluation model is established to calculate the key economic performance index of the module.Based on the module index,a composite weighted comprehensive score method and a fuzzy comprehensive evaluation model are put forward to give overall evaluation results.Evaluation results are basically consistant with real-time coal consumption,which proves the rationality of two evaluation methods.