| With the proposal of “Medium and Long-term Energy Development Plan” and“Carbon Peaking and Carbon Neutrality Goals”,the energy saving and emission reduction performance of coal-fired power units is facing significant pressure.The operation state of the unit largely depends on the combustion state of the boiler.And the combustion state of the boiler is difficult to control due to the influence of many other factors.We can find out the repeatable operation modes from many combustion experiments and operation modes,then weigh the advantages and disadvantages of these operation modes.When the working conditions are matched,we can adjust the combustion state to the optimal mode by consulting those typical and repeatable operation modes.This method is an effective way to achieve the goal of energy saving and emission reduction.Eliminating interference and identifying the typical operation modes with superior performance and repeatability are the core steps of this method.However,the previous pattern recognition method for energy saving and emission reduction performance of thermal power units does not take into account the impact of disturbance on unit parameters,as well as the random factors in the evaluation process.Because of the superiority of evidence theory in many information fusion methods,this paper studies the operation mode recognition of the coal-fired power units based on evidence theory.Firstly,a coal properties identification method based on evidence theory is proposed to identify the coal properties state.Then,considering the influence of coal properties disturbance on units’ energy saving and emission reduction performance,a pattern recognition method of coal-fired power units’ energy saving and emission reduction performance based on cloud model and improved evidence theory is proposed.The specific work and innovation of this paper are as follows.(1)Because the fluctuation of coal properties will not only affect the energy saving and emission reduction performance of the unit,but also disturb the selection of typical operation modes,a coal properties identification method based on evidence theory is proposed.The operation state of the unit is divided into two modes: “normal coal properties”and “poor coal properties”.Then,three kinds of parameters which can characterize the current coal properties are constructed as evidence.Finally,these three kinds of evidence are integrated by evidence theory.This method successfully realizes the on-line monitoring of coal properties without additional equipment.(2)The randomness existing in the process of unit operation pattern recognition is taken into consideration.A pattern recognition method for energy saving and emission reduction performance of units based on cloud model and improved evidence theory is proposed.First of all,a reasonable evaluation index system of energy saving and emission reduction performance is established.Then,the cloud mode is used to solve the fuzziness and randomness in the evaluation process.Finally,the information fusion is realized through the improved evidence theory,which can increase the credibility of the results and reduce the uncertainty of the whole system.The experimental results show that this method can correctly identify the energy saving and emission reduction performance of the coal-fired power units while considering the coal properties disturbance and the randomness in the process.The result can provide guidance and reference for the improvement of the coal-fired power units. |
