| The thermal process control system for thermal power units is designed in accordance with certain performance requirements,and most of them are in a good operating state at the initial stage of completion and commissioning.However,over time,the performance of the control system deteriorated due to poor tuning effects of the controller,lack of maintenance of the control loop,equipment failure,and changes in the characteristics of the production process.If the control system is in a bad state for a long time,it will not only affect the economic efficiency of the unit,but also threaten the safe operation of the unit.At the same time,with the development of technology,the degree of automation of generator sets is getting higher and higher.The development of generator sets towards large capacity and high parameters has also put forward higher requirements for the control quality of thermal systems.Therefore,assessing the performance of the control system and improving its control quality are of great significance to the safety and economics of the unit.This paper first studies the performance assessment methods of thermal process control systems.For single-loop control systems,the minimum variance is selected as the assessment index.Taking the furnace pressure control system and primary air pressure control system of a thermal power unit as the research object,the historical operating data were selected,and the control performance was assessed using the assessment method based on the minimum variance index.The cascade control system was regarded as A "quasi-multivariable" system with single input and double output,combined with the control characteristics of the cascade control system,proposes an assessment index combining primary and secondary variable covariance and primary variable variance-a mixed variance index.Taking the superheated steam temperature control system and the drum water level control system of a thermal power unit as the research object,historical operating data was selected,and the control performance was assessed by using the assessment method based on the mixed variance index.Then,the method of obtaining engineering practical object model is studied.These include model identification methods based on disturbance test data and model identification methods based on historical operating data.The superheated steam temperature control system is taken as the research object,and its object model is identified.Finally,based on the identified superheated steam temperature object model,the controller parameter optimization method was researched,and the conventional ant colony algorithm was improved.A continuous domain ant colony algorithm with adaptive adjustment of pheromone was proposed,which achieved good optimization effect. |
