Evaluation Of The Operating State Of Thermal Systems And Its Key Equipment Based On Fuzzy Comprehensive Evaluation Method

With the rapid development of renewable energy installed capacity connected to the grid,thermal power units will be more in a low-load operating state,and the stable operation of the unit under low load is the main challenge facing the power plant in the next step,so the monitoring and fault diagnosis technology of the power plant thermal system and key equipment operation status is crucial to the safety and economy of power plant operation.Based on the operating parameters of thermal power plant equipment,and based on the simulation modeling results of thermal system,the operation status evaluation model of thermal system and key equipment is finally obtained.First,the simulated modeling of the studied unit is carried out.According to the design parameters of each working condition of the unit provided by the factory,the thermal system simulation modeling software EBSILON is used to model the thermal system,and a simulation model of the main operating conditions that meets the research requirements is constructed.Based on the historical operation data of the power plant,the historical operation database of the operation status evaluation index of each major equipment is calculated by EBSILON.Then,according to the simulated model,the influence curve of each equipment evaluation index on the heat consumption of the unit is calculated.EBSILON was used to simulate and calculate the impact of the operating state evaluation index of each cylinder of the steam turbine,condenser and each return heater on the heat consumption of the unit,including,and obtain the heat consumption sensitivity of each equipment evaluation index under different conditions.After that,the operation status evaluation model of key equipment and thermal system is obtained by the secondary fuzzy comprehensive evaluation.The Particle Swarm Optimization algorithm is used to compress the normal operation data,the function relationship between the evaluation index and the boundary condition is obtained.Taking the function relationship as the running benchmark under different boundary conditions,the operating status of the evaluation index is divided by the selection of the confidence interval of the normal distribution,then the membership function of each evaluation index related to the boundary condition is determined,and the state evaluation matrix is established by combining the historical data;the dynamic weight matrix is determined according to the heat consumption sensitivity of the evaluation index,and the weight matrix is combined with the evaluation matrix to obtain the key equipment state evaluation model.According to the heat consumption sensitivity of the equipment,the dynamic weight matrix of the equipment is determined,and the obtained weight matrix is combined with the state evaluation model of each equipment to obtain the state evaluation model of the thermal system.Finally,the fault source of each system warning status is located by principal component analysis method.Selecting the main measurement point information of the research object,and the principal component analysis method is used to reduce the dimensionality of the measurement point information.The T~2 and SPE statistics control limits of each system are calculated according to the warning state confidence,and then the fault data is entered for verification and analysis,and the fault source information is determined according to the contribution of each measurement point to the statistics.The unit and equipment state evaluation and fault diagnosis model are completed finally.