Research On The Performance Calculating Models In Reheating Units

Models of on-line performance monitoring arithmetic in reheating units are researched in the thesis. The main contents and achievements of the thesis are as follows:1. Introduction of general algorithm models in on-line performance monitoring system. Such as data pre-handling models, boiler arithmetic models, turbine arithmetic models, auxiliaries arithmetic models, etc.2. Research on performance calculating for thermal system. The power-efficiency distributing matrix equation for thermal system is developed in this thesis, which is developed on the basis of conventional thermodynamic methods, q ?γ?τmatrix and equivalent enthalpy drop. Each matrix of the power-efficiency distributing matrix equation has a explicit physical sense and strong regular form, which makes it simple, clear, objective for global calculation and local analysis in thermal system. Samples show that the power-efficiency distributing matrix model has the characteristic of clear physical meaning, sententious mathematic description and convenient to parallel calculation.3. Research on calculating enthalpy of the exhaust steam released by turbine with BP neural network. The steam released by turbine is in the wet steam area, so its enthalpy is hard to fix for the lack of means to measure the steam humidity in the local area. Its calculation has always been one of the difficulties of on-line analysis and diagnosis of the thermal economy for thermal power generation units. Many factors determine the exhaust steam enthalpy, which have typical nonlinear relationship with the exhaust steam enthalpy. Considering the better nonlinear mapping ability of BP neural network, the method of establishing the model to calculate the exhaust steam enthalpy with BP neural network is introduced in this thesis, which takes Ping Wei N600-165/535/535 units as an example.