The Research Of Modeling, Monitoring And Optimizing For Thermal System Of Thermal Power Plant Based On Ebsilon

Current has seen the increasing requirements of energy saving, which means real-time monitoring and operation optimizationhasbecome the key to thehigh operation-efficiency of thermal power units..This paper taking a 1000 MW ultra-supercritical thermal power unit in Huadian Zouxian as the research object, based on the software platform which designed by a German company called Steag, successfully achieved the goal ofmonitoring the full condition of the thermal power system and providing a simulation tool for optimizing operation through sliding pressure.First,as a fundamental tool, a comprehensive study of Ebsilon was made,this was focused on analysing and achieving its main functions.Secondly, we successfully simulate the research object-a power unit through following steps: 1.Modeling its main components based on the physical mechanism which was named as the "dominant factor" method that combined with the method of actual data mining: the key to this approach is to determing “characteristic parameters” that can tell the operating characteristics of each component module and their “dominant factors”by theoretical analysis based onphysical mechanisms and characteristic equation; the essential part to this approach is to determing the relationship between the two mentioned above by data digging; and the difficulty is to making this selection agree with Ebsilon. 2Modeling other components according to Ebsilon built guidelines.3.Simplify inputs by finding the conection between the mass flow of main stem and other negligible steams.Then, we made a test to the simulation model through its design values: the results show that the relative error of each simulation value and design value for almost every measuring point were less than 1%, which means the simulation model has high accuracy.On this basis, we then use this simulation model in reality, and solve arising problems through correcting original model according to realistic, and the results shown by corrected simulation model tells its practicability in real-time monitoring.In addition, through analysingthe obtained data, we ultimately build our own turbine control stage through self-modeling,associating the valve opening level to the load, making the sliding pressure curve used for operation optimization to be formed through tentative calculation.Eventually, through the whole process mentimoned above, conclusions in various aspects were given as well as providing ideas onmodeling and optimization for the development of our own software platform.