Flame Visualization And Combustion Intelligent Diagnostic Studies

The strict environment laws in our country demand the reduction of the pollutant from power plant boilers. It is the optimism of the combustion in furnaces that can effectively approach this goal. What's more, combustion optimism can not only save the fuel, but also avoid tube explosion in furnaces. In this way, the efficiency and safety of power plant boilers can be ensured. In these years, a lot of studies have been done on the alteration of combustor, the numerical simulation in air dynamical field in furnaces and so on, which have played the important roles in improving combustion. However, the research on the monitoring and control of combustion in furnaces is behindhand. The purpose of this thesis is to establish a new combustion monitoring and control system based on the image processing and artificial intelligence technology to carry out the combustion visualization and diagnosis, and then give the instruction information for power plant staff.First, this thesis studied the performance of the image processing system, such as the color principle of CCD camera, the radiation transfer principle of system, geometry optics model of system and so on. Then the system hardware frame was established.Second, based on the flame images, the thesis used the medical CT (computerized Tomography) principle to develop the flame section temperature field reconstruction model and algorithm, which was testified by the experiments. But this model can't satisfy the requirement of real time control in power plant boilers, because the reconstruction time is too long. However, based on this reconstruction model and algorithm, we established the real time"monitoring model for section temperature center. As a result, this model can give the important information to judge the position of the combustion circle, and to change the combustion state to avoid the accident in furnaces.Of course, the prediction for combustion state is also the emphasis in this thesis. A quantitative index for combustion state was defined based on the Kohonen neural network. Then with this index, the thesis used the flame images and back propagation neural network to predict the combustion state.Then, the thesis also did some effective research on the monitoring of the NOX emission from the power plant boiler. A monitoring model was developed to trace the change of the NOX emission by the flame images.Finally, the thesis illustrated the primary application of flame image processing system to the 300MWe power plant boiler. The design idea, hardware system, and software function was described in details. With this system, the research on the combustion visualization and diagnosis will be testified and advanced.