Research Of The Supervisory Control Methods Based On Video Fusion And Its Application For Boiler Combustion;research Of The Supervisory Control Methods Based On Video Fusion And Its Application For Boiler Combustion

Power system is the security system of nation-building, and in the sustainable power generation technologies, coal power generation plays a major role. Coal-fired power generation technologies from the perspective of energy consumption, sub-critical unit is constantly replaced by supercritical units and ultra-supercritical unit because of low-power zero-emission and high power generation efficiency on equivalent coal consumption. However, in the angle of power generation technology, they are all coal-fired power generation. Coal-fired power generation includes the "link of the provide-powder," "burning part," "power generation" and other auxiliary links. The all power generation process is a Multi-energy conversion process with high capacity, large delay, physical and chemical nonlinear model parameters, which is real-time monitored by a power plant distributed control system ( DCS ).Plant DCS plays an important role in the plant production, is a widely used monitoring system in plant production. However, in the monitoring process of the power plant DCS, the basis of control strategy does not come from the most important technological object of generation- the furnace flame, but the relevant extension parameters, such as main steam temperature, main steam pressure, which are indirectly measured. In fact, the furnace flame is linked to all the technical parameters and associated with all aspects of power generation and the security, stability and reliability of production.Research of furnace flame and control of flame combustion involve a lot of technologies, such as thermal dynamics, combustion, graphics, signal processing and control. In the large coal-fired boilers, state of the furnace flame burning is a very complex suspension combustion, and its condition is unstable. Combustion flame safety depends on the stability of combustion. If the combustion is unstable, the furnace temperature field would be uneven, which may cause some major accidents. Therefore, the research of furnace flame has important significance and significant academic value.Flame signal acquisition is based on the two-dimensional CCD sensor video signals. In this paper, on the basis of grasping the image processing methods and analyzing the flame basic information and preprocessing algorithm, conducted a comprehensive study of flame image features by using image processing technology. Meanwhile, explored the diagnosis technology and real-time monitoring method of Flame, the specific tasks are as follows:Analyzed the flame image noise and the suppression, and took the arithmetic mean filter and median filter algorithm to realize the image denoising; studied the intensity characteristics, temperature characteristics and related coal properties of furnace flame. Comprehensively studied the temperature and temperature field measurement technique of furnace flame, proposed the error correction method based on the depth of field temperature field. Divided the flame region based on characteristics, determined the features of region edge and image characteristics presented in feature area and did a comprehensive study of main diagnosis based on furnace flame characteristics. Got the temperature of characteristics areas, the average flame temperature, high temperature area ratio, roundness rate of high temperature and the rate of flame centroid through Comparative Analysis, and the real-time data could be used to do the furnace flame monitoring.Researched the fire detection technology and its control. Comprehensively studied the furnace flame image processing algorithms and provided effective and accurate Numerical to control strategies. Studied the flame combustion control strategies, which includes the control object identification model, control strategy and its implementation. Proposed the integration method of flame characteristics based on two-dimensional image, cluster characteristics and time-based flame detector combustion state.Summarized the operation, preprocessing, conversion, processing, feature calculation, strategy and temperature - color transform, pseudo-color display technology of the furnace flame image, designed flame image information platform. Based on image acquisition and image processing, calculation, determine and control , developed the real-time monitoring DCS and the underlying flame monitoring base stations, which can complete the wireless image transmission function and ensure the real-time delivery. DCS constructed by the technology of this paper can meet the real-time control of furnace flame, and make plants produce more secure, stable and reliable.