Research On The Coal Type Identification And Boiler Optimizing Control Technique Based On The Flame Spectrum Detection In Furnace

With the "Paris climate change agreement" to come into effect,the current energy-saving emission reduction environment has a great impact on the development pattern of the energy industry,as the greenhouse gas emissions of coal-fired power plants,the urgent need to further improve the operating efficiency of the boiler to alleviate the pressure to reduce emissions.In recent years,the energy saving potential of the process improvement of unit main equipment and auxiliary system has been almost exhausted,the most promising research direction is the use of advanced control technology to realize the optimal operation of the boiler,and the technology bottleneck of the boiler optimizing control is the on-line measurement of furnace parameters and the adaptability of advanced control algorithms for complicated process.In this paper,the following research work is carried out according to the two technical requirements:1)Based on the research of basic theory and measurement mechanism of flame emission spectroscopy,and the identifying function of the content and proportion of Na,K,Li and other alkali metal elements in coals,a coal flame spectrum identification method using the strength relationship characteristics of alkali metal atoms emission spectra in the coal flame was proposed.After eliminating and compensating the measurement and environmental factors by the mechanism,the stable characteristics were obtained,and the experimental results show that the characteristics can be reproduced,easily distinguished and the adaptability for different conditions meet coal identification requirements.2)According to the mechanism characteristics of coal identification,combined with a series of auxiliary features,SVM algorithm was adopted to realize the algorithm design,system development and testing application of coal identification methods,and the accuracy of coal identification methods based on the furnace flame emission spectral characteristics was verified.3)To carry out the research of soft measurement of boiler efficiency based on the least squares support vector machine(LSSVM)algorithm,and the boiler efficiency optimizing control algorithm based on T-S fuzzy modeling and genetic algorithms,the following technical route is proposed and implemented:Using DCS and real-time coal identification system to obtain coal parameters and boiler combustion parameters of sufficient quantity,selecting parameters in stable conditions to calculate discrete boiler efficiency with positive balance method,which is used to implement the soft sensor modeling for boiler efficiency,and carry out the continuous boiler efficiency measurement and optimizing control.4)Put forward the generalized predictive control with the reference model and disturbance model(RDM-GPC)algorithm and the predictive function control with the leading disturbance model(LDM-PFC)algorithm,the simulation test show that there are obvious advantages in the model adaptability,response speed and control precision,the optimizing control scheme can be used for the boiler important performance parameters control such as main steam temperature and pressure etc..5)Develop the thermal optimization control platform(TOP)system for the integrated application of optimizing control algorithms,put forward the progressive safety control mechanism,and the flexible configuration as upper station carrying out advanced algorithms and lower station operating real-time control,create the good conditions for flexible application of optimizing control system.