A novel time-span input neural network was developed to accurately predict the trend of the main steam temperature of a 750-t/d waste incineration boiler.Its historical operating data were used to retrieve sensitive p...A novel time-span input neural network was developed to accurately predict the trend of the main steam temperature of a 750-t/d waste incineration boiler.Its historical operating data were used to retrieve sensitive parameters for the boiler output steam temperature by correlation analysis.Then,the 15 most sensitive parameters with specified time spans were selected as neural network inputs.An external testing set was introduced to objectively evaluate the neural network prediction capability.The results show that,compared with the traditional prediction method,the time-span input framework model can achieve better prediction performance and has a greater capability for generalization.The maximum average prediction error can be controlled below 0.2°C and 1.5°C in the next 60 s and 5 min,respectively.In addition,setting a reasonable terminal training threshold can effectively avoid overfitting.An importance analysis of the parameters indicates that the main steam temperature and the average temperature around the high-temperature superheater are the two most important variables of the input parameters;the former affects the overall prediction and the latter affects the long-term prediction performance.展开更多
A trajectory imaging based method for measuring the velocity and diameter of coal particles was presented.By using an industrial charge-coupled device(CCD)camera and a low power semiconductor laser,the images of coal ...A trajectory imaging based method for measuring the velocity and diameter of coal particles was presented.By using an industrial charge-coupled device(CCD)camera and a low power semiconductor laser,the images of coal particles under relatively long exposure time were recorded and then processed to yield both the velocities and sizes.Fundamental research on this method with special attention to recording parameters,e.g.,magnification factor and exposure time,was carried out.For most of the test cases,the results agree with those obtained by particle image velocimetry(PIV)and shadow imaging method.Measurements with good accuracy can be obtained when the imaging magnification factor and exposure time are set appropriately,making N be larger than 3.5,and R between 5-7,where N and R are the number of pixels occupied by the average width and the ratio of length to width of particle trajectory on the image,respectively.The work indicates the feasibility and potential application of the present measurement method for online measurement of coal powder in pipes in industrial power plants.展开更多
基金Project supported by the National Key Research and Development Program of China(No.2018YFC1901300)the Research Project of Multi-data Fusion and Strategy of Intelligent Control and Optimization for Large Scale Industrial Combustion System,China。
文摘A novel time-span input neural network was developed to accurately predict the trend of the main steam temperature of a 750-t/d waste incineration boiler.Its historical operating data were used to retrieve sensitive parameters for the boiler output steam temperature by correlation analysis.Then,the 15 most sensitive parameters with specified time spans were selected as neural network inputs.An external testing set was introduced to objectively evaluate the neural network prediction capability.The results show that,compared with the traditional prediction method,the time-span input framework model can achieve better prediction performance and has a greater capability for generalization.The maximum average prediction error can be controlled below 0.2°C and 1.5°C in the next 60 s and 5 min,respectively.In addition,setting a reasonable terminal training threshold can effectively avoid overfitting.An importance analysis of the parameters indicates that the main steam temperature and the average temperature around the high-temperature superheater are the two most important variables of the input parameters;the former affects the overall prediction and the latter affects the long-term prediction performance.
基金Project supported by the National Basic Research Program(973Program)of China(No.2011CB201500)the National Science & Technology Pillar Program of China(No.2012BAB09B03)the National High-Tech R&D Program(863 Program)of China(No.2012AA063505)
基金Project supported by the National Natural Science Foundation of China (Nos. 51176162 and 51276164)the National Basic Research Program (973) of China (No. 2009CB219802)+2 种基金the Zhejiang Provincial Science and Technology Project (No. 2012C21077)the Zhejiang Provincial Natural Science Foundation of China (No. Y1110642)the Program of Introducing Talents of Discipline to University (No. B08026),China
文摘A trajectory imaging based method for measuring the velocity and diameter of coal particles was presented.By using an industrial charge-coupled device(CCD)camera and a low power semiconductor laser,the images of coal particles under relatively long exposure time were recorded and then processed to yield both the velocities and sizes.Fundamental research on this method with special attention to recording parameters,e.g.,magnification factor and exposure time,was carried out.For most of the test cases,the results agree with those obtained by particle image velocimetry(PIV)and shadow imaging method.Measurements with good accuracy can be obtained when the imaging magnification factor and exposure time are set appropriately,making N be larger than 3.5,and R between 5-7,where N and R are the number of pixels occupied by the average width and the ratio of length to width of particle trajectory on the image,respectively.The work indicates the feasibility and potential application of the present measurement method for online measurement of coal powder in pipes in industrial power plants.
