Analysis And Mining Of Anode Voltage Distribution Data

With the development of the aluminum reduction pot control system,the overall control of the aluminum reduction pot is increasingly perfect,but the refined management of the aluminum reduction pot is still in its infancy.Aluminum electrolysis produces a large amount of data in the production process,such as anode voltage,cathode voltage,temperature of cathode steel rod,and temperature of pot shell etc.Through analysis and processing of these data,the state of the aluminum reduction pot can be accurately understood in a timely and accurate manner,the tedious operations of the technologists can be simplified,and can use the relevant technology of data mining to mine useful information from a large number of process parameters to improve the intelligent production management and control of aluminum electrolysis.This article first briefly introduces the application research of data mining technology in aluminum electrolysis production process.The collected raw anode voltage data is preprocessed,and the preprocessed data is summarized,analyzed and displayed at different particle sizes.Developed a real-time monitoring screen of the anode voltage of the electrolyzer,acquired life-cycle voltage data for each anode of each location from historical production data,and compared anode-voltage lifetime data at different locations,different life-cycle data at the same location and fluctuations of data at different phases of voltage rise at the same location.A new clustering evaluation index was proposed.The cluster analysis was performed using data features such as anode-to-pole rise phase length,number of speed rises,number of speed rises,number of speed drops,speed drop amplitude,and impact factor,to achieve an assessment of the anode changed quality,and compare the data of various parameters under different quality.An anode key-polarity identification algorithm for electrolytic cells is designed.Piecewise Aggregate Approximation(PAA)is used to reduce the dimension of voltage raw data,and the symbolic aggregation approximation(SAX)algorithm is used to symbolize the anode voltage time series data,so as to realize the mining of the voltage curve variation pattern,and perform similarity analysis on the pattern of changes in anode fluctuations at different locations.Based on the above analysis method and the application of corresponding algorithms,an anode voltage intelligent analysis and mining system was developed to provide basis for intelligent production and decision-making of aluminum electrolytic cells.