Research On Process Data Adaptive Compression Technology Based On Power Station Big Data

At present,with the development of sensor and communication technology,the number of various signal measuring points in the power station is also increasing.At the same time,with the high sampling rate of signals,the amount of data collected in the power station system is increasing exponentially.In order to solve the storage problem of massive data in power stations,and in view of the fact that the current power station compression technology cannot solve the problem of balance between storage efficiency and compression accuracy and the problem of automatic adjustment of compression parameters,this paper proposes an adaptive compression based on power station process data.Algorithms,and improve the problems that arise in the improved algorithms.The main work of this paper is as follows:First of all,in view of the shortcomings of the current power station compression technology,this paper proposes an adaptive compression algorithm for power station process data.At the same time,in order to realize the self-adaptive adjustment of different data compression processes,this paper combines the BP neural network to design a BP-PID controller to adjust the compression parameters online,and use an improved genetic algorithm to optimize the BP-PID controller.,to further improve the efficiency of the controller;in addition,according to the fluctuation of the original data,an automatic acquisition method of compression accuracy based on data variance is proposed.After simulation tests,under the same compression accuracy,the compression of the adaptive revolving door algorithm is 1-4 times higher than that of the standard revolving door algorithm,and the absolute difference between the expected compression accuracy and the actual compression accuracy is in the order of 10-3,and the improved algorithm is verified.feasibility.Secondly,in the process of testing the adapti ve compression algorithm,it is found that the noise signal has a serious impact on the compression efficiency.In order to solve the influence of noise on compression efficiency,and considering that wavelet transform can perform multi-scale detailed analysis of signals,this paper chooses a wavelet threshold denoising algorithm based on wavelet transform to denoise power plant signal data.At the same time,in order to improve the accuracy of denoising,an adaptive wavelet threshold denoising algorithm is proposed in this paper.The ratio of approximate coefficient energy ratio and wavelet energy entropy is used to determine the optimal wavelet function,and the noise whitening test is used to determine the number of wavelet decomposition layers.The optimal wavelet threshold is determined by adaptive iteration using the new threshold function.The algorithm realizes the autonomous decision of wavelet type,decomposition level and threshold and other parameters.Finally,the feasibility of the adaptive wavelet threshold denoising algorithm is verified through the simulation test.Finally,according to the adaptive revolving door compression algorithm and the adaptive wavelet threshold denoising algorithm,combined with the conventional compression method,and according to the different characteristics of the power station process data,the overall compression strategy of the power station process data is proposed.The compression strategy designs different compression methods according to different data types and signal characteristics in the power station,and solves the problems of poor pertinence and low compression efficiency of the current single compression method,and has significant practical significance.Finally,the field data test verifies the rationality and effectiveness of the compression strategy.