State Recognition Of Vibration Signals Of Hydropower Units Based On Multifractal And Gravity Search Algorithm

As the core equipment to convert water energy into electric energy,the operation state of hydropower unit is of great significance to the power plant and even the society,and it is also the basic guarantee for the effective development and efficient production of hydropower energy.Therefore,timely and accurate monitoring and identification of the operation status of the unit to ensure the healthy operation of the unit is of great significance to the safe operation of the power plant,the stability of the power grid and the development of the society.Due to the non-linear and non-stationary characteristics of the vibration signals of hydropower units,it is difficult to accurately extract the characteristics of hydropower units and correctly identify the status of units according to the characteristics,which makes it a hot and difficult research topic.In order to accurately extract the vibration signal feattures which can represent the state of hydropower units,the vibration signal processing and feature extraction methods are studied in this paper.By comparing the signal features extracted by EMD fuzzy entropy method and multifractal detrend fluctuation analysis(MFDFA)method,several classifiers are used to select the feature extraction method with better representation ability.Through the research of vibration signal feature extraction method and classifier algorithm,it is found that there are many kinds of multifractal features of vibration signal extracted by MFDFA method,and the classifier parameters also need to be set artificially.In order to solve this problem,this paper uses gravity search algorithm to reduce the dimension of feature.At the same time,the algorithm is used to find the optimal classifier parameters,and the bearing fault data set is used for example analysis and verification.According to the analysis results,this paper puts forward a state identification method of vibration signals of hydropower units based on Multifractal and gravity search algorithm,and compares and verifies the vibration signals measured by hydropower units.The results show that compared with EMD fuzzy entropy feature,the multifractal feature of hydropower vibration signal can better characterize the operation state of hydropower units.After the optimization of gravity search algorithm,the performance of the classifier is given full play,and a better recognition effect is obtained,which has effectiveness and practical value.