Intelligent Opening and Closing Control Method for Large Hydropower Stations Based on Time-Frequency Feature Subset

In order to solve the problem of difficult opening and closing of cylinder valves caused by imbalance., looseness., wear; etc., a large hydropower station cylinder valve opening and closing control method based on time-frequency feature subset and relay position is studied. Collect vibration signals of large hydropower station cylinder valves., use slice wavelet transform method to extract time-frequency features from the vibration signals of cylinder valves., and construct a time-frequency feature set of cylinder valves. Using the K-means clustering algorithm to cluster the time-frequency characteristics of vibration signals and historical working conditions of large hydropower station cylinder valves., the clustering effect and the number of clustering centers are used to characterize the operating status of cylinder valves and servos. By accurately controlling the position of the servos, the opening and closing process of the cylinder valves is optimized. Combined with the top-down principle, the root node of the decision tree is selected as the initial classification., and the splitting attribute is selected in each node of the decision tree to accurately classify the working conditions., determine whether they belong to the opening/closing attribute., and output the opening and closing control result of the cylinder valves. The experimental results show that this method can use the extracted time-frequency characteristics of the cylinder valve to dynamically analyze the out of step and jamming conditions of the cylinder valve. The mean square error value of the clustering center is less than 0.1., which improves the opening and closing control effect of the cylinder valve.