Research On Remote Monitoring And Fault Diagnosis Methods For Hydraulic Supports

China has always been a large coal country,the economic development can not be separated from the energy power provided by coal,hydraulic bracket as one of the important coal mining equipment,it provides a safe working space for underground operators,is an important support equipment for modern coal mine safety.However,due to the complex underground working environment and the high working intensity of the hydraulic bracket itself,the hydraulic bracket often fails.The traditional diagnosis method is time-consuming and laborious,which seriously interferes with the process of coal mine development,increases the development cost and reduces the economic benefit,and the failure repair may also cause coal mine safety accidents in time.",by means of remote monitoring,to a certain extent,reduce the labor intensity of workers and the risk of casualties.In view of the current situation,this paper conducts a study on remote monitoring and fault diagnosis of hydraulic supports,and the specific work is as follows:(1)Firstly,we explain the working principle and structural composition of hydraulic bracket,analyze the parts where hydraulic bracket often fails and the causes of failure,and then propose a fault diagnosis method based on two-channel convolutional neural network and two-way long and short-term memory network,channel one is a convolutional neural network channel,which first uses wavelet transform to convert the time-frequency map of one-dimensional signal,and then uses convolutional neural network and attention mechanism module for feature extraction,channel two is a two-way long and short-term memory network channel,which directly mines the time-series features of one-dimensional signal,and finally fuses spatial features and time-series features,which can quickly locate the fault of hydraulic system.(2)Due to the influence of coal mining environment,the collected signals are often disturbed,and signal denoising as an important link in signal monitoring and processing,this paper proposes a new denoising method for signal monitoring of hydraulic supports,fully adaptive noise ensemble empirical modal decomposition and improved wavelet threshold fusion denoising method.Firstly,the fully adaptive noise ensemble empirical modal decomposition is performed on the monitoring signal,then the high frequency intrinsic modal function is processed by the improved wavelet threshold algorithm,and finally the processed signal is reconstructed with the low frequency intrinsic modal function and the residual function to obtain the denoised signal,which is more accurate than the one before denoising and also reduces the risk caused by inaccurate monitoring information to some extent.(3)Develop a set of hydraulic support remote monitoring and fault diagnosis system,firstly determine the signal transmission scheme,then select the hardware and software,then use Python language as the basis,Py Qt5 as the framework,then combine with My SQL database technology to realize the development of the system,the system to achieve the functions are mainly user management module,hydraulic support real-time status monitoring module,fault diagnosis module The system mainly includes user management module,realtime status monitoring module,fault diagnosis module,historical data query module,etc.Through the development of the system,the labor intensity of workers is reduced to a certain extent,and it also has a certain reference value in related fields,which has good application prospects.