Research On Intelligent Online Fault Detection System Of Working Pneumatic Solenoid Valve

Testing and health management of components at work in industry has a very important role.If there is a detection device that can alert the component to an abnormal situation or predict a possible failure of the component,it would be possible to arrange maintenance in advance to reduce loss.Therefore,the purpose of this article is to develop a detection system to detect the fault of pneumatic solenoid valves.Therefore,the idea of obtaining the dynamic characteristics of the pneumatic solenoid valve by asynchronous method first and predicting the failure of the solenoid valve based on the change of the dynamic characteristics is proposed.In order to reduce the harm caused by erroneous operation during detection,a common voltage signal replaced by a current signal is used as an input signal for indirect testing of dynamic characteristics,and a processing algorithm for a DC current signal and an AC current signal is proposed respectively.Based on this,in order to complete the management of the solenoid valve’s health status,a solenoid valve failure risk prediction algorithm based on LSTM prediction is proposed.Its core idea is to predict the abnormality by referring to the predicted value.In order to conveniently use the detection system in engineering applications,we realize the informationization of the detection system by building a wireless communication network with a star topology based on Bluetooth 4.0 technology,Android smart device and cloud database technology.Android smart device is the BLE master,which can communicate with multiple Bluetooth slaves at the same time and upload the data to the cloud library in time.Because the energy cost of the detection equipment designed in this paper is extremely low,a lithium batterty is embedded for power supply,and a generation is proposed to use the energy of the nearest gas path to charge the battery.In this paper,a tiny pneumatic turbine power generation is designed as an auxiliary part of the detection device,and the optimal design is made using the Gr EA multi-objective optimization algorithm to improve the power generation,efficiency and battery life.The APP developed based on the Android operating system is the core software of human-computer interaction in this system,which is not only the gateway of the wireless sensor network,but also the operation panel of the detection system.It can control and run multiple test projects at the same time and perform data visualization..After the software and hardware of the detection system are constructed,the performance of the transient response time calculation algorithm of the detection system is evaluated.The actual data is used to verify the fault risk prediction algorithm,and the performance of the microturbocharged system is tested.The experiments show that the detection system designed in this paper can realize fault detection and health management of the solenoid valve in working condition.