Design And Implementation Of Intelligent Detection System For Cone Valve Based On SCADA

As the wave of the fourth industrial revolution is sweeping across the entire industry,manufacturing informatization and intelligent manufacturing models have come into people’s field of vision.Cone valve inspection is a typical inspection of precision devices.Traditional inspections are done through decentralized processes and the cooperation of multiple people.This detection method is mainly based on manual detection,which is time-consuming and labor-intensive,and cannot be accurately detected for a long time and high intensity,and cannot grasp the operating status of the detection equipment during the detection process and realize the traceability of the cone valve product.For this reason,this paper proposes an intelligent detection scheme of cone valve based on expert diagnosis system and video sequence detection technology.The scheme is mainly researched from the following aspects:First,the key parameters for poppet valve detection are determined.Establish the mathematical model of the cone valve,use AMEsim to conduct simulation experiments according to the mathematical model,analyze the dynamic and static characteristics of the cone valve,and plan the key indicators,detection parameters and test environment requirements of the cone valve.Second,construct the inference rules of the expert system.According to the analysis of the key indicators of the cone valve detection,combined with the pressure-flow characteristic curve in the detection process of the cone valve and the valve opening pressure characteristic relationship in the static characteristic detection process,an expert diagnosis rule based on reasoning is proposed.Through the establishment of pressure-flow rules,the knowledge base,inference engine and inference rules for the quality judgment of the cone valve are constructed.Then,investigate the seal detection method.According to the sealing detection requirements of the cone valve,a video sequence diagnosis method based on blisters is proposed to detect the possible leakage points of the cone valve,such as the end cover,the main airway,and the air control port.Through on-site sampling and analysis of the defect characteristics of the leak point,a target detection model based on YOLOv3 was built.Through the data collection,sample making and model training of each leakage point,the cone valve sealing detection model is obtained.Through the verification of the test set,the average accuracy of the model is as high as 95.5%,and the rate is as high as 15 FPS.Finally,build a SCADA system for cone valve detection.The first step is to analyze the actual needs of the cone valve detection system,and formulate the functional modules of the cone valve detection system according to the requirements.The second step is to design the overall scheme,hardware scheme and software scheme of the SCADA-based cone valve intelligent detection system,and realize the architecture of the SCADA system and the deployment of the main frame of the system.The third step,according to the design plan,adopts the OPC UA data acquisition protocol and the Modbus TCP communication protocol.The Open SCADA system design on the testing platform mainly includes 7 important human-computer interaction interfaces such as the main interface,workstation,parameter setting and alarm list.The expert diagnosis rule and video sequence detection method are applied in the SCADA system to realize the effective detection of the quality of the cone valve.The debugging results show that the detection accuracy of the system can reach99.96%,realizing the traceability of the cone valve detection results.