Design And Optimization Of Intelligent Inspection System For Cryogenic Insulated Cylinder

In recent years,the Internet and information technology industries have developed rapidly,and all walks of life are constantly combining with Internet technologies.With the development of low-temperature storage and transportation technology,the use of cryogenic insulated cylinder has shown an increasing trend year by year.However,the high usage of cryogenic insulated cylinder and the low inspection efficiency of domestic inspection stations have created contradictions.The traditional information transmission methods of inspection stations are inefficient and slow,which seriously affects the inspection efficiency.This paper combines the actual needs of a province's special equipment inspection center,using the technologies of Web,cloud server,database and machine learning,the intelligent inspection system for cryogenic insulated cylinder was designed,and the optimization problem in the static evaporation rate inspection of cylinders was studied.This paper first investigates the information transmission mode and information management status of gas cylinder inspection stations,clarifies the significance and purpose of inspection system development,and analyzes the current status of domestic gas cylinder inspection industry management.Combined with engineering application examples,the functional requirements,feasibility and overall system architecture design of the Web-based intelligent inspection system for cryogenic insulated cylinder were analyzed.Then,the specific functions of user information management module,basic information collection subsystem,internal and external inspection subsystem and security attachment inspection subsystem are designed respectively,and the database of each subsystem is designed.Finally,the design and implementation of the intelligent inspection system for cryogenic insulated cylinder was completed,and the design process of the implementation interface and background management part of each sub-module was elaborated.The basic functions of the design include querying,adding,deleting and modifying the information of the gas cylinder.After the network function of the overall function is completed,the system basically realizes that the gas cylinder information system has passed the electronic stage and realized the paper quality stage.The transformation of the informationization stage.In order to shorten the inspection time and improve the inspection efficiency,this paper takes the gas cylinder data collected in the field as the research object,adopts the BP neural network and Long Short-term Memory neural network model,and uses the stochastic gradient descent as the training method for the instructor to learn the gas cylinder.The static evaporation rate is predicted.After constructing the BP neural network and Long Short-term Memory neural network,the gas data collected in the field is used as the training data to predict the trend of the future cylinder weight change,and the results predicted by Long Short-term Memory neural network model are based on analysis of BP network model prediction results with adaptive learning rate.The results show that the BP network based on adaptive learning rate has certain feasibility in short-term prediction.The Long Short-term Memory neural network model has good effects in both short-term and long-term prediction.