Medium-low Gas Regulator Safety Warning System Application And Optimization

Medium-low gas regulator(hereinafter referred to as gas regulator)safety warning technology can diagnose the health status of a component to perform its function in advance,early warn and maintenance of the fault symptom,to avoid the damage of the components.Reasonable safety warning system and gas regulator with good performance,are the key to ensure the normal operation of gas regulator,are of great significance to the safe operation of the whole city gas system,life and property safety of gas users.In this paper,through the application evaluation and data analysis of safety warning system,the accuracy of safety warning system now available is not high and unrealizes intelligent early warning were obtained.In order to optimize the safety warning system.Firstly,transfer the pen position of regulator station,using the form of questionnaire to obtain the judgment result of the operation state of gas regulator,at the same time,the corresponding energy moment values in SCADA system were obtained form sample data.Next,through the comparative analysis of several commonly used machine learning algorithm,the applicability of the support vector machine(SVM)algorithm was determined.in this paper,three SVM safety warning models based on grid search and 10-cross validation to optimize modle parameter were built,through the training and validation of the sample data,the best model with the highest accuracy was confirmed,the accuracy of that model is 76%.The SVM algorithm has a good generalization ability,can be used for the optimization of safety warning technology.Finally,this article refers to the “City gas pressure regulators” specification requirements,detection and analysis the static characteristic of gas regulators,by comparing the performance parameters of the new gas regulator and gas regulator after adjusting,obtain that it is necessary to adjust the regulator before it is put into use.Based on the project of existing off-line and on-line safety warning system,the optimization method of hybrid programming between Matlab and C# language was proposed,to optimize the interface and program of the safety warning system,the system's popularization and application are improved.