Design Of Multi-functional Inspection Well Pipe Network Online Monitoring Systems Based On NB-IoT

Along with the growth and improvement of state financial condition,science and technology,the requirements of urban infrastructure management in the process of modernization are constantly improving.Due to the wide distribution,wide variety and complex internal environment of inspection wells in cities,it is difficult to centralize management and there are certain potential safety hazards,while the traditional management methods are lagging behind.Internet of things technology was proposed as early as the 1990 s.After decades of exploration,it has been applied to all fields of life.NB-Io T technology is an emerging Internet of things technology that has matured little by little in last several years.It has the strong point of low power consumption,wide coverage and access to a great quantity of connections.It fully meets the needs of municipal inspection well pipe network for wireless communication technology.The existing on-line monitoring system for municipal inspection well pipe network generally has some disadvantages,such as unstable data transmission,poor applicability,single function of monitoring system and so on.Therefore,this paper comprehensively uses wireless communication technology,sensor technology,embedded system and computer technology to realize the multi-functional inspection well pipe network on-line monitoring system based on NB-Io T.The system can monitor the temperature and humidity of municipal inspection well pipe network,well water depth,underground harmful gas and well cover.It also provides host monitoring platform with various functions,such as data visualization,abnormal data warning,and rainwater well liquid level prediction and so on.The main contents of this paper are as follows:Aiming at the internal environment and structure of the manhole pipe network,a data acquisition terminal satisfying the demand is designed.Based on STM32L452,the central module and peripheral circuit is devised,and the low-power hardware of the equipment is also be devised.On the basis of the devised hardware circuit,the programming of each module,wireless communication algorithm and low-power software are designed.In view of the complex internal environment of inspection well pipe network and the need to ensure the acquisition accuracy of sensor module as much as possible.The signal conversion circuit of some sensors is designed to meet the low-power and make the digital quantity collected by ADC interface better reflect the change of gas concentration;Write ADC filtering algorithm aimed to reduce random interference and raise the accuracy of signal collection;A temperature and humidity compensation model suitable for the combustible gas sensor module is designed to improve the acquisition accuracy of the combustible gas sensor.The traditional liquid level prediction model can not consider the impact of time series change on liquid level prediction.LSTM neural network is used to build the liquid level prediction model of rainwater well.Through the data collected by the data acquisition terminal and the data collected on the official website of China Meteorological Administration,the LSTM neural network is used to build the liquid level prediction model of rainwater wells.Finally,the internal liquid level of rainwater well is predicted through the liquid level prediction model of rainwater well.Aiming at the problems of difficult unification,informatization and convenient management of acquisition terminals,the Remote inspection well monitoring platform based on MVC architecture is designed.Mongo DB is selected as the database to save the relevant data obtained from the cloud platform;The design of system server is based on Node.js;mongoose is used to operate the database,interact with the cloud platform,and provide various API interfaces for the remote inspection well monitoring page;The remote inspection well monitoring web page is designed based on Java Script to realize the functions of data visualization,abnormal data warning,rainwater well liquid level detection and so on.Finally,the system is tested by a combination of laboratory and field.It mainly includes system operation test,data acquisition terminal test and remote inspection well monitoring platform test.Finally,it is concluded that the whole online monitoring system has the advantages of low power consumption,accurate sensor data collection,and can meet the information and intelligent management needs of municipal inspection well pipe network.