| Currently,the main methods of waste disposal are landfill and incineration,which can effectively reduce the accumulation of garbage,but can cause undetectable pollution to the surrounding environment,especially groundwater.Composting wastewater treatment technology can,on the one hand,eliminate garbage,decompose harmful substances such as leachate,and solve the problem of non-point source pollution of water resources caused by waste.On the other hand,sewage can be used as raw material to provide the required water and nutrients required by microorganisms for composting.However,the application of composting wastewater treatment in China has just started,and problems such as low industrial level,high cost,low efficiency,and low compost quality make it difficult to meet domestic demand.With the continuous integration and development of Internet of Things technology and various fields,the application of intelligent control for composting wastewater treatment can effectively solve the problem of composting promotion.Based on the analysis of the development status of the Internet of Things monitoring technology and the analysis of composting wastewater treatment process,this paper combines the Internet of Things monitoring technology with composting technology,builds a RBF real-time prediction model,and designs and implements an intelligent control system for composting wastewater treatment.The system consists of four parts: an offline data acquisition system,an MQTT communication system,a platform monitoring system,and a data prediction system.The system implements real-time monitoring and prediction functions for temperature,humidity,oxygen concentration,carbon dioxide content,and other influencing factors in the composting wastewater treatment process,data collection,storage,and sharing functions,wastewater treatment process management,system maintenance,and other functions.After the development of this system,it can operate stably for a long time in the experimental base,and its functional operation can meet practical needs.It effectively improves the efficiency and quality of composting wastewater treatment,provides convenience for personnel management and maintenance,and plays a positive role in protecting water resources and reducing wastewater treatment costs.The main achievements achieved during the research process include:(1)Build an offline information collection system.Clarify the principle of composting wastewater treatment technology,select suitable sensors,control equipment,and controllers based on the required indicators and control parameters,and design an offline data acquisition system using multisensor fusion technology,achieving real-time and stable data acquisition and process instruction issuance.(2)Devise MQTT communication system.Research and analyze the meaning of the MQTT protocol fields and the message push process.Use EMQ to build a lightweight MQTT proxy server.After testing,the average response time is 75 ms,and the average throughput is 50 bit/s,which meets practical requirements.(3)Build a RBF neural network time series prediction model.Analyze and preprocess the data collected during the composting wastewater treatment process,using humidity,oxygen concentration,and carbon dioxide content as the input feature vectors of the neural network model,and temperature as the predicted output value.Compare the collected temperature data to determine the operation of the entire composting wastewater treatment process.The absolute mean square error of the neural network test set is 0.0081 and can be effectively predicted in experiments.(4)Design and implement an intelligent monitoring system for online composting wastewater treatment process.Collect user needs.The platform system uses the front and rear end separation technology of DRF and VUE to implement a wastewater treatment process monitoring system.Users can manage the basic information of the system,set up and distribute composting wastewater treatment processes,and the response time of the entire website is within 3s.It can operate normally and stably during the two-month experiment. |
