Font Size: a A A

Design And Implementation Of Grain Situation Measurement And Control System Based On LoRa And Neural Network

Posted on:2022-03-19Degree:MasterType:Thesis
Country:ChinaCandidate:D C GeFull Text:PDF
GTID:2481306542462534Subject:Electronics and Communications Engineering
Abstract/Summary:PDF Full Text Request
Food security has always been an important support point for the security of national military strategy and the basic livelihood of farmers,and it has an extraordinary impact on the maintenance of social order,market rationalization and national security.At present,the problems of grain situation measurement and control in my country mainly include the high cost of temperature and humidity collection hardware,and the inaccurate measure and control changes in the granary environment,resulting in damage and waste of grain,and seriously affecting the safety of grain storage and harming the national interests.Therefore,it is necessary to apply wireless sensor technology,low-power and long-distance radio transmission technology(Long Range,LoRa)and neural network knowledge to the field of grain condition for the existing grain condition measurement and control problems,so the main work of this thesis is as follows :(1)Aiming at the problem of the sensor is easy to burn out in bad weather and the root layer number is easy to be disordered,the information collection subsystem with an automatic recovery mechanism for the root layer flight number is designed.It optimized the flash storage algorithm,that designed an automatic recovery plan for the root flight of sensor,it's data processing mechanism is optimized to realize rapid locating and troubleshooting of sensor damage;it's measurement and control processing mechanism is optimized to shorten the average effective time and improve real-time efficiency;it's read-write separation and multi-level drive amplification mechanism is optimized to increase the transmission distance of serial data and increase the number of single-channel mounted sensors.In order to reduce the risk of burning the measurement and control cable when the equipment is working abnormally,the isolation protection module is designed to reduce the damage rate.(2)Aiming at the problems of high transmission power consumption and low system capacity of system,an information transmission subsystem with low power consumption and low packet loss rate is designed.It mainly completed the interaction of the acquisition instructions,control instructions and configuration instructions in the information control and management subsystems,and it designed low power software and hardware schemes of LoRa node and LoRa gateway for the measurement and control of grain conditions.It conducts specific research on the TDMA realization principle and CDMA realization principle of LoRa transmission mechanism,and it designed a new LoRa network CSMA/CA transmission scheme.Through the test,it can be known that reducing the number of receiving and listening modes,improving the success rate of communication,and keeping sleep as much as possible can reduce power consumption and extend the use time;turning on the adaptive rate and perform inter-frequency processing on the gateway,which can effectively reduce the packet loss rate and increase the system capacity.(3)Aiming at the large error of the current granary temperature prediction,an information control and management subsystem is designed,that combines the intelligent measurement and control of the internal and external environmental data of the granary and the user's reverse control.Comparing and analyzing the average error,and establishing a temperature prediction model based on neural network of LSTM,when when the optimizer is adamax,the number of training cycles is 150,and the loss function is MAE,the average error is the lowest,and the temperature prediction results are more accurate,so as to realize the intelligent storage,analysis,processing and prediction of the environmental data inside and outside the granary,and realize the intelligent interaction between the machine and the user.In summary,the grain condition measurement and control system based on LoRa and neural network is designed in this thesis,it is composed of hardware and software parts of each subsystem,it has the advantages of long distance,large system capacity,low power consumption,strong load capacity,safety,rapid fault location,small temperature prediction error,and long-term stable operation and so on,and it is of special significance for the precise measurement and control of the current granary environment and national food security.
Keywords/Search Tags:LoRa, Neural network, Grain situation measurement and control, Low power consumption, System capacity, Temperature prediction
PDF Full Text Request
Related items