Research On Operation Information Monitoring And Jam Fault Warning Method Of Combine Harvester Group

With the trend of agricultural intensification,scale and industrialization,the cooperative operation mode of multiple combine harvesters began to appear in the crop harvesting task,which is called combine harvester group mode.At present,the research on the combine harvester group is mainly related to navigation and group control,and there is less research on the operation information monitoring under the group operation mode.Combine harvester,a system with complex mechanical transmission structure,is prone to component blockage and other faults,which heavily reduce the efficiency of the whole combine harvester group.Therefore,it is necessary to monitor the operation information of the combine harvester group in real time,and warn the fault early in case occurrence of fatal fault.Therefore,according to the operation characteristics and group operation scenarios of combine harvesters,this paper designed and studied an operation information monitoring and jam fault warning method of combine harvester group with some related technologies.The main research contents are as follows.(1)Main scheme of operation information monitoring and jam fault warning method under combine harvester group mode.After studying the multi-machine operation mode of combine harvester and the literature related to combine harvester monitoring system,this paper designed an operation information monitoring system of combine harvester group.The MCU system installed on combine harvester collects the operation parameter information,and summarizes their data to the ARM system through LoRa technology.Then,the QT program running on the ARM system is responsible for managing and monitoring the operation status of the combine harvester group,and uploading all data to the database of the network server.(2)Research on combine harvester group wireless communication system based on 2.4GHz LoRa technology.This paper studied the communication network structure in the scenario of combine harvester group.Firstly,the communication model is established.With the help of network simulation software,NS-3,the influence of application parameters in LoRa technology is simulated and analyzed.According to the simulation experiments,the best allocation scheme of spreading factor in LoRa network is selected.The results show that the comprehensive performance is better with specific allocation proportion;The number of nodes that can be supported by LoRa communication network corresponding to different transmission intervals is analyzed.It is concluded that about 25 nodes can be supported to communicate at the transmission interval of 1s under the bandwidth of 1625khz;The RSSI and delay performance of LoRa communication network in dynamic scenario are analyzed.The minimum overall delay in communication is about 10 ms.(3)Hardware design of the system.According to the overall design scheme of the system,the hardware modules are selected,including MCU module,ARM module,satellite positioning module,rotation speed sensor,LoRa communication module,human-computer interaction module,etc.According to the functional requirements of the MCU system,the peripheral circuits are designed,including power supply circuit,rotation speed acquisition circuit,CAN bus conversion circuit,RS232 conversion circuit and so on.(4)Research on the jam fault warning method of combine harvester.By analyzing the existing diagnosis and warning methods for the rotating parts blockage,the GRU neural network method is selected to alarm the jam fault of combine harvester.After preprocessing and data enhancement of the collected data set,the GRU neural network is trained,and the neural network results are analyzed according to different performance evaluation criteria.The test set results show that the GRU neural network can well complete the warning task of jam fault,and the prediction accuracy can reach99% when using the window sequence with length of 10.(5)Software design of the system.The software development is divided into the program development of MCU system and ARM system.The software design of MCU system is developed with C,mainly including rotation speed measurement program,satellite positioning program,different communication bus processing program,LoRa communication program and human-computer interaction program design.For ARM system,QT Quick development framework is adopted,QML and C++ are used for development,mainly including GUI design,communication program with hardware terminal,group management program and network communication program design.(6)Operation information monitoring system field experiments of combine harvester group.The experiments include operation data acquisition test,group communication quality test and jam fault warning model test.The test results show that the system can realize the function of monitoring the operation information of the combine harvester group,that is,the operation data of the combine harvester is completely collected and successfully transmitted to the database of the network server.The communication quality test results show that the average success rate of data packet reception of LoRa network in report mode is 99.3%,the average success rate of communication in response mode is 92.5%,the average response time is 122.07 ms,and the response time to the server is 126.6ms.The test of jam fault warning model shows that the trained GRU neural network can effectively predict the blockage degree.