Research Of Remote Monitoring System Of Piggery Environment Based On LoRa And Prediction Model Of Ammonia Concentration

With the scale and intensive development of pig breeding,the cultivation density is increasing,and the culture environment is very important for the healthy growth of the pig.It is an important aspect that affects the growth of animals and the economic benefits of production.The external factors that affect pigs include temperature,humidity,ammonia and ventilation.At present,the environmental monitoring system has short transmission distance,poor noise immunity and high power consumption,so it can not be applied to the monitoring of large-scale farms.Based on the Internet of things and grey neural network,this paper designs a remote monitoring system for piggery environment based on LoRa,which can monitor the environment parameters such as temperature,humidity,ammonia and so on,and control the functions of the fan and radiator.The structure of the system includes environmental data acquisition module,wireless data transmission module and remote monitoring platform.The hardware part of data acquisition includes terminal environment collector,concentrator and control mechanism,which realizes acquisition,transmission and control of environmental parameters.The wireless transmission module uses LoRa technology and GPRS module to connect with the server to achieve low power and long distance transmission.The monitoring server receives the environment parameters,uses the grey neural network algorithm,designs the prediction model,takes the ammonia forecast as an example,and carries out the simulation prediction analysis through the Matlab software.The results show that the prediction model is reliable.The remote monitoring platform includes the central server and the Android client.The C/S interface is applied to design the man-machine interaction interface to monitor the piggery at any time.The software part of the system introduces the data receiving and sending program of the lower computer,the function module program of the server,the embedded Matlab interface to realize the prediction function and the design of the Android client function.The system not only collect,transmit,process,store and predict all kinds of sensor information,but also carry out remote control to the field actuators according to the parameters.The system is installed in a breeding base,and the performance of the system is tested.The data transmission of the server and the client is accurate and the human-computer interface is friendly.The user can monitor the present situation of the piggery at any time in the network coverage area and control the switch of the equipment;the LoRa module has a long distance and strong penetration,and The prediction accuracy is 94%,which can save half of the Internet card traffic and battery power.Experiments show that the system has stable operation and high real-time performance,which is of great significance for the application of Internet of things and the development of intelligent pig breeding.