Design Of Calf Feeding And Health Monitoring System Based On Neural Network PID Control

With the improvement of living standards,people pay more and more attention to the balanced intake of nutrients,and dairy products are popular because they are rich in nutrients.The increase in demand for dairy products has in turn driven the boom of the livestock industry,due to the low efficiency and high cost of traditional manual feeding,farms have begun to develop in the direction of semi-automation and automation,among which calf feeding equipment is highly sought-after.A complete set of calf feeding equipment can not only improve the survival rate of calves,but also shorten the growth cycle of calves and improve the economic efficiency of enterprises.In this paper,the development status of calf feeding system at home and abroad is compared and combined with the development of local breeding industry in Ningxia,a calf feeding and health monitoring system based on neural network PID control is designed.The system can automatically monitor the health status of calves and automatically adjust the feeding plan according to the health status to meet the calves’ requirements for constant temperature and rationing of milk.The use of this system can effectively ensure the healthy growth of calves,shorten their growth cycle,reduce the workload of staff,and improve the economic efficiency of the farm.The system adopts modular design ideas,and the algorithm and software and hardware design of calf identification system,health monitoring system and calf feeding system are respectively,and finally the three are combined to form calf feeding and health monitoring system.The calf identification system uses RFID identification technology and adopts EPC coding rules to encode different calves,which can accurately identify the numbers of different calves,which provides great convenience for calf health monitoring and accurate feeding.The health monitoring system uses neural networks to train the mapping between calf age,daily feeding volume,weight,daily weight gain,body temperature,daily exercise and health.The real-time calf condition data collected by STM32 microcontroller can be fed into the neural network to predict its health score and realize calf health monitoring.The calf feeding system uses the PID control algorithm to control the milk at a constant temperature,and the BP neural network PID control algorithm is used to accurately feed calves with different health scores to improve their health.The system uses I.MX6 ULL chip as the control unit,4.3-inch capacitive touch screen for human-computer interaction to assist staff in feeding work,and NB-Io T module for data communication with the server.Finally,the recognition accuracy of the calf identification system is tested.The neural network of the health monitoring system was trained using MATLAB and the effect analysis was carried out.The constant temperature control and precise feeding control of the calf feeding system were simulated and tested respectively.Finally,the function of the host computer is tested whether it is well implemented.A series of test results show that the system basically meets the needs of enterprises for calf feeding and management.