Design Of Dairy Goat Behavior Monitoring And Management System

With the development of animal husbandry in China toward the trend of scale and precision,it requires the modern breeding industry to have a scientific management method and a healthy housing environment.In order to achieve the above requirements,this article has designed a monitoring system for captive dairy goats as a research object to realize the monitoring of its behavior and culture environment during the dairy goat breeding process.The system has been divided into three parts(front-end wearable devices Partially,coordinating node parts and PC-side software systems),and based on the BP neural network training behavior recognition classifier,the milk goats' four kinds of behaviors were classified.Finally,a PC-side milk goat behavior monitoring and management system software was designed and implemented.The main research contents of this article are:(1)Overall system design.Analyze the requirements and functions of the entire project,select the wireless sensor network technology design system,and divide the entire system into three parts: the front-end wearable device part,the coordination node part,and the PC-side software system.The front-end wear portion includes a ZigBee module of the MPU6050 triaxial acceleration gyroscope.The device is worn on a dairy goat to collect three-axis acceleration data of the dairy goat.The coordination node is composed of a ZigBee module integrating the CC2530 chip and a TQ2440 motherboard.The main board is composed of data received from the front-end wearable device and the sheephouse environment sensor and sent to the PC software system.The PC-side software system processes,stores,and classifies and displays the collected data.(2)Identification of milk goat behavioral classification.Due to the complex and changeable daily behaviors of dairy goats,in order to reduce the difficulty of modeling the system,dairy goats were divided into four states in this article: walking,jogging,jumping and standing.After preprocessing the collected acceleration data,three eigenvalues,standard deviation and kurtosis were selected.The BP neural network was used to establish a dairy goat behavior recognizer to classify its behavior.The experimental results show that the overall recognition rate of the four kinds of exercise behavior of milk goats can reach 89.8%,which can meet the actual needs.(3)Design and implementation of dairy goat behavior monitoring management system software.According to the requirements research,this paper uses C# and Matlab language programming,uses MySQL as the system database,uses the Visual Studio2010 development platform to design and develop a C/S framework of milk goat behavior monitoring and management system,complete data receiving and analysis,behavior data Real-time display,classification of dairy goat behavior,query of historical data,threshold setting and alarm of environmental parameters,design of functional nodes such as sensor node management and user management,to realize the classification of behaviors in dairy goat breeding and the real-time supervision function of the breeding environment,and provide information for aquaculture management support.