Design And Implementation Of Dynamic Weighing System Based On LabVIEW

With the increasing consumption level,the public’s demand for meat products is also increasing,and animal farming has changed from the free family-style farming to the large-scale and barn type.In recent years,smart pastures combining Internet of Things technology and large-scale farming technology have developed rapidly,and animal weight is an important basis for intelligent management such as judging the growth status of animals and grouping according to weight.The body weight information is got in the animal weighing grouping,and the movement of the animal greatly interferes with the weighing result.It is necessary to reduce the error through the dynamic weighing algorithm.The currently widely used dynamic weighing algorithm based on the average filtering algorithm is difficult to improve the weighing speed and accuracy of the system,it is necessary to study more efficient dynamic weighing algorithms.The dissertation designs and implements the dynamic weighing grouping system,develops the data acquisition platform,identity recognition module,temperature acquisition module,detection control module and data acquisition and processing software based on Lab VIEW based on load cells and weight transmitters.The neural network dynamic weighing algorithm is mainly studied by taking the system dynamic weighing data as the object..By comparing the performance of various algorithms,the BP neural network algorithm is finally selected to process the weighing data and the particle swarm algorithm is used to optimize the algorithm.The dissertation completed the static and dynamic weighing experiment.The static weighing test showed that the static weighing error of the system was less than 0.5%.The dynamic weighing test verified that the PSO-BP neural network dynamic weighing algorithm combined with the de-extreme average filtering algorithm can be the most effective.The dynamic weighing error is reduced and the weighing error is within 1kg.The grouping test results show that the system’s grouping accuracy rate is close to 100%,which can satisfy the system requirements.