Research And Development Of The Control System For Fruit Dynamic Weighing And Automatic Sorting Lines

The production of high added-value farming products is an important direction of agriculture development in the 21th century. High quality fruit grading can greatly increase the fruit production tail added value, and contribute to the fruits'international competitiveness and promote the fruit production industrialization and internationalization. The fruit grading based on weighing takes a distinctive position in the fruit sortation because of its fast speed and high accuracy of packing characteristics. The study of the dynamic weighing and automatic sorting has extremely practical economic significance in many fields, such as food, mining, transportation and measurement.However, it is the fact that there is a certain gap between the domestic fruit weighing and sorting system and the imports, not only consisting in weighing precision and sorting speed, but also in the software usability. What's more, the high cost of foreign equipment limits the further development of the domestic fruit industry. It's obvious that the development of suitable systems for domestic market becomes particularly urgent.According to the actual requirements, the thesis carried out the research of dynamic weighing strategy design, automatic sorting strategy design and completed the control system development. The main work we did is as follows:(?) The dynamic weighing signal processing techniques are studied. And based on the prediction of weighing signals, a weighing strategy is proposed. Firstly, the whole process of dynamic weighing signals is explained in detail, and the physical significance of each part of signals is expounded. To reduce the system noises, a noise cancellation method based on adaptive algorithms is adopted. The formula of normalized LMS is derived, and the influences of the order and iteration step on the filtering effect are discussed. Experimental results show that the proposed method has a good effect on restraining of the system noise. In high-speed separation condition, a strategy based on AR model is proposed. It predicted steady state values through self-regression method based on the rising section of the weighing signal. Experimental results shown that the model could predict the steady state value quickly and accurately.(?) The automatic sorting technology is studied and the sorting strategy is designed. According to the moving and synchronous signal features of the fruit weighing and automatic sorting systems, a fruit positioning synchronous tracking method is designed based on the shift operation of the instruction, which has good performance of fruit positioning synchronous tracking and expansibility. As the unequal opportunity of different classification outlets which has the same grading level, a load equilibrium method is proposed based on the token-ring technique. To improve the accuracy of packing and separation efficiency, a method which can be self-correcting according to the cumulative error is proposed.(?) The PCS (PLC-Control-System) based on the three-layer C/S network system was designed. The system composed of operation station and control station. The key programming and development technology of the operation station software and control station software is studied, and relative work including installation, debugging and trial operation is implemented. The results of the system application show that the speed of fruit weighing classification can be up to 14 fruits per second and weighing error less than 2%, which could meet the requirements of actual producing efficiency.