Direct falling weightless type material feeding method based on neural network

A neural network and material technology, applied in the field of quantitative blanking, can solve the problems of only continuous blanking, without considering the effect of weight loss, and unable to meet high-precision blanking, so as to reduce the total error and reduce the fluctuation of blanking rate. Effect
CN107601083AActive Publication Date: 2018-01-19CHINA JILIANG UNIV

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
CHINA JILIANG UNIV
Publication Date
2018-01-19

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a direct falling weightless type material feeding method based on a neural network. A neural network module is established in a controller, and the four input quantities of thematerial level, the material falling rate, the material density and the opening diameter of a feeding valve are mapped as a fed material weightlessness value by the adoption of the neural network; atraining sample is obtained, and offline training is conducted on the neural network; and when feeding control is conducted online, based on signals collected by a bin level sensor in a feeding bin and a weighing module bearing the feeding bin, the neural network predicts the material falling weightlessness value, a processing module corrects the feeding quantity based on the predicted value and then adjusts the closing time of the feeding valve. According to the direct falling weightless type material feeding method based on the neural network, the material falling weightlessness values at different material falling states are predicted by the adoption of the neural network, direct accurate feeding can be achieved, and the direct falling weightless type material feeding method based on the neural network is suitable for small-batch production; the material accumulation state in the feeding bin is detected and adjusted by the adoption of the bin level sensor and a stirrer, so that thematerial falling rate fluctuation is reduced; and the batch feeding total error is reduced through control over the feeding accumulative error.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the field of quantitative cutting, in particular to a neural network-based method for cutting materials in a straight-fall and weightless manner. Background technique

[0002] In industrial and agricultural manufacturing and commodity packaging, there are a large number of powder materials, such as iron concentrate, coal powder and other ironmaking raw materials, polypropylene, polystyrene, polyvinyl chloride, light methyl cellulose, polypropylene nitrile, environmental protection, etc. Chemical raw materials such as oxygen resin powder coatings, building materials such as quartz sand and cement, household chemical products such as washing powder, agricultural products such as millet and soybeans, or processed foods such as powder, slag, and granules, agricultural production such as feed, chemical fertilizers, and pesticides Materials, as well as powdered health care products, Chinese and Western medicines, condiments, etc., al...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More