Research On Product Quality Online Control In The Precision Injection Molding

Plastic injection molding is an important processing technology and is suitable forproducing plastic products with complex shape and accurate dimensional request. High partquality demands are requested with the application of proudest injection molding in variablefields. The quality of injection-molded part is deeply influenced by the material properties,mold design, setting up of the plastic processing parameters, and the environmental conditionin injection molding process. In molding process, process variables directly impact flow Stateof melt in cavity and final quality of part. So, precondition of improving part quality is to getand keep process variables better. But the influence of disturbance makes it more difficult toget the stability of product quality. Therefore, a real-time control of the process parameters ofinjection molding to improve the processing stability of the product and produce high-qualityproducts has an important significance in the injection molding process.This paper mainly studies the relationship between quality indexes and processingparameters, builds the relational model and regulation algorithm between injection moldingprocess parameters and product quality indicators and designs a closed-loop control methodof the injection products quality.At the injection site, setting up the orthogonal experiment which chooses the weight,length and width of product as quality indicators and melt temperature, injection speed, V/Pswitch position and holding pressure as the study factors, studying the impact of processparameters on the product quality indicators and finding that all of the orders that impact theproduct weight, length and width are followed by holding pressure> V/P switch position>injection speed> melt temperature under the existing experimental range and conditions; Inaddition, this paper also increase the single variable experiments to study the impact ofvarious process parameters on product quality indicators trend, and these data is being used toprepare for the model of the relationship between injection molding process parameters andproduct quality indicators and online-control of process parameters.According to the experimental data and the principle of BP neural network, the relationalmodel between main process parameters and product quality indicators is established inMatlab platform. Then the accuracy and generalization ability of BP neural network model is verified by experiment, the BP neural network mode provides the basis for online control ofthe process parameters in injection molding.At last, the impacts of each process parameter on the quality indicators which determinedby the range analysis of orthogonal test are treated as dynamic rules about parameteradjustment order, adjustment step of the process parameters provided by the expertknowledge and the empirical knowledge are treated as empirical rule, then, online control ofprocess parameters is built based on dynamic rules, empirical rules, and neural network model.In the molding process, when the detecting device detects the quality indexes exceeding adefined range, the online adjustment module of injection molding products quality is switchedon, the injection parameters are revised based on the regulatory strategy formulation, the newinjection molding parameters are acted on the injection molding machine to keep the productquality within the required quality range.