Research On Data-driven Model Analysis Method Of Raw Material Standards Of Dumpling

The research content of this thesis comes from the sub-project of the National Key R & D Program "Modern Food Processing and Grain Collection,Storage and Transportation Technology and Equipment": "R & D and Industrialization Demonstration of Key Processing Technology Equipment for Nutritional Rice and Special Rice.One of the research tasks is to provide a method of raw material standardization for China's rice products enterprises.This paper starts with the problem of formulating special raw material standards for quick-frozen dumplings.Using the existing production data of quick-frozen dumplings,a three-stage data-driven model analysis method is established.The method can also be applied to similar production scenarios in theory,and can be used to formulate standards for raw materials of other industries and other products.This article contains two parts.The first part is the design of the three-stage data-driven model analysis method,and the second part is the system integration of the three-stage data-driven model analysis method.The three-stage data-driven model analysis method is summarized as follows: First,based on the experimental data of the production of quick-frozen dumplings made from glutinous rice flour raw materials in the "prediction" stage,a suitable regression model is sought to predict the production of quick-frozen dumplings.Then,in the "modeling" stage,an optimization model is established based on the regression model combined with the product quality constraints.The optimization model uses the regression model to transfer the constraints of quick-frozen glutinous rice balls to the glutinous rice flour raw material constraints,and obtain the feasible range of glutinous rice flour raw material quality standards.Finally,the “regulation” step obtains the optimal solution in the feasible region—the solution of the quality standard range of glutinous rice flour raw materials with the characteristics of various quality indicators required to produce quick-frozen dumplings.The above steps have formed a three-stage data-driven model analysis method.Based on this method,experiments on the formulation of special raw material standards for quick-frozen glutinous rice balls were carried out.From the comparison results of the modeling solution conclusion and the analysis conclusion of the experimental sample data,the model method can solve the problem of formulating the quality standard of glutinous rice flour raw materials from the quality standard of quick-frozen rice dumplings.In this paper,after establishing a three-stage data-driven model analysis method,the system integration work is carried out.According to the experimental operation flow,the system processing flow made by the standard of glutinous rice flour raw material for quick-frozen glutinous rice balls was established to improve the applicability of the method.From the previous research literature,the data analysis methods used by the Research Institute of Glutinous Rice Products stay at the research of the correlation of basic indicators.The core content and innovation of this article are: using the idea of optimization modeling to make the formulation of raw material standards as the optimization target of the optimization model,and using the range of target product quality indicators as constraints to establish the optimization model.And the model can search for a better solution in the feasible domain that meets the requirements of the index.