Study On Wine Quality Class Classification Based On Adaptive Strategy Whale Optimization Algorithm And Neural Network

With the progress of science and technology,the study of classifying mixtures by quality class according to the nature of compounds in the mixture such as type and content has received more and more attention from scholars and experts.Alcohol products,as a typical mixture with the properties of quality class,naturally attract more people to study.Since 2020,the Chinese wine industry has faced multiple challenges against the background of domestic and international epidemic situations,complex and changing economic trends,and slowing consumption,and the strategic transformation and upgrading of the industry has become inevitable.In particular,the quality classification of wine products is crucial.In the current standard evaluation techniques,the physical and chemical characteristics of alcoholic beverages are less considered and more often relied on manual tasting,and there is not yet a unified and effective quality classification model for alcoholic beverage products.Due to the lack of an assessment scheme for the physical and chemical properties of alcoholic beverages,it is imperative to study a scientific and reasonable model to classify the quality of alcoholic beverages.To address the problem that the above methods cannot accurately reflect the quality of alcoholic beverage products,this paper implements an improved model based on the whale optimization algorithm and BP neural network to explore the feasibility of the WOA-BP model in the classification of alcoholic beverage product quality levels.The model is built on a BP neural network,and after establishing the topology of the network,the wine data set is fed into the network for training,and then the completed network model is used to classify the quality level of wine products.In this paper,we choose the real number encoding method to encode the initial weights and thresholds of the network into the whale population,and optimize the selection operation to adjust the search ability of the whale optimization algorithm for the global optimal solution to improve the problem that the traditional BP neural network converges slowly and is easily trapped in the local optimal solution.The adaptive adjustment strategy is introduced to enhance the global search ability of individual whales and the ability to jump out of the local optimal solution while maximizing the diversity of the population and strengthening the algorithm in the initial exploration breadth and the later local development process to accelerate the convergence of the algorithm.After experimental comparison,the HAWOA-BP model implemented in this paper has a higher classification accuracy compared with the original network model and other traditional classification models for classifying the quality level of alcoholic beverage products,and has a strong practical application value.