Design And Implementation Of Recommendation System For The Maintenance Of Equipment In Drying Stage Of Cigarette Factory

In the tobacco industry,drying is the most important process in the production of tobacco,and the stable operation of the drying stage equipment has a decisive impact on the quality of the produced tobacco,so it is crucial to maintain the production equipment in a timely manner.Currently,in the cigarette production is mainly through the manual regular inspection of the way to check the operation of the equipment,which have some disadvantages such as poor timeliness and low efficiency.With the increasing degree of integration and scale of production equipment,maintenance personnel may take different repair measures for the same fault due to differences in individual technical level or professional direction when facing complex production equipment.During the maintenance process,there is a lack of corresponding reference schemes,resulting in inadequate maintenance,low maintenance efficiency and excessive reliance on manual experience.In order to improve the stability and maintenance efficiency of the tobacco drying process and ensure the quality of tobacco products,based on an project,this study has researched the fault diagnosis and maintenance plan recommendation methods for tobacco drying equipment in cigarette factories.and finally develops a maintenance program recommendation system for the drying stage equipment in cigarette factories.The main work of this paper is as follows:1.In response to issues such as poor timeliness and low efficiency of manual periodic inspections,this paper implements real-time monitoring of production data by integrating with the underlying data acquisition interface of the drying stage,using the Websocket protocol and visualization technology.And a local sensitive hash algorithm based on fuzzy C-mean clustering is proposed and an equipment fault diagnosis model is constructed.The experimental results show that the average accuracy of the model for fault diagnosis of the equipment in the drying stage is over 90%(under top3 search),which is 8.73% higher than that of the fault diagnosis model based on the Exact Euclidean Local Sensitive Hash algorithm,and achieves self-diagnosis of equipment faults and meets the project design indicators.2.Regarding the problem of maintenance personnel lacking reference solutions during equipment maintenance.This paper constructs an inference rule base based on the historical maintenance records of the equipment in the drying stage of the cigarette factory and the experience of domain experts,and uses the probabilistic soft logic inference framework to construct a recommendation model for equipment maintenance solutions.When a fault occurs in the equipment of the drying stage,the fault data diagnosed by the equipment fault diagnosis model are combined to make program recommendations for the maintenance personnel.The model is evaluated by Normalized Depreciation Cumulative Gain(NDCG)index,and the experimental results show that its NDCG average is higher than 0.8,which achieves the design index of the project.3.According to the project requirements,based on the idea of developing based on front-end and back-end separation,using Vue.js,Spring Boot and other frameworks to design and implement a cigarette factory drying stage equipment maintenance program recommendation system,to achieve real-time data monitoring of the drying stage equipment,fault diagnosis and maintenance program recommendations and other functions and through the corresponding system testing,to meet the project design requirements.