Design And Implementation Of Expert System For Smart Tea Garden

China as the origin of the tea culture,tea production should not only meet the domestic demand,but also positive external supply,the use of information technology production,the processing of tea plantations links such as integrated management,with the help of Internet resources to provide accurate agricultural extension service,is to improve the overall production capacity of tea garden,an important means to promote the tea industry structure optimization.Considering that the current information management system used in tea gardens is mostly based on hardware devices,there are problems such as lack of universality and difficulty in promotion.In this regard,this paper designed and implemented a comprehensive and versatile intelligent tea garden expert system to strengthen the orderly management of the whole chain of the tea industry and provide precise guidance tools for relevant personnel in the tea industry.This paper first analyzes the domestic and foreign research status of agricultural expert system,the question answering system,respectively summarizes application technology and development trend,and combining the actual demand of tea plantation operation,wisdom tea garden expert system for the overall design,system's overall function is divided into two aspects: the core decision-making functions of guidance,full service.The specific work is as follows:(1)Adopt the form of automatic question and answer to build the core decision guidance function.In order to ensure the timeliness and accuracy of the automatic question-answer function,this paper trained the Sentence-Bert model on the question-answer data set built by the author in the tea industry,which was used to generate semantic embedding representation of sentences,and combined with the FAISS similarity retrieval database,completed the semantic similarity matching of QA.This method,which uses twin network structure to learn sentence semantic information and obtain sentence embedding,solves the problem of excessive time cost when BERT is applied to semantic similarity retrieval.Compared with Siamese-LSTM method with similar structure,the accuracy of this method is significantly improved.Finally,the model was encapsulated into the system for effect testing,and the test results showed that the automatic question-and-answer function could help the field stakeholders to solve the problems encountered in tea planting and production,which was of great significance to improve the tea yield.(2)According to the comprehensive demand analysis,to achieve a full service management function.To ensure the rationality of the system function,analysis of tea garden in the process of operating the business process,and from the perspective of tea farmers,the administrator,the domain experts,functional requirements,secondly determine for tea cultivation,processing,the integration of production management functions,including farming,postharvest management,procurement management,traceability management,system management,etc.Finally,based on the Spring Boot,Vue,such as technology,implement omni-directional service management expert system for the wisdom of the tea garden.After system deployment and testing,the overall performance of the system is good,and the operation of each functional module is stable.It can realize second-level response for a single user,and can carry a large amount of visits.