Research On Key Technology Of Knowledge-based Intelligent Expert System Of Water And Fertilizer Integration In Orchard

Orchard water and fertilizer integration technology involves irrigation system,agriculture,resource management and automation,and has a unique regional and professional.There are many factors to consider in response to the changes of soil characteristics and environment of fruit trees,and the lack of timely response of intelligent fertilization strategy will easily cause serious economic losses and other problems.Moreover,due to the high heterogeneity of multidisciplinary knowledge data structure in orchard water and fertilizer field,it is difficult to realize reasonable knowledge storage,knowledge expression and knowledge management,resulting in the waste of data knowledge resources,data can not be shared and information islands.In view of the above problems,this paper based on knowledge management and expert system and other related theories and technologies,database and expert system and other theories and technologies,to achieve the management of orchard domain knowledge.An orchard water and fertilizer domain ontology model was constructed to describe the relationship between orchard domain concepts in a formal and explicit way.An algorithm for similarity calculation of fertilization strategy in orchard water and fertilizer field was proposed,and an intelligent expert system was constructed by combining reasoning mechanism with orchard water and fertilizer rule base,which improved the accuracy of reasoning strategy.The management of irrigation and fertilization strategy based on orchard water and fertilizer knowledge is realized,and the specific work is shown as follows:（1）System scheme design and knowledge management.Firstly,the lack of fertilization strategy guidance of cloud platform is analyzed,the overall scheme of intelligent expert system is determined,and the key technologies are analyzed.Combined with database technology,structured,unstructured and semi-structured knowledge storage,knowledge expression and knowledge call management were designed to provide support for domain ontology modeling and expert system.（2）The orchard water and fertilizer ontology model and strategy similarity calculation algorithm were constructed.Part-of-speech recognition was carried out through the system dictionary and custom dictionary.The HMM model combined with Viterbi algorithm was used for part-of-speech recognition of undefined words in the dictionary,and Text Rank algorithm was used for extraction.According to the domain knowledge,the extracted concept of orchard water and fertilizer was defined by attribute and relationship,and the domain ontology of orchard water and fertilizer was constructed by Protege.On this basis,the semantic structure similarity of orchard domain ontology,the similarity of orchard water and fertilizer irrigation attribute elements and the similarity of fertilization irrigation instance were calculated.The similarity calculation method of fertilization strategy for orchard water and fertilizer field was proposed,and the fertilization strategy with high similarity was selected for inference,which provided a more comprehensive and targeted similarity calculation algorithm for inference mechanism of expert system.（3）The integrated intelligent expert decision-making system of orchard water and fertilizer was established.The rule base of fertilization strategy of orchard water and fertilizer was established,including fertilization strategy corresponding to different attributes,fertilization model based on nutrient balance method,testing formula of soil nutrients,fertilizer unit conversion formula,etc.The similarity calculation formula and Drools reasoning mechanism were combined to complete the integration of intelligent fertilization strategy.At the same time,the integrated cloud platform and hardware system of orchard water and fertilizer were realized to ensure accurate water and fertilizer operation.