Research On Intelligent Irrigation Strategy Of Tea Garden Based On Interval Type-2 Fuzzy Algorithm

At present,China’s agricultural cultivation area is extensive,but the irrigation method of crops is still mainly diffuse irrigation.This experience-based irrigation method is difficult to meet the requirements of normal crop growth,and the main problem it brings is the waste of water resources.The purpose of this paper is to carry out research on intelligent irrigation strategy for tea plantations,which aims to guarantee normal crop growth while reducing water demand during irrigation.Through unified analysis of environmental information in tea plantations and expert knowledge of tea plantations,a control strategy for intelligent irrigation in tea plantations is developed using Interval type-2 fuzzy algorithm.The main research contents include.(1)Judgment of abnormal data and extraction of key attributes.In order to reduce the impact of such data on the intelligent irrigation strategy of tea plantations,the Shawville criterion and the before-and-after mean filling method were used to determine the difference repair.In order to reduce the problem of high complexity of the system due to too many input attributes,the CART(Classification And Regression Trees)algorithm was used to analyze the correlation between various attributes and irrigation and evaluate the CART model by classification evaluation index.The experimental results show that the CART model established in this paper has good results in the evaluation of accuracy,precision,recall and F1 value,and the input attributes are reduced from 12 to 4,which reduces the number of samples input by the Interval type-2 fuzzy algorithm and reduces the complexity of the intelligent irrigation system.(2)Interval type-two fuzzification process based on fuzzy clustering.In Interval type-2fuzzy aggregation FOU(Foot print Of Uncertainty;uncertain footprint)ensures that the aggregation has higher uncertainty,but the acquisition of FOU often relies on expert experience.In order to reduce the difficulty of FOU acquisition as well as to improve the readability of tea garden data,a fuzzy clustering algorithm is used to cluster and analyze the exact data collected in the tea garden and perform semantic segmentation using expert knowledge,thus completing the system’s type-one fuzzification process.Multiple homotypic one-type fuzzy sets are fitted using Gaussian function to obtain FOU among homotypic one-type fuzzy sets,and finally complete the process of interval two-type fuzzification.(3)Acquisition of fuzzy rules and reduced-order defuzzification method.In the analysis of tea garden data using CART algorithm to obtain multiple sets of CART model branches,and subsequently convert the multiple sets of branches to semantic segmentation accordingly,so as to extract the corresponding fuzzy rules among them.The reduced-order defuzzification of the type-2 fuzzy sets after fuzzy inference is performed using the EIASC(Enhanced Iterative Algorithm with Stop Condition;Enhanced Iterative Algorithm with Stop Condition).The experimental results show that 68 fuzzy rules can be effectively extracted using the CART algorithm.The Interval type-2 fuzzy irrigation system completed using the above three steps showed good stability in the system simulation,and the practicality of the method was also verified in the subsequent field experiments.