Desigin And Implementation Of Rule Engine Module Of Intelligent Pension Service Platform

Recently,with the arrival of aging society and the rapid development of information and communication technology in China,the intelligent pension service platform developed by using information and communication technologies such as artificial intelligence,big data,Internet of things,cloud computing,edge computing,mobile Internet and 5g has become a hot spot.The intelligent pension service platform will have a large number of devices connected to the Internet of things platform,and the problems of complex linkage relationship between devices,high business logic and device data coupling need to be solved.In order to solve the above problems,this thesis is based on the existing intelligent pension service platform architecture according to the technical characteristics of the intelligent pension service platform network architecture and rule engine,using edge computing technology,the functions of rule control and distribution and alarm plug-in management are designed and implemented in the cloud,and the functions of streaming rule processing and rule offline autonomy are designed and implemented in the edge.This thesis deeply discusses the rule cloud edge collaboration mechanism,edge rule engine and edge list watch mechanism,and puts forward a design and implementation scheme of rule engine for intelligent pension service platform.The main work is as follows:1.In order to solve the problem of computing pressure and data loss of intelligent pension cloud service platform,this thesis designs and implements a rule engine architecture of cloud side collaboration.The test results show that the architecture designed in this thesis effectively reduces the cloud computing bandwidth by 59.98%.And a regular cloud side collaboration mechanism is proposed,and the mechanism proposed in this thesis is verified from three levels:reliability,order and timing synchronization of data communication.The test results show that the mechanism proposed in this thesis can provide data consistency guarantee for the intelligent pension cloud platform,and control the packet loss rate below 1%.2.In order to solve the problem of large emergency response delay of intelligent pension cloud service platform,this thesis designs and implements a lightweight rule engine deployed at the edge based on Rete algorithm.The rule engine is encoded in go language and runs in binary form.The test results show that the memory occupation of the rule engine is only about 10m when running,realizing lightweight deployment;In order to further improve the execution speed of the rule engine,the Rete algorithm is optimized through the combination of rule preprocessing and changing the rule serial matching to parallel,which improves the number of data processing per unit time and reduces the response delay from second level to millisecond level,making it more suitable for rule processing in case of emergency for the elderly in the intelligent pension service platform.3.In order to improve the intelligence of the intelligent pension Internet of things platform,this thesis proposes an edge list watch mechanism,which can provide the ability of regular edge autonomy of inter device linkage.This mechanism ensures that the linkage between devices can not be completely controlled through the cloud to realize the linkage work at the edge.The test results show that compared with the cloud control mechanism,this mechanism improves the number of device linkage rules successfully processed in unit time,and improves the intelligence and autonomy of the edge device linkage of the intelligent pension platform.4.Design and implement the rule management module in the cloud based on the spring cloud framework.The system background module mainly includes sub modules such as rule dynamic configuration,alarm plug-in management and historical exception statistics.At the same time,it realizes the functions of sending email,SMS alarm and log storage,which is convenient for the operation and maintenance personnel to uniformly configure the rules in the cloud,and quickly understand and deal with the alarm triggered by the rules.It can be used as a reference for the rule engine design of intelligent elderly care scenario.