Design And Implementation Of Smart Water Platform Based On Big Data Technology

Water sector from the process of informatization to the exploration of wisdom,wisdom water platform with the development of the Internet of things and big data technology arises at the historic moment and is becoming an important part of how to effectively manage water construction,mining,predict a large number of water quality data acquisition device has become a bottleneck which restrict the development of the intelligent water platform.Therefore,it is necessary to design a reasonable technical scheme to develop intelligent water platform.Based on the above background,this thesis designs and implements a smart water platform based on big data technology.Based on the detailed analysis of the business process and function of the water platform,the architecture design and database design are carried out in this thesis.Aiming at the deficiency of single water application,the design of each module of the platform is completed with the idea of micro-service.Kafka message middleware is used to buffer water data from different water plants for downstream consumption.The data end uses Hadoop and Spark frameworks to perform offline and real-time calculation of water data.The platform uses The Influx DB timing library to store the timing data of the acquisition device,and uses Redis to cache the data after real-time calculation.In this thesis,according to the functional characteristics and business boundary of the business layer,the detailed design and implementation of the module includes:platform management module,water plant management module,water plant data acquisition module,water plant data analysis module,water plant data retrieval module,water plant equipment inspection order module and water quality prediction module.The platform management module includes user,role,permission management,configuration management and log management.Water plant management module includes water plant organization,object,equipment configuration,operation function.The data acquisition module of water plant realizes the business logic of equipment start-stop control,gateway data acquisition,data entry,equipment status monitoring and so on.The data analysis module of water plant realizes the statistical analysis function of real-time data and off-line data.Real-time analysis mainly includes index analysis of region,water plant and water station.Offline analysis mainly includes data report analysis and report management.The water plant data retrieval module realizes the water quality retrieval function.Water plant equipment inspection single module to achieve equipment management,inspection management and other functions.Water quality prediction module uses DBN-BP and LSTM algorithm to predict water quality parameters in the short term.The platform has completed the development of core functions.The platform has been connected to the data of various water plants in a city,and the function and performance tests have been carried out.The test results show that the design scheme based on big data architecture can solve the performance problems of storage and computing to a certain extent,and has good scalability.After operation,the platform can meet the daily needs of water operation personnel and provide convenient and efficient services for water operation personnel.The platform has reached the expected goals,which has certain reference significance.