Development Of Mariculture Optimized Feeding Monitoring System Based On Data Mining

With the improvement of people’s living standards, the quantity of seafood that the market needs also is being improved significantly, which puts forward higher requirements for mariculture. Traditional manual breeding is unable to carry out large-scale cultivation with high-quality due to the labor-intensive, low level of automation, restricting human disturbance and other factors. In order to solve such problems, we are commissioned by the related company and design an aquaculture feeding monitoring system which includes detection technology, control technology, image processing technology and data mining and optimization technology. Currently, the system has been initially applied in aquaculture, and summary or perfection is carried out constantly for future spread.This system is made up of on-site monitoring station and the host station. The microcontroller is the core of the on-site monitoring station which contains control module, feeding monitors control module, water quality testing module, UV sterilization control module, circulating water filtration module, communication module and image acquisition module, used for multi-parameter detection and control. PC monitoring station includes a database and an open monitoring platform. Among them, the database is created by the SQL Server 2000 Enterprise Manager, being used to store data such as history of feeding, water quality and operational records. The open monitoring platform is developed by Delphi7 platform and its main functions include displaying field data, issuing feeding instructions parameters, setting the limit of alarm parameters, Remote control of field devices, dynamic monitoring of fish growth state, recording evaluation data. This system can not only detect and control temperature, dissolved oxygen concentration, PH value and salinity of water the fish survive in in real-time, but also process and analysis images of fish collected at the scene using digital image processing techniques to get the fish the current state of the group’s activity and then complete evaluation of the amount of feeding fish demand for. In addition, we can find scientific laws of fish growth from massive data by adopting data mining technology, and then establish a cost-effective strategies aiming at fish aquaculture to increase fish aquaculture production, quality and production efficiency.