Research On Muti-source Data Model Aided Precise Aeration Control For Factory Reciculating Aquaculture

Traditional aquaculture technology has been proved to be difficult to adapt to the needs of social development.The combination of modern information technology and traditional aquaculture industry is the inevitable trend of scientific development of aquaculture industry.In factory recirculating aquaculture,the control of dissolved oxygen plays an important role in cost control and production efficiency.However,nowadays factories often have excessive aeration in low density aquaculture and insufficient aeration in high density aquaculture due to backward technology.Therefore,industrial transformation of aquaculture system,the use of Internet of things technology and artificial intelligence technology to improve it,real-time monitoring of dissolved oxygen information,remote control of aeration equipment,these are of great significance to the intelligent and information of industrial recirculating aquaculture.This study is mainly attached to a small factory recirculating aquaculture system in a university in Chongqing.The main research contents and conclusions are as follows:（1）Construction of aeration monitoring and control system in factory recirculating aquaculture system.Aeration precise control system based on aquaculture of factory recirculating water.The data of aquaculture system can be obtained by camera and dissolved oxygen probe,which can monitor water quality and fish condition in real time.The aeration precision control system based on multi-source data model is completed,which can assist or even replace the artificial aeration decision.The intelligent control of aeration equipment is carried out through the system,which provides a new idea for the construction of monitoring system of factory recirculating aquaculture.（2）Design and implementation of aeration precise control algorithm based on multi-source data model.In this paper,three sub-models are designed to solve the precise aeration control problem.In combination with computer vision,neural network and other technologies,the calculation of fish hypoxia state,fish oxygen consumption rate,aeration and oxygen increase rate is realized,and the optimal aeration strategy is finally decided.In terms of aeration method,the wall type aeration plate was designed,which was combined with the bottom type aeration plate to enhance the aeration efficiency and raise the aeration upper limit.In the algorithm design,we also optimized and solved the problems such as the long operation time of VGG16 model and slow convergence of BP neural network.（3）Accurate aeration control model verification.In this paper,a comparative experiment is conducted in a small factory recirculating aquaculture system in a university in Chongqing.The results show that the energy consumption of the aeration strategy controlled by the precise control algorithm decreases by 26.3% compared with that of the manually controlled aeration strategy,and 12.8% compared with that of the aeration strategy with threshold alarm mode.At the same time,it was put into trial operation in a factory circulating water cleaning breeding plant.In this plant,the operation effect is good.Under the condition of ensuring the health of fish,the stable water quality conditions accelerate the growth of fish,and the breeding cycle with the average weight of fish 400 g as the target is shortened from 5-6 months to 3-4 months.