Optimal Design And Simulation Of Mobile Sensor Monitoring Network In Marine Ranch

In recent years,the coastal fishery is suffering from the maritime pollution and fisheries transition has become a task of top priority.By introducing diversified cultivation modes,the modern marine ranch could reduce the cultivation density and protect marine ecology,so as to expand the space for cultivation and boost the traditional fishery.Ministry of Agriculture and Rural Affairs has issued many files(e.g.,"The Research on Development Strategy of China's Marine Ranch")to support the development of marine ranch.Marine ranch could promote the economic growth and the development of marine ecology and fishery,and it is of great significance to monitor the environment of marine ranch for scientific and sustainable cultivation.First of all,a mobile sensor monitoring network(MSMN)model for marine ranch environmental monitoring is proposed in this thesis.The model could monitor the underwater environment of the marine ranch in the real time,and could extend the nearshore network to realize the monitoring of the high seas.Furthermore,MSMN introduces MS to balance the network load and prolong the network life.Secondly,in order to reasonably deploy the network element nodes in the sea area,the MSMN deployment problem is modeled mathematically.Taking the connectivity and reliability of the network,the distance and capacity of the network element nodes as the constraints,the lowest energy consumption of network communication as the optimization objective,the gurobi optimizer is used to optimize the network deployment problem and obtain the network deployment scheme.The validity of the mathematical model is verified with a series of small-scale network,and the scalability of the mathematical model is tested on a large-scale network.Thirdly,in order to optimize the mobile path of MS,a hybrid algorithm of imperialist competition algorithm and ant colony algorithm(ICA-ACA)is proposed in this thesis.The original ant colony algorithm requires multiple iterations to highlight the shortest path through pheromone accumulation,resulting in poor convergence of the algorithm.The assimilation competition of the imperialist competitive algorithm can effectively mine the optimal solution.Combining the characteristics of the two algorithms,a new MS path planning algorithm based on a hybrid algorithm of imperialist competition algorithm and ant colony algorithm is proposed to improve the convergence and the ability of mining new solutions.Finally,a MSMN visualization demo platform is built.The demo platform not only displays the diverse farming areas and the corresponding cultivation mode in three-dimensional way,but also displays the workflow of MSMN and visualizes the network performance parameters.The MSMN dynamic demoplatform omni-directionally presents the dynamic changes of the mobile underwater sensor network in the marine ranch in the real time and has practical significance for the design optimization,performance testing,verification display and the actual deployment of the MSMN network.