Interval Multi-Objective Underwater Wireless Sensor Networks Scheduling Optimization Based On The Integration Of Preference And Evolutionary Optimization

In recent years,underwater wireless sensor networks(UWSNs)have comprehensive application prospects in marine environment monitoring,data collection,disaster prevention and control,cruise assistance,and military defense.Due to the need to monitor the marine environment,increasing the coverage of UWSNs has become an important goal of optimizing the network.In addition to coverage,because it is difficult to replace the battery of sensor nodes in the marine environment,it is particularly important to save energy.In addition,reducing the energy span of the UWSNs is also a key to ensuring the efficient operation of the network,prolonging the network survival time.Simultaneous optimization of three goals in UWSNs is essentially a multi-objective optimization problem,and in a complex ocean environment,due to the influence of environmental factors such as ocean currents and tides,the uncertainty of the location of sensor nodes in the network makes the network contains uncertain parameters,so the network model needs to be established as an interval multi-objective optimization model with uncertain parameters.In view of this,this article studies the following:(1)According to the environmental characteristics in the ocean and the main performance indicators of underwater wireless sensor networks,describe the network conditions in actual application scenarios and establishe uncertain interval multi-objective scheduling optimization model with the non-coverage rate,energy consumption and energy span as the network performance evaluation indicators,uses “ Memetic Algorithm for Multi-objective Optimization Problems with Interval Parameters” to simulate the model.Finally,the experimental results are compared and analyzed to verify the superiority of the underwater wireless sensor network scheduling optimization model based on interval multi-objective proposed in this paper in solving practical engineering applications.(2)In some special cases,the decision maker may only focus on a part of the target space,is not interested in other target areas.At this time,the preference of the decision maker should be incorporated as an important factor to guide the population search.In addition,traditional interval multi-objective evolutionary optimization algorithms tend to be difficult for the population to converge due to a sharp increase in non-domination individuals and a reduction in selection pressure in the later stages of evolution.Therefore,based on the network model proposed in the previous section,we combines the preference information provided by the decision maker and propos an interval multi-objective evolutionary optimization algorithm based on reference points and angle preferences.Through experiments,the proposed algorithm proved its effectiveness in the benchmark test function and the interval multi-objective underwater wireless sensor network scheduling optimization model.The thesis includes 30 figures,4 tables and 91 references.