Swarm Intelligence-based Multi-objective Optimization Issue Triggered By Dynamic Events

It has innate skills in multi-objective optimization due to the introduction of the notion of distributed task processing in swarm intelligence optimization algorithm(SIOA),combined with the perfect fit of its own multi-agent cluster features and distributed computing principles.Advantage.As a result,the application of swarm intelligence algorithms to multi-objective optimization problems has been a research focus.We proposed to improve the ability of distributed processing of dynamic multitask objectives through dynamic adaptive grouping of swarm intelligence algorithm to solve multi-objective problems in order to apply the swarm intelligence algorithm and the concept of distributed task processing to the research of dynamic multi-objective optimization problems.Target dispersion causes sluggish convergence and local optimization traps.First,the mixed-level data fusion method is applied to the problem of agent multisource information processing in this study.The state information suitable for system decision-making and control is gained by the rapid and complete analysis of multi-source data obtained by the agent’s multi-sensors,in order to guide and update the agent’s behavior.This method efficiently minimizes the multi-sensor complicated information processing time of the agent by combining the advantages of feature-level data fusion to reduce data volume and good real-time and flexibility of decision-level data fusion.Second,in order to reduce the frequency of control triggers in multi-objective optimization,the dynamic threshold allocation in multi-objective optimization is defined as the game relationship between the task object and the distance by using the concept of game theory,and a dynamic event-triggered mechanism(DETMs)strategy suitable for multi-source information fusion data decision-making is proposed.And the algorithm is applied to the inspection robot control in the substation inspection system to address the issue that manual inspection cannot gather data and enter the system in real time in the actual complicated substation environment,resulting in low inspection efficiency.To secure the stability of the power grid system,the mixed-level data fusion and dynamic multi-objective optimization methods are utilized to handle substation information.Finally,simulation and physical items are used to validate the method’s effectiveness.Furthermore,a Mesh dynamic networking approach based on robot position topology is provided to divide the multi-agent cluster into groups,and a distributed partition job processing strategy is implemented on this basis.Swarm intelligence replaces single-agent control.While assuring particle initiative in adjacent node partitions,it is not constrained by distance and range limits under a fixed topology,and it eliminates task assignment duplication through information exchange within the group.Grouping decreases the scope of global distributed cooperative control and simplifies multi-agent cluster control problem.This topic,in particular,provides a mixed-level information fusion method for single-agent multi-sensor systems and examines its dynamic event triggering mechanism for multi-task objectives.Finally,the data processing technique is utilized to build a distributed collaborative control network of multi-agent clusters in a dynamic networking fashion in order to address the issue that multi-objective optimization is prone to falling into local optimal solutions.The experimental results show that the data processing and swarm intelligence distributed grouping control algorithm implemented under the dynamic event trigger mechanism can effectively deal with dynamic nonlinear multi-objective optimization problems while being low in time complexity and fast in convergence.