Memory-Enhanced Dynamic Multi-Objective Evolutionary Algorithm Based on Lp Decomposition

Abstract

Decomposition-based multi-objective evolutionary algorithms provide a good framework for static multi-objective optimization. Nevertheless, there are few studies on their use in dynamic optimization. To solve dynamic multi-objective optimization problems, this paper integrates the framework into dynamic multi-objective optimization and proposes a memory-enhanced dynamic multi-objective evolutionary algorithm based on L p decomposition (denoted by dMOEA/D- L p ). Specifically, dMOEA/D- L p decomposes a dynamic multi-objective optimization problem into a number of dynamic scalar optimization subproblems and coevolves them simultaneously, where the L p decomposition method is adopted for decomposition. Meanwhile, a subproblem-based bunchy memory scheme that stores good solutions from old environments and reuses them as necessary is designed to respond to environmental change. Experimental results verify the effectiveness of the L p decomposition method in dynamic multi-objective optimization. Moreover, the proposed dMOEA/D- L p achieves better performance than other popular memory-enhanced dynamic multi-objective optimization algorithms.

Study on Optimization Parameters of Solid-Liquid Separator

“…Take out a part of the roller screen as the research object, establish a coordinate system as shown in Fig. 4, and decompose the received gravity, inertia force, screen support force, and friction force along the radial direction r, axis direction z, and tangential direction [9], [10]. Then the force of the particles in the three coordinate directions is     …”

“…In the equation, θ--the angle between the radial and vertical directions of the particles at a point on the screen, rad; δ--angle between ellipse long axis direction and screen axis, rad; Where a r is negative, because we select the upward direction as the positive direction [10]- [13].…”

A Review of Multi-objective Evolutionary Algorithms for Information Retrieval System

“…The algorithm is structured using a consistent state structure.Although seeking to lighten the scalarization-related problems (generally faced with comparison heading-based strategies), unity between respectable variety and union is preserved using a conspired simple preemptive separation association.A dynamic multi-objective, memory-improved transformative algorithm focused on depletion of Lp (indicated by dMOEA / D-Lp) is proposed in Xu et al . [16] . In specific, dMOEA / D-Lp decays and at the same time advances a complex multi-object optimization question into numerous efficient scalar advancement subproblems.…”

“…The double end configuration allows for distinctly dreadful display of an algorithm of quality to even now have the option of winning everything. Xu et al [48] proposed a MOEA / HD, which layers subproblems into various progressions and adapts the inquiry heading for each subproblem of the lower-pecking order according to the progressive framework and the Xu et al [16]examines systematically the use of EMO for multi-clustering (i.e. simultaneous scanning of numerous bunching).A powerful bi-target model is manufactured in which the quantity of clusters and the whole square distance(SSD) between information centers and their cluster centroids are considered destinations.…”

An Improved Brain Storm Optimization for a Hybrid Renewable Energy System

“…In addition, multi-objective optimization algorithms have also been developed. The typical applications of the multi-objective optimization algorithms including the MOEA/D method [63], the dynamic decomposition method [64], the cognitive-radiobased Internet of Things [65], the hybrid flow shop problems [66], the financial loss problems [67], the rescheduling congestion management problems [68], the multi-attribute group decision problems [69], the stochastic nonlinear systems [70], the flexible job shop problems [71], the lotstreaming flow shop problems [72], the blocking flow shop problems [73]- [74], and other applications [75][76][77][78][79][80][81][82][83][84]. The BSO algorithm was proposed in 2011, and it simulates a kind of collective brainstorming behavior.…”