微粒群优化算法及其改进形式综述

微粒群优化算法是一类新的基于群体智能的启发式全局优化技术,群体中的每一个微粒代表待解决问题的一个候选解,算法利用微粒之间的相互作用发现复杂问题解空间的最优候选区域。该文综述了算法的基本形式及其多种改进形式,并给出了未来可能的研究方向。     

克服恋食行为的PSO算法改进研究

基本粒子群优化算法(PSO)存在易陷入局部极值的缺点.为此,研究鸟群迁徙觅食中的行为习惯,以加强PSO的鸟群社会模型和对鸟群行为的模拟.在所提出的改进算法中,历史飞行速度在实际觅食中不作为判断因子,只有发生位置重复时粒子才发生变异或摄动,以此增强粒子群优化算法跳出局部最优解的能力.实验结果表明,新算法的全局搜索能力有了显著提高.     

基于实数编码的自适应粒子群优化算法

提出了一种新的自适应粒子群优化算法(AMPSO)。该算法在运行过程中根据粒子群多样性的度量指标大小和当前最优解的大小来确定最优粒子的变异概率以对算法进行自适应变异,从而有效地增强了粒子群优化(PSO)算法跳出局部最优解的能力,使PSO算法既摆脱了后期易陷入局部最优点的束缚,又保持了其前期搜索速度快的优点。对几个典型函数的测试结果表明,该算法是非常有效的。     

随机微粒群优化算法

微粒群优化算法是继蚁群算法之后又一种新的基于群体智能的启发式全局优化算法,其概念简单、易于实现,而且具有良好的优化性能,目前已在许多领域得到应用。但在求解高维多峰函数寻优问题时,算法易陷入局部最优。该文结合模拟退火算法的思想,提出了一种改进的微粒群优化算法——随机微粒群优化算法,该算法在运行初期具有更强的探索能力,可以避免群体过早陷入局部极值点。基于典型高维复杂函数的仿真结果表明,与基本微粒群优化算法相比,该混合算法具有更好的优化性能。     

基于粒子群优化和禁忌搜索的混合调度算法

提出了用于解决作业车间调度问题的离散版粒子群优化算法。该算法采用基于先后表编码方案和新的位移更新模型,使具有连续本质的粒子群优化算法直接适用于车间调度问题。同时,利用粒子群优化算法的全局搜索能力和禁忌搜索算法的自适应优点,将粒子群优化算法和禁忌搜索结合起来,设计了广义粒子群优化算法和粒子群—禁忌搜索交替算法两种混合调度算法。实验结果表明,两种混合调度算法能够有效地、高质量地解决作业车间调度问题。     

二阶微粒群算法

为了提高标准微粒群算法的全局收敛性,提出了一种新的微粒群算法——二阶微粒群算法.首先,介绍了二阶微粒群算法的引入,分析了其收敛性,并且研究了其参数的选择范围.其次,在分析二阶微粒群算法的进化方程的基础上,引出了具有随机惯性权重的标准微粒群算法.再次,在二阶微粒群算法中加入振荡因子来调整微粒的速度变化率,更好地使二阶微粒群算法收敛于全局最优.最后,利用这几种改进方法对典型测试函数进行仿真,实验结果表明,这些方法能够有效克服早熟问题,在全局收敛性和收敛速度方面均优于标准微粒群算法.     

基于寿命的粒子群算法研究

针对粒子群算法易陷入局部最优的缺陷,提出了一种具有寿命的PSO(LS-PSO),算法赋予gbest有限的寿命,并且根据其引导能力对寿命进行自适应调整。当gbest耗尽其寿命时,它将失去领导能力,并被一个新产生并经测试具有足够引导能力的粒子所代替,继续引导群体搜索解空间的不同区域,并在两个单峰标准测试函数和六个多峰标准测试函数上对算法进行了测试。结果表明,LS-PSO比传统PSO及改进算法CLPSO有更好的求解精度和收敛速度。     

An improved particle swarm optimiser based on swarm success rate for global optimisation problems

Inertia weight is one of the control parameters that influences the performance of particle swarm optimisation (PSO) in the course of solving global optimisation problems, by striking a balance between exploration and exploitation. Among many inertia weight strategies that have been proposed in literature are chaotic descending inertia weight (CDIW) and chaotic random inertia weight (CRIW). These two strategies have been claimed to perform better than linear descending inertia weight (LDIW) and random inertia weight (RIW). Despite these successes, a closer look at their results reveals that the common problem of premature convergence associated with PSO algorithm still lingers. Motivated by the better performances of CDIW and CRIW, this paper proposed two new inertia weight strategies namely: swarm success rate descending inertia weight (SSRDIW) and swarm success rate random inertia weight (SSRRIW). These two strategies use swarm success rates as a feedback parameter. Efforts were made using the proposed inertia weight strategies with PSO to further improve the effectiveness of the algorithm in terms of convergence speed, global search ability and improved solution accuracy. The proposed PSO variants, SSRDIW_PSO and SSRRIW_PSO were validated using several benchmark unconstrained global optimisation test problems and their performances compared with LDIW-PSO, CDIW-PSO, RIW-PSO, CRIW-PSO and some other existing PSO variants. Empirical results showed that the proposed variants are more efficient.     

Binary particle swarm optimisation with quadratic transfer function: A new binary optimisation algorithm for optimal scheduling of appliances in smart homes

Demand response (DR) is the response of electricity consumers to time-varying tariffs or incentives awarded by the utility. Home energy management systems are systems whose role is to control the consumption of appliances under DR programs, in a way that electricity bill is minimised. While, most researchers have done optimal scheduling only for non-interruptible appliances, in this paper, the interruptible appliances such as electric water heaters are considered. In optimal scheduling of non-interruptible appliances, the problem is commonly formulated as an optimisation problem with integer decision variables. However, consideration of interruptible appliances leads to a binary optimisation problem which is more difficult than integer optimisation problems. Since, the basic version of binary particle swarm optimisation (PSO) does not perform well in solving binary engineering optimisation problems, in this paper a new binary particle swarm optimisation with quadratic transfer function, named as quadratic binary PSO (QBPSO) is proposed for scheduling shiftable appliances in smart homes. The proposed methodology is applied for optimal scheduling in a smart home with 10 appliances, where the number of decision variables is as high as 264. Optimal scheduling is done for both RTP and TOU tariffs both with and without consideration of consumers’ comfort. The achieved results indicate the drastic effect of optimal scheduling on the reduction of electricity bill, while consumers’ comfort is not much affected. The results testify that the proposed QBPSO outperforms basic binary PSO variant and 9 other binary PSO variants with different transfer functions.     

Novel algorithms of particle swarm optimisation with decision criteria

The particle swarm optimisation (PSO) is a stochastic, optimisation technique based on the movement and intelligence of swarms. In this paper, three new effective optimisation algorithms BPSO, HPSO and WPSO, by incorporating some decision criteria into PSO, have been proposed and analysed both in terms of their efficiency, resistance to the problem of premature convergence and the ability to avoid local optima. In the new algorithms, for each particle except position, two sets of velocities are generated and the profit matrix is constructed. Using the decision criteria the best strategy is selected. Simulations for benchmark test nonlinear function show that the algorithms in which the decision criteria have been applied, are beneficial over classical PSO in terms of their performance and efficiency.     

Elucidating multiprocessors flow shop scheduling with dependent setup times using a twin particle swarm optimization

Particle swarm optimization (PSO) is a novel metaheuristic, which has been applied in a wide variety of production scheduling problems. Two basic characteristics of this algorithm are its efficiency and effectiveness in providing high-quality solutions. In order to improve the traditional PSO, this study proposes the incorporation of a local search heuristic into the basic PSO algorithm. The new, hybrid, metaheuristic is called “twin particle swarm optimization (TPSO)”. The proposed metaheuristic scheme is applied to a flow shop with multiprocessors scheduling problem, which can be considered a real world case regarding the production line. This study, as far as the multiprocessors flow shop production system is concerned, utilizes sequence dependent setup times as constraints. Finally, simulated data confirm the effectiveness and robustness of the proposed algorithm. The data test results indicate that TPSO has potential to replace PSO and become a significant heuristic algorithm for similar problems.     

车辆路径问题的改进混合粒子群算法研究

针对各种启发式算法在求车辆路径问题(VRP)中的缺陷,提出了改进的混合粒子群算法(MHPSO)的求解方法.分析了基于速度-位置更新策略传统粒子群算法在解决离散的和组合优化问题的不足.考虑到算法在求解过程中种群多样性的损失过快,引进了种群的多样性测度参数-平均粒距,以保持种群的多样性.同时利用混沌运功的随机性、遍历性和规律性等特性,采用混沌初始化粒子编码.详细讨论了该算法在车辆路径问题中的求解策略.针对同一个实例,将改进的混合粒子群算法与遗传算法从多个角度进行比较.仿真结果表明,论文所提出的算法性能较好,可以快速、有效求得车辆路径问题的优化解或近似优化解.     

带扩展记忆的粒子群优化算法仿真分析

从心理学的角度提出带扩展记忆的粒子群优化算法（PSOEM）,以克服标准粒子群优化算法（PSO）在优化多维函数过程中粒子搜索方向性差、目的性弱的缺陷.采用扩展记忆存储粒子的历史信息,并引入参数表征扩展记忆的重要性.利用经典离散控制理论分析其定值算法的稳定范围.此算法与标准算法是同源异构的,可以与已改进的PSO算法结合使用.基准测试函数的仿真结果验证了所提出算法的有效性.     

基于混合粒子群优化算法的现金流优化

以最大化现金流净现值为优化目标的多模式资源约束调度问题MMRCPSP(Multi-mode Resource-Constrained Project Scheduling Problem)是一类带有复杂非线性特征的NP-hard问题,传统粒子群算法在解决该类离散问题上具有一定局限性。从粒子群算法的优化原理出发,结合遗传算法,在粒子群算法中引入交叉和变异操作,得出一种应用于MMRCPSP现金流优化的快速、易实现的混合粒子群算法,拓宽了粒子群优化算法在离散优化领域的应用。仿真实验结果验证了算法的有效性和高效性。     

Self-organising swarm (SOSwarm)

This paper introduces a novel version of the particle swarm optimisation (PSO) algorithm which we call self-organising swarm SOSwarm. SOSwarm can be used for unsupervised learning. In the algorithm, input vectors are projected into a lower-dimensional map space producing a visual representation of the input data in a manner similar to a self-organising map (SOM). In SOSwarm, particles react to input data during the learning process by modifying their velocities using an adaptation of the PSO velocity update function. SOSwarm is successfully applied to ten benchmark problems drawn from the UCI Machine Learning repository. The paper also demonstrates how the canonical SOM can be explored within the PSO paradigm. Illustrating this linkage between the heretofore distinct literatures of SOM and PSO opens up several new avenues of research for the development of novel self-organising algorithms.     

A two-stage quantum-behaved particle swarm optimization with skipping search rule and weight to solve continuous optimization problem

Quantum-behaved particle swarm optimization (QPSO) is a recently developed heuristic method by particle swarm optimization (PSO) algorithm based on quantum mechanics, which outperforms the search ability of original PSO. But as many other PSOs, it is easy to fall into the local optima for the complex optimization problems. Therefore, we propose a two-stage quantum-behaved particle swarm optimization with a skipping search rule and a mean attractor with weight. The first stage uses quantum mechanism, and the second stage uses the particle swarm evolution method. It is shown that the improved QPSO has better performance, because of discarding the worst particles and enhancing the diversity of the population. The proposed algorithm (called ‘TSQPSO’) is tested on several benchmark functions and some real-world optimization problems and then compared with the PSO, SFLA, RQPSO and WQPSO and many other heuristic algorithms. The experiment results show that our algorithm has better performance than others.     

A simultaneous facility location,vehicle routing and dynamic pricing in a distribution network

This paper incorporates location, pricing and routing decisions by the goal of maximizing profit in a distribution network. In this problem, multiple consecutive time periods are considered in the decision of depot locations at the beginning of the planning horizon, pricing, and routing during each period. According to the varying willingness to pay (w.t.p) of the consumers across different regions and time periods, dynamic regional pricing techniques were incorporated into this problem. In this study, a non-linear mixed integer model is proposed for solving the problem. This model is then converted into a mixed integer quadratic constrained problem that can be solved with the CPLEX solver. Due to the inability of the exact algorithm to solve certain medium and all large instances, and in order to improve the obtained upper bounds for medium test problems, lagrangian relaxation (LR) was introduced. Two pure and hybrid heuristic algorithms are proposed for tackling this problem. The heuristic algorithm includes price optimization and location-routing steps. In the hybrid heuristics, these steps are embedded in the particle swarm optimization (PSO) and self-learning PSO (SLPSO) algorithms framework. Computational experiments illustrate the efficiency of the proposed algorithms. Sensitivity analysis indicates the necessity of switching from the pure heuristic to the hybrid version for scarce capacity settings.     

广义粒子群优化模型

粒子群优化算法提出至今一直未能有效解决的离散及组合优化问题．针对这个问题,文中首先回顾了粒子群优化算法在整数规划问题的应用以及该算法的二进制离散优化模型,并分析了其缺陷．然后,基于传统算法的速度一位移更新操作,在分析粒子群优化机理的基础上提出了广义粒子群优化模型（GPSO）,使其适用于解决离散及组合优化问题．GPSO模型本质仍然符合粒子群优化机理,但是其粒子更新策略既可根据优化问题的特点设计,也可实现与已有方法的融合．该文以旅行商问题（TSP）为例,针对遗传算法（GA）解决该问题的成功经验,使用遗传操作作为GPSO模型中的更新算子,进一步提出基于遗传操作的粒子群优化模型,并以Inverover算子作为模型中具体的遗传操作设计了基于GPSO模型的TSP算法．与采用相同遗传操作的GA比较,基于GPSO模型的算法解的质量与收敛稳定性提高,同时计算费用显著降低．     

基于粒子群优化的蚁群算法在TSP中的应用

结合粒子群算法的问题,提出用混合蚁群算法来求解著名的旅行商问题.问题的核心是应用粒子群算法对蚁群算法的控制参数:启发式因子、信息素挥发系数、随机性选择阈值进行优化,以及运用蚁群系统算法寻找最短路径.新算法对于蚂蚁算法中的参数调整大大减低,减少了大量盲目的实验,力求在开发最优解和探究搜索空间上找到平衡点.对旅行商问题的仿真实验表明,新算法的优化质量和效率都优于传统蚁群算法和遗传算法,接近理论最佳值.新算法也可推广用于其他NP问题的求解.     

改进微粒群算法求解模糊交货期Flow-shop调度问题

针对模糊交货期Flow-shop调度问题的特点,论文提出用微粒群这种具有快速收敛、全局性能好的迭代优化算法进行求解,并使用惩罚函数、增加数据记忆库和自适应变异机制等方法对微粒群算法进行改进,减少了算法陷入局部极值的可能性。通过仿真实例,改进微粒群算法的全局寻优、收敛性和克服早熟的能力均优于遗传、启发式算法。