多智能体粒子群算法在配电网络重构中的应用

结合多智能体的学习、协调策略及粒子群算法,提出了一种基于多智能体粒子群优化的配电网络重构方法。该方法采用粒子群算法的拓扑结构来构建多智能体的体系结构,在多智能体系统中,每一个粒子作为一个智能体,通过与邻域的智能体竞争、合作,能够更快、更精确地收敛到全局最优解。粒子的更新规则减少了算法不可行解的产生,提高了算法效率。实验结果表明,该方法具有很高的搜索效率和寻优性能。     

一种基于多智能体进化的广义图染色算法

基于对广义图染色问题的研究,提出了一种求解广义图染色问题的多智能体进化算法(multiagent evolutionary algorithm for T-coloring problem,简称MAEA-TCP),并将该算法应用到实际中的频率分配问题上,取得了良好的效果.该方法中每个智能体作为一个候选解被固定在智能体网格上,为了增加自身能量而与邻域当中的智能体展开竞争或者合作,同时智能体也可以利用自身的知识进行自学习来增加能量.根据广义图染色问题的特点,为智能体设计了3种算子:竞争算子、自学习算子和变异算子,以引导其进化,并用进化的方式来控制各算子,以协调智能体之间的相互作用.在实验中,分别使用大规模的随机图实例和费城实例来测试算法性能,同时给出参数测试结果和最佳取值区间.比较结果表明,该算法优于其他方法,具有良好的收敛性和实用价值.     

具有轮盘反转算子的多Agent算法用于线性系统逼近

针对John Holland的反转算子在数值优化中的不合理性, 提出了一种轮盘反转算子来克服这种不合理性,并结合该算子提出了一种多Agent进化算(RAER), 证明了算法的全局收敛性. 无约束优化仿真实验表明, 该算法性能好于其他算法. 在求解线性系统逼近工程优化问题时, 无论在固定区域还是动态扩展区域搜索, 算法都能得到更好的模型, 较其他算法能够对搜索区域进行更为充分的探索和求精. RAER算法是实际有效的.     

用蛙跳算法优化RBF神经网络参数的研究

针对径向基函数(Radial Basis Functions,RBF)神经网络结构参数确定问题,提出了一种基于蛙跳算法优化RBF神经网络参数的新方法。将RBF神经网络参数组成一个多维向量,作为蛙跳算法中的参数进行优化。以适应度函数为标准,在可行解空间中搜索最优解,并对蛙跳算法进行了改进。非线性函数逼近实验结果表明,该优化算法相对标准遗传优化算法、粒子群优化算法有较小的均方误差,具有更好的逼近能力。     

基于异步优势执行器评价器的自适应PID控制

自适应PID较好地解决了传统PID无法自整定参数的问题,已成为控制领域内的研究热点。研究基于异步优势执行器评价器(Asynchronous Advantage Actor-Critic,A3C)算法设计了一种新的自适应PID控制器。该控制器利用A3C结构的多线程异步学习特性,并行训练多个执行器评价器(Actor-Critic,AC)结构的智能体,每个智能体采用多层前馈神经网络逼近策略函数和值函数实现在连续动作空间中搜索最优的参数整定策略,以达到最佳的控制效果。与已有的多种自适应PID控制器性能对比分析结果表明该方法具有收敛速度快,自适应能力强的特点。     

一类新颖的粒子群优化算法

粒子群优化(PSO)是一类有效的随机全局优化技术。它利用一个粒子群搜索解空间,每个粒子表示一个被优化问题的解,通过粒子间的相互作用发现复杂搜索空间中的最优区域。提出一类新颖的PSO算法,该算法在基本PSO算法的粒子位置更新公式中增加了一个积分控制项。积分控制项根据每个粒子的适应值决定粒子位置的变化,改善了PSO算法摆脱局部极小点的能力。另外,该算法增加了限制搜索空间范围的机制,这对某些函数优化问题是必需的。用5个基准函数做的对比实验结果显示,该算法优于基本PSO算法以及自适应修改惯性因子的PSO算法。     

基于Bloch 球面坐标编码的量子粒子群算法及应用

为提高粒子群算法的优化效率,在分析粒子群优化算法的基础上,提出了一种基于Bloch球面坐标编码的量子粒子群优化算法。该算法每个粒子占据空间三个位置,每个位置代表一个优化解。采用传统粒子群优化方法的搜索机制调整量子位的两个参数,可以实现量子位在Bloch球面上的旋转,从而使每个粒子代表的三个优化解同时得到更新,并快速逼近全局最优解。标准测试函数极值优化和模糊控制其参数优化的实验结果表明,与同类算法相比,该算法在优化能力和优化效率两方面都有改进。     

一种新型的非线性系统模型参数辨识方法

针对传统模型参数辨识方法和遗传算法用于模型参数辨识时的缺点,提出了一种基于微粒群优化(PSO)算法的模型参数辨识方法,利用PSO算法强大的优化能力,通过对算法的改进,将过程模型的每个参数作为微粒群体中的一个微粒,利用微粒群体在参数空间进行高效并行的搜索来获得过程模型的最佳参数值,可有效提高参数辨识的精度和效率.     

蚁群算法求解分布式系统任务分配问题

蚁群算法是受自然界蚂蚁觅食过程中,基于信息素的最短路径搜索食物行为的启发提出的一种智能优化算法.研究表明,在求解复杂优化问题方面该算法具有一定的优越性.任务分配问题是一类典型的组合优化问题.应用蚁群算法来解决多处理器分布式系统上的任务分配问题,一个任务只能分配给一个处理器处理,而一个处理器可以处理多个任务,其中每个处理器都有固定成本和能力限制.仿真结果表明,该算法比禁忌搜索和随机方法具有更好的求解能力.     

一种基于进化泛函网络的建模与函数逼近方法

提出了一种进化泛函网络的建模与函数逼近方法,该方法把泛函网络建模过程转变为结构和泛函参数的优化搜索过程,利用遗传规划设计泛函网络神经元函数,对网络结构和参数共存且相互影响的复杂解空间进行全局最优搜索,实现泛函网络结构和参数的共同学习,并用混合基函数实现目标函数的逼近,改变了人们通常用同类型基函数来实现目标函数逼近的方式.数值仿真结果表明,提出的网络建模与逼近方法具有较高的逼近精度.     

A multiagent genetic algorithm for global numerical optimization.

In this paper, multiagent systems and genetic algorithms are integrated to form a new algorithm, multiagent genetic algorithm (MAGA), for solving the global numerical optimization problem. An agent in MAGA represents a candidate solution to the optimization problem in hand. All agents live in a latticelike environment, with each agent fixed on a lattice-point. In order to increase energies, they compete or cooperate with their neighbors, and they can also use knowledge. Making use of these agent-agent interactions, MAGA realizes the purpose of minimizing the objective function value. Theoretical analyzes show that MAGA converges to the global optimum. In the first part of the experiments, ten benchmark functions are used to test the performance of MAGA, and the scalability of MAGA along the problem dimension is studied with great care. The results show that MAGA achieves a good performance when the dimensions are increased from 20-10,000. Moreover, even when the dimensions are increased to as high as 10,000, MAGA still can find high quality solutions at a low computational cost. Therefore, MAGA has good scalability and is a competent algorithm for solving high dimensional optimization problems. To the best of our knowledge, no researchers have ever optimized the functions with 10,000 dimensions by means of evolution. In the second part of the experiments, MAGA is applied to a practical case, the approximation of linear systems, with a satisfactory result.     

An Effective Way to Neighborhood Construction for MAGA

The MAGA is an effective algorithm used for global numerical optimization problems. Drawbacks, however, still existed in the neighborhood selection part of the algorithm. Based on the social cooperate mechanism of agents, an effective neighborhood construction mode is proposed. This mode imports an acquaintance net which describes the relation of agents, and uses that to construct the local environment (neighborhood) for agents. This strategy makes the new mode more reasonable than that of MAGA. The Multi-Agent Social Evolutionary Algorithm (MASEA) based on this construction mode is introduced, and some standard testing functions are tested. In the first experiments, two dimensional, 30 dimensional and 20-1000 dimensional functions are tested to prove the effectiveness of this algorithm. The experimental results show MASEA can find optimal or close-to-optimal solutions at a low computational cost, and its solution quality is quite stable. In addition, the comparative results indicate that MASEA performs much better than the CMA-ES and MAGA in both quality of solution and computational complexity. Even when the dimensions reach 10,000, the performance of MASEA is still good.     

Robust cooperative output regulation of heterogeneous Lur'e networks

In this paper, we study robust cooperative output regulation problems for a directed network of Lur'e systems that consist of a nominal linear dynamics with an unknown static nonlinearity around it through negative feedback. We assume that the linear part of each agent is identical, but the nonlinearities are allowed to be different for distinct agents. In this sense, the network is heterogeneous. As is common in the context of Lur'e systems, the unknown nonlinearities are assumed to be sector bounded within one given sector. The interconnection graph among these agents is assumed to contain a directed spanning tree. Similar to classical output regulation problems, there is a virtual exosystem generating a reference signal in which all the agents are required to track cooperatively. Our designed distributed dynamic state/output feedback protocol makes a copy of the reference signal at each agent asymptotically, and then the robust cooperative output regulation problem becomes a robust tracking problem that can be handled by each agent via local information. It turns out that our cooperative protocols are fully distributed. Sufficient conditions on the existence of output synchronization protocols are given along with some discussions on these conditions. Finally, two simulation examples illustrate our design. Copyright © 2016 John Wiley & Sons, Ltd.     

Coordinating Agents in Organizations Using Social Commitments

One of the main challenges faced by the multi-agent community is to ensure the coordination of autonomous agents in open heterogeneous multi-agent systems. In order to coordinate their behaviour, the agents should be able to interact with each other. Social commitments have been used in recent years as an answer to the challenges of enabling heterogeneous agents to communicate and interact successfully. However, coordinating agents only by means of interaction models is difficult in open multi-agent systems, where possibly malevolent agents can enter at any time and violate the interaction rules. Agent organizations, institutions and normative systems have been used to control the way agents interact and behave. In this paper we try to bring together the two models of coordinating agents: commitment-based interaction and organizations. To this aim we describe how one can use social commitments to represent the expected behaviour of an agent playing a role in an organization. We thus make a first step towards a unified model of coordination in multi-agent systems: a definition of the expected behaviour of an agent using social commitments in both organizational and non-organizational contexts.     

智能体遗传算法空间维数的比较研究

将智能体引入到遗传算法构成一个局部环境,可有效保持种群的多样性从而获得优良的优化性能。但是这个局部环境的空间维数一直未得到研究。根据智能体遗传算法的工作原理,空间维数越小,越能避免过早收敛现象发生。基于此,提出一种维数为1的链式智能体遗传算法(CAGA),并针对函数优化问题将其与文献[4]提出的维数为2的网络式智能体遗传算法(MAGA)进行了比较。实验采用了多个多维复杂函数进行优化实验,结果表明,该遗传算法比二维网格式遗传算法可获得更优的优化结果。     

Finding influential agent groups in complex multiagent software systems based on citation network analyses

Current complex engineering software systems are often composed of many components and can be built based on a multiagent approach, resulting in what are called complex multiagent software systems. In a complex multiagent software system, various software agents may cite the operation results of others, and the citation relationships among agents form a citation network; therefore, the importance of a software agent in a system can be described by the citations from other software agents. Moreover, the software agents in a system are often divided into various groups, and each group contains the agents undergoing similar tasks or having related functions; thus, it is necessary to find the influential agent group (not only the influential individual agent) that can influence the system outcome utilities more than the others. To solve such a problem, this paper presents a new model for finding influential agent groups based on group centrality analyses in citation networks. In the presented model, a concept of extended group centrality is presented to evaluate the impact of an agent group, which is collectively determined by both direct and indirect citations from other agents outside the group. Moreover, the presented model addresses two typical types of agent groups: one is the adjacent group where agents of a group are adjacent in the citation network, and the other is the scattering group where agents of a group are distributed separately in the citation network. Finally, we present case studies and simulation experiments to prove the effectiveness of the presented model.     

Pareto optimization for the two-agent scheduling problems with linear non-increasing deterioration based on Internet of Things

The Internet of Things (IoT) enables these objects to collect and exchange data and it is an important character of smart city. Multi-agent scheduling is one necessary part of Internet of Things. In this paper, we investigate the Pareto optimization scheduling on a single machine with two competing agents and linear non-increasing deterioration, which is Multi-agent scheduling problems often occurred in the Internet of Things. In the scheduling setting, each of the two competing agents wants to optimize its own objective which depends on the completion times of its jobs only. The assumption of linear non-increasing deterioration means that the actual processing time of a job will decrease linearly with the starting time. The objective functions in consideration are the maximum earliness cost and the total earliness. Two Pareto optimization scheduling problems are studied in this paper. In the first problem, each agent has the maximum earliness cost as its objective function. In the second problem, one agent has the maximum earliness cost as its objective function and the other agent has the total earliness as its objective function. The goal of a Pareto optimization scheduling problem is to find all Pareto optimal points and, for each Pareto optimal point, a corresponding Pareto optimal schedule. In the literature, the two corresponding constrained optimization scheduling problems are solved in polynomial time under the assumption that the inverse cost function of each job is available. In this paper, we extend these results to the setting without the availability assumption. Furthermore, by estimating the number of Pareto optimal points, we show that the above two Pareto optimization scheduling problems are solved in polynomial time. Hence, our results have much more theoretically meaningful constructs. Experimentation results show that the algorithms presented in this paper are efficient.     

Autonomous Adaptive Agents for Single Seller Sealed Bid Auctions

In developing open, heterogeneous and distributed multi-agent systems researchers often face a problem of facilitating negotiation and bargaining amongst agents. It is increasingly common to use auction mechanisms for negotiation in multi-agent systems. The choice of auction mechanism and the bidding strategy of an agent are of central importance to the success of the agent model. Our aim is to determine the best agent learning algorithm for bidding in a variety of single seller auction structures in both static environments where a known optimal strategy exists and in complex environments where the optimal strategy may be constantly changing. In this paper we present a model of single seller auctions and describe three adaptive agent algorithms to learn strategies through repeated competition. We experiment in a range of auction environments of increasing complexity to determine how well each agent performs, in relation to an optimal strategy in cases where one can be deduced, or in relation to each other in other cases. We find that, with a uniform value distribution, a purely reactive agent based on Cliff’s ZIP algorithm for continuous double auctions (CDA) performs well, although is outperformed in some cases by a memory based agent based on the Gjerstad Dickhaut agent for CDA.     

Minimizing total weighted completion time under makespan constraint for two-agent scheduling with job-dependent aging effects

In this paper, we consider a two-agent single-machine scheduling problem with linear position-based aging effects and job-dependent aging ratios. The objective is to minimize the total weighted completion time of all jobs for two agents, where the makespan for one agent is constrained under an upper bound. After showing that this problem is at least NP-hard, we develop two solution algorithms: First, we devise a branch-and-bound algorithm to find an optimal solution through the establishment of several dominance and feasibility properties, and a lower bound. Second, we propose efficient simulated annealing algorithms, using three different methods to generate an initial solution. Through a numerical experiment, we demonstrate that the suggested algorithms can be applied to efficiently find near-optimal solutions within a reasonable amount of CPU time. In particular, we show that the initial solution method (arranging the jobs for one agent in non-increasing order of aging ratio, and scheduling the jobs for the other in the weighted shortest normal processing time order) is superior to others. Moreover, through scalability testing, we verify its consistent and relatively outstanding performance for larger systems with many processing jobs.     

Evolutionary asymmetric games for modeling systems of partially cooperative agents

Based on the tenet of Darwinism, we propose a general mechanism that guides agents (which can be partially cooperative) in selecting appropriate strategies in situations of complex interactions, in which agents do not have complete information about other agents. In the mechanism, each participating agent generates many instances of itself to help it find an appropriate strategy. The generated instances adopt alternative strategies from the agent's strategy set. While all instances generated by different agents meet randomly to complete a task, every instance adapts its strategy according to the difference between the average utilities of its current strategy and all its strategies. We give a complete analysis of the mechanism for the case with two agents when each agent has two strategies, and show that by the tenet of Darwinism, agents can find their appropriate strategies through evolution and adaptation: 1) if dominant strategies exist, then the proposed mechanism is guaranteed to find them; 2) if there are two or more strict Nash equilibrium strategies, the proposed mechanism is guaranteed to find them by using different initial strategy distributions; and 3) if there is no dominant strategy and no strict Nash equilibrium, then agents will oscillate periodically. Nevertheless, the mechanism allows agent designers to derive the appropriate strategies from the oscillation by integration. For cases with two agents when each agent has two or more strategies, it is shown that agents can reach a steady state where social welfare is optimum.