基于改进混合遗传细菌觅食算法的贝叶斯结构学习算法

针对利用启发式学习算法学习贝叶斯网络时容易陷入局部最优和寻优效率低的问题,提出一种改进的混合遗传细菌觅食优化算法的贝叶斯网络结构学习算法。该算法首先通过遗传算法求得较优种群并作为细菌觅食算法的初始种群;然后利用交叉和变异策略改进细菌觅食算法的复制行为,增加种群多样性,扩大搜索空间;最后通过改进细菌觅食算法的迁移行为的初始化操作更新种群,防止精英个体的丢失。通过种群的迭代搜索最终获得最优的贝叶斯网络结构。实验仿真结果表明,与其他算法相比,该算法的收敛精度和效率有所提升。     

一种针对水泥回转窑故障诊断的贝叶斯网络模型

为了实现水泥回转窑的故障诊断,采用贝叶斯网络建立了水泥回转窑故障智能诊断模型。在模型建立 过程中,提出了一种基于数据样本、不依赖先验知识的贝叶斯网络结构学习改进算法。在利用改进结构学习算法 建立诊断模型贝叶斯网络的基础上,利用MLE算法和变量消除法完成了模型的参数学习和诊断推理。为了验证 水泥回转窑故障诊断贝叶斯网络模型的准确率以及可行性,利用现场数据进行了大量的测试实验。     

基于分类优化贝叶斯结构算法的篦冷机参数状态分析及其算法收敛性分析

针对种群算法建立贝叶斯结构存在参数多、易陷入局部最优的问题,提出一种改进贝叶斯结构学习算法。该算法将候选结构分为优劣解集,利用师生交流机制优化优解集保留精英个体,利用变异机制优化劣解集来增加结构多样性,从而加快算法收敛速度,并在准确率和运行时间上达到平衡。最后不仅利用马尔科夫链证明该算法是全局收敛的,而且通过仿真实验验证了所提出算法的性能。将该算法应用到水泥篦冷机的实际数据中,构建水泥篦冷机工艺参数的贝叶斯网络结构,并完成篦冷机参数状态分析。     

基于跳点搜索-遗传算法的自主移动机器人路径规划

为了解决采用遗传算法解析最优路径中存在的转折点较多、易陷入局部最优解、迭代次数较多以及寻优时间过长等问题，引入自适应交叉算子和变异算子，将改进后的跳点搜索（jump point search）算法与改进遗传算法融合，得到跳点搜索-遗传（jump point search-genetic,JPSG）算法。JPSG算法利用JPS算法的高效局部搜索能力来提高整体搜索能力，加速算法整体收敛趋势；利用改进遗传算法的全局搜索能力改变JPS算法不能在复杂障碍物状况下解析最优路径的状态，提高算法对动态环境的适应性。在栅格矩阵中的路径规划仿真表明，相比于改进遗传算法、传统遗传算法，JPSG算法可以有效缩短寻优执行时间，提高寻优准确率，减少运算执行次数，在稳定性、准确性、快速性上具有明显的优势。     

基于改进免疫遗传优化蚁群算法的移动机器人路径寻优研究

针对移动机器人路径规划中使用蚁群算法(ACO)易陷入局部最优和收敛速度慢的问题,提出了一种适用于机器人静态路径寻优的改进免疫遗传优化蚁群算法(IMGAC)。该算法可以根据实际情况自动调整变异概率和变异方式,以及自动调节个体免疫位的长度,将通过改进的变异算子和免疫算子嵌入蚁群算法来提高全局寻优能力与收敛速度。仿真及实验表明:相比于经典ACO算法以及最大最小蚂蚁系统,IMGAC算法收敛速度更快,全局寻优能力更强。利用该算法寻找移动机器人最优路径,提高了静态路径寻优的效果和效率。     

改进QPSO算法识别结构模态参数

为了进一步增强量子粒子群优化算法的全局寻优能力,提高粒子寻优效率,改善其容易陷入局部最优的缺陷,首先在引入同化和竞争思想的基础上提出一种改进的量子粒子群算法。该改进算法将民族间的同化竞争思想引入粒子寻优过程,以全局最优粒子作为中心粒子,不断同化其余粒子,使粒子之间保持不断竞争关系,以改进粒子的进化方式,提高粒子的寻优性能。接着将改进算法应用于结构模态参数识别,并采用简支梁数值模型对该算法的有效性进行验证,结果表明,改进算法较量子粒子群算法的识别精度和抗噪性都有显著的提高。最后通过三层框架试验验证改进算法在实际工程应用中的有效性。     

基于改进拟态物理学算法的配电网故障区段定位

配电网故障定位的本质是一个离散域二进制寻优问题,因此找到一种全局寻优能力强的二进制算法来解决配电网故障定位是十分困难的。本文针对拟态物理学算法(APO,ArtificialPhysicsOptimization)陷入局部最优的缺点,通过引入反向学习原理改进算法初始解生成过程,并在局部最优时利用混沌无序的特点保持算法的多样性,最后构建了配电网故障定位的数学模型,利用改进后的APO算法对配电网故障进行定位处理。仿真结果表明,采用改进类APO算法进行配电网故障区段定位具有较高容错性,能够实现单点和多点故障的准确定位,通过与遗传算法、蚁群算法比较,本文算法在定位准确和容错性方面有较大优势。     

一种改进果蝇算法及其在风电机组MPPT中的应用

为提高果蝇优化算法(FOA)的寻优效率和精度,针对标准果蝇算法在全局范围内搜索能力不均匀导致的问题,提出一种步长改进策略。该策略在运行过程中根据当前果蝇群体中最优个体位置,动态地对果蝇前进步长进行调整,使果蝇算法能够平衡在全局范围内的搜索能力,增强初期收敛速度和后期收敛精度。通过经典测试函数对改进算法进行仿真研究,结果表明:在保证寻优成功率的同时,该文所提出改进算法的收敛精度和速度均得到显著提高。风电机组滑模控制器参数寻优中的应用实例也表明该算法的有效性。     

基于圆周割线改进型粒子群优化算法的叶片临界颤振辨识方法研究

针对风力机叶片在颤振风速下的临界颤振现象,创新性地结合几何圆周割线和传统粒子群优化算法,首次设计了一种圆周割线改进型粒子群优化算法,应用于叶片临界颤振系统的参数辨识。该方法利用圆周上移动点的割线距离来动态调节全局学习因子和局部学习因子,针对优化辨识提高全局搜索和局部搜索的动态平衡性,避免陷入局部最优,提高算法的整体寻优性能和优化效率。仿真试验中,将该方法与多种先进粒子群优化算法(如改进型粒子群优化(MPSO)算法、基于线性递减惯性权重的粒子群优化(LDIW-PSO)算法、基于动态学习因子的免疫粒子群优化(IPSODCLF)算法)的辨识结果相比较,结果表明该辨识方法在辨识精度、计算时间和鲁棒性方面均具有显著的优越性。     

基于贝叶斯网络的项目组合风险度量

项目组合的交互效应特性使得项目组合风险不能通过单个项目风险的线性叠加获得。基于贝叶斯网络建模提出了一种项目组合风险度量的新方法。该方法通过将专家知识与K2算法相结合,求得项目组合风险的贝叶斯网络结构,并通过度量交互效应对项目风险的影响计算网络中每个节点的条件概率表,实现项目组合风险的贝叶斯网络推理。为了得到K2算法所需的有序节点输入,计算项目风险间的互信息,并基于互信息与条件独立检验求得项目节点的顺序。最后通过一个高新技术企业项目组合的应用实例说明该方法的实用性和有效性。     

Methodology for the optimal component selection of electronic devices under reliability and cost constraints

The objective of this paper is to present an efficient computational methodology for the reliability optimization of electronic devices under cost constraints. The system modeling for calculating the reliability indices of the electronic devices is based on Bayesian networks using the fault tree approach, in order to overcome the limitations of the series–parallel topology of the reliability block diagrams. Furthermore, the Bayesian network modeling for the reliability analysis provides greater flexibility for representing multiple failure modes and dependent failure events, and simplifies fault diagnosis and reliability allocation. The optimal selection of components is obtained using the simulated annealing algorithm, which has proved to be highly efficient in complex optimization problems where gradient‐based methods can not be applied. The reliability modeling and optimization methodology was implemented into a computer program in Matlab using a Bayesian network toolbox. The methodology was applied for the optimal selection of components for an electrical switch of power installations under reliability and cost constraints. The full enumeration of the solution space was calculated in order to demonstrate the efficiency of the proposed optimization algorithm. The results obtained are excellent since a near optimum solution was found in a small fraction of the time needed for the complete enumeration (3%). All the optimum solutions found during consecutive runs of the optimization algorithm lay in the top 0.3% of the solutions that satisfy the reliability and cost constraints. Copyright © 2007 John Wiley & Sons, Ltd.     

基于改进遗传算法的DBN结构自适应学习算法

针对动态贝叶斯网络(dynamic bayesian network, DBN)是NP困难问题,提出基于改进遗传算法的DBN结构自适应学习算法。该算法计算最大互信息和时序互信息完成DBN结构搜索空间的初始化。在此基础上设计改进遗传算法,引入评分标准差构建交叉概率和变异概率的自适应调节函数,以降低结构学习过程陷入局部最优解的概率。仿真结果表明,该算法在无先验知识的情况下,相比其他优化算法,汉明距离和运行时长平均减少了30%, 37.3%,评分值平均增大了18.0%。     

A Bayesian network approach for fixture fault diagnosis in launch of the assembly process

Verification and correction of faults related to tooling design and tooling installation are important in the auto body assembly process launch. This paper introduces a Bayesian network (BN) approach for quick detection and localisation of assembly fixture faults based on the complete measurement data set. Optimal sensor placement for effective diagnosis of multiple faults, structure learning of the Bayesian network and the diagnostic procedure are incorporated in the proposed approach. The effective independence sensor placement method is used to reach the desired number of optimal sensor locations, which provide the concise and effective sensor nodes to build the diagnostic Bayesian network. A new algorithm based on conditional mutual information tests is put forward to learn the Bayesian network structure. The body side assembly case was used to illustrate the suggested method and the simulation analysis was performed to evaluate the effectiveness of the diagnostic network. The work demonstrated that the proposed methodology composes a feasible and powerful tool for fixture fault diagnosis in launch of the assembly process.     

基于改进狮群算法的汽轮机热耗率模型预测

针对汽轮机热消耗率模型难以精准预测的问题,提出一种基于改进的狮群算法和快速学习网综合建模的方法。首先,针对传统狮群算法易早熟收敛以及在迭代后期寻优速度缓慢导致算法陷入局部最优的缺陷,通过引入禁忌搜索、非线性扰动因子以及黄金正弦策略进行改进;其次,对改进后的狮群算法进行数值验证,结果证明其具有更高的收敛精度和收敛速度;最后,采用某热电厂汽轮机的运行数据建立汽轮机热消耗率预测模型,并将改进狮群算法优化的快速学习网对其进行热耗率预测,将实验结果与其他优化策略进行对比验证,实验结果表明,基于改进狮群算法的快速学习网预测模型具有更高的泛化能力,提高了汽轮机热耗率的预测精度。     

基于贝叶斯优化与改进LeNet-5的滚动轴承故障诊断

考虑到卷积神经网络在滚动轴承故障诊断中存在网络结构难以确定、训练次数过多、时间过长等问题,设计了一种贝叶斯优化改进LeNet-5算法,以及采用该算法构建的轴承故障诊断模型。采用贝叶斯优化训练过程中学习率等超参数,多种故障轴承的振动信号直接作为改进LeNet-5网络的输入,对池化输出采用批归一化处理和改进池化层激活函数防止过拟合,利用全局平均池化层替代全连接层提高改进LeNet-5网络的泛化能力,用Softmax分类器实现滚动轴承故障的分类。通过轴承数据库开展实验,实验表明,该算法构建的轴承故障诊断模型在训练集上准确率为99.94%,验证集上的准确率为99.89%,测试集准确率也达到99.65%,与一维卷积神经网络和二维卷积神经网络对比分析,基于贝叶斯优化改进LeNet-5算法构建的轴承故障诊断模型在滚动轴承的故障诊断模型具有更高的准确率,更少的训练次数和训练时间。     

Forward and backward risk assessment throughout a system life cycle using dynamic Bayesian networks: A case in a petroleum refinery

In this paper, risk modeling was conducted based on the defined risk elements of a conceptual risk framework. This model allows for the estimation of a variety of risks, including human error probability, operational risk, financial risk, technological risk, commercial risk, health risk, and social and environmental risks. Bayesian network (BN) structure learning techniques were used to determine the relationships among the model variables. By solving a bi-objective optimization problem applying the genetic algorithm (GA) with the Pareto ranking approach, the network structure was learned. Then, risk modeling was performed for a petroleum refinery focusing on HydroDeSulfurization (HDS) technology throughout its life cycle. To extend the model horizontally and make it possible to evaluate the risk trend throughout the technology life cycle, we developed a dynamic Bayesian network (DBN) with three-time slices. A two-way forward and backward approach was used to analyze the model. The model validation was performed by applying the leave-one-out cross-validation method.     

Hybrid BWO-IACO Algorithm for Cluster Based Routing in Wireless Sensor Networks

Wireless Sensor Network (WSN) comprises a massive number of arbitrarily placed sensor nodes that are linked wirelessly to monitor the physical parameters from the target region. As the nodes in WSN operate on inbuilt batteries, the energy depletion occurs after certain rounds of operation and thereby results in reduced network lifetime. To enhance energy efficiency and network longevity, clustering and routing techniques are commonly employed in WSN. This paper presents a novel black widow optimization (BWO) with improved ant colony optimization (IACO) algorithm (BWO-IACO) for cluster based routing in WSN. The proposed BWO-IACO algorithm involves BWO based clustering process to elect an optimal set of cluster heads (CHs). The BWO algorithm derives a fitness function (FF) using five input parameters like residual energy (RE), inter-cluster distance, intra-cluster distance, node degree (ND), and node centrality. In addition, IACO based routing process is involved for route selection in inter-cluster communication. The IACO algorithm incorporates the concepts of traditional ACO algorithm with krill herd algorithm (KHA). The IACO algorithm utilizes the energy factor to elect an optimal set of routes to BS in the network. The integration of BWO based clustering and IACO based routing techniques considerably helps to improve energy efficiency and network lifetime. The presented BWO-IACO algorithm has been simulated using MATLAB and the results are examined under varying aspects. A wide range of comparative analysis makes sure the betterment of the BWO-IACO algorithm over all the other compared techniques.     

An Adjust Duty Cycle Method for Optimized Congestion Avoidance and Reducing Delay for WSNs

With the expansion of the application range and network scale of wireless sensor networks in recent years, WSNs often generate data surges and delay queues during the transmission process, causing network paralysis, even resulting in local or global congestion. In this paper, a dynamically Adjusted Duty Cycle for Optimized Congestion based on a real-time Queue Length (ADCOC) scheme is proposed. In order to improve the resource utilization rate of network nodes, we carried out optimization analysis based on the theory and applied it to the adjustment of the node’s duty cycle strategy. Using this strategy to ensure that the network lifetime remains the same, can minimize system delay and maximize energy efficiency. Firstly, the problems of the existing RED algorithm are analyzed. We introduce the improved SIG-RED algorithm into the ADCOC mechanism. As the data traffic changes, the RED protocol cannot automatically adjust the duty cycle. A scheduler is added to the buffer area manager, referring to a weighted index of network congestion, which can quickly determine the status of network congestion. The value of the weighting coefficient W is adjusted by the Bayesian method. The scheduler preferably transmits severely urgent data, alleviating the memory load. Then we combined improved data fusion technology and information gain methods to adjust the duty cycle dynamically. By simulating the algorithm, it shows that it has faster convergence speed and smaller queue jitter. Finally, we combine the adjusted congestion weight and the duty cycle growth value to adjust the data processing rate capability in the real-time network by dynamically adjusting it to adapt to bursts of data streams. Thus, the frequency of congestion is reduced to ensure that the system has higher processing efficiency and good adaptability.