基于改进的粒子群优化的FastSLAM方法

提出了一种基于改进的粒子群优化(IPSO)的快速同时定位和地图创建(FastSLAM)方法--IPSO FastSLAM算法.该算法在粒子预估过程中引入观测信息,调整了粒子的提议分布,增强了位置预测的准确性.改进的粒子群优化采用两步优化策略,即首先通过种群速度自适应调整惯性权重,有效地克服了粒子退化问题,改善了算法的实...     

Cross-docking truck scheduling with product unloading/loading constraints based on an improved particle swarm optimisation algorithm

Cross-docking is a very useful logistics technique that can substantially reduce distribution costs and improve customer satisfaction. A key problem in its success is truck scheduling, namely, decision on assignment and docking sequence of inbound/outbound trucks to receiving/shipping dock doors. This paper focuses on the problem with the requirement of unloading/loading products in a given order, which is very common in many industries, but is less concerned by existing researches. An integer programming model is established to minimise the makespan. An improved particle swarm optimisation (ωc-PSO) algorithm is proposed for solving it. In the algorithm, a cosine decreasing strategy of inertia weight is designed to dynamically balance global and local search. A repair strategy is put forward for continuous search in the feasible solution space and a crossover strategy is presented to prevent the algorithm from falling into local optimum. After algorithm parameters are tuned using Taguchi method, computational experiments are conducted on different problem scales to evaluate ωc-PSO against genetic algorithm, basic PSO and GLNPSO. The results show that ωc-PSO outperforms other three algorithms, especially when the number of dock doors, trucks and product types is great. Statistical tests show that the performance difference is statistically significant.     

PWM optimisation for three-level voltage inverter based on clonal selection algorithm

A clonal selection algorithm (CSA) is first applied to determine the switching time for a three-level voltage pulse width modulation (PWM) inverter. To obtain an optimal PWM switching pattern, CSA should calculate a set of nonlinear transcendental equations with nonlinear equality and inequality constraints. Unlike the general CSA by penalising infeasible solutions, this new CSA deals with constraints and guides the search to the true optimum solution by an infeasibility degree with a random disturbance (IFDD) selection operation. The proposed method minimise the total harmonic distortion and generates a waveform with adjustable amplitude of the fundamental component. Calculation results verify that the proposed method is efficient to obtain the optimal PWM switching pattern and is superior to natural sampling and genetic algorithm-based PWM switching pattern.     

An improved ant colony optimisation algorithm for three-tier supply chain scheduling based on networked manufacturing

This paper presents a study on supply chain scheduling from the perspective of networked manufacturing (NM). According to feature analysis of supply chain scheduling based on NM, we comprehensively consider the combined benefits of cost, time, and satisfaction level for customised services. In order to derive a scheduling strategy among supply chain members based on NM, we formulate a three-tier supply chain scheduling model composed of manufacturer, collaborative design enterprise and customer. Three objective functions – time function, cost function and delay punishment function – are employed for model development. We also take into account multi-objective optimisation under the constraint of product capacity. By using an improved ant colony optimisation algorithm, we add different pheromone concentrations to selected nodes that are obtained from feasible solutions and we confine pheromone concentrations τ within the minimum value τ _min and the maximum value τ _max, thus obtaining optimal results. The results obtained by applying the proposed algorithm to a real-life example show that the presented scheduling optimisation algorithm has better convergence, efficiency, and stability than conventional ant colony optimisation. In addition, by comparing with other methods, the output results indicate that the proposed algorithm also has better solutions.     

Hybrid constrained genetic algorithm/particle swarm optimisation load flow algorithm

A hybrid constrained genetic algorithm and particle swarm optimisation (PSO) method for the evaluation of the load flow in heavy-loaded power systems is developed. The new algorithm is applied to find the maximum loading points of three IEEE test systems. The experimental determination of the best values of the parameters for use in the PSO part of the hybrid algorithm is also reported.     

A hybrid particle swarm based method for process planning optimisation

A process planning (PP) problem is defined as to determine a set of operation-methods (machine, tool, and set-up configuration) that can convert the given stock to the designed part. Essentially, the PP problem involves the simultaneous decision making of two tasks: operation-method selection and sequencing. This is a combinatorial optimisation problem and it is difficult to find the best solution in a reasonable amount of time. In this article, an optimisation approach based on particle swarm optimisation (PSO) is proposed to solve the PP problem. Due to the characteristic of discrete process planning solution space and the continuous nature of the original PSO, a novel solution representation scheme is introduced for the application of PSO in solving the PP problem. Moreover, two kinds of local search algorithms are incorporated and interweaved with PSO evolution to improve the best solution in each generation. The numerical experiments and analysis have demonstrated that the proposed algorithm is capable of gaining a good quality solution in an efficient way.     

Precise identification of moving vehicular parameters based on improved glowworm swarm optimization algorithm

This paper proposes an indirect method for the identification of moving vehicular parameters using the dynamic responses of the vehicle. The moving vehicle is modelled as 2-DOF system with 5 parameters and 4-DOF system with 12 parameters, respectively. Finite element method is used to establish the equation of the coupled bridge–vehicle system. The dynamic responses of the system are calculated by Newmark direct integration method. The parameter identification problem is transformed into an optimization problem by minimizing errors between the calculated dynamic responses of the moving vehicle and those of the simulated measured responses. Glowworm swarm optimization algorithm (GSO) is used to solve the objective function of the optimization problem. A local search method is introduced into the movement phase of GSO to enhance the accuracy and convergence rate of the algorithm. Several test cases are carried out to verify the efficiency of the proposed method and the results show that the vehicular parameters can be identified precisely with the present method and it is not sensitive to artificial measurement noise.     

Clamping-sequence optimisation based on heuristic algorithm for sheet-metal components

The traditional clamping-sequence optimisation of sheet-metal parts requires many complicated finite element analyses, and clamping-sequence planning does not account for the springback from clamp-release. Therefore, this paper proposes a new optimisation method based on a heuristic algorithm. We first propose a new contact model of parts, clamps and supporting locators to analyse assembly deformation. Then, we use the distance between the actual and nominal positions to evaluate the clamp layout. Finally, we apply the heuristic algorithm to optimise the clamping sequence. We illustrate the proposed method with a case study of a taillight bracket, whose results show that the method of clamping-sequence optimisation can effectively decrease the deformation of sheet metal from clamping.     

Recurrent wavelet neural network controller with improved particle swarm optimisation for induction generator system

A recurrent wavelet neural network (RWNN) controller with improved particle swarm optimisation (IPSO) is proposed to control a three-phase induction generator (IG) system for stand-alone power application. First, the indirect field-oriented mechanism is implemented for the control of the IG. Then, an AC/DC power converter and a DC/AC power inverter are developed to convert the electric power generated by a three-phase IG from variable frequency and variable voltage to constant frequency and constant voltage. Moreover, two online trained RWNNs using backpropagation learning algorithm are introduced as the regulating controllers for both the DC-link voltage of the AC/DC power converter and the AC line voltage of the DC/AC power inverter. Furthermore, an IPSO is adopted to adjust the learning rates to further improve the online learning capability of the RWNN. Finally, some experimental results are provided to demonstrate the effectiveness of the proposed IG system.     

基于改进型能量守恒SOC估算法的电动汽车三段式智能充电方式研究

为研究电动汽车安全、快速的智能充电方式,基于传统能量守恒法,考虑电池容量衰减和电池内阻损失对荷电状态(state of charge,SOC)估算的影响,提出改进型能量守恒SOC估算方法。对比分析几种传统充电方式,根据马斯定理确定最佳充电电流,提出以改进型能量守恒SOC估算法得到的SOC值作为判断依据的电动汽车三段式(小电流充电、脉冲充电、恒压充电)智能充电方式,并建立其仿真模型。结果表明:改进型能量守恒SOC估算法得到的SOC值要小于传统能量守恒法,其更加接近真实SOC值;三段式智能充电方式能根据电池组SOC值的变化智能地选择具体充电方式,实现了对电池的安全、快速充电。提出的基于改进型能量守恒SOC估算的三段式智能充电方式对当前电动汽车充电方式的研究提供了一定的参考价值,为智能充电方式研究效能的提升提供了一种可行方法,也为智能充电理论应用于工程实践打下基础。     

Modified Elman neural network controller with improved particle swarm optimisation for linear synchronous motor drive

A modified Elman neural network controller is proposed to control the mover of a permanent magnet linear synchronous motor (PMLSM) servo drive to track periodic reference trajectories. First, the dynamic model of the PMLSM drive system is derived. Next, a modified Elman neural network is proposed to control the PMLSM. Moreover, the connective weights of the modified Elman neural network are trained online by back-propagation (BP) methodology. However, the learning rates of the online-training weights are usually selected by trial-and- error method, which is time-consuming. Therefore an improved particle swarm optimisation (IPSO) is adopted in this study to adapt the learning rates in the BP process of the modified Elman neural network to improve the learning capability. Finally, the control performance of the proposed modified Elman neural network controller with IPSO is verified by the simulated and experimental results.     

采用改进的基于余量的算法实现人脸识别中最佳Gabor特征的选择

基于2DGabor变换的人脸特征描述已经受到了很多人的关注。然而现有的Gabor特征维数较高,而且具有冗余性,因此选择最佳的Gabor特征用于人脸识别显得尤为的重要。利用最大余量原理的特征选择算法在目前的机器学习研究中已经占据了重要的地位。本文在基于余量的迭代搜索法(Simba)的基础上,引入了一种新的选择算法:基于余量的共轭梯度法(Cgmba),它只需较少次迭代就可以找到最佳解。我们在IMM人脸库上进行了实验,实验结果表明:尽管只使用了一半不到的特征,但Cgmba和Simba的识别率却分别提高了3.75和1.25个百分点,同时也证实了我们提出的Cgmba明显优于Simba。最后我们对Cgmba选择的Gabor特征的分布情况进行了分析,可以看出较大尺度的特征相对于较小尺度的特征对于分辩人脸的细微差别具有同等的重要性,而且在垂直,135°方向的特征具有更强的分辩能力。     

A bioinspired path planning approach for mobile robots based on improved sparrow search algorithm

In this paper, a bioinspired path planning approach for mobile robots is proposed. The approach is based on the sparrow search algorithm, which is an intelligent optimization algorithm inspired by the group wisdom, foraging, and anti-predation behaviors of sparrows. To obtain high-quality paths and fast convergence, an improved sparrow search algorithm is proposed with three new strategies. First, a linear path strategy is proposed, which can transform the polyline in the corner of the path into a smooth line, to enable the robot to reach the goal faster. Then, a new neighborhood search strategy is used to improve the fitness value of the global optimal individual, and a new position update function is used to speed up the convergence. Finally, a new multi-index comprehensive evaluation method is designed to evaluate these algorithms. Experimental results show that the proposed algorithm has a shorter path and faster convergence than other state-of-the-art studies.The full text can be downloaded at https://link.springer.com/article/10.1007/s40436-021-00366-x     

一种改进粒子群算法的立体阵列优化方法

利用传统粒子群算法对立体阵列所有臂的阵元分布形式同时进行优化时,不仅耗时,而且易于收敛到局部解。为了解决这个问题,提出了一种改进粒子群算法(Improved Particle Swarm Optimization,IPSO)。改进算法采用并行计算思想,同时初始化多个粒子群,每个粒子通过优化一个臂(优化臂)的阵元参数达到"降维"的目的,使用线性递减惯性权重,对多个粒子群同时进行预优化,获得中间解。利用中间解构建一个"升维"的新粒子,使用最小惯性权重对新粒子继续优化,满足停止条件后输出。通过对5臂星形立体阵列进行优化设计,发现改进算法不仅耗时短,而且能够得到更优的结果,最后通过6个仿真实验讨论了所设计的阵列的指向特性。     

An improved subspace selection algorithm for meshless collocation methods

Choosing data points is a common problem for researchers who employ various meshless methods for solving partial differential equations. On the one hand, high accuracy is always desired; on the other, ill‐conditioning problems of the resultant matrices, which may lead to unstable algorithms, prevent some researchers from using meshless methods. For example, the optimal placements of source points in the method of fundamental solutions or of the centers in the radial basis functions method are always unclear. Intuitively, such optimal locations will depend on many factors: the partial differential equations, the domain, the trial basis used (i.e. the employed method itself 1pt), the computational precisions, some user‐defined parameters, and so on. Such complexity makes the hope of having an optimal centers placement unpromising. In this paper, we provide a data‐dependent algorithm that adaptively selects centers based on all the other variables. Copyright © 2009 John Wiley & Sons, Ltd.     

Bi-objective mathematical model and improved algorithm for optimisation of welding shop scheduling problem

This paper addresses a bi-objective welding shop scheduling problem (BWSSP) aiming to minimise the total tardiness and the machine interaction effect. The BWSSP is a special flow-shop scheduling problem (FSP) which is characterised by the fact that more than one machine can process on one job at a certain stage. This study analyses the operation of a structural metal manufacturing plant, and includes various aspects such as job sequence, machine-number-dependent processing time, lifting up time, lifting down time and different delivery time. A novel mixed-integer programming model (MIPM) is established, which can be used to minimise the delayed delivery time and the total machine interaction effect. One machine interaction effect formula is given in this paper. In order to solve this BWSSP, an appropriate non-dominated sorting Genetic Algorithm III (NSGAIII), embedded with a restarted strategy (RNSGAIII), is proposed. The restarted strategy, which can increase the diversity of the solutions, will be triggered with a restart probability. Following the iterative process, an effective strategy is applied to reduce the interaction effect penalty, on the premise that the makespan will remain unchanged. Total five algorithms, namely NSGAII, NSGAIII, harmony search algorithm (HSA), strength Pareto evolutionary algorithm (SPEA2), and RNSGAIII are utilised to solve this engineering problem. Numerical simulations show that the improved RNSGAIII outperforms the other methods, and the Pareto solution distribution and diversity, in particular, are significantly improved.     

A multi-improved genetic algorithm for facility layout optimisation based on slicing tree

For the facility layout optimisation problem, we use the slicing tree structure based on the order of traversal to form a new chromosome encoding system demonstrating facilities’ order, the relationship and the location. We generate the initial solution based on two principles namely the facilities’ adjacency and random generation. The structure of chromosome is made up with three sections in the research so that we can do the genetic operations to these three sections respectively, and we use dynamic and feedback mechanisms to improve the penalty function. As a result, the analysis of typical cases shows that there are certain improvements to this algorithm both in effectiveness and efficiency.     

Cloud manufacturing service composition based on QoS with geo-perspective transportation using an improved Artificial Bee Colony optimisation algorithm

Cloud Manufacturing (CMfg) ambitions to create dedicated manufacturing clouds (i.e. virtual enterprises) for complex manufacturing demands through the association of various service providers’ resources and capabilities. In order to insure a dedicated manufacturing cloud to match the level of customer’s requirements, the cloud service selection and composition appear to be a decisive process. This study takes common aspects of cloud services into consideration such as quality of service (QoS) parameters but extend the scope to the physical location of the manufacturing resources. Unlike the classic service composition, manufacturing brings additional constraints. Consequently, we propose a method based on QoS evaluation along with the geo-perspective correlation from one cloud service to another for transportation impact analysis. We also insure the veracity of the manufacturing time evaluation by resource availability overtime. Since the composition is an exhaustive process in terms of computational time consumption, the proposed method is optimised through an adapted Artificial Bee Colony (ABC) algorithm based on initialisation enhancement. Finally, the efficiency and precision of our method are discussed furthermore in the experiments chapter.     

Analysis of closed loop supply chain using genetic algorithm and particle swarm optimisation

This paper describes a relational database system for semi-generative process planning for sheet metal parts that emulates expert system capabilities. The system integrates a feature-based relational database for the parts, a forward chaining rule-based strategy for machine selection, both global and feature-specific execution of the rules and a graph theoretic cost optimization model for optimal process plan selection. This system, which is currently being developed for a sheet metal fabrication company, suggests that, using the experience of shopfloor personnel, an efficient integration of feature-based process planning and expert system strategies can be accomplished.