Workspace analysis to generate a collision-free torch path for a ship welding robot

We propose a method to determine the optimal initial location and to generate torch paths for a ship welding robot with 6 degrees of freedom (DOF). The optimal initial location is determined using an objective function, which is set up by combining constraints on the torch posture, manipulability, and the range of each joint angle to avoid collisions. A genetic algorithm (GA) is used to optimize the objective function because it does not require additional derivatives. After the initial location is determined, torch paths are generated by interpolating the starting point, endpoints and torch postures using inverse kinematics. Our method can be applied to automate the welding job for each block during ship building, irrespective of the shape of the robots, by changing the objective function.     

基于遗传算法优化模糊控制器的船舶航向控制

针对模糊控制器中的量化因子、比例因子、积分系数、模糊规则之间互相耦合,人工整定困难的问题,提出了一种基于遗传优化的船舶航向模糊PID自动控制算法。仿真对比试验表明,经遗传优化后的船舶航向模糊PID控制性能得到了极大的提高,系统无超调,上升快,工作稳定,具有较强的鲁棒性。     

船舶柴油机调速器参数整定的多目标优化和多属性决策研究

综合考虑系统在输入阶跃扰动和负荷扰动两个工况时的超调量、稳定时间和ITAE指标,研究了PID控制器参数寻优的问题。采用多目标遗传算法求出Pareto最优解,使用信息熵法和逼近理想解的排序方法（TOPSIS）对Pareto最优解给出了排序。计算了一艘水面船舶柴油机调速器控制的数值算例,结果表明本文方法有效,所设计的PID性能优异,可以工程实际应用。     

智能冲击压路机路基压实度自动检测方法

为了提高路基压实度检测的准确性,提出了路基压实度自动检测的瑞雷波法。采用落锤法敲击地表产生瑞雷波,分别通过检波器和地震仪采集和记录瑞雷波信号,应用-τp变换提取经过信息处理后的瑞雷波,获取频散曲线,利用瑞雷波反演横波速度的方法计算压实度,分析了压实度分布剖面图信息,对比了用瑞雷波法和灌砂法计算的路基压实度。计算结果表明:两条压实度曲线几乎重叠,最大绝对误差为-0.300%,相对误差为0.308%,说明用瑞雷波法检测路基压实度数据可靠,可以用于控制冲击碾压施工。     

互通式立体交叉方案综合评价方法

基于投影寻踪算法的基本原理和求解过程,提出互通立交方案设计综合评价方法,该方法直接从待评样本本身寻求各评价指标的客观权重,避免了主观评价法评价指标权重的确定存在着主观依赖性、评价结果受到人为赋权的干扰问题.采用实码加速遗传算法对投影指标函数进行优化,求解最优投影方向向量,可有效地解决高维数据全局寻优的难题.     

物流中心选址与规模的层次优化模型

同时考虑生产商和分销商的利益,采用双层规划模型来描述物流中心选择以及规模问题.上层规划从生产商的角度出发,通过调整物流中心位置及规模使生产商的生产成本、运输成本和经营成本最小；下层规划从分销商的角度出发,通过调整从各个物流中心进货的数量使自身的平均运输成本最低.设计了基于遗传算法的求解方法,并用数值算例验证了模型和算法的有效性.     

基于谐波平衡法的Broyden网格分区迭代算法

在融合Broyden算法与网格分区思想的基础上,提出一种基于谐波平衡法且适用于对间隙型非线性动力系统进行频响特性分析的新型Broyden网格迭代算法.阐述了Broyden网格迭代算法的基本思想与计算流程.对决定该算法有效性的Broyden算法发散特性及不完整变量循环迭代过程等关键特性进行了分析与讨论.通过与相关文献采用的二分法所得的同等条件下的计算结果进行对比,证明该算法比二分法具有更好引导性与更全面的求解能力.通过网格分区的引导,该方法避免了数值迭代方法中普遍面临的迭代初值的盲目选择问题,为基于谐波平衡法全面考察间隙型非线性动力系统的频响特性提供了新途径.     

Multi-objective optimization of train routing problem combined with train scheduling on a high-speed railway network

Based on train scheduling, this paper puts forward a multi-objective optimization model for train routing on high-speed railway network, which can offer an important reference for train plan to provide a better service. The model does not only consider the average travel time of trains, but also take the energy consumption and the user satisfaction into account. Based on this model, an improved GA is designed to solve the train routing problem. The simulation results demonstrate that the accurate algorithm is suitable for a small-scale network, while the improved genetic algorithm based on train control (GATC) applies to a large-scale network. Finally, a sensitivity analysis of the parameters is performed to obtain the ideal parameters; a perturbation analysis shows that the proposed method can quickly handle the train disturbance.     

TRIP DISTRIBUTION MODELLING USING FUZZY LOGIC AND A GENETIC ALGORITHM

This article examines possibilities for the application of soft computing techniques for the prediction of travel demand. The model, based on fuzzy logic and a genetic algorithm, successfully solves the trip distribution problem. The possibilities of using the proposed model in solving trip generation, modal split and route choice problems have also been indicated. The model has been tested on a real numerical example. Exceptionally good correspondences between estimated and real values of passenger flows have been obtained.