AN EFFICIENT META-HEURISTIC ALGORITHM FOR ROUTING PRODUCT COLLECTING VEHICLES OF DEHYDRATION PLANTS. II. ALGORITHM PERFORMANCE AND CASE STUDIES

Routing of vehicle fleet for collecting newly cropped raw materials for multi-product dehydration plants is a component of plant production schedule of utmost significance. A meta-heuristic algorithm for efficiently solving the collecting vehicle routing problem was developed and analyzed in detail in Tarantilis and Kiranoudis (2000). Meta-heuristic algorithms are broadly characterized by a stochastic nature in producing tender solution configurations in linear search terms, which sweep the huge solution space in a guided and rational way. Algorithm performance is examined through an analysis of the impact of model parameters on solution procedure during the execution of typical routing problems. The most important model parameter examined was found to be the value of the initial threshold as well as the way that the value of this actual parameter is appropriately adjusted during the optimization process. The main characteristic of the algorithm is the way that threshold is not only lowered but also raised, or backtracked, depending on the success of the inner loop iterations to provide for an acceptable new solution that would replace an older one. An important feature of the algorithm is the fact that appearance of better configurations within a process run is distributed according to the Poisson probability distribution. The suggested algorithm is tested against typical literature benchmarks as well against real-world problem encountered in the production planning procedures of dehydration plants in Greece.     

AN EFFICIENT META-HEURISTIC ALGORITHM FOR ROUTING PRODUCT COLLECTING VEHICLES OF DEHYDRATION PLANTS. I. ALGORITHM DEVELOPMENT

Dehydration plants are characterized by a multi-product nature chiefly attributed to the utilization of different raw materials to be processed in parallel so that demand constraints are met. The Just-In-Time production planning policy of these plants require the collection of raw materials to be dehydrated shortly before the actual processing, immediately after harvesting. One of the most important aspects in collection of plant fresh products, is routing of collecting vehicles, so that total collection time is minimized. The aim of this study is to describe a new stochastic search meta-heuristic algorithm for solving the Vehicle Routing Problem (VRP), termed as the Backtracking Adaptive Threshold Accepting (BATA) algorithm. Our effort focuses on developing an innovative method, which produces reliable and high quality solutions, requiring reasonable computing effort. The main innovation of the algorithm, toward a typical threshold accepting algorithm, is that during the optimization process the value of the threshold is not only lowered, but also raised or backtracked according to how effective a local search is. This adaptation of the value of the threshold, plays an important role in finding high quality routing solutions. BATA is described in detail while its performance and characteristic case studies are presented by Tarantilis and Kiranoudis (2000).     

AN EFFICIENT META-HEURISTIC ALGORITHM FOR ROUTING PRODUCT COLLECTING VEHICLES OF DEHYDRATION PLANTS. I. ALGORITHM DEVELOPMENT

Dehydration plants are characterized by a multi-product nature chiefly attributed to the utilization of different raw materials to be processed in parallel so that demand constraints are met. The Just-In-Time production planning policy of these plants require the collection of raw materials to be dehydrated shortly before the actual processing, immediately after harvesting. One of the most important aspects in collection of plant fresh products, is routing of collecting vehicles, so that total collection time is minimized. The aim of this study is to describe a new stochastic search meta-heuristic algorithm for solving the Vehicle Routing Problem (VRP), termed as the Backtracking Adaptive Threshold Accepting (BATA) algorithm. Our effort focuses on developing an innovative method, which produces reliable and high quality solutions, requiring reasonable computing effort. The main innovation of the algorithm, toward a typical threshold accepting algorithm, is that during the optimization process the value of the threshold is not only lowered, but also raised or backtracked according to how effective a local search is. This adaptation of the value of the threshold, plays an important role in finding high quality routing solutions. BATA is described in detail while its performance and characteristic case studies are presented by Tarantilis and Kiranoudis (2000).     

A THRESHOLD-ACCEPTING METAHEURISTIC METHOD FOR SCHEDULING THE OPERATIONS OF DEHYDRATION PLANTS

Dehydration plants are broadly characterized by a multi-product nature chiefly attributed to the utilization of different raw materials to be processed sequentially so that demand constraints are met. Processing of raw materials is implemented through a series of preprocessing operations that together with drying constitute the production procedure of a pre-specified programme. The core of the manufacturing system that a typical dehydration plant involves, is scheduling of operations so that demand is fulfilled within a pre-determined time horizon imposed by production planning. The typical scheduling operation that dehydration plants involve can be formulated as a general job shop scheduling problem. The aim of this study is to describe a new metaheuristic method for solving the job shop scheduling problem of dehydration plants, termed as the Backtracking Adaptive Threshold Accepting (BATA) method. Our effort focuses on developing an innovative method, which produces reliable and high quality solutions, requiring reasonable computing effort. The main innovation of this method, towards a typical threshold accepting algorithm, is that during the optimization process the value of the threshold is not only lowered, but also raised or backtracked according to how effective a local search is. BATA is described in detail while a characteristic job shop scheduling case study for dehydration plant operations is presented.     

A THRESHOLD-ACCEPTING METAHEURISTIC METHOD FOR SCHEDULING THE OPERATIONS OF DEHYDRATION PLANTS

ABSTRACT

Dehydration plants are broadly characterized by a multi-product nature chiefly attributed to the utilization of different raw materials to be processed sequentially so that demand constraints are met. Processing of raw materials is implemented through a series of preprocessing operations that together with drying constitute the production procedure of a pre-specified programme. The core of the manufacturing system that a typical dehydration plant involves, is scheduling of operations so that demand is fulfilled within a pre-determined time horizon imposed by production planning. The typical scheduling operation that dehydration plants involve can be formulated as a general job shop scheduling problem. The aim of this study is to describe a new metaheuristic method for solving the job shop scheduling problem of dehydration plants, termed as the Backtracking Adaptive Threshold Accepting (BATA) method. Our effort focuses on developing an innovative method, which produces reliable and high quality solutions, requiring reasonable computing effort. The main innovation of this method, towards a typical threshold accepting algorithm, is that during the optimization process the value of the threshold is not only lowered, but also raised or backtracked according to how effective a local search is. BATA is described in detail while a characteristic job shop scheduling case study for dehydration plant operations is presented.     

全局优化搜索新算法──列队竞争算法(Ⅱ)解网络综合问题

本文给出了列队竞争算法解组合优化问题的框架和确定变异领域的两条原则 .对管路网络综合问题和换热网络综合问题确定了相应的变异领域 ,用列队竞争算法分别解这两个网络综合问题 ,所得到的最优解优于文献报道的结果 .     

基于优进策略的差分进化算法及其化工应用

鉴于常规差分进化算法容易“早熟”,全局寻优能力有待提高,提出了基于优进策略的差分进化算法,利用种群繁衍的有用信息改进子代分布,并引入确定性寻优操作,以提高寻优性能.设计了单纯形寻优操作和重布操作,并调整有关概率等.测试表明新算法的全局寻优性能有明显改善,已成功地应用于铯-铷-钒系低温硫酸催化剂上SO2氧化反应模型参数的估计,效果良好.     

随机需求VRP的一个算法

研究了随机需求VRP问题,在服务仅能路由失败一次和不允许部分服务的情况下,给出了两阶段Clarke-Wright模拟退火算法。对需求为均匀分布的VRP问题做了数值试验,给出了具体的方案。数值结果表明这个算法比简单的两阶段模拟退火算法好,而且算法简单,容易实现。这个算法在第Ⅱ阶段模拟退火算法开始时给出了一个比较合理的初始解,较好地处理了大范围交叉的问题,从而使得模拟退火算法快速收敛到近似最优解。     

The heterogeneous vehicle routing and truck scheduling problem in a multi-door cross-dock system

Cross-docking is a logistics technique applied by many industrial firms to get substantial savings in two warehousing costly functions like storage and order picking. Incoming shipments are unloaded from inbound trucks on a cross-dock terminal with minimal storage space and directly transferred to outbound vehicles that carry them to their destinations. The major decisions at the operational level are the vehicle routing and scheduling, the dock door assignment and the truck scheduling at the cross-dock. Because such decisions are interdependent, all of them are simultaneously considered in the so-called vehicle routing problem with cross-docking (VRPCD). Previous contributions on VRPCD assume that pickup and delivery tasks are accomplished by a homogeneous vehicle fleet, and they mostly ignore the internal transportation of goods through the cross-dock. This work introduces a new rigorous mixed-integer linear programming (MILP) formulation for the VRPCD problem to determine the routing and scheduling of a mixed vehicle fleet, the dock door assignment, the truck docking sequence and the travel time required to move the goods to the assigned stack door all at once. To improve the computational efficiency of the branch-and-cut search, an approximate sweep-based model is developed by also considering a set of constraints mimicking the sweep algorithm for allocating nodes to vehicles. Numerous heterogeneous VRPCD examples involving up to 50 transportation requests and a heterogeneous fleet of 10 vehicles with three different capacities were successfully solved using the proposed approaches in acceptable CPU times.     

应用遗传算法求解含化学反应体系的相平衡

基于Gibbs自由能最小原理 ,提出一种求解含化学反应体系相平衡的方法。采用反应进度和相分率消去等式约束 ,将含化学反应体系的相平衡计算转化为无约束的最优化问题 ,并应用遗传算法求解。采用本文提出的方法 ,能够避免计算中负数的组分摩尔数或摩尔分数的出现 ,确保得到组分的摩尔数或摩尔分数为非负的全局最优解。以聚甲醛反应体系的汽液平衡计算为例 ,表明本文所提出的方法是可行和有效的     

低温多效蒸发海水淡化系统性能估算方法及公式

Low temperature multi-effect evaporation (LT-MEE) seawater desalination systems with top brine temperature lower than 70℃ have attracted attention in recent years.A method for performance estimation of LT-MEE is proposed, and a formula is built for a typical LT-MEE configuration.Excluding complex properties of working fluids and details of heat and mass transfer processes, this formula involves only simple algebraic operations, thus provides a convenient way to evaluate the water production of LT-MEE for different motive heat source conditions and design parameters.Comparing the formula prediction with the data from both rigorous models and references, it indicates that the formula can give good results in wide parameter ranges.Although this method is proposed based on a typical LT-MEE configuration, it is also applicable to other LT-MEE configurations and thermal desalination systems.     

路径长度受限的随机需求VRP的模型和算法

路径长度受限的随机需求VRP在生产、生活中有着广泛的应用。给出了路径长度受限的随机需求VRP问题的线性整数规划模型,在不允许部分服务和仅能服务失败一次的策略下设计了一个启发式算法。该算法以离散优化中广泛应用的扫描算法为基础生成服务的一个初始可行方案,然后利用模拟退火算法改进得到近似最优解。对需求为二项分布的50个结点、1个服务中心的问题进行了数值试验,数值结果表明该算法对求解路径长度受限的随机需求VRP是有效的。     

火电厂和IGCC及煤气化SOFC混合循环减排CO2的分析

二氧化碳减排已经成为缓解全球气候变化的一个重要议题 .目前 ,火电厂排放的 CO2约占中国 CO2 排放量的 1 /3左右 ,减少其 CO2 排放可以通过提高能量转化效率和回收封存 CO2 两种主要方式 .常规锅炉汽机电厂、IGCC以及煤气化 -固体氧化物燃料电池 ( SOFC)混合循环分别代表了现在、近期及未来燃煤电厂的典型配置 ,超临界及超超临界电厂效率可以达到 40 %以上 ,采用GEH型等先进燃气轮机的 IGCC可提高到 5 0 %以上 ,而混合循环电厂的效率则有望达到 60 %以上 .利用 Aspen Plus TM对这三种电厂进行了模拟 ,考察了三者在回收 CO2 前后性能的变化 .在此基础上 ,分析了减排 CO2 及征收排放税等措施对各电厂发电成本的影响 ,进而就未来如何促进电厂减排 CO2 进行了探讨 .     

平面正交各向异性体材料参数识别算法及软件设计

基于边界元法,本文提出了平面正交各向异性体材料参数识别算法.通过构造以测量位移与边界元计算的相应位移之差的平方和作为目标函数,采用Levenberg-Marquardt方法迭代极小化目标函数,从而把参数识别问题转化为极小化目标函数的问题.同时在识别过程中考虑了测量位移含正态分布的误差,研究测量误差对参数识别结果的影响.算例表明本文提出的参数识别方法是有效的.     

A QUASI-NEWTON ALGORITHM FOR SOLVING MULTIPHASE EQUILIBRIUM FLASH PROBLEMS

A new algorithm has been developed for solving the multicomponent vapor-liquid-liquid equilibrium Mash problem. The algorithm is an extension of the “inside-out” approach proposed by Boston and Britt for the vapor-liquid equilibrium flash problem.

Conventional flash algorithms use temperature, pressure, composition, and phase fraction as the problem independent variables, In the inside-out approach a new set of independent variables is introduced in place of the conventional variables. The new variables are chosen to be as independent as possible of the conventional variables and as free as possible of mutual interaction. Complex phase equilibrium models are used only to generate parameters for a simple equilibrium ratio model. These parameters become the problem independent variables. The Quasi-Newton method of Broyden is employed to promote convergence of these variables.

The algorithm first obtains a solution for the vapor-liquid equilibrium flash. By examining the liquid phase, a heuristic algorithm is employed which quickly locates a two liquid phase composition region of reduced total system free energy when the original liquid is unstable. The solution of the vapor-liquid-liquid equilibrium flash is initiated only when this occurs.

The performance of the algorithm is demonstrated by a number of problems which exhibit varying degrees of nonideality.     

用于多产品间歇化工过程排序的模拟退火算法

提出了一种适用于多产品间歇化工过程排序的改进模拟退火算法 ,该算法采用了非平衡、多次退火策略及多个相邻解产生方法 .通过对无限中间储罐 (UIS)类问题的研究表明 ,同已有的方法相比 ,该方法求解精度高 ,计算速度快 ,适合于求解多产品间歇化工过程的排序问题 .     

A QUASI-NEWTON ALGORITHM FOR SOLVING MULTIPHASE EQUILIBRIUM FLASH PROBLEMS

A new algorithm has been developed for solving the multicomponent vapor-liquid-liquid equilibrium Mash problem. The algorithm is an extension of the “inside-out” approach proposed by Boston and Britt for the vapor-liquid equilibrium flash problem.

Conventional flash algorithms use temperature, pressure, composition, and phase fraction as the problem independent variables, In the inside-out approach a new set of independent variables is introduced in place of the conventional variables. The new variables are chosen to be as independent as possible of the conventional variables and as free as possible of mutual interaction. Complex phase equilibrium models are used only to generate parameters for a simple equilibrium ratio model. These parameters become the problem independent variables. The Quasi-Newton method of Broyden is employed to promote convergence of these variables.

The algorithm first obtains a solution for the vapor-liquid equilibrium flash. By examining the liquid phase, a heuristic algorithm is employed which quickly locates a two liquid phase composition region of reduced total system free energy when the original liquid is unstable. The solution of the vapor-liquid-liquid equilibrium flash is initiated only when this occurs.

The performance of the algorithm is demonstrated by a number of problems which exhibit varying degrees of nonideality.     

基于改进双种群遗传算法的复合材料层合板铺层优化设计

针对复合材料层合板铺层优化设计过程中易出现的算法早熟及收敛慢的问题,结合多种群遗传算法和自适应遗传算法,提出一种引入自适应算子的改进双种群遗传算法。在算法过程中以对称复合材料层合板为例进行验证,通过Patran建立有限元初始模型,采用Matlab编写遗传算法主程序及数据传递程序,实现对Nastran的输入输出文件的读写,并在以Tsai-Wu准则为基础确立的适应度函数下,对复合材料层合板的铺层顺序进行优化。对比传统遗传算法,结果表明该改进算法能够明显提高优化效率,并能够有效收敛于全局最优解,对解决复合材料结构优化设计问题有一定的指导意义。     

基于GA优化的非线性岭回归方法及其在软测量中的应用

提出一种基于遗传算法的非线性岭回归建模方法(GA-NLRR).该方法的核心是先通过RBF的转换实现输入样本的非线性映射,然后用岭回归方法进行线性建模,并采用遗传算法优化岭参数k.该建模方法能很好弥补常规岭回归方法的不足,即无法处理复杂非线性问题和岭参数难确定的问题.将该方法应用于溶剂脱水塔的软测量建模中,仿真研究表明:使用GA-NLRR建立的模型具有很好的预测精度.     

湿磨干烧水泥生产线分解炉温度控制

湿磨干烧水泥生产线的分解炉中生料量波动大,利用HoneywellC200系统自带的带规PID无法满足温度控制工艺要求,本工作利用线性神经元参数自学习PID控制和前馈控制,通过Control Build编制成控制模块,下载到底层控制器实现温度控制,与利用API函数编制用户算法在上层计算机上运行相比,该方案克服了采样和控制周期受计算机负菏影响而变化的缺陷,而不受通讯影响,实际应用结果表明了本文提供出的控制方法的有效性。