Multi-objective optimization of rolling schedule based on cost function for tandem cold mill

In terms of tandem cold mill productivity and product quality, a multi-objective optimization model of rolling schedule based on cost fimction was proposed to determine the stand reductions, inter-stand tensions and rolling speeds for a specified product. The proposed schedule optimization model consists of several single cost fi.mctions, which take rolling force, motor power, inter-stand tension and stand reduction into consideration. The cost function, which can evaluate how far the rolling parameters are from the ideal values, was minimized using the Nelder-Mead simplex method. The proposed rolling schedule optimization method has been applied successfully to the 5-stand tandem cold mill in Tangsteel, and the results from a case study show that the proposed method is superior to those based on empirical formulae.

Multi-objective Function Optimization for Environmental Control of a Greenhouse Based on a RBF and NSGA-Ⅱ

To better meet the needs of crop growth and achieve energy savings and efficiency enhancements,constructing a reliable environmental model to optimize greenhouse decision parameters is an important problem to be solved.In this work,a radial-basis function(RBF)neural network was used to mine the potential changes of a greenhouse environment,a temperature error model was established,a multi-objective optimization function of energy consumption was constructed and the corresponding decision parameters were optimized by using a non-dominated sorting genetic algorithm with an elite strategy(NSGA-Ⅱ).The simulation results showed that RBF could clarify the nonlinear relationship among the greenhouse environment variables and decision parameters and the greenhouse temperature.The NSGA-Ⅱ could well search for the Pareto solution for the objective functions.The experimental results showed that after 40 min of combined control of sunshades and sprays,the temperature was reduced from 31℃to 25℃,and the power consumption was 0.5 MJ.Compared with tire three days of July 24,July 25 and July 26,2017,the energy consumption of the controlled production greenhouse was reduced by 37.5%,9.1%and 28.5%,respectively.

A Penalty Function Algorithm with Objective Parameters and Constraint Penalty Parameter for Multi-Objective Programming

In this paper, we present an algorithm to solve the inequality constrained multi-objective programming (MP) by using a penalty function with objective parameters and constraint penalty parameter. First, the penalty function with objective parameters and constraint penalty parameter for MP and the corresponding unconstraint penalty optimization problem (UPOP) is defined. Under some conditions, a Pareto efficient solution (or a weakly-efficient solution) to UPOP is proved to be a Pareto efficient solution (or a weakly-efficient solution) to MP. The penalty function is proved to be exact under a stable condition. Then, we design an algorithm to solve MP and prove its convergence. Finally, numerical examples show that the algorithm may help decision makers to find a satisfactory solution to MP.

Optimality for Multi-Objective Programming Involving Arcwise Connected d-Type-I Functions

This paper deals with the optimality conditions and dual theory of multi-objective programming problems involving generalized convexity. New classes of generalized type-I functions are introduced for arcwise connected functions, and examples are given to show the existence of these functions. By utilizing the new concepts, several sufficient optimality conditions and Mond-Weir type duality results are proposed for non-differentiable multi-objective programming problem.

A vague-set-based fuzzy multi-objective decision making model for bidding purchase

A vague-set-based fuzzy multi-objective decision making model is developed for evaluating bidding plans in a bid- ding purchase process. A group of decision-makers (DMs) first independently assess bidding plans according to their experience and preferences, and these assessments may be expressed as linguistic terms, which are then converted to fuzzy numbers. The resulting decision matrices are then transformed to objective membership grade matrices. The lower bound of satisfaction and upper bound of dissatisfaction are used to determine each bidding plan’s supporting, opposing, and neutral objective sets, which together determine the vague value of a bidding plan. Finally, a score function is employed to rank all bidding plans. A new score function based on vague sets is introduced in the model and a novel method is presented for calculating the lower bound of sat- isfaction and upper bound of dissatisfaction. In a vague-set-based fuzzy multi-objective decision making model, different valua- tions for upper and lower bounds of satisfaction usually lead to distinct ranking results. Therefore, it is crucial to effectively contain DMs’ arbitrariness and subjectivity when these values are determined.

Overview of multi-objective optimization methods

To assist readers to have a comprehensive understanding, the classical and intelligent methods roundly based on precursory research achievements are summarized in this paper. First, basic conception and description about multi-objective (MO) optimization are introduced. Then some definitions and related terminologies are given. Furthermore several MO optimization methods including classical and current intelligent methods are discussed one by one succinctly. Finally evaluations on advantages and disadvantages about these methods are made at the end of the paper.

Multi-Objective Redundancy Optimization of Continuous-Point Robot Milling Path in Shipbuilding

The 6-DOF manipulator provides a new option for traditional shipbuilding for its advantages of vast working space,low power consumption,and excellent flexibility.However,the rotation of the end effector along the tool axis is functionally redundant when using a robotic arm for five-axis machining.In the process of ship construction,the performance of the parts’protective coating needs to bemachined tomeet the Performance Standard of Protective Coatings(PSPC).The arbitrary redundancy configuration in path planning will result in drastic fluctuations in the robot joint angle,greatly reducing machining quality and efficiency.There have been some studies on singleobjective optimization of redundant variables,However,the quality and efficiency of milling are not affected by a single factor,it is usually influenced by several factors,such as the manipulator stiffness,the joint motion smoothness,and the energy consumption.To solve this problem,this paper proposed a new path optimization method for the industrial robot when it is used for five-axis machining.The path smoothness performance index and the energy consumption index are established based on the joint acceleration and the joint velocity,respectively.The path planning issue is formulated as a constrained multi-objective optimization problem by taking into account the constraints of joint limits and singularity avoidance.Then,the path is split into multiple segments for optimization to avoid the slow convergence rate caused by the high dimension.An algorithm combining the non-dominated sorting genetic algorithm(NSGA-II)and the differential evolution(DE)algorithm is employed to solve the above optimization problem.The simulations validate the effectiveness of the algorithm,showing the improvement of smoothness and the reduction of energy consumption.

Multi-objective forest harvesting under sustainable and economic principles

Selective logging is well-recognized as an effective practice in sustainable forest management.However,the ecological efficiency or resilience of the residual stand is often in doubt.Recovery time depends on operational variables,diversity,and forest structure.Selective logging is excellent but is open to changes.This may be resolved by mathematical programming and this study integrates the economic-ecological aspects in multi-objective function by applying two evolutionary algorithms.The function maximizes remaining stand diversity,merchantable logs,and the inverse of distance between trees for harvesting and log landings points.The Brazilian rainforest database(566 trees)was used to simulate our 216-ha model.The log landing design has a maximum volume limit of 500 m3.The nondominated sorting genetic algorithm was applied to solve the main optimization problem.In parallel,a sub-problem(p-facility allocation)was solved for landing allocation by a genetic algorithm.Pareto frontier analysis was applied to distinguish the gradientsα-economic,β-ecological,andγ-equilibrium.As expected,the solutions have high diameter changes in the residual stand(average removal of approximately 16 m^(3) ha^(-1)).All solutions showed a grouping of trees selected for harvesting,although there was no formation of large clearings(percentage of canopy removal<7%,with an average of 2.5 ind ha^(-1)).There were no differences in floristic composition by preferentially selecting species with greater frequency in the initial stand for harvesting.This implies a lower impact on the demographic rates of the remaining stand.The methodology should support projects of reduced impact logging by using spatial-diversity information to guide better practices in tropical forests.

A Study on the Multi-Objective Optimization Method of Brackets in Ship Structures

The shape and size optimization of brackets in hull structures was conducted to achieve the simultaneous reduction of mass and high stress,where the parametric finite element model was built based on Patran Command Language codes.The optimization procedure was executed on Isight platform,on which the linear dimensionless method was introduced to establish the weighted multi-objective function.The extreme processing method was applied and proved effective to normalize the objectives.The bracket was optimized under the typical single loads and design waves,accompanied by the different proportions of weights in the objective function,in which the safety factor function was further established,including yielding,buckling,and fatigue strength,and the weight minimization and safety maximization of the bracket were obtained.The findings of this study illustrate that the dimensionless objectives share equal contributions to the multi-objective function,which enhances the role of weights in the optimization.

Multi-objective robust controller synthesis for discrete-time systems with convex polytopic uncertain domain

Multi-objective robust state-feedback controller synthesis problems for linear discrete-time uncertain systems are addressed. Based on parameter-dependent Lyapunov functions, the Gl2 and GH2 norm expressed in terms of LMI (Linear Matrix Inequality) characterizations are further generalized to cope with the robust analysis for convex polytopic uncertain system. Robust state-feedback controller synthesis conditions are also derived for this class of uncertain systems. Using the above results, multi-objective state-feedback controller synthesis procedures which involve the LMI optimization technique are developed and less conservative than the existing one. An illustrative example verified the validity of the approach.

Multi-objective Particle Swarm Optimization Algorithm Based on Performance and Reliability of Discrete System Resources Configuration

Considering research on multi-objective optimization for reliability and performance suffering cost constraints in digital circuits,an improved multi-objective optimization algorithm based on performance and reliability was proposed to solve the problem of discrete system resources configuration in this paper. This algorithm used the particle-swarm optimization( PSO) to evaluate the tradeoffs configuration of the system resources between reliability and performance and proved the feasibility through the simulation.Finally, the information of resources configuration from optimization algorithm was used to effectively guide the system design so as to mitigate soft errors caused by single event effect( SEE).

Searching for an Optimized Single-objective Function Matching Multiple Objectives with Automatic Calibration of Hydrological Models

In the calibration of hydrological models, evaluation criteria are explicitly and quantitatively defined as single-or multi-objective functions when utilizing automatic calibration approaches.In most previous studies, there is a general opinion that no single-objective function can represent all important characteristics of even one specific hydrological variable(e.g., streamflow).Thus hydrologists must turn to multi-objective calibration.In this study, we demonstrated that an optimized single-objective function can compromise multi-response modes(i.e., multi-objective functions) of the hydrograph, which is defined as summation of a power function of the absolute error between observed and simulated streamflow with the exponent of power function optimized for specific watersheds.The new objective function was applied to 196 model parameter estimation experiment(MOPEX) watersheds across the eastern United States using the semi-distributed Xinanjiang hydrological model.The optimized exponent value for each watershed was obtained by targeting four popular objective functions focusing on peak flows, low flows, water balance, and flashiness, respectively.Results showed that the optimized single-objective function can achieve a better hydrograph simulation compared to the traditional single-objective function Nash-Sutcliffe efficiency coefficient for most watersheds, and balance high flow part and low flow part of the hydrograph without substantial differences compared to multi-objective calibration.The proposed optimal single-objective function can be practically adopted in the hydrological modeling if the optimal exponent value could be determined a priori according to hydrological/climatic/landscape characteristics in a specific watershed.

一种改进蚁群算法的路径规划研究

针对机器人在复杂环境中的路径规划问题,本文提出了一种改进蚁群算法的路径规划研究方法.首先,在启发函数中引入一种自适应调整的放大因子,以提高相邻节点的启发信息差异,使蚂蚁朝着最优路径的方向搜索;其次,采用一种奖惩机制对路径上的信息素进行更新,使算法的收敛速度得到有效的提高;然后,通过对信息素挥发因子进行动态调整,提高蚁群的搜索速度,使算法快速收敛.最后,在最优路径的基础上,采用拐点优化算法与分段B样条曲线相结合的方法来进行路径优化,有效的改善了路径的平滑性.仿真结果表明,所提的研究方法具有更好的收敛性和搜索能力,更符合机器人运动的实际要求.

基于深度强化学习的SCR脱硝系统协同控制策略研究

针对选择性催化还原(SCR)脱硝系统大惯性、多扰动等特点,提出了一种基于多维状态信息和分段奖励函数优化的深度确定性策略梯度(DDPG)协同比例积分微分(PID)控制器的控制策略。针对SCR脱硝系统中存在部分可观测马尔可夫决策过程(POMDP),导致DDPG算法策略学习效率较低的问题,首先设计SCR脱硝系统的多维状态信息;其次,设计SCR脱硝系统的分段奖励函数;最后,设计DDPG-PID协同控制策略,以实现SCR脱硝系统的控制。结果表明:所设计的DDPG-PID协同控制策略提高了DDPG算法的策略学习效率,改善了PID的控制效果,同时具有较强的设定值跟踪能力、抗干扰能力和鲁棒性。

ACR-MLM:a privacy-preserving framework for anonymous and confidential rewarding in blockchain-based multi-level marketing

Network marketing is a trading technique that provides companies with the opportunity to increase sales.With the increasing number of Internet-based purchases,several threats are increasingly observed in this field,such as user privacy violations,company owner(CO)fraud,the changing of sold products’information,and the scalability of selling networks.This study presents the concept of a blockchain-based market called ACR-MLM that functions based on the multi-level marketing(MLM)model,through which registered users receive anonymous and confidential rewards for their own and their subgroups’sales.Applying a public blockchain as the ACR-MLM framework’s infrastructure solves existing problems in MLM-based markets,such as CO fraud(against the government or its users),user privacy violations(obtaining their real names or subgroup users),and scalability(when vast numbers of users have been registered).To provide confidentiality and scalability to the ACR-MLM framework,hierarchical identity-based encryption(HIBE)was applied with a functional encryption(FE)scheme.Finally,the security of ACR-MLM is analyzed using the random oracle(RO)model and then evaluated.

改进MADDPG算法的非凸环境下多智能体自组织协同围捕

针对多智能体在非凸环境下的围捕效率问题,提出基于改进经验回放的多智能体强化学习算法。利用残差网络(ResNet)来改善网络退化问题,并与多智能体深度确定性策略梯度算法(MADDPG)相结合,提出了RW-MADDPG算法。为解决多智能体在训练过程中,经验池数据利用率低的问题,提出两种改善经验池数据利用率的方法;为解决多智能体在非凸障碍环境下陷入障碍物内部的情况(如陷入目标不可达等),通过设计合理的围捕奖励函数使得智能体在非凸障碍物环境下完成围捕任务。基于此算法设计仿真实验,实验结果表明,该算法在训练阶段奖励增加得更快,能更快地完成围捕任务,相比MADDPG算法静态围捕环境下训练时间缩短18.5%,动态环境下训练时间缩短49.5%,而且在非凸障碍环境下该算法训练的围捕智能体的全局平均奖励更高。

实时功能磁共振成像神经反馈在肥胖症中的应用进展

肥胖症及减重后不能维持健康体质量的核心因素多为食物成瘾,食物成瘾在神经影像学中表现为奖赏网络与认知控制网络间神经环路的失衡。实时功能磁共振成像神经反馈(real time functional magnetic resonance imaging neurofeedback,rtfMRI-NF)作为一种新型生物反馈技术,已被应用于其他物质成瘾领域的临床研究和治疗中。在食物成瘾肥胖症中,rtfMRI-NF同样具有重塑异常脑功能、改善摄食行为并达到减重效果的潜力。本综述总结了肥胖患者食物成瘾的功能磁共振脑成像模型,探讨应用rtfMRI-NF作为其潜在治疗工具的可行神经靶点,并回顾了rtfMRI-NF在肥胖应用中的最新研究进展,为未来rtfMRI-NF在肥胖中的治疗策略和临床指导提供参考。

融合IMR-WGAN的时序数据修复方法

工业数据由于技术故障和人为因素通常导致数据异常,现有基于约束的方法因约束阈值设置的过于宽松或严格会导致修复错误,基于统计的方法因平滑修复机制导致对时间步长较远的异常值修复准确度较低.针对上述问题,提出了基于奖励机制的最小迭代修复和改进WGAN混合模型的时序数据修复方法.首先,在预处理阶段,保留异常数据,进行信息标注等处理,从而充分挖掘异常值与真实值之间的特征约束.其次,在噪声模块提出了近邻参数裁剪规则,用于修正最小迭代修复公式生成的噪声向量.将其传递至模拟分布模块的生成器中,同时设计了一个动态时间注意力网络层,用于提取时序特征权重并与门控循环单元串联组合捕捉不同步长的特征依赖,并引入递归多步预测原理共同提升模型的表达能力;在判别器中设计了Abnormal and Truth奖励机制和Weighted Mean Square Error损失函数共同反向优化生成器修复数据的细节和质量.最后,在公开数据集和真实数据集上的实验结果表明,该方法的修复准确度与模型稳定性显著优于现有方法.

用于移动机器人路径规划的改进强化学习算法

针对传统Q-learning算法规划出的路径存在平滑度差、收敛速度慢以及学习效率低的问题,本文提出一种用于移动机器人路径规划的改进Q-learning算法。首先,考虑障碍物密度及起始点相对位置来选择动作集,以加快Q-learning算法的收敛速度;其次,为奖励函数加入一个连续的启发因子,启发因子由当前点与终点的距离和当前点距地图中所有障碍物以及地图边界的距离组成;最后,在Q值表的初始化进程中引入尺度因子,给移动机器人提供先验环境信息,并在栅格地图中对所提出的改进Q-learning算法进行仿真验证。仿真结果表明,改进Q-learning算法相比传统Q-learning算法收敛速度有明显提高,在复杂环境中的适应性更好,验证了改进算法的优越性。

基于蒙特卡罗策略梯度的雷达观测器轨迹规划

在目标跟踪过程的雷达观测器轨迹规划(OTP)中,针对马尔可夫步进规划智能决策问题,在离散动作空间上,提出了一种基于蒙特卡罗策略梯度(MCPG)算法的雷达轨迹规划方法.首先,联合目标跟踪状态、奖励机制、动作方案和雷达观测器位置,将OTP过程建模为一个连续的马尔可夫决策过程(MDP),提出基于MCPG的全局智能规划方法.其次,将跟踪幕长内的每个时间步作为单独一幕来进行策略更新,提出基于MCPG目标跟踪中观测器轨迹的步进智能规划方法,并深入研究目标的跟踪估计特性,构造以跟踪性能优化为目的的奖励函数.最后,对最优非线性目标跟踪过程中基于强化学习的智能OTP决策仿真实验,表明了所提方法的有效性.