近优解 meaning in Chinese
near-optimum solution
Examples
- Considering the np - complete problem , how to get the approximate optimized scheme of job - shop scheduling , and aimed at improving the efficiency of products and taking good advantage of concurrence , asynchronism , distributing and juxtaposition in multi - products and devices processing , we could divide the working procedures into the attached one which has the only precursor and subsequence and unattached one by analyzing working flow chart of job - shop , that is the working procedures are divided into two types , then the bf and the ff methods about memory scheduling in os are applied , therefore a new approximate optimized scheme is presented in the paper which could solve the common job - shop scheduling . namely , the acpm and the bfsm are applied to the classified and grouped working procedures considering the compact of the procedures and practical examples approved it . the results we analyzing and tested show that it is better than the heuristic algorithm common used , for less restriction terms , more satisfying algorithm complexity and better optimized results
针对job - shop调度问题求最优解算法这一npc问题，本文以充分发挥多产品、多设备加工所具有并发性、异步性、分布性和并行性的加工优势，从而提高产品的加工效率为目标，对job - shop调度问题的工艺图进行适当分解，使工序在一定时间段或是为具有唯一紧前、紧后相关工序或是为独立工序，即将工序分两类，再结合操作系统中内存调度的最佳适应( bf )调度方法和首次适应( ff )调度方法的先进思想，通过分析提出了一种解决一般job - shop调度问题的全新近优解方案：在考虑关键设备上工序尽量紧凑的前提下，将工序分类、对这两类工序分批采用拟关键路径法( acpm )和最佳适应调度方法( bfsm )安排工序的算法，用实例加以验证，并给出结果甘特图。 - When the approximate optimized scheme of common job - shop scheduling discussed in the paper is applied to the practice , it could make some parts of working procedures of processing product be the tail end of the tree that working procedures of this product makes , that is , these parts of working procedures still makes a tree like the whole process . during the processing of one product , if another product needed to be processed , we could process it with the mentioned method solving static job - shop scheduling . therefore , a new method to solve dynamic job - shop scheduling is put forward and validated by practice
同时采用本文提出的解决一般job - shop调度问题的全新近优解方案，可以使得产品所加工的部分工序是产品的加工工艺图（加工树）某些枝杈的末端，即产品未加工的剩余部分工序的加工工艺图仍然是一棵加工树，这样对于正在加工的产品，如果有另外需要加工的产品，可一并按上述解决静态job - shop调度问题的方法处理，于是本文又提出了一个解决动态job - shop调度问题的新方法，并通过实例加以说明。 - Abstract : in this paper , we propose an improved lagrangian relaxation algorithm to solve job - shop scheduling problems . besides the addition of augmented objective , we expand the search scope of near - optimal solutions and improve the computational efficiency greatly by restricting the solution scope of sub - problems and modifying the search method of dual problem . at the same time , we develop a genetic algorithm combining with the lr ( lagrangian relaxation ) method . using the numerous useful solutions we get in the lagrangian relaxation as the original genes , we can improve the solution further . test results show that these methods achieve satisfied outcome for job - shop problems . they can also be applyed to other programming problems with constraints
文摘:针对车间调度问题，提出了一种改进的拉氏松弛算法.在增加辅助目标函数的基础上，通过对子问题的限制和搜索策略的改变，使拉氏算法的计算量减少，近优解的搜索能力有很大改善.本文还提出了一种基因优化算法，充分利用拉氏算法得到的多个近优解，进一步优化结果.仿真结果表明对车间调度问题得到了较好的结果.本方法也可用于其它有约束的规划问题 - The number of the hidden layers of mul - tilayer perceptrons ( mlps ) is analyzed , and three - layer perceptrons neural network is adopted ; by analyzing the mechanism of the neural cells in hidden layer , a method for combining genetic algorithm and bp algorithm to optimize the design of the neural networks is presented , and it solves the defects of getting into infinitesimal locally and low convergence efficiently , it can also solve the problem that it can usually obtain nearly global optimization solution within shorter time through using genetic algorithm method lonely ; several examples validate that this algorithm can simplify the neural networks effectively , and it makes the neural networks solve the practical problem of fault diagnosis more effectively
对多层感知器隐层数进行了分析,确定采用三层感知器神经网络;通过对隐层神经元作用机理的分析,引入了遗传算法与bp算法相结合以优化设计神经网络的方法,有效地解决了bp算法收敛速度慢和易陷入局部极小的弱点,还可以解决单独利用遗传算法往往只能在短时间内寻找到接近全局最优的近优解的问题;通过实例验证了这种算法能够有效地简化神经网络,使神经网络更加有效地解决实际的故障诊断问题。