模块化多电平变换器优化调度调制策略研究

由于MMC电源的模块电容电压存在波动,一般认为它较为适用于定频运行。首先,定义了MMC电源的2种工作模式,即高效放电的工作模式和高效充电的工作模式;然后在此基础上提出了MMC电源的优化调度调制方法,该方法能够实现MMC变频变压运行时的低电容电压波动和高效控制;最后仿真结果验证了所提方法的有效性。     

多排料方案下的加工排程优化研究

结合工程机械中金属结构件的加工特点,针对金属结构件生产过程中多任务混合下料与加工排程优化问题,提出多种排料方案下的带工艺约束多目标加工排程优化模型,并设计一种蚁群--递阶遗传算法来求解该模型。该算法首先利用蚁群算法选择一组优异的排料方案,然后利用递阶遗传算法的选择、交叉、变异等操作求解作业加工顺序及机器选择问题。通过对加工排程的优化,缩短金属结构件产品生产周期,降低其在制品数量和提高设备利用率,并通过实例分析证明了该方法的有效性。     

面向工艺规划与车间调度的集成方法研究

面向工艺规划与车间调度的集成优化,构建了基于虚拟制造单元的资源调度组织模型。通过技术分析、资源分析和详细设计三阶段来进行工艺与调度的集成:技术分析阶段完成粗工艺路线的制定;在资源分析阶段对满足条件的制造资源进行优化集成;而详细设计阶段则引入了调度因子的概念,通过调度因子来进行工序与对应制造资源的匹配选择,制定最优的工艺路线和生产计划的排产。最后通过实例验证了该方法的有效性和可行性。     

Rate monotonic scheduling re-analysed

In this paper, we re-analyse the rate monotonic scheduler. Traditionally, the schedulability condition was obtained from the greatest lower bound of utilisation factors over all the task sets that (are schedulable and) fully utilise the processor. We argue that full utilisation is not very appropriate for this purpose. We re-establish Liu and Layland's classic schedulability theorem by finding the greatest lower bound of utilisation factors over all the unschedulable task sets instead. The merits of our approach include: Firstly, the fact that the bound is both sound and tight for schedulability follows directly from definition; Secondly, our proof is simpler technically.     

Minimum Moment Method for Resource Leveling Using Entropy Maximization

The minimum moment method for resource leveling is revisited and restated as an entropy-maximization problem. The minimum moment method assumes that the moment of the daily resource demands about the horizontal axis of a project’s resource histogram is a good measure of the resource utilization and that the optimal resource allocation exists when the total moment is at a minimum, thus when the resource histogram is of rectangular shape. The entropy-maximization method proposed in this paper makes use of the general theory of entropy and two of its principal properties (subadditivity and maximality) to revisit the minimum moment method for resource leveling. The entropy-maximization method presented allows for activity stretching and provides resource allocation solutions that show improvement over previous approaches. A case study is also presented that validates the results.     

自然崩落法放矿排产品位连续预测模型

简单介绍了自然崩落法的放矿原理,阐述了劳伯斯特品位估值法的原理,针对目前该方法不能计算任意放矿阶段出矿品位的缺点,提出了连续品位预测模型,并采用VB语言和Access数据库技术实现了排产品位预测软件。     

Rate regulated TDMA system for cochannel interference limited wireless ad hoc networks

The time slotted system with rate regulation, which adapts the transmission rate to the interference for each time slot, is proposed to enhance the throughput while meeting the SINR requirement for cochannel interference limited wireless ad hoc networks. We show that the throughput of the rate regulated TDMA system is larger than that of the rate regulated TDMA/CDMA system. Furthermore, we also propose a scheduling algorithm to maximize the total throughput of the rate regulated TDMA system.     

Petri-net based scheduling strategy for semiconductor manufacturing processes

Semiconductor manufacturing is a highly automated and capital-intensive industrial process. The operating cost of a wafer processing plant is in general closely related to the design and management of its process flows. Traditionally, the task of production scheduling is performed manually on the basis of past experiences. There are thus real incentives to develop a systematic approach to construct a mathematical programming model in order to reduce the chance of human errors and to ensure operational efficiency in implementing the resulting schedules. To this end, the Petri nets are adopted in this work to accurately model the semiconductor manufacturing activities. The token movements in a Petri net are represented with the well-established scheduling model for batch chemical processes, and the optimal schedule of the given semiconductor process can then be determined accordingly. The feasibility and effectiveness of this scheduling strategy is demonstrated in the present paper with three examples, i.e., the final test process, the re-entrant flow process, and the photolithography-etching process.     

Detection probability of Campylobacter

A rapid presence/absence test for Campylobacter in chicken faeces is being evaluated to support the scheduling of highly contaminated broiler flocks as a measure to reduce public health risks [Nauta, M. J., & Havelaar, A. H. (2008). Risk-based standards for Campylobacter in the broiler meat chain. Food Control, 19, 372–381]. Although the presence/absence test is still under development, an example data set of test results is analysed to illustrate the benefit of the detection probability concept. The detection probability of Campylobacter increases with the logarithm of the Campylobacter concentration in faeces according to an S-shaped curve which stretches about 2–3 log units. The detection probability is 50% at a Campylobacter concentration of 7.4 × 10⁶ cfu/g. The uncertainty in the detection probability is 32% at the most for a 90% confidence interval. This type of information allows for realistic calculations on the Campylobacter status of different food processing paths after splitting. Usable quantitative estimates on detection probability await a data set of test results from a test that is ready for use or has similar properties.     

A note on “A heuristic method for two-stage hybrid flow shop with dedicated machines”

In a recent paper by Wang and Liu (A heuristic for two-stage hybrid flow shop with dedicated machines. Computers & Operations Research 2013;40:438–50), two elimination rules are proposed for the hybrid flow shop with dedicated machines. This note shows that one of them is not valid and that the other is dominated by an other elimination rule.