粒子群优化算法及其应用

粒子群优化（PSO）算法是一种新颖的演化算法，它属于一类随机全局优化技术，PSO算法通过粒子间的相互作用在复杂搜索空间中发现最优区域。PSO的优势在于简单而又功能强大。介绍了基本的PSO算法、研究现状及其应用，并讨论将来可能的研究内容。     

On finding optimal and near-optimal lineal spanning trees

The search for good lineal, or depth-first, spanning trees is an important aspect in the implementation of a wide assortment of graph algorithms. We consider the complexity of findingoptimal lineal spanning trees under various notions of optimality. In particular, we show that several natural problems, such as constructing a shortest or a tallest lineal tree, are NP-hard. We also address the issue of polynomial-time, near-optimization strategies for these difficult problems, showing that efficient absolute approximation algorithms cannot exist unlessP = NP.This author's research was supported in part by the Sandia University Research Program and by the National Science Foundation under Grant M IP-8603879.This author's research was supported in part by the National Science Foundation under Grants ECS-8403859 and MIP-8603879.     

The Lazy Cook Problem, or Scheduling Two Parallel Machines to Optimize Vehicle Utilization

A cook has to prepare n cakes using an oven with two racks. According to the recipe, the i-th cake has to be baked for exactly a _i minutes. Cakes to be cooked are taken from a table and carried to the oven, and once cooked are carried back to the table by means of a trolley that can carry two cakes at a time. What is the minimum number q ^* of round trips required of the cook? This problem has application to the operation scheduling of transportation systems and to material cutting. A different problem arises according to whether the cook accepts or not to stay near the oven for awhile with the trolley. If the trolley cannot be idle at the oven, an optimum schedule with no oven idle-time always exists: consequently, the trolley schedule is trivial, and the problem is transformed into a set packing. For this case, we propose and test a heuristic method which generates all of the promising columns of the set packing, and solves the resulting problem by branch-and-bound. Instead, if the trolley can be idle at the oven for a limited amount of time, a problem arises to find an optimal schedule of the trolley: in this case we show how to use a scaling technique in order to obtain a very good feasible solution by the method above.     

Experiences with a parallel algorithm for data flow analysis

We have designed a family of parallel data flow analysis algorithms for execution on distributed-memory MIMD machines, based on general-purpose, hybrid algorithms for data flow analysis [Marlowe and Ryder 1990]. We exploit a natural partitioning of the hybrid algorithms and explore a static mapping, dynamic scheduling strategy. Alternative mapping-scheduling choices and refinements of the flow graph condensation used are discussed. Our parallel hybrid algorithm family is illustrated on Reaching Definitions, although parallel algorithms also exist for many interprocedural (e.g., Aliasing) and intraprocedural (e.g., Available Expressions) problems [Marlowe 1989]. We have implemented the parallel hybrid algorithm for Reaching Definitions on an Intel iPSC/2. Our empirical results suggest the practicality of parallel hybrid algorithms.An earlier version of this paper was presented at Supercomputing '90.The research reported here was supported, in part, by the New Jersey Commission on Science and Technology and the CAIP Center's Industrial Members, by Siemens Research Corporation and by National Science Foundation grant CCR-8920078.     

Application of parallelized analogical planning to engineering design

Analogical planning provides a means of solving engineering problems where other machine learning methods fail. Unlike many machine learning paradigms, analogy does not require numerous previous examples or a rich domain theory. Instead, analogical planners adapt knowledge of solved problems in similar domains to the current problem. Unfortunately, the analogical planning task is an expensive one. While the process of forming correspondences between a known problem and a new problem is complex, the problem of selecting a base case for the analogy is virtually intractable.This paper addresses the issue of efficiently forming analogical plans. The Anagram planning system is described, which takes advantage of the massively parallel architecture of the Connection Machine to perform base selection and map formation. Anagram provides a tractable solution to analogical planning, with a complexity that is sublinear in the size of the plans.This paper describes the Anagram system and its parallel algorithms. The paper also presents theoretical analyses and empirical results of testing the system on a large database of plans from the domain of automatic programming.     

Surveillance test policy optimization through genetic algorithms using non-periodic intervention frequencies and considering seasonal constraints

In order to maximize systems average availability during a given period of time, it has recently been developed a non-periodic surveillance test optimization methodology based on genetic algorithms (GA). The fact of allowing non-periodic tests turns the solution space much more flexible and schedules can be better adjusted, providing gains in the overall system average availability, when compared to those obtained by an optimized periodic test scheme. This approach, however, turns the optimization problem more complex. Hence, the use of a powerful optimization technique, such as GA, is required.Considering that some particular features of certain systems can turn it advisable to introduce other specific constraints in the optimization problem, this work investigates the application of seasonal constraints for the set of the Emergency Diesel Generation of a typical four-loop pressurized water reactor in order to planning and optimizing its surveillance test policy. In this analysis, the growth of the blackout accident probability during summer, due to electrical power demand increases, was considered. Here, the used model penalizes surveillance test interventions when the blackout probability is higher.Results demonstrate the ability of the method in adapting the surveillance test policy to seasonal constraints. The knowledge acquired by the GA during the searching process has lead to test schedules that drastically minimize test interventions at periods of high blackout probability. It is compensated by more frequent redistributed tests through the periods of low blackout probability in order to improve on the overall average availability at the system level.     

Partitioning multiple objective optimal solutions with applications in radiotherapy design

The optimal partition for linear programming is induced by any strictly complementary solution, and this partition is important because it characterizes the optimal set. However, constructing a strictly complementary solution in the presence of degeneracy was not practical until interior point algorithms became viable alternatives to the simplex algorithm. We develop analogs of the optimal partition for linear programming in the case of multiple objectives and show that these new partitions provide insight into the optimal set (both pareto optimality and lexicographic ordering are considered). Techniques to produce these optimal partitions are provided, and examples from the design of radiotherapy plans show that these new partitions are useful. A. Holder. Research conducted at Trinity University, TX, and The University of Mississippi, MS. This research was partially supported by ONR Grant N00014-01-1-0917.     

基于遗传算法的工作辊温度场参数优化模型

针对热连轧机工作辊热辊形计算中温度场模型的热交换等参数难以确定的问题 ,建立了基于遗传算法的参数优化模型 ,可以解决复杂条件下的热参数的求解问题。利用优化参数计算的轧辊温度场与实际测量结果一致。