基于IsoMap和MBFO-SVR的瓦斯涌出量动态预测研究

为了能够实现高精度与实时性的动态预测煤矿绝对瓦斯涌出量,本文提出了等容特征映射IsoMap（Isometric feature Mapping）与改进细菌觅食优化算法MBFO（Modified Bacteria Foraging Optimization）优化支持向量回归机SVR（Support Vector Regression）相结合的预测方法。瓦斯涌出是在多种影响因子共同作用下的结果,并且这些因素之间是复杂的非线性关系,因此本文中提出采用流形学习方法IsoMap对其进行降维特征提取,该方法用测地距离（geodesic distace）取代了普遍采用的欧氏距离,有利于对高维特征内在关系的挖掘,取得了优于传统的主成分分析（PCA）的结果;将MBFO算法对SVR的相关参数进行寻优;将IsoMap分析结果输入预测模型。仿真表明,与PSO算法比较,本文提出的预测方法预测精度较高,更加有利于对瓦斯涌出量预测。     

基于细菌觅食优化算法的城市轨道交通调度优化

为了合理高效地制定城市轨道交通调度方案,实现客流与车次的优化配置,提出了一种基于细菌觅食优化算法的城市轨道交通调度优化策略。兼顾乘客与运营企业双方利益,以发车间隔为决策变量,乘客平均候车时间最短和发车次数最少为优化目标,建立调度优化模型,并对细菌觅食优化算法求解该调度模型的过程进行分析。结合某城市轨道交通一号线实际运营数据进行仿真实验,并与其他算法的优化结果进行对比分析,实验表明该算法和模型能有效解决城市轨道交通调度优化问题。     

一种群体智能融合算法及其在应急设施选址的应用

针对粒子群优化算法早熟及细菌觅食算法收敛慢的问题,提出了将量子粒子群优化与细菌觅食算法融合的一种群体智能融合算法。该算法将细菌觅食、量子计算理论及粒子群优化的优点进行融合,以细菌觅食算法为主体,将量子进化算法及粒子群优化算法嵌入其中,从而极大地提高了算法的性能。通过对三个标准函数求解和验证,结果表明该算法提高了收敛精度及速度。最后用该算法求解公共卫生应急服务设施点选址问题,取得了较好的效果,说明了该算法的有效性。     

电网供电系统短期电力负荷预测优化仿真

对电网供电系统短期电力负荷预测模型进行优化,能提升预测结果的准确性和鲁棒性.虽然现有预测模型可以满足预测速度的要求,但预测结果的精确性和稳定性却无法保证.为了得到更加准确和稳定的预测结果,提出了细菌觅食算法优化极限学习机预测模型.首先在电力负荷样本数据中形成训练样本和预测样本集,利用细菌觅食优化算法对极限学习机预测模型中的不确定参数进行优化,然后利用改进后的模型进行电力负荷预测.新模型的优化仿真结果显示,利用细菌觅食算法优化极限学习机预测模型的预测精度和稳定性均优于传统预测模型的预测结果,该算法具有很好地实用性.     

基于翻筋斗觅食策略的灰狼优化算法

灰狼优化算法(grey wolf optimization,GWO)存在收敛的不合理性等缺陷,目前对GWO算法的收敛性改进方式较少,除此之外,当GWO迭代至后期,所有灰狼个体都逼近α狼、β狼、δ狼,导致算法陷入局部最优。针对以上问题,提出了一种增强型的灰狼优化算法(disturbance and somersault foraging-grey wolf optimization,DSF-GWO)。首先引入一种扰动因子,平衡了算法的开采和勘探能力;其次引入翻筋斗觅食策略,在后期使其不陷入局部最优的同时也使得前期的群体多样性略有提升。对DSF-GWO算法的寻优性能进行验证,选取14个单/多峰目标函数进行实验,在相同的参数设置下,结果表明DSF-GWO算法在寻优性能上较GWO算法有明显优势。     

Aviation maintenance technician scheduling with personnel satisfaction based on interactive multi-swarm bacterial foraging optimization

This study focuses on the challenges of aviation maintenance technician (AMT) scheduling and constructs a model based on personnel satisfaction and the parallel execution of aircraft maintenance tasks. To obtain the scheduling scheme from the constructed NP-hard model, an interactive multi-swarm bacterial foraging optimization (IMSBFO) algorithm is proposed using multi-swarm coevolution, structural recombination, and three information interactive mechanisms among individuals. Moreover, considering the distributed feature of the AMT scheduling problem, a specific mechanism is designed to convert continuous solution to a binary AMT scheduling scheme. Finally, a series of comparative experiments highlight the efficiency and superiority of our proposed IMSBFO algorithm, and the optimal scheduling scheme owns the delicate balance between the work and rest time.     

Mathematical modeling and a hybridized bacterial foraging optimization algorithm for the flexible job-shop scheduling problem with sequencing flexibility

The flexible job-shop scheduling problem (FJSP) is an extension of the classical job-shop scheduling problem (JSP) in which operations can be performed by a set of candidate capable machines. An extended version of the FJSP, entitled sequencing flexibility, is studied in this work, which considers precedence between the operations in the form of a directed acyclic graph instead of a sequential order. In this work, a mixed integer linear programming (MILP) formulation is presented to minimize weighted tardiness for the FJSP with sequencing flexibility. Due to the NP-hardness of the problem, a novel biomimicry hybrid bacterial foraging optimization algorithm (HBFOA) is developed, which is inspired by the behavior of E. coli bacteria in its search for food. The developed HBFOA search method is hybridized with simulated annealing (SA). Additionally, the algorithm has been enhanced by a local search method based on the manipulation of critical operations. Classical dispatching rules have been employed to create the initial swarm of HBFOA, and a new dispatching rule named minimum number of operations has been devised. The developed approach has been packaged in the form of a decision support system (DSS) developed on top of Microsoft Excel—a tool most small and mid-range enterprises (SME) use heavily for planning. A case study with local industry is presented to validate the proposed HBFOA and MILP. Additional numerical experiments using literature benchmarks are further used for validation. The results demonstrate that the HBFOA outperformed the classical dispatching rules and the best integer solution of MILP when minimizing the weighted tardiness and offered comparable results for the makespan instances.     

Locating calories: Does the high-calorie bias in human spatial memory influence how we navigate the modern food environment?

Human memory appears to be adaptively “biased” towards remembering the locations of (fitness-relevant) high-calorie nutritional resources. It remains to be investigated whether this high-calorie bias in human spatial memory influences how individuals navigate the modern food environment, and whether it is proximally associated with attentional processes. 60 individuals completed computer-based food eye-tracking and spatial memory tasks in a lab setting, as well as a food search and covert food choice task in an unfamiliar supermarket. The high-calorie spatial memory bias was replicated, as individuals more accurately recalled locations of high-calorie relative to low-calorie foods, regardless of hedonic evaluations or familiarity with foods. Although individuals were faster at (re)locating high-calorie (versus low-calorie) items in the supermarket, the bias did not predict a lower search time for high-calorie foods, or a higher proportion of high-calorie food choice. Rather, an enhanced memory for high-calorie food locations was associated with a lower perceived difficulty (i.e. greater ease) of finding high-calorie items in the supermarket, which may potentiate later choice of a high-calorie food. The high-calorie spatial memory bias was also found to be expressed independently of the amount of visual attention individuals allocated to high-calorie versus low-calorie foods. Findings further substantiate the notion that human spatial memory shows sensitivity to the caloric content of a potential resource and automatically prioritizes those with greater energy payoffs. Such a spatial mechanism that was adaptive for energy-efficient foraging within fluctuating ancestral food environments could presently yield maladaptive “obesogenic” consequences, through altering perceptions of food search convenience.     

Foraging Behavior and Microhabitat Use by Spiny Mice, Acomys cahirinus and A. russatus, in the Presence of Blanford's Fox (Vulpes cana) Odor

We investigated the responses of common and golden spiny mice (Acomys cahirinus and A. russatus, respectively) to the fecal odor of Blanford's fox (Vulpes cana), a predator of Acomys, which overlaps in habitat use with the mice. Neither species of mouse showed a significant response to the presence of fox odor compared with the presence of the fecal odor of a local herbivore (Nubian ibex, Capra ibex nubia). One explanation is that the impact of predation from V. cana may be sufficiently low that the cost of avoidance, in terms of missed feeding opportunities, conveys little selective advantage. Alternatively, fecal odor may not provide a focused cue of immediate danger for spiny mice. The diurnal A. russatus showed a stronger (near significant) response than the nocturnal A. cahirinus to fecal odor of this nocturnal predator. This may be a legacy of the underlying nocturnal activity rhythm of A. russatus or may indicate a generally more cautious response to predator odors, as A. russatus has a much stronger preference for sheltered microhabitats than A. cahirinus.     

Change in Behavioral Response to Herbivore-induced Plant Volatiles in a Predatory Mite Population

Damage by herbivorous spider mites induces plants to produce volatiles that attract predatory mites that consume the spider mites. A clear attraction to volatiles from Lima bean plants infested with the spider mite Tetranychus urticae has been consistently reported during more than 15 years for the predatory mite Phytoseiulus persimilis. We have monitored the response to volatiles from spider-mite infested Lima bean plants for a laboratory population of the predatory mite from 1991 to 1995 on a regular basis. A reduction in the level of attraction in the laboratory population of P. persimilis was recorded in mid-1992. The attraction of the laboratory population was weaker than that of a commercial population in the latter part of 1992, but the responses of these two populations were similarly weak in 1994 and 1995. Therefore, a behavioral change has also occurred in this commercial population. Experiments were carried out to address the potential causes of this change in attraction. The attraction of predators from a commercial population with a strong response decreased after being reared in our laboratory. Within a predator population with a low degree of attraction, strongly responding predators were present and they could be isolated on the basis of their behavior: predators that stayed on spider-mite infested plants in the rearing set-up had a strong attraction, while predators that had dispersed from the rearing set-up were not attracted to prey-infested bean plants. From our laboratory population with a low degree of attraction, isofemale lines were initiated and maintained for more than 20 generations. All isofemale lines exhibited a consistently strong attraction to spider mite-induced plant volatiles, similar to the attraction recorded for several populations in the past 15 years. Neither in a population with a strong attraction nor in two with a weak attraction was the response of the predators affected by a starvation period of 1–3 hr. Based on these results, possible causes for the observed reduction in predator attraction to spider mite-induced bean volatiles are discussed. The predatory mite P. persimilis is a cornerstone of biological control in many crops worldwide. Therefore, the change in foraging behavior recorded in this predator may have serious consequences for biological control of spider mites.