Self-adaptive extreme learning machine

In order to overcome the disadvantage of the traditional algorithm for SLFN (single-hidden layer feedforward neural network), an improved algorithm for SLFN, called extreme learning machine (ELM), is proposed by Huang et al. However, ELM is sensitive to the neuron number in hidden layer and its selection is a difficult-to-solve problem. In this paper, a self-adaptive mechanism is introduced into the ELM. Herein, a new variant of ELM, called self-adaptive extreme learning machine (SaELM), is proposed. SaELM is a self-adaptive learning algorithm that can always select the best neuron number in hidden layer to form the neural networks. There is no need to adjust any parameters in the training process. In order to prove the performance of the SaELM, it is used to solve the Italian wine and iris classification problems. Through the comparisons between SaELM and the traditional back propagation, basic ELM and general regression neural network, the results have proven that SaELM has a faster learning speed and better generalization performance when solving the classification problem.     

高深宽比金属光栅制备及中红外波段传感特性

利用纳米压印结合溅射和反应离子刻蚀工艺制备了具有高深宽比的金光栅,使用傅里叶变换红外光谱仪测得了反射谱线.测量结果显示,只在p偏振光垂直于光栅矢量方向入射条件下才存在共振反射峰,证明了“伪表面等离子体激元波”的存在.基于严格耦合波分析理论计算了金属光栅的反射率,研究了其作为中红外波段波长调制型表面等离子体共振传感器的可行性.数值计算表明负级次衍射光波对应的共振反射峰的移动能获得较高的波长灵敏度.对于深宽比为10的金光栅结构,+1级次和-3级次衍射光波对应的波长灵敏度分别为1600 nm/RIU和5000 nm/RIU,品质因子分别为20 RIU-1和60RIU-1.     

Chaotic cuckoo search

This study proposes a novel chaotic cuckoo search (CCS) optimization method by incorporating chaotic theory into cuckoo search (CS) algorithm. In CCS, chaos characteristics are combined with the CS with the intention of further enhancing its performance. Further, the elitism scheme is incorporated into CCS to preserve the best cuckoos. In CCS method, 12 chaotic maps are applied to tune the step size of the cuckoos used in the original CS method. Twenty-seven benchmark functions and an engineering case are utilized to investigate the efficiency of CCS. The results clearly demonstrate that the performance of CCS together with a suitable chaotic map is comparable as well as superior to that of the CS and other metaheuristic algorithms.     

Hybrid krill herd algorithm with differential evolution for global numerical optimization

In order to overcome the poor exploitation of the krill herd (KH) algorithm, a hybrid differential evolution KH (DEKH) method has been developed for function optimization. The improvement involves adding a new hybrid differential evolution (HDE) operator into the krill, updating process for the purpose of dealing with optimization problems more efficiently. The introduced HDE operator inspires the intensification and lets the krill perform local search within the defined region. DEKH is validated by 26 functions. From the results, the proposed methods are able to find more accurate solution than the KH and other methods. In addition, the robustness of the DEKH algorithm and the influence of the initial population size on convergence and performance are investigated by a series of experiments.     

Solving IIR system identification by a variant of particle swarm optimization

A variant of particle swarm optimization (PSO) is represented to solve the infinitive impulse response (IIR) system identification problem. Called improved PSO (IPSO), it makes significant enhancement over PSO. To begin with, the population initialization step makes use of golden ratio to segment solution space so as to obtain high-quality solutions. It is followed by all particles using different inertia weights in velocity updating step, which is beneficial for preserving the balance between global search and local search. Subsequently, IPSO uses normal distribution to disturb the global best particle, which enhances its capacity of escaping from the local optimums. The above three operations cannot only guarantee high-quality solutions, strong global search capacity, and fast convergence rate, but also avoid low diversity, excessive local search, and premature stagnation. These properties of IPSO make it much better suited for IIR system identification problems. IPSO is applied on 12 examples. The experimental results amply demonstrate the capability of IPSO toward obtaining the best objective function values in all the cases. Compared with the other four PSO approaches, IPSO has stronger convergence and higher stability which clearly points out its desirable performance in search accuracy and identifying efficiency.

     

Krill herd algorithm based on cuckoo search for solving engineering optimization problems

This paper presents a hybrid krill herd (CSKH) approach to solve structural optimization problems. CSKH improved the Krill herd algorithm (KH) by combining KU/KA operator originated from cuckoo search algorithm (CS) with KH. In CSKH, a greedy selection scheme is used and often overtakes the original KH and CS. In addition, in order to further enhance the assessment of CSKH, a fraction of the worst krill is thrown away and substituted with newly randomly generated ones by KA operator at the end of each generation. The CSKH is applied to five real engineering problems to verify its performance. The experimental results have proven that CSKH algorithm is well capable of solving constrained engineering design problems more efficiently and effectively than the basic CS and KH algorithm.

     

真空预冷毛豆冷藏期间色泽和失重率变化动力学模型构建

目的探究真空预冷毛豆贮藏过程中色泽和失重率变化规律及其动力学模型。方法将真空预冷毛豆贮藏在0、5、10℃温度条件下,研究其色泽L*、a*、b*、?E和失重率随贮藏时间的变化规律,采用动力学模型对指标进行分析,并结合Arrhenius方程建立基于色泽和失重率变化的预测模型。结果在实验条件下,贮藏温度越低,真空预冷毛豆的色泽和失重率变化速率越慢。不同贮藏温度下毛豆色泽和失重率变化符合零级动力学规律和Arrhenius方程,色泽L*、a*、?E和失重率的活化能Ea分别为27.78、49.70、30.55、54.32 kJ/mol,建立依赖于温度-时间的色泽和失重率动力学模型,验证表明其模型拟合度良好(r20.9)。结论动力学模型能够预测真空预冷毛豆贮藏期间色泽和失重率的变化,为毛豆冷藏期间色泽和失重率的变化提供理论依据。     

Solving 0–1 knapsack problem by a novel binary monarch butterfly optimization

Neural Computing and Applications - This paper presents a novel binary monarch butterfly optimization (BMBO) method, intended for addressing the 0–1 knapsack problem (0–1 KP). Two...     

A novel fruit fly framework for multi-objective shape design of tubular linear synchronous motor

This paper addresses a novel multi-objective fruit fly optimization algorithm (MOFOA) for solving multi-objective optimization problems. The essence of MOFOA lies in its having two characteristic features. For the first feature, a population of random fruit flies initializes the algorithm. During this initialization phase, the dominated fruit fly is replaced by the nearest non-dominated one. Subsequently, the fruit flies undergo evolution by flying randomly around the non-dominated solution or around the reference point, i.e., the best location of the individual objectives. Afterwards, the fruit flies are updated according to the nearest location whether from the reference point or the previous non-dominated location. For the second feature, the weighted sum method is incorporated to update the previous best locations of fruit flies and the reference point to emphasize the convergence of the non-dominated solutions. To prove the capability of the proposed MOFOA, two standard benchmark problems in addition to the real world application, namely, multi-objective shape design of tubular linear synchronous motor (TLSM) are checked. The corresponding TLSM objective functions aims to maximize operating force and to minimize the flux saturation. The outcomes clearly demonstrate the effectiveness of the proposed algorithm for finding the non-dominated solutions.