Hybrid Differential Evolution Algorithm Research Articles

A new hybrid differential evolution algorithm with self-adaptation for function optimization

In this study, a new hybrid algorithm, hDEBSA, is proposed with the aid of two evolutionary algorithms, Differential Evolution (DE) and Backtracking Search Optimization Algorithm (BSA). The control parameters of both algorithms are simultaneously considered as a self-adaptation basis such that the values of the parameters update automatically during the optimization process to improve performance and convergence speed. To validate the proposed algorithm, twenty-eight CEC2013 test functions are considered. The performance results of hDEBSA are validated by comparing them with several state-of-the-art algorithms that are available in literature. Finally, hDEBSA is applied to solve four real-world optimization problems, and the results are compared with the other algorithms, where it was found that the hDEBSA performance is better than that of the other algorithms.

Evolving Spatial Clusters of Genomic Regions From High-Throughput Chromatin Conformation Capture Data.

High-throughput chromosome-conformation-capture (Hi-C) methods have revealed a multitude of structural insights into interphase chromosomes. In this paper, we elucidate the spatial clusters of genomic regions from Hi-C contact maps by formulating the underlying problem as a global optimization problem. Given its nonconvex objective and nonnegativity constraints, we implement several evolutionary algorithms and compare their performance with non-negative matrix factorization, revealing novel insights into the problem. In order to obtain robust and accurate spatial clusters, we propose and describe a novel hybrid differential evolution algorithm called HiCDE, which adopts non-negative matrix factorization as local search according to each candidate individual provided by differential evolution algorithm. Based on the fitness value of each individual, the population is partitioned into three subpopulations with different sizes; each subpopulation is equipped with a specific mutation strategy for either exploitation or exploration. Finally, all control parameters in the pool have equal probability to be selected for generating trial vectors. The effectiveness and robustness of HiCDE are supported by real-world performance benchmarking on chromosome-wide Hi-C contact maps of yeast and human, time complexity analysis, convergence analysis, parameter analysis, and case studies.

A hybrid differential evolution and particle swarm optimization algorithm for numerical kinematics solution of remote maintenance manipulators

In the ITER or the future DEMO reactor systems, due to the neutron activation, the remote handling tasks such as inspection, repair and/or maintenance of in-vessel and ex-vessel components must be carried out using a wide variety of special tailored automatic manipulators. These manipulators can be designed as a pure serial articulated arm or a pure parallel mechanism, but for the sophisticated remote handling tasks requiring large workspace and high payload/weight ratio, it would be designed as a hybrid structure combined by these two mechanisms. The complex structure to obtain special remote operation makes the maintenance very time-consuming. To obtain the forward and/or the inverse kinematics of these manipulators, a global optimization method combined by differential evolution (DE) and modified particle swarm optimization (MPSO), here we call it as DEMPSO, is developed to save solution time. The general case studies focus on serial, parallel manipulators and the 10-DOF hybrid redundant serial-parallel robot designed for ITER vacuum vessel remote maintenance. The results would be extrapolated to solve the kinematic problems of the ITER or the future DEMO remote handling manipulators. The proposed optimization algorithm can also be used for the static and/or dynamic parameter identification in robot calibration systems.

Scheduling with job-splitting considering learning and the vital-few law

This research, which is motivated by real cases in labor-intensive industries where learning effects and the vital-few law take place, integrates learning and job splitting in parallel machine scheduling problems to minimize the makespan. We propose the lower bound of the problem and a job-splitting algorithm corresponding to the lower bound. Subsequently, a heuristic called SLMR is proposed based on the job-splitting algorithm with a proven worst case ratio. Furthermore, a branch-and-bound algorithm, which can obtain optimal solutions for very small problems, and a hybrid differential evolution algorithm are proposed, which can not only solve the problem, but also serve as a benchmark to evaluate the solution quality of the heuristic SLMR. The performance of the heuristic on a large number of randomly generated instances is evaluated. Results show that the proposed heuristic has good solution quality and calculation efficiency.

Application of complete ensemble intrinsic time-scale decomposition and least-square SVM optimized using hybrid DE and PSO to fault diagnosis of diesel engines

Targeting the mode-mixing problem of intrinsic time-scale decomposition (ITD) and the parameter optimization problem of least-square support vector machine (LSSVM), we propose a novel approach based on complete ensemble intrinsic time-scale decomposition (CEITD) and LSSVM optimized by the hybrid differential evolution and particle swarm optimization (HDEPSO) algorithm for the identification of the fault in a diesel engine. The approach consists mainly of three stages. First, to solve the mode-mixing problem of ITD, a novel CEITD method is proposed. Then the CEITD method is used to decompose the nonstationary vibration signal into a set of stationary proper rotation components (PRCs) and a residual signal. Second, three typical types of time-frequency features, namely singular values, PRCs energy and energy entropy, and AR model parameters, are extracted from the first several PRCs and used as the fault feature vectors. Finally, a HDEPSO algorithm is proposed for the parameter optimization of LSSVM, and the fault diagnosis results can be obtained by inputting the fault feature vectors into the HDEPSO-LSSVM classifier. Simulation and experimental results demonstrate that the proposed fault diagnosis approach can overcome the mode-mixing problem of ITD and accurately identify the fault patterns of diesel engines.

Synthesis of Some Low Sidelobe Linear Arrays Using Hybrid Differential Evolution Algorithm Integrated With Convex Programming

In this letter, some low sidelobe linear arrays are synthesized by a hybrid differential evolution (DE) algorithm integrated with convex programming. In the hybrid algorithm, the DE algorithm is used to optimize the array layouts only. Once the array layouts are fixed, the subproblems for minimizing the peak sidelobe level of the array pattern with constraints on its beamwidth become convex. A readily available MATLAB toolbox CVX is then used to solve the convex subproblems. Based on the optimal solution of each convex subproblem, an appropriate fitness function is defined in the DE algorithm. Some low sidelobe linear array synthesis examples are given to demonstrate the effectiveness of the proposed algorithm.

Hybrid real-code population-based incremental learning and differential evolution for many-objective optimisation of an automotive floor-frame

In this paper, a many-objective hybrid real-code population-based incremental learning and differential evolution algorithm (MnRPBILDE) is proposed based on the concept of objective function space reduction. The method is then implemented on real engineering design problems. The topology, shape and sizing design of a simplified automotive floor-frame structure are formulated and used as test problems. A variety of well-established multi-objective evolutionary algorithms (MOEAs) including the original version of MnRPBILDE are employed to solve the test problems while the results are compared based on hypervolume and C indicators. The results indicate that our proposed algorithm outperforms the other MOEAs. The proposed algorithm is effective and efficient for many-objective optimisations of a car floor-frame structure.

Hybrid Differential Evolution Optimisation for Earth Observation Satellite Scheduling with Time-Dependent Earliness-Tardiness Penalties

We study the order acceptance and scheduling (OAS) problem with time-dependent earliness-tardiness penalties in a single agile earth observation satellite environment where orders are defined by their release dates, available processing time windows ranging from earliest start date to deadline, processing times, due dates, sequence-dependent setup times, and revenues. The objective is to maximise total revenue, where the revenue from an order is a piecewise linear function of its earliness and tardiness with reference to its due date. We formulate this problem as a mixed integer linear programming model and develop a novel hybrid differential evolution (DE) algorithm under self-adaptation framework to solve this problem. Compared with classical DE, hybrid DE employs two mutation operations, scaling factor adaptation and crossover probability adaptation. Computational tests indicate that the proposed algorithm outperforms classical DE in addition to two other variants of DE.

A hybrid differential evolution algorithm with estimation of distribution algorithm for reentrant hybrid flow shop scheduling problem

This paper proposes a reentrant hybrid flow shop scheduling problem where inspection and repair operations are carried out as soon as a layer has completed fabrication. Firstly, a scheduling problem domain of reentrant hybrid flow shop is described, and then, a mathematical programming model is constructed with an objective of minimizing total weighted completion time. Then, a hybrid differential evolution (DE) algorithm with estimation of distribution algorithm using an ensemble model (eEDA), named DE–eEDA, is proposed to solve the problem. DE–eEDA incorporates the global statistical information collected from an ensemble probability model into DE. Finally, simulation experiments of different problem scales are carried out to analyze the proposed algorithm. Results indicate that the proposed algorithm can obtain satisfactory solutions within a short time.

Fuzzy Decision Making Approach to Identify Optimum Enzyme Targets and Drug Dosage for Remedying Presynaptic Dopamine Deficiency.

Model-based optimization approaches are valuable in developing new drugs for human metabolic disorders. The core objective in most optimal drug designs is positive therapeutic effects. In this study, we considered the effects of therapeutic, adverse, and target variation simultaneously. A fuzzy optimization method was applied to formulate a multiobjective drug design problem for detecting enzyme targets in the presynaptic dopamine metabolic network to remedy two types of enzymopathies caused by deficiencies of vesicular monoamine transporter 2 (VMAT2) and tyrosine hydroxylase (TH). The fuzzy membership approach transforms a two-stage drug discovery problem into a unified decision-making problem. We developed a nested hybrid differential evolution algorithm to efficiently identify a set of potential drug targets. Furthermore, we also simulated the effects of current clinical drugs for Parkinson’s disease (PD) in this model and tried to clarify the possible causes of neurotoxic and neuroprotective effects. The optimal drug design could yield 100% satisfaction grade when both therapeutic effect and the number of targets were considered in the objective. This scenario required regulating one to three and one or two enzyme targets for 50%–95% and 50%–100% VMAT2 and TH deficiencies, respectively. However, their corresponding adverse and target variation effect grades were less satisfactory. For the most severe deficiencies of VMAT2 and TH, a compromise design could be obtained when the effects of therapeutic, adverse, and target variation were simultaneously applied to the optimal drug discovery problem. Such a trade-off design followed the no free lunch theorem for optimization; that is, a more serious dopamine deficiency required more enzyme targets and lower satisfaction grade. In addition, the therapeutic effects of current clinical medications for PD could be enhanced in combination with new enzyme targets. The increase of toxic metabolites after treatment might be the cause of neurotoxic effects of some current PD medications.

An interactive possibilistic programming approach for a multi-objective hub location problem: Economic and environmental design

This paper presents a new multi-objective mathematical model for a multi-modal hub location problem under a possibilistic-stochastic uncertainty. The presented model aims to minimize the total transportation and traffic noise pollution costs. Furthermore, it aims to minimize the maximum transportation time between origin-destination nodes to ensure a high probability of meeting the service guarantee. In order to cope with the uncertainties and the multi-objective model, we propose a two-phase approach, including fuzzy interactive multi-objective programming approach and an efficient method based on the Me measure. Due to the NP-hardness of the presented model, two meta-heuristic algorithms, namely hybrid differential evolution and hybrid imperialist competitive algorithm, are developed. Furthermore, a number of sensitivity analyses are provided to demonstrate the effectiveness of the presented model. Finally, the foregoing meta-heuristics are compared together through different comparison metrics.

Hybrid differential evolution algorithm combined with heuristic correction and chaotic search for online energy-efficient optimization of server cluster

Server cluster has been widely used to improve the performance of various servers. How to optimize the deployment of each server in a cluster so as to minimize cluster’s power consumption while satisfying load capacity requirement is an urgent problem to be resolved. In this paper, we first formula te a typical energy-efficient optimization problem of server cluster as a constrained MIP (Mixed Integer Programming) problem. Then, aimed at the traits of the problem, we propose a hybrid DE (Differential Evolution) algorithm combined with heuristic correction and chaotic search to solve the problem. We introduce a correction operation to traditional DE algorithm, which bases on greedy idea to correct the individuals that do not satisfy constraints. Besides, we generate the initial population based on chaotic sequence so as to enhance population diversity, and implement chaotic search around the best individual of each generation so as to improve solving accuracy and accelerate convergence speed. The algorithm has high solving efficiency and can run online even when applied in large-scale server clusters. Simulation results verify the feasibility and effectiveness of the algorithm.

A hybrid differential evolution for optimal multilevel image thresholding

Image thresholding is a process for separating interesting objects within an image from their background. An optimal threshold’s selection can be regarded as a single objective optimization problem, where obtaining a solution can be computationally expensive and time-consuming, especially when the number of thresholds increases greatly. This paper proposes a novel hybrid differential evolution algorithm for selecting the optimal threshold values for a given gray-level input image, using the criterion defined by Otsu. The hybridization is done by adding a reset strategy, adopted from the Cuckoo Search, within the evolutionary loop of differential evolution. Additionally a study of different evolutionary or swarm-based intelligence algorithms for the purpose of thresholding, with a higher number of thresholds was performed, since many real-world applications require more than just a few thresholds for further processing. Experiments were performed on eleven real world images. The efficiency of the hybrid was compared to the cuckoo search and self-adaptive differential evolution, the original differential evolution, particle swarm optimization, and artificial bee colony where the results showed the superiority of the hybrid in terms of better segmentation results with the increased number of thresholds. Since the proposed method needs only two parameters adjusted, it is by far a better choice for real-life applications.

Optimal operation of microgrid using hybrid differential evolution and harmony search algorithm

This paper proposes the generation scheduling approach for a microgrid comprised of conventional generators, wind energy generators, solar photovoltaic (PV) systems, battery storage, and electric vehicles. The electrical vehicles (EVs) play two different roles: as load demands during charging, and as storage units to supply energy to remaining load demands in the MG when they are plugged into the microgrid (MG). Wind and solar PV powers are intermittent in nature; hence by including the battery storage and EVs, the MG becomes more stable. Here, the total cost objective is minimized considering the cost of conventional generators, wind generators, solar PV systems and EVs. The proposed optimal scheduling problem is solved using the hybrid differential evolution and harmony search (hybrid DE-HS) algorithm including the wind energy generators and solar PV system along with the battery storage and EVs. Moreover, it requires the least investment.

A novel hybrid differential evolution algorithm with modified CoDE and JADE

JADE and CoDE are two well-known state-of-the-art DE algorithms for solving global optimization problems (GOPs). JADE is found to be suitable for solving unimodal and simple multimodal functions as an exploitation mutation strategy, i.e.“DE/current-to-pbest/1”, is employed, while CoDE is shown to fit for handling complicated multimodal functions due to its exploration mutation strategies, such as “DE/rand/1/bin”, “DE/currant-to-rand/1”, and “DE/rand/2/bin”. To combine their merits for tackling different types of GOPs, a novel hybrid framework is designed based on the modified JADE (MJADE) and modified CoDE (MCoDE), named HMJCDE. Different from the simple combination of MJADE and MCoDE, they are operated alternatively according to the improvement rate of the fitness value. To assess the performance of HMJCDE, 30 benchmark problems taken from CEC2014 competition on real parameter optimization are employed. When compared with JADE, CoDE, other state-of-the-art DE variants and non-DE heuristic algorithms, HMJCDE performs better than all the competitors on most of test problems. Moreover, the sensitivity analysis of some parameters in HMJCDE is conducted and the effectiveness of our proposed hybrid framework is also justified experimentally.

Option pricing and hedging for optimized Lévy driven stochastic volatility models

This paper pays attention to Ornstein-Uhlenbeck (OU) based stochastic volatility models with marginal law given by Classical Tempered Stable (CTS) distribution and Normal Inverse Gaussian (NIG) distribution, which are subclasses of infinite activity Lévy processes and are compared to finite activity Barndorff-Nielsen and Shephard (BNS) model. They are applied to option pricing and hedging in capturing leptokurtic features in asset returns and clustering effect in volatility that are consistently observed phenomena in underlying asset dynamics. The analytical formula of option pricing can be obtained through use of characteristic functions and Fast Fourier Transform (FFT) technique. Additionally, we introduce two hybrid optimization techniques such as hybrid Particle Swarm optimization (PSO) algorithm and hybrid Differential Evolution (DE) algorithm into parameters calibration schemes to improve the calibration quality for newly constructed models. Finally, we conduct experiments on Chinese emerging option markets to examine the performance of proposed models exploiting hybrid optimization techniques.

Lifetime Improvement in Wireless Sensor Networks using Hybrid Differential Evolution and Simulated Annealing (DESA)

Abstract The major concerns in Wireless Sensor Networks (WSN) are energy efficiency as they utilize small sized batteries, which can neither be replaced nor be recharged. Hence, the energy must be optimally utilized in such battery operated networks. One of the traditional approaches to improve the energy efficiency is through clustering. In this paper, a hybrid differential evolution and simulated annealing (DESA) algorithm for clustering and choice of cluster heads is proposed. As cluster heads are usually overloaded with high number of sensor nodes, it tends to rapid death of nodes due to improper election of cluster heads. Hence, this paper aimed at prolonging the network lifetime of the network by preventing earlier death of cluster heads. The proposed DESA reduces the number of dead nodes than Low Energy Adaptive Clustering Hierarchy (LEACH) by 70%, Harmony Search Algorithm (HSA) by 50%, modified HSA by 40% and differential evolution by 60%.

A Hybrid Differential Evolution Algorithm Based on Dynamic Variable Neighborhood Search for Permutation Flowshop Scheduling Problem

Permutation flowshop scheduling problem (PFSP) is a classical NP-hard combinatorial optimization problem, which provides a challenge for evolutionary algorithms.Since it has been shown that simple evolutionary algorithms cannot solve the PFSP efficiently, local search methods are often adopted to improve the exploitation ability of evolutionary algorithms. In this paper, a hybrid differential evolution algorithm is developed to solve this problem. This hybrid algorithm is designed by incorporating a dynamic variable neighborhood search (DVNS) into the differential evolution. In the DVNS, the neighborhood is based on multiple moves and its size can be dynamically changed from small to large so as to obtain a balance between exploitation and exploration. In addition, a population monitoring and adjusting mechanism is also incorporated to enhance the search diversity and avoid being trapped in local optimum.Experimental results on benchmark problems illustrated the efficiency of the proposed algorithm.

Multiobjective multiarea unit commitment using hybrid differential evolution algorithm considering import/export and tie-line constraints

Recent power system networks are characterized by large proportions, high interconnections, and high nonlinearities. Challenge of supplying the nation with high-quality, reliable electrical energy at a reasonable cost converted government policy into deregulation and restructuring environment. To achieve significant cost-savings, multiarea unit commitment strategies are employed, whose intention is to establish the optimal commitment stratagem for power generating units situated in numerous areas which are interconnected through tie-lines, and combined operation of generation resources can result in considerable operational cost-savings. Differential evolution is a population-based stochastic search algorithm, whose simple yet powerful and straightforward features make it gorgeous for optimization. Differential evolution uses somewhat greedy and less stochastic approach for optimization problem solution. Although global exploration ability of differential evolution (DE) algorithm is adequate, its local exploitation ability is feeble and convergence velocity is too low, and it suffers from the problem of untimely convergence for multimodal objective function, in which search process may be trapped in local optima and it loses its diversity. Also, it suffers from the stagnation problem, where the search process may infrequently stop proceeding toward the global optimum even though the population has not converged to a local optimum or any other point. To improve the exploitation ability and global performance of DE algorithm, a novel and hybrid version of differential evolution algorithm combined with random search algorithm is presented in the proposed research to solve multiobjective and multiarea unit commitment problem of electric power system. The performance of the proposed hybrid algorithm is tested with benchmark of three-area interconnected system, which consist of IEEE-30 Bus system. Experimental results show that proposed technique has the prospective for the solution of multiobjective and multiarea unit commitment problem and power generation scheduling in deregulated electricity market with import and export constraints.

A hybrid differential evolution algorithm approach towards assisted history matching and uncertainty quantification for reservoir models

Abstract History matching is an important process in the reservoir model development. In the process of history matching, the most significant uncertain model parameters are identified and adjusted to get an acceptable match between the simulated production with the historical field production data. In the past decade, many population based algorithms have been applied for history matching. In this paper, a novel population based stochastic algorithm called hybrid differential evolution (HDE) is applied for the assisted history matching process. An adaptive mechanism for the control parameters is incorporated in the algorithm which automatically adjusts the control parameters according to the problem. The performance of the algorithm is tested on a 3-D reservoir model called PUNQ-S3 which is a benchmark model for the comparison of different history matching and uncertainty quantification techniques. Since history matching is an inverse problem, multiple models can give good match. So, prediction using a single history matched model involves more risk because of the parameter uncertainty. One of the methods to solve this problem is to quantify the uncertainty in the predictions. In this paper, the neighbourhood approximation Bayes (NAB) algorithm is applied to quantify the uncertainty in reservoir forecast which is a Bayesian extension of neighbourhood algorithm. The NAB algorithm quantifies the uncertainty in the predictions using multiple models generated during history matching phase and this does not require additional simulations. The main focus of this paper is to study about how HDE algorithm can be used when coupling with the NAB algorithm in predicting the true forecast with minimum uncertainty range under limited number of simulations. The influence of population size on the performance of the algorithm in history matching and forecast is analyzed. The HDE provides wide sampling of the search space and the truth case was comfortably included within the predicted confidence bounds. The results show that HDE can be used as a promising tool for assisted history matching of the reservoir models.