Earticle

In order to improve the global searching ability of differential evolution algorithm in solving complex optimization problem, an improved differential evolution (SMDE) algorithm based on the self-adaptive method and multi-population is proposed in this paper. In the proposed SMDE algorithm, the population is divided into multi-populations in order to keep the diversity, then the self-adaptive method is used to control the parameters of differential evolution algorithm in order to balance the local search and global search ability. Finally, several complex benchmark functions are selected to validate the efficiency of the SMDE algorithm. The experiment results show that the proposed SMDE algorithm is better at the global convergence ability and the searching precision.

For the shortcomings of differential evolution algorithm(DE), such as the low convergence rate in the late evolution, easy to trap into the local optimal solution, and weak situation of the global search ability and the stability of optimization, an improved differential evolution algorithm based on multi-population and dynamic local search(MPDLSDE) is proposed in this paper. In the MPDLSDE algorithm, different populations select different mutation operation model in order to obtain superiority reciprocity between different models in the process of evolution. And the random selected method and small probability perturbation are used to increase the diversity of population and balance the exploitation ability and exploration ability of the algorithm. Then dynamic local search method is used to solve the current optimal solution in order to speed up the convergence rate. Several well-known benchmark functions are selected to validate the efficiency of the MPDLSDE algorithm. The simulation experiment and comparative analysis results show that the MPDLSDE algorithm can enhance the global convergence ability and get the high accuracy solution in high dimensional complex optimization problems.

DNA microarray technique can detect tens of thousands of genes activity in cells and has been widely used in clinical diagnosis. However, microarray data has the characteristics of high dimension and small samples, moreover many irrelevant and redundant genes also decrease performance of classification algorithm. Feature gene selection is an effective method to solve this problem. This paper proposes a hybrid feature gene selection method. Firstly, a lot of irrelevant genes from original data were eliminated by using reliefF algorithm, and the candidate feature genes subset is obtained; Secondly, Fuzzy neighborhood rough set with information entropy which deals directly with continuous data is proposed to reduce redundant genes among genes subset above. Here, differential evolution algorithm is used to optimize radius before reduction by using fuzzy neighborhood rough set, because radius of neighborhood greatly affects reduction performance. The simulation results on six microarray datasets indicate that our method can obtain higher classification accuracy by using as few genes as possible, especially feature genes selected are important for understanding microarray data and identifying the pathogenic genes. The results demonstrated that this method is effective and efficient for feature genes selection.

The joint replenishment problem with deterministic resource restriction is considered. We present a differential evolution (DE) algorithm that uses indirect grouping strategy to solve constrained joint replenishment. The procedure and structure of the DE algorithm is proposed. Extensive computational experiments are performed to compare the performances of the DE algorithm with results of genetic algorithm (GA) and heuristic algorithm CRAND. The experimental results indicate that the DE algorithm performs relative to CRAND and superior to GA.

To solve the slow convergence speed, low precision in later period and tedious parameter setting of differential evolution when applied to complex optimization functions, an improved differential evolution algorithm (dn-DADE) based on dynamic adaptive strategy is proposed. Firstly, the elite solutions of current population are utilized in the new mutation strategy (DE/current-to-dnbest/1) to guide the search direction, and then these optional elite solutions tend to the global optimal solution in the late stage of evolution to balance the diversity of population and convergence speed. Secondly, the adaptive update strategies of scaling factor and crossover factor are designed for control parameter values self-adapting at different search stages, thus improve the stability and robustness of the algorithm. A set of 14 benchmark functions is adopted to test the performance of the proposed algorithm. The results show that dn-DADE algorithm has the advantages of remarkable optimizing ability, higher search precision, faster convergence speed and outperforms several state-of-the-art improved differential evolution algorithms in terms of the main performance indexes.

In order to improve the weak situation of the global search ability, the stability and time consuming of optimization of differential evolution(DE) algorithm in solving high dimensional optimization problem, an improved differential evolution algorithm with multi- population and multi-strategy(MPMSIDE) is proposed to solve high dimensional optimization problem. Firstly, the different DE mutation strategies are studied. Then the MPMSIDE algorithm divides the population into several sub-populations, which evolve independently and communicate with each other at regular intervals by using different DE strategies, in order to save the computation time. And the improved mutation strategy and local optimization strategy are introduced to raise and balance the global searching ability and local searching ability, and improve the optimization efficiency. The selfadaptive update strategy is used to adjust the scaling factor and crossover factor for making the parameter sensitivity of DE algorithm and improving the stability and robustness. Finally, the proposed MPMSIDE algorithm is applied to standard test function optimization for verifying the effectiveness. The experimental results show that the proposed MPMSIDE algorithm has a relatively better optimization performance for solving complex optimization problem, and takes on remarkable optimizing ability, higher searching accuracy and faster convergence speed.

In order to improve the problem of premature convergence and computational efficiency of traditional differential evolution algorithm in solving high-dimensional problems, an improved differential evolution (HMSDE) algorithm based on combing elite synergy strategy, multi-population strategy and dynamic adaptive strategy is proposed in this paper. In the proposed HMSDE algorithm, the population is dynamically divided into multi-populations in order to keep the diversity of the population, elite synergy strategy is used to achieve information exchange among different sub-populations, and dynamic adaptive strategy is used to dynamically control the parameter values of scaling factor and crossover factor in order to improve the stability and robustness of the HMSDE algorithm. In order to test the performance of the HMSDE algorithm, a set of 10 benchmark functions are selected in here. The results show that the HMSDE algorithm takes on remarkable optimized ability, faster convergence speed and higher search accuracy. And the HMSDE algorithm can avoid the premature convergence and outperforms several state-of-the-art performances.

In this paper we propose a novel differential evolution algorithm based on self-learning, in order to improve the environment adaptive ability of the population in dynamic optimization. The proposed algorithm can monitor the environment changes using re-evaluation of individuals. We direct the population evolution based on the current best individual and another two random individuals, so that the convergence speed is faster and the diversity of the population is maintained. In this way we may reduce the influence from the frequent environment changes. Testing on six dynamic functions, we study the influences caused by period and dimensions. We also compared the proposed algorithm with existing algorithms, the experimental results show that our algorithm has a better environment adaptive ability and achieves better optimization result.

The optimal test sequence design for fault diagnosis is a challenge NP-complete problem. An improved Differential Evolution algorithm with additional inertial weighting item (inertial velocity) is proposed to solve the Optimal Test sequence Problem (OTP) in complicated electronic system. The proposed algorithm called Inertial Velocity Differential Evolution (IVDE) is constructed based on an adaptive differential evolution algorithm. IVDE combined with a new individual fitness function optimizes the test sequence sets with the index of fault isolation rate satisfied in top-down to generate diagnostic decision tree to decrease the test cost and the number of tests used. The simulation results show that IVDE algorithm can cut down the test cost under the satisfied fault isolation rate requirement. Compared with the other algorithm such as PSO (particle swarm optimizer)and GA(genetic algorithm), IVDE can get better solution of the OTP.

Synergetic neural network (SNN) is a top-down network to explain the phase transition and self-organization in non-equilibrium system. The network parameters have a crucial impact on the recognition performance of synergetic neural network. At present, there is no good way to control and adjust the network parameters. To solve these problems, an improved parameters optimization algorithm based on differential evolution algorithm is proposed and implemented in this paper. There are two main works in this paper. Firstly, a semantic analysis model based on synergetic neural network is presented. Secondly, differential evolution algorithm is used to search the global optimum of network parameters in the corresponding parameter space. The experiments showed that the optimization algorithm can improve the synergetic recognition performance.

In allusion to such problems as low accuracy and long convergence time in traditional two-echelon vehicle routing optimization algorithm, a Memetic algorithm (QDEMA) based on Q learning theory and differential evolution is proposed in this article to solve above problems. Firstly, it is necessary to research the two-echelon vehicle routing optimization problem and adopt the optimal segmentation method to obtain the relatively reasonable distribution plan for the first-echelon SDVRP problem in order to accordingly determine the distribution quantity of the transfer stations; secondly, the second-echelon MDVRP distribution scheme is solved to obtain the total distance and the total number of the distribution vehicles for the two-echelon optimization problem; thirdly, in allusion to the solution of the second-echelon MDVRP distribution scheme, Q learning theory and the differential evaluation algorithm are adopted to design new Memetic algorithm in order to globally optimize MDVRP distribution scheme; finally, the simulation experiment is carried out to verify the algorithm effectiveness.

A novel hybrid Bat Algorithm (BA) with the Differential Evolution (DE) strategy using the feasibility-based rules, namely BADE is proposed to deal with the constrained optimization problems. The sound interferences induced by other things are inevitable for the bats which rely on the echolocation to detect and localize the things. Through integration of the DE strategy with BA, the insects’ interferences for the bats can be effectively mimicked by BADE. Moreover, the bats swarm’ mean velocity is simulated as the other bats’ effects on each bat. Having considered the living environments the bats inhabit, the virtual bats can be lifelike. Experiments on some benchmark problems and engineering designs demonstrate that BADE performs more efficient, accurate, and robust than the original BA, DE, and some other optimization methods.

Localization as a key problem of wireless sensor network technology, has been the subject of widespread attention. This paper presents an improved positioning algorithm of wireless sensor network based on differential evolution(DE). Firstly, the algorithm improves on each jump distance measurement method in the traditional DV-hop algorithm, thereby reducing distance estimation error between anchor nodes and the unknown nodes, and then through the DE algorithm to obtain better positioning accuracy. Simulation results show that improved positioning algorithm has improved significantly in positioning accuracy compared to the previous DE positioning algorithm and least-squares algorithm, under different error factor and the density of anchor nodes. The algorithm has features of good robustness, global search capability and the ability to inhibit the accumulation of errors, suitable for a variety of applications in wireless sensor networks localization.

To solve the "soft field" effect and the ill-posed problem in electrical capacitance tomography technology, a novel image reconstruction algorithm based on population entropy and adaptive differential evolution for Electrical Capacitance Tomography is proposed in this study. The algorithm uses all the gray pixels as the initial population’s individual. After finite iterations, the algorithm mutates and makes crossover of the population in order to obtain the optimal species populations. That is the optimal value for the ECT imaging pixels. The population entropy and the variation factor make the range of each searching generation decreasing. In the simulation, the improved adaptive differential evolution algorithm will be compared with the LBP algorithm. The result shows that the new algorithm has better image quality and more stable boundary than the LBP Algorithm, which provides a new way to reconstruct images for ECT.

Job shop scheduling is well-known as one of the hardest combinatorial optimization problems and has been demonstrated to be NP-hard problem. In the past decades, several researchers have devoted their effort to develop evolutionary algorithms such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) for job shop scheduling problem. Differential Evolution (DE) algorithm is a more recent evolutionary algorithm which has been widely applied and shown its strength in many application areas. However, the applications of DE on scheduling problems are still limited. This paper proposes a one-stage differential evolution algorithm (1ST-DE) for job shop scheduling problem. The proposed algorithm employs random key representation and permutation of m-job repetition to generate active schedules. The performance of proposed method is evaluated on a set of benchmark problems and compared with results from an existing PSO algorithm. The numerical results demonstrated that the proposed algorithm is able to provide good solutions especially for the large size problems with relatively fast computing time.

Optimal Power dispatch is the short-term decision of the optimal output of a number of power generation facilities, to meet the system demand, with the objective of Power dispatching at the lowest possible cost, subject to transmission lines power loss and operational constraints. The operational constraint includes power balance constraint, generator limit constraint, and emission dispatch constraint and valve point effects. In this paper, Opposition based Differential Evolution Algorithm (ODEA) has been proposed to handle the objective function and the operational constraints simultaneously. Furthermore, the valve point loading effects and transmission lines power loss are also considered for the efficient and effective Power dispatch. The ODEA has unique features such as self tuning of its control parameters, self acceleration and migration for searching. As a result, it requires very minimum executions compared with other searching strategies. The effectiveness of the algorithm has been validated through four standard test cases and compared with previous studies. The proposed method out performs the previous methods.

Distribution system is a critical link between customer and utility. The control of power loss is the main factor which decides the performance of the distribution system. There are two methods such as (i) distribution system reconfiguration and (ii) inclusion of capacitor banks, used for controlling the real power loss. Considering the improvement in voltage profile with the power loss reduction, later method produces better performance than former method. This paper presents an advanced evolutionary algorithm for capacitor inclusion for loss reduction. The conventional sensitivity analysis is used to find the optimal location for the capacitors. In order to achieve a better approximation for the current candidate solution, Opposition based Differential Evolution (ODE) is introduced. The effectiveness of the proposed technique is validated through 10, 33, 34 and85-bus radial distribution systems.

Generation Expansion Planning (GEP) is one of the most important decision-making activities in electric utilities. Least-cost GEP is to determine the minimum-cost capacity addition plan (i.e., the type and number of candidate plants) that meets forecasted demand within a pre specified reliability criterion over a planning horizon. In this paper, Differential Evolution (DE), and Opposition-based Differential Evolution (ODE) algorithms have been applied to the GEP problem. The original GEP problem has been modified by incorporating Virtual Mapping Procedure (VMP). The GEP problem of a synthetic test systems for 6-year, 14-year and 24-year planning horizons having five types of candidate units have been considered. The results have been compared with Dynamic Programming (DP) method. The ODE performs well and converges faster than DE.

A sliding mode control (SMC) for servo motors based on the differential evolution (DE) algorithm, called DE-SMC, is proposed in this study. The parameters of SMC should be designed exactly to improve the robustness, realize the precision positioning, and reduce the steady-state speed error of the servo drive. The main parameters of SMC are optimized using the DE algorithm according to the speed feedback information of the servo motor. The most significant influence factor of the DE algorithm is optimization iteration. A suitable iteration can be achieved by the tested optimization process profile of the main parameters of SMC. Once the parameters of SMC are optimized under a convergent iteration, the system realizes the given performance indices within the shortest time. The experiment indicates that the robustness of the system is improved, and the dynamic and steady performance achieves the given performance indices under a convergent iteration when motor parameters mismatch and load disturbance is added. Moreover, the suitable iteration effectively mitigates the low-speed crawling phenomenon in the system. The correctness and effectiveness of DE-SMC are verified through the experiment.

In this paper, we present a new morphology-based homomorphic filtering technique for feature enhancement in medical images. The proposed method is based on decomposing an image into morphological subbands. The homomorphic filtering is performed using the morphological subbands. The differential evolution algorithm is applied to find an optimal gain and structuring element for each subband. Simulations show that the proposed filter improves the contrast of the features in medical images.