Parallel Multi-population Research Articles

РЕШЕНИЕ ЗАДАЧИ РАЗМЕЩЕНИЯ ЭЛЕМЕНТОВ ЭВА НА ОСНОВЕ ИНТЕГРАЦИИ МЕТОДОВ ЭВОЛЮЦИОННОГО ПОИСКА И НЕЧЕТКОГО УПРАВЛЕНИЯ

The problem of elements placement in digital computing equipment are consider in the article.The analysis of the current state of research on this topic is carried out, the relevance of theproblem under consideration is noted. The importance of developing new effective methods forsolving engineering design problems is emphasized. The prospects of developing and using hybrid approaches and models for solving complex semi-formalized design and optimization problems arenoted. The statement of the problem of placement of circuit elements of digital computing equipmentis given. The importance of a qualitative solution of the placement problem from the point of view ofthe successful implementation of the subsequent stages of design is noted. The analysis of variousapproaches and algorithms for solving the placement problem is carried out. Options for choosingvarious criteria for assessing the quality of placement are given. A complex additive criterion forassessing the quality of placement is proposed. The objective function and limitations of the consideredplacement problem as an optimization problem are given. A hybrid approach to solving theplacement problem is proposed. To increase the efficiency and reduce the running time of the algorithm,a model of a parallel multipopulation genetic algorithm is proposed. To synchronize evolutionaryprocesses in a multipopulation genetic algorithm, a modified migration operator has beendeveloped. An analysis of the efficiency of the proposed migration operator has been carried out andrecommendations for its use have been formulated. In order to increase the speed of the algorithm forsolving the placement problem, a model for organizing parallel evolutionary computations throughthe use of multithreading at the local level is proposed. The principles of operation of the fuzzy controlmodule are described. The procedure of logical inference using the rule base is described.The structure of a multilayer neural network that implements the Gaussian function is proposed.A model of a fuzzy logic controller for dynamically changing the values of control parameters of agenetic algorithm is proposed. The control parameters of the fuzzy logic controller are determined.The proposed hybrid algorithm is implemented as an application program. A series of computationalexperiments were carried out to determine the efficiency of the developed algorithm and to select theoptimal values of the control parameters.

Multi-block synchronous prediction of concrete dam displacements using MIMO machine learning paradigm

The integration of numerous monitoring points poses a significant challenge to the efficient modeling of dam displacement behavior, and multi-point synchronous prediction is an effective solution. However, traditional approaches usually construct site-specific data-driven models for each monitoring point individually, which focus on single-target regression and discard the underlying spatial correlation among different displacement monitoring points. This study therefore proposes a multi-input multi-output (MIMO) machine learning (ML) paradigm based on support vector machine (SVM) for synchronous modeling and prediction of multi-point displacements from various dam blocks. In this method, a novel multi-output data-driven model, termed as multi-target SVM (MSVM), is formulated through a deep hybridization of classical SVM architecture and multi-target regression. During the initialization of MSVM, the intercorrelation of multiple target variables is fully exploited by decomposing and regulating the weight vectors. The proposed MSVM is designed to capture the complex MIMO mapping from influential factors to multi-block displacements, while taking into account the correlation between multi-block displacement outputs. Additionally, in order to avoid obtaining the unreliable prediction results due to the empirical selection of parameters, an efficient optimization strategy based on the parallel multi-population Jaya (PMP-Jaya) algorithm is used to adaptively tune the hyperparameters involved in MSVM, which contains no algorithm-specific parameters and is easy to implement. The effectiveness of the proposed model is verified using monitoring data collected from a real concrete gravity dam, where its performance is compared with conventional single-target SVM (SSVM)-based models and state-of-the-art ML-based models. The results indicate that our proposed MSVM is much more promising than the SSVM-based models because only one prediction model is required, rather than constructing multiple site-specific SSVM-based models for different dam blocks. Moreover, MSVM can achieve better performance than other ML-based models in most cases, which provides an innovative modeling tool for dam multi-block behavior monitoring.

About Selecting the Number of Processors for Parallel Multipopulation Genetic Algorithm

Introduction. The paper considers some features of the parallel implementation of a multipopulation genetic algorithm, as well as approaches to its optimization. The results of experiments with the use of a different number of processors and different methods of generating initial populations are presented in order to optimize the algorithm according to several criteria (assessment of the use of computational and time resources). On the example of a specific test problem, estimates are given for choosing the optimal number of processors to obtain the desired result. The purpose of this work is to conduct experiments with a given test problem with a different number of processors and alternative methods for generating the initial population to evaluate the effectiveness of the algorithm. Results. For the test problem, to obtain a result of 90–94 % of the optimum, the most efficient in terms of computing resources is the use of 4 processors with an algorithm for uniform scanning of the space of factor values. To achieve a result exceeding 94 % and optimize by K1 (computational resources), 8 processors and an algorithm for uniform scanning of the space of factor values showed the best result. If we also take into account the criterion of time resources K2, then to achieve 90–98 % of the optimum, it is necessary to use 8 processors, for 99–100 % 12 or 16 processes, depending on C1 and C2 (cost of computational and time resources respectively). Conclusions. Performed experiments show that the algorithm of uniform scanning of the space of factor values is more efficient than the random method of generating the initial population. Experiments also showed that in order to achieve the maximum efficiency of PMGA, the number of processors must be chosen depending on the desired result precision. Keywords: parallel genetic algorithm, initial population generation, choice of the number of processors (populations), algorithm optimization.

Auto-Design of Multi-Pass Cell With Small Size and Long Optical Path Length Using Parallel Multi-Population Genetic Algorithm

In tunable diode laser absorption spectroscopy (TDLAS), the multipass cell (MPC) that comprises two identical and coaxial spherical mirrors can provide long optical path length (OPL) and reduce the limit of system detection. We proposed an intelligent method to design MPC with long optical path length (OPL) and small size using parallel multi-population genetic algorithm (PMPGA). The developed MPCs presented the maximum mirror utilization in the same size of spherical mirrors with the diameter of 25.4 mm and 40 mm. The performance of the developed MPCs were compared with other teams. We verified the results of PMPGA and obtained a micro MPC design with an OPL of 6.3 m. The diameter of the mirror was 25.4 mm, and the distance between the two mirrors was only 30.5 mm. And the developed MPC was verified by using TDLAS, which can improve the development of portable gas sensors.

Distribution State Estimation using Multiple Stages considering Asynchronous Measurement Data by Dependable Parallel Multi-population Global-best Brain Storm Optimization with Differential Evolution Strategies

Distribution State Estimation using Multiple Stages considering Asynchronous Measurement Data by Dependable Parallel Multi-population Global-best Brain Storm Optimization with Differential Evolution Strategies

Parallel multipopulation optimization for belief rule base learning

As a rule-based expert system, the belief rule base (BRB) exhibits tremendous advantages in modeling nonlinearity for complex systems. Present BRB learning studies can be classified into three categories: BRB structure learning, BRB parameter learning, and BRB joint optimization but only in an iterative and separate fashion. In this study, a novel Parallel Multipopulation optimization approach for BRB, i.e., PMP-BRB, is proposed that simultaneously optimizes the structure and parameters of a BRB. In the optimization model of PMP-BRB, the structure of BRB, i.e., the number of rules, is used as another decisive variable. In the optimization algorithm of PMP-BRB, multiple populations are initialized and subsequently optimized, with different populations representing BRBs of varied sizes. Furthermore, a “completion and deletion” strategy is proposed, wherein individual BRBs in different populations are completed with additional genes that only engage in the optimization operations but not in fitness calculations. Moreover, a trade-off analysis is conducted for decision-makers to identify the final optimal configuration of a BRB based on their preference. The proposed PMP-BRB approach is validated by four cases, namely a numerical case, two practical cases, and a case study with five classification benchmarks. Four evolutionary algorithms are tested and compared. With the structure and parameters optimized simultaneously, all the four cases yield competitive results in comparison with those of previous studies.

Prediction, monitoring, and interpretation of dam leakage flow via adaptative kernel extreme learning machine

The magnitude of leakage in the dam body and its foundation can be used as an important indicator in dam risk management. This study presents a data mining and monitoring framework for safety control of the dam leakage flow. First, the influencing factors in dam leakage flow are investigated. Second, a kernel extreme learning machine (KELM) is trained to predict dam leakage, where the parameters are optimized adaptively by parallel multi-population Jaya algorithm. Finally, a novel global sensitivity analysis is proposed to evaluate the relative importance of each input variable based on the KELM. Monitoring data of leakage flow from the concrete face rockfill dam in a pumped-storage power station is used for modeling and verification. The simulated results of the case study reveal that KELM achieves a satisfactory prediction of the leakage flow. It is also found that the water level fluctuation and rainfall have a significant impact on leakage magnitude. The sensitivity analysis provides a useful qualitative metric of dam leakage, which is of great value for dam safety monitoring and operation.

Machine learning for downscaling: the use of parallel multiple populations in genetic programming

In the implementation of traditional GP algorithm as models are evolved in a single deme (an environment in which a population of models is evolved) it may tend to produce sub-optimal models with poor generalisation skills due to lack of model diversity. As a solution to above issue, in this study the potential of evolving models in parallel multiple demes with different genetic attributes (parallel heterogeneous environments) and subsequent further evolution of some of the fittest models selected from each deme in another deme called the master deme was investigated, in relation to downscaling of large-scale climate data to daily minimum temperature (Tmin) and daily maximum temperature (Tmax). It was discovered that independent of the climate regime (i.e. warm or cold) and the geographic location of the observation station, a fraction of the fittest models (e.g. 25%) obtained from the last generation of each deme alone are sufficient for the formulation of a diverse initial population of models for the master deme. Also, independent of the climate regime and the geographic location of the observation station, both daily Tmin and Tmax downscaling models developed with the parallel multi-population genetic programming (PMPGP) algorithm showed better generalisation skills compared to that of models developed with the traditional single deme GP, even when the amount of redundant information in the data of predictors was high. The models developed for daily Tmin and Tmax with the PMPGP algorithm simulated fewer unphysically large outliers compared to that of models developed with the GP algorithm.

Multi-population genetic algorithm for peak-to-average power ratio suppression in an optical OFDM transmission system.

We proposed and experimentally demonstrated the parallel multi-population genetic algorithm (MPGA)-based selected mapping (SLM) technique to reduce the peak-to-average power ratio of the optical orthogonal frequency-division multiplexing system. By introducing the parallel genetic algorithm (GA), the MPGA-SLM can effectively escape the local optimization of the conventional GA and attains fast convergence on the global optimization. Compared with the conventional SLM method, we experimentally achieved 1-dB performance improvement at a 3.8×10-3 bit error ratio by utilizing the MPGA-SLM in the back-to-back case and that of 0.7dB in the 30-km single-mode fiber transmission case.

Parallel Multipopulation Differential Evolutionary Particle Swarm Optimization for Voltage and Reactive Power Control

SUMMARYThis paper presents parallel multipopulation differential evolutionary particle swarm optimization (DEEPSO) for voltage and reactive power control (VQC). The problem can be formulated as a mixed integer nonlinear optimization problem and various evolutionary computation techniques have been applied to the problem including PSO, differential evolution (DE), and DEEPSO. Since VQC is one of the online controls, speed‐up of computation is required. Moreover, there is still room for improvement in solution quality. This paper applies parallel multipopulation DEEPSO in order to speed up the calculation and improve solution quality. The proposed method is applied to IEEE 30, 57, and 118 bus systems. The results indicate that the proposed method can realize fast computation and minimize more active power losses than the conventional evolutionary computation techniques.

電圧・無効電力制御に対する複数集団を用いた並列DEEPSOの適用

電圧・無効電力制御に対する複数集団を用いた並列DEEPSOの適用

Optimization of High-Speed Train Control Strategy for Traction Energy Saving Using an Improved Genetic Algorithm

A parallel multipopulation genetic algorithm (PMPGA) is proposed to optimize the train control strategy, which reduces the energy consumption at a specified running time. The paper considered not only energy consumption, but also running time, security, and riding comfort. Also an actual railway line (Beijing-Shanghai High-Speed Railway) parameter including the slop, tunnel, and curve was applied for simulation. Train traction property and braking property was explored detailed to ensure the accuracy of running. The PMPGA was also compared with the standard genetic algorithm (SGA); the influence of the fitness function representation on the search results was also explored. By running a series of simulations, energy savings were found, both qualitatively and quantitatively, which were affected by applying cursing and coasting running status. The paper compared the PMPGA with the multiobjective fuzzy optimization algorithm and differential evolution based algorithm and showed that PMPGA has achieved better result. The method can be widely applied to related high-speed train.

A parallel multi-population biased random-key genetic algorithm for a container loading problem

This paper presents a multi-population biased random-key genetic algorithm (BRKGA) for the single container loading problem (3D-CLP) where several rectangular boxes of different sizes are loaded into a single rectangular container. The approach uses a maximal-space representation to manage the free spaces in the container. The proposed algorithm hybridizes a novel placement procedure with a multi-population genetic algorithm based on random keys. The BRKGA is used to evolve the order in which the box types are loaded into the container and the corresponding type of layer used in the placement procedure. A heuristic is used to determine the maximal space where each box is placed. A novel procedure is developed for joining free spaces in the case where full support from below is required. The approach is extensively tested on the complete set of test problem instances of Bischoff and Ratcliff [1] and Davies and Bischoff [2] and is compared with 13 other approaches. The test set consists of 1500 instances from weakly to strongly heterogeneous cargo. The computational experiments demonstrate that not only the approach performs very well in all types of instance classes but also it obtains the best overall results when compared with other approaches published in the literature.

Concurrent Societies Based on Genetic Algorithm and Particle Swarm Optimization

A parallel multi-population based metaheuristic optimization framework, called Concurrent Societies, inspired by human intellectual evolution, is proposed. It uses population based metaheuristics to evolve its populations, and fitness function approximations as representations of knowledge. By utilizing iteratively refined approximations it reduces the number of required evaluations and, as a byproduct, it produces models of the fitness function. The proposed framework is implemented as two Concurrent Societies: one based on genetic algorithm and one based on particle swarm optimization both using k -nearest neighbor regression as fitness approximation. The performance is evaluated on 10 standard test problems and compared to other commonly used metaheuristics. Results show that the usage of the framework considerably increases efficiency (by a factor of 7.6 to 977) and effectiveness (absolute error reduced by more than few orders of magnitude). The proposed framework is intended for optimization problems with expensive fitness functions, such as optimization in design and interactive optimization.

The Application of Parallel Multipopulation Genetic Algorithms to Dynamic Job-Shop Scheduling

This paper describes the use of parallel multipopulation genetic algorithms (GAs) to meet the dynamic nature of job-shop scheduling. A modified genetic technique is adopted by using a specially formulated genetic operator to provide an efficient optimisation search. The proposed technique has been success fully implemented using the programming language MATrix LABoratory (MATLAB), providing a powerful tool for job-shop scheduling. Comparisons indicate that the proposed genetic algorithm has successfully improved upon the solution obtained from conventional approaches, particularly in coping with jobshop scheduling.