Synonyms and the strong association of semantic information increase the dimension text feature vectors, and greatly affect the efficiency and accuracy of text classification. In order to reduce the dimension of the t...
详细信息
ISBN:
(纸本)9781784660543
Synonyms and the strong association of semantic information increase the dimension text feature vectors, and greatly affect the efficiency and accuracy of text classification. In order to reduce the dimension of the text feature vectors, this paper presents an improved parallel genetic algorithm to solve the text feature clustering problem. Firstly, a K-means algorithm is used to perform thick-granularity clustering for feature words. Successively, a parallel genetic algorithm is used to perform thin-granularity clustering for feature words. In the process of applying geneticalgorithms, the crossover operator is improved so that the algorithm has a global search ability and local search capability, and reduces the dependence on the initial cluster centers. Finally, feature words in each cluster are analyzed and compressed to form a feature word set which reflects the feature of text classes and semantic information. The experiments validate that our method for text feature extraction is effective.
Accurately obtaining the radiation properties of insulating ceramic materials is essential for engineering applications. This article obtained the bidirectional transmittance and reflectance of thermal insulation cera...
详细信息
Accurately obtaining the radiation properties of insulating ceramic materials is essential for engineering applications. This article obtained the bidirectional transmittance and reflectance of thermal insulation ceramics, and introduced a new scattering phase function to establish a radiation transfer model based on Monte Carlo method. Combined with parallel genetic algorithm inversion, radiation properties such as extinction coefficient, scattering albedo, and scattering phase function are obtained. Firstly, the experimental optical path is simulated and analyzed, which has little effect on the measurement results due to slight deflection of strong extinction material samples and detectors. For the measurement of bidirectional transmittance, a larger spot radius and detector radius will increase the measurement bidirectional transmittance. Secondly, through parallel genetic algorithm inversion, >5 measurement points are required to obtain their radiative properties, however, the radiation properties of backscattering materials cannot be precisely obtained using bidirectional transmittance for inversion. The required inversion accuracy can be achieved when the bidirectional transmittance and reflectance ratio measurement angle step is <4 degrees . Finally, this study determined the radiation properties of ceramic insulating materials that show little wavelength variation, their spectral extinction coefficients are above 9000m(-1), and spectral scattering albedo are greater than 0.9. It is difficult to characterize scattering features using isotropic scattering phase functions because materials have both forward and backward scattering characteristics. The scattering characteristics of insulation ceramics described using the newly proposed scattering phase function have higher accuracy.
The treatment of waste streams is an important research topic for the sustainable development of chemical process. Recently, reactive distillation-based (RD) processes with ethylene oxide hydrolysis have been develope...
详细信息
The treatment of waste streams is an important research topic for the sustainable development of chemical process. Recently, reactive distillation-based (RD) processes with ethylene oxide hydrolysis have been developed to separate water-containing azeotropic mixtures. To separate Serafimov's class 2.0-2b mixtures containing methanol/methyl acetate, a methyl acetate transesterification reaction between propylene glycol monomethyl ether is introduced to convert methyl acetate to methanol and coproduce produce propylene glycol monomethyl ether acetate as advanced solvents. Three-column reactive-extractive distillation (TCRED) and extractivereactive distillation (TCERD), and double-column pre-separation-reactive distillation (DCPSRD) processes, are proposed. Then, we use a parallel genetic algorithm to optimize processes to maximize total net revenue. The case study is methanol/methyl acetate/ethyl acetate. Though TCRED is not suitable due to the occurrence of ethyl acetate transesterification reactions, to compare economic performances of three RD-based processes, TCRED is regarded as a pseudo process. The total net revenue of three RD-based processes is relatively larger than (similar to 20 % increase) that of the conventional extractive distillation process. Compared with TCRED process, TCERD and DCPSRD can achieve $582336 and $1045995 increase in TNR, 27.43 % and 49.99 % TAC reduction, respectively. In summary, proposed RD-based processes are promising to separate Serafimov's class 2.0-2b mixtures containing methanol/methyl acetate, especially TCRED and DCPSRD.
The performance of glue laminated bamboo (glubam) members is governed by the nonlinear response at their joints, where high deformation levels and stress concentrations are developed. Numerous phenomenological models ...
详细信息
The performance of glue laminated bamboo (glubam) members is governed by the nonlinear response at their joints, where high deformation levels and stress concentrations are developed. Numerous phenomenological models are presently employed to describe the hysteresis behavior of these joints, while these models always have an excessive number of parameters, and the physical interpretation of these parameters is often challenging. Moreover, some hysteresis models cannot capture all hysteresis features such as asymmetry, pinching, and damage. Consequently, this paper introduces a novel phenomenological-based hysteretic model named Asymmetric Pinching Damaged (APD) model, and implemented it in Abaqus by combining connector and spring elements in series or parallel. This model encompasses asymmetry, pinching, and strength degradation for bamboo joint components, with parameters that possess clear physical meanings and are readily comprehensible. This study also presented a parameter identification framework coupling the parallel genetic algorithm (PGA) and Bayesian Neural Network (BNN). By merging the FE modeling and optimizing algorithms with the interactive application of ABAQUS and Python software platforms, the integrated identification framework is capable of performing multi-threaded parallel computation of finite element models considering the BNN-based uncertainty quantification, thus greatly improving the efficiency of parameter identification.
In this work, we show an effective approach for application of reduced order model (ROMs) constructed with proper orthogonal decomposition (POD) using liquid production rates for snapshot generation. These ROMs are us...
详细信息
In this work, we show an effective approach for application of reduced order model (ROMs) constructed with proper orthogonal decomposition (POD) using liquid production rates for snapshot generation. These ROMs are used to maximise oil production while controlling associated water production. Using ROMs, we perform a parallel genetic algorithm (PGA) and consider liquid production rates as decision variables. The balanced rates are used for injection wells, based on open-flow potential and total amount of produced reservoir volume. The net present value (NPV) is selected as objective function to ensure project profitability and to penalise water production. The NPV by ROM optimisation is approached to within 98% of the NPV obtained by optimisation using the full-order model demonstrating acceptable accuracy. A synthetic model and a real field sector are used for evaluation. The optimisation runtime reduces by 55% in the synthetic model and 71% for optimisation with ROM in the sector case. [Received: April 4, 2022;Accepted: July 14, 2022]
Interpreting high resolution nuclear magnetic resonance (NMR) spectra of complex samples is investigated by the use of real valued parallel and nonparallel genetic algorithm based on stochastic search procedure. The p...
详细信息
Interpreting high resolution nuclear magnetic resonance (NMR) spectra of complex samples is investigated by the use of real valued parallel and nonparallel genetic algorithm based on stochastic search procedure. The population-centric crossover operators were used in real coded geneticalgorithm (RCGA) based on some probability distributions. This paper also presents parallel genetic algorithms computations with different genetic immigration operators. Different results were found with respect to the problems, even at the different stages of the genetic process in the same problem. It is observed that the grid and centralized type genetic immigration (island models) were effective on the global optimization. The parallel and non-parallelalgorithms were also applied for solving some multi-modal test problems. It is found that the island models achieve superior performance on multi-modal test problems and on deconvolution of complex NMR spectra. (C) 2013 Elsevier B.V. All rights reserved.
This paper presents a parallel genetic algorithm (GA) called the cellular compact geneticalgorithm (c-cGA) and its implementation for adaptive hardware. An adaptive hardware based on the c-cGA is proposed to automate...
详细信息
This paper presents a parallel genetic algorithm (GA) called the cellular compact geneticalgorithm (c-cGA) and its implementation for adaptive hardware. An adaptive hardware based on the c-cGA is proposed to automate real-time classification of ECG signals. The c-cGA not only provides a strong search capability while maintaining genetic diversity using multiple GAs but also has a cellular-like structure and is a straight-forward algorithm suitable for hardware implementation. The c-cGA hardware and an adaptive digital filter structure also perform an adaptive feature selection in real time. The c-cGA is applied to a block-based neural network (BbNN) for online learning in the hardware. Using an adaptive hardware approach based on the c-cGA, an adaptive hardware system for classifying ECG signals is feasible. The proposed adaptive hardware can be implemented in a field programmable gate array (FPGA) for an adaptive embedded system applied to personalised ECG signal classifications for long-term patient monitoring.
This paper presents a performance evaluation between hardware and software implementation of a probabilistic parallel genetic algorithm. The compact geneticalgorithm is extended to support parallel implementation. Th...
详细信息
ISBN:
(纸本)9781467353229
This paper presents a performance evaluation between hardware and software implementation of a probabilistic parallel genetic algorithm. The compact geneticalgorithm is extended to support parallel implementation. The parallelized compact geneticalgorithm is implemented in FPGA hardware and parallelized software version running on multi-core processors for performance evaluation using standard benchmark functions. The experimental results show that the hardware implementation of the parallel compact geneticalgorithm delivers speedup of between 100-fold to 500-fold depending on problems size and number of generations.
In general efficient way of routing method is used to transfer the data. This routing problem is solved by using Different types of routing algorithms, here we use Coarse-Grained parallel GA-Based shortest path algori...
详细信息
ISBN:
(纸本)9781479939251
In general efficient way of routing method is used to transfer the data. This routing problem is solved by using Different types of routing algorithms, here we use Coarse-Grained parallel GA-Based shortest path algorithm. Time computation is the vital parameter in all routing methods. The very shortest path routing algorithm involves reduce time of the transferring data. Using geneticalgorithm we can find an efficient path. This algorithm found by using the nature of the genetic operation. geneticalgorithm is used to change the genes from one sub-population to another sub-population in a proper manner. In this paper discussion is going through both simple geneticalgorithm and parallel genetic algorithm and compares performance of both. Here Migration strategy is used to replace the genes. There are four types of strategies that are used to change the genes. These are: Best replace Worst, Best replace Random, Random replace Random, Random replace Worst, Random replace Random. Among these four types of Strategies worst replace best gives the better performance.
The probabilistic minimum spanning tree (PMST) problem is NP-complete and is hard to solve. However, it has important theoretical significance and wide application prospect. A parallel genetic algorithm based on coars...
详细信息
ISBN:
(纸本)9781479925483
The probabilistic minimum spanning tree (PMST) problem is NP-complete and is hard to solve. However, it has important theoretical significance and wide application prospect. A parallel genetic algorithm based on coarse-grained model is proposed to solve PMST problem in this paper. Firstly, we discuss several problems of determinant factorization encoding, and develop repairing method for illegal individuals. Secondly, a coarsegrained parallel genetic algorithm, which combines message passing interface (MPI) and geneticalgorithm, is designed to solve probabilistic minimum spanning tree problems. Finally, the proposed algorithm is used to test several probabilistic minimum spanning tree problems which are generated by the method introduced in the literature. The statistical data of the test results show that the expectation best solution and average best solution obtained by the proposed algorithm are better than those provided in the literature.
暂无评论