Feature selection has been widely used in classification for improving classification accuracy and reducing computational complexity. Recently, evolutionary computation (EC) has become an important approach for solving feature selection problems. However, firstly, as the datasets processed by classifiers become
Feedforward neural network (FNN) is one of the most widely used and fastest-developed artificial neural networks. Much evolutionary computation (EC) methods have been used to optimize the weights of FNN. However, as the dimension of datasets increases, the number of weights also increases dramatically. On high-dimensional datasets, if EC methods are used directly to optimize the weights of FNN, it is impossible to obtain the optimal weights of the FNN by EC methods in an acceptable time. Feature selection is a method that can effectively reduce the computational complexity of FNN by reducing irrelevant and redundant features. It may be practical to optimize the FNN by EC methods if we first employ the feature selection for the large-scale datasets. In this paper, we present a self-adaptive parameter and strategy-based particle swarm optimization (SPS-PSO) algorithm to optimize FNN with feature selection. First, we propose an optimization model for FNN by transforming the designing of FNN into a weights optimization problem. Simultaneously, we present a feature selection optimization model. Second, we present an SPS-PSO algorithm. In this algorithm, we use the parameter and strategy self-adaptive mechanism. In addition, five candidate solution generating strategies (CSGS) are used. The experiments are divided into two groups. In the first group, SPS-PSO and three other EC methods are used to directly optimize the weights of FNN on eight datasets without any modification. In the second group, we first employ SPS-PSO-based feature selection on the original datasets and obtain eight relatively smaller datasets with the k-nearest neighbor (KNN) which is used as the evaluation function for saving time. Then, we use the new datasets as the inputs for FNN. We optimize the weights of FNN again by SPS-PSO and three other EC methods to investigate whether we can get similar or even better classification accuracy by comparing the results with that of the first group. The experimental results show that SPS-PSO has the advantage in optimizing the weights of FNN compared with the other EC methods. Meanwhile, the SPS-PSO-based feature selection can reduce the solution size and computational complexity while ensuring the classification accuracy when it is used to preprocess the datasets for FNN. In this method, a solution with an originally higher than 700 000 dimensions can be even reduced to hundreds of dimensions. INDEX TERMS Classification, evolutionary computation, feature selection, feedforward neural networks, self-adaptive, parameter adaptation, particle swarm optimization. I. INTRODUCTION Artificial neural network (ANN) [1], [2] is a hot research topic in the field of artificial intelligence [3] since the 1980s The associate editor coordinating the review of this manuscript and approving it for publication was Yongming Li. and it has been widely used as an effective classification method. ANN is a nonlinear and adaptive information processing system composed of a large number of interconnected processing units. It abstra...
The long-term use of a piezoelectric smart structure make it difficult to judge whether the structure or piezoelectric lead zirconate titanate (PZT) is damaged when the signal changes. If the sensor fault occurs, the cases and degrees of the fault are unknown based on the electromechanical impedance method. Therefore, after the principal component analysis (PCA) of six characteristic indexes, a two-component solution that could explain 99.2% of the variance in the original indexes was obtained to judge whether the damage comes from the PZT. Then LibSVM was used to make an effective identification of four sensor faults (pseudo soldering, debonding, wear, and breakage) and their three damage degrees. The result shows that the identification accuracy of damaged PZT reached 97.5%. The absolute scores of PCA comprehensive evaluation for structural damages are less than 0.5 while for sensor faults are greater than 0.6. By comparing the scores of the samples under unknown conditions with the set threshold, whether the sensor faults occur is effectively judged; the intact and 12 possible damage states of PZT can be all classified correctly with the model trained by LibSVM. It is feasible to use LibSVM to classify the cases and degrees of sensor faults.
Structural damage recognition is always the concerned focus in many fields like aerospace, petroleum and petrochemical industry, industrial production and civil life. For damage recognition in complex structure or structural interior, especially somewhere sensors can't go, minor damage is often hard identified by not only traditional nondestructive testing methods like ultrasonic testing, radiographic testing, magnetic particle testing, penetrant testing, eddy current testing, but also the current popular ultrasonic guided wave based on the piezoelectric wafer, electromagnetic acoustic transducer or magnetostrictive sensor, which is mainly because the response signals are always affected by many structural features. In this article, the advanced global search algorithm, quantum particle swarm optimization algorithm is first combined with the finite element method to accurately recognize the structural damage based on the conductance-frequency spectrum resulted from electromechanical impedance method. Meanwhile, the objective function is designed to compare the difference of peak frequency variations in the experiment and finite element calculation respectively. By adopting the stiffness reduction method of the elements near the structural damage, the identification efficiency is largely improved for no need to repeatedly partition the model grid. And after multiple iteration optimization of the artificial intelligence algorithm-quantum particle swarm optimization algorithm QPSO, the identification error of damage parameters including location and degree can be reduced to below 4 percent. Therefore, the combination of finite element method and quantum particle swarm optimization algorithm is quite effective for guaranteeing high accuracy and efficiency for damage parameters' recognition in complex structures.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.