Percussion and PSO-SVM-Based Damage Detection for Refractory Materials

Abstract

Refractory materials are basic materials widely used in industrial furnaces and thermal equipment. Their microstructure is similar to that of many heterogeneous high-performance materials used in micro/nanodevices. The presence of damage can reduce the mechanical properties and service life of refractory materials and even cause serious safety accidents. In this paper, a novel percussion and particle swarm optimization-support vector machine (PSO-SVM)-based method is proposed to detect damage in refractory materials. An impact is applied to the material and the generated sound is recorded. The percussion-induced sound signals are fed into a mel filter bank to generate time–frequency representations in the form of mel spectrograms. Then, two image descriptors—the local binary pattern (LBP) and histogram of oriented gradient (HOG)—are used to extract the texture information of the mel spectrogram. Finally, combining both HOG and LBP features, the fused features are input to the PSO-SVM algorithm to realize damage detection in refractory materials. The results demonstrated that the proposed method could identify five different degrees of damage of refractory materials, with an accuracy rate greater than 97%. Therefore, the percussion and PSO-SVM-based method proposed in this paper has high potential for field applications in damage detection in refractory material, and also has the potential to be extended to research on damage detection methods for other materials used in micro/nanodevices.

Temperature Compensation of SAW Winding Tension Sensor Based on PSO-LSSVM Algorithm

“…The LSSVM algorithm is a supervised learning technique that offers a model designed specifically for addressing regression and classification problems [27]. Compared to the SVM model, LSSVM transforms the quadratic programming problem of SVM into a linear equation-solving problem, thereby increasing computational complexity [28]. This model exhibits exceptional nonlinear modeling capabilities, enabling it to adapt to complex relationships by mapping data into high-dimensional spaces through kernel tricks.…”

“…But in the LSSVM model, parameter σ, γ , as a fixed parameter, greatly affects the accuracy of the whole model. Therefore, this paper will optimize it through the PSO algorithm to make the LSSVM effect more accurate [27][28][29].…”

“…In this paper, the output frequency shift ∆ f of the SAW winding tension sensor at 34 • C in Table 2 is taken as the expected output value of the LSSVM algorithm, and this algorithm is used to train the output frequency shift ∆ f under other temperature conditions in Table 3. After temperature compensation, the values of the frequency shift ∆ f of the SAW winding tension sensor under different temperature conditions were obtained, as shown in Table 3 and Figure 8; ∆T = 50 − 30 = 20 • C, ∆d max = 10912 − 10326 = 586 Hz (when F = 0.8 N), and ∆ f (FS) = 10912 Hz, so we have the temperature sensitivity coefficient Therefore, this paper will optimize it through the PSO algorithm to make the LSSVM effect more accurate [27][28][29].…”

A pipeline corrosion detecting method using percussion and residual neural network

“…Pan and Yang [38] proposed a dual convolutional neural networkbased approach for post-disaster image detection and repair cost estimation of damage in reinforced concrete buildings to support building owners and decision makers in rapid risk management and resource allocation after earthquakes. Mel spectrograms, as a popular acoustic feature, are widely used in loss identification for knock detection methods because human sensitivity to frequency is taken into account in the extraction process [39,40]. In addition, the combination of traditional CNN and Mel spectrogram can help to classify the damages based on the percussion sound after data processing [41][42][43].…”