Least Squares Support Vector Machine Research Articles

Study on detecting main ingredients of silicone rubber based on terahertz spectrum

AbstractThe authors investigated the ingredient detection technique of silicone rubber based on the Terahertz spectrum. For this purpose, 18 diverse high‐temperature vulcanised silicone rubber (HTVSR) formulations were customised, 8 of which are used as calibration set while the rest 10 as prediction set. Based on the Beer‐Lambert Law, the partial‐least‐square (PLS) regression model and the least‐squares support‐vector machines (LS‐SVM) regression model were used to yield the relationships between the absorption spectrums and the content percentages of polydimethylsiloxane (PDMS), alumina trihydrate (ATH), and silica in HTVSR. The results showed that for the formulations tested, the prediction accuracy of all three main ingredients by the PLS regression model could be improved by changing the spectrum range from 0.2–4 to 0.5–2 THz. If the data were pre‐processed by the Savitzky–Golay smoothing method or multiplicative scatter correction method, the prediction accuracy of PDMS could be further enhanced. However, this would lead to a slight decrease in the prediction accuracy of ATH. For the LS‐SVM regression model, the radial basis function (RBF) kernel and the linear kernel were studied. It was found that the prediction accuracy of both kernels was better than that of the PLS regression model. With the LS‐SVM regression model using the RBF kernel, the correlated coefficients of PDMS and ATH in the prediction set could be up to 0.9915 and 0.9742, respectively.

New insights into culvert scour depth calculation by soft computing models using multi-model ensemble approach and uncertainty analysis

ABSTRACT The accurate prediction of scour depth in culverts is crucial to ensure public safety and due to the uncertainties in classical empirical/deterministic equations, it remains challenging. This study presents new ensemble multi-models to predict scour depth and quantify model bias and uncertainty. In this study, culvert scour was predicted by gene expression programming (GEP) and least square support vector machine (LSSVM), firstly based on dimensionless parameters. In the multi-model ensemble strategy, four ensemble approaches, namely SM, WM, LSSVM, and BMA were applied. The LSSVM showed overall best skill for the multi-model approach, that reduced the error by 61.3% and increased the coefficient of determination by 52.1%. The results showed the proposed approach can satisfy the requirements of simplicity, applicability, and accuracy at the same time. For application purposes, we provided a simple code to obtain scour depth with great accuracy using LSSVM. BMA and LSSVM were used to quantify the scour culvert uncertainty. The results of the uncertainty analysis showed that the LSSVM predicts scour depth quite well and generates high-quality prediction intervals. The results revealed that a multi-model ensemble strategy improved accuracy, certainty and reliability for culvert scour compared to empirical equations.

Underwater Sound Speed Field Forecasting Based on the Least Square Support Vector Machine

Underwater sound speed is one of the most significant factors that affects high-accuracy underwater acoustic positioning and navigation. Due to its complex temporal variation, the forecasting of the underwater sound speed field (SSF) becomes a challenging task. Taking advantage of machine learning methods, we propose a new method for SSF forecasting based on the least square support vector machine (LSSVM) and a multi-parameter model, aiming to enhance the forecasting accuracy of underwater SSF with hourly resolution. We first use a matching extension method to standardize profile data and train the LSSVM with the parameters of observation time, temperature, salinity, and depth. We then employ radial basis function kernels to construct the forecasting model of SSF. We validate the feasibility and effectiveness of the LSSVM model by comparing it with the polynomial fitting (PF) and back propagation neural network (BPNN) methods, using hourly data obtained from the measured data and open data. The results show that the means of the root mean square for the LSSVM based on the observation time parameter and the LSSVM based on the multi-parameter model achieve 0.51 m/s and 0.45 m/s, respectively, presenting a significant improvement compared with the PF (0.82 m/s) and BPNN (0.76 m/s) methods.

Hybrid Metaheuristic Optimization Algorithms with Least-Squares Support Vector Machine and Boosted Regression Tree Models for Prediction of Air-Blast Due to Mine Blasting

Hybrid Metaheuristic Optimization Algorithms with Least-Squares Support Vector Machine and Boosted Regression Tree Models for Prediction of Air-Blast Due to Mine Blasting

Enhanced complex wire fault diagnosis via integration of time domain reflectometry and particle swarm optimization with least square support vector machine

AbstractUrban power systems rely on intricate wire networks, known as the power grid, which form the essential electric infrastructure in cities. While these networks transmit electricity from power plants to consumers, they are vulnerable to faults caused by manufacturing errors and improper installation, posing risks to system integrity. Thus, accurate identification and assessment of these faults are crucial to prevent damage and maintain system reliability. The objective of this research is to present an innovative and efficient methodology for diagnosing complex wire networks through the application of time domain reflectometry (TDR) combined with the particle swarm optimization (PSO) and least squares support vector machine (LSSVM) algorithm. This research addresses the imperative need to accurately locate and assess breakage faults within wire networks, emphasizing their role in both power transmission and communication infrastructure. To model the TDR answer of a specific complex wire network, a forward model is established utilizing resistance, inductance, capacitance and conductance (RLCG) parameters and the finite difference time domain (FDTD) method. Subsequently, the PSO‐LSSVM approach is used to solve the inverse problem of localizing faults in complex wire networks. The experimental results validate the practicality of this approach in real‐world systems.

Research on SPAD Estimation Model for Spring Wheat Booting Stage Based on Hyperspectral Analysis.

With the rapid progression of agricultural informatization technology, the methodologies of crop monitoring based on spectral technology are constantly upgraded. In order to carry out the efficient, precise and nondestructive detection of relative chlorophyll (SPAD) during the booting stage, we acquired hyperspectral reflectance data about spring wheat vertical distribution and adopted the fractional-order differential to transform the raw spectral data. After that, based on correlation analysis, fractional differential spectra and fractional differential spectral indices with strong correlation with SPAD were screened and fused. Then, the least-squares support vector machine (LSSSVM) and the least-squares support vector machine (SMA-LSSSVM) optimized on the slime mold algorithm were applied to construct the estimation models of SPAD, and the model accuracy was assessed to screen the optimal estimation models. The results showed that the 0.4 order fractional-order differential spectra had the highest correlation with SPAD, which was 9.3% higher than the maximum correlation coefficient of the original spectra; the constructed two-band differential spectral indices were more sensitive to SPAD than the single differential spectra, in which the correlation reached the highest level of 0.724. The SMA-LSSSVM model constructed based on the two-band fractional-order differential spectral indices was better than the single differential spectra and the integration of both, which realized the assessment of wheat SPAD.

Ultra-Wideband Ranging Error Mitigation with Novel Channel Impulse Response Feature Parameters and Two-Step Non-Line-of-Sight Identification.

The effective identification and mitigation of non-line-of-sight (NLOS) ranging errors are essential for achieving high-precision positioning and navigation with ultra-wideband (UWB) technology in harsh indoor environments. In this paper, an efficient UWB ranging-error mitigation strategy that uses novel channel impulse response parameters based on the results of a two-step NLOS identification, composed of a decision tree and feedforward neural network, is proposed to realize indoor locations. NLOS ranging errors are classified into three types, and corresponding mitigation strategies and recall mechanisms are developed, which are also extended to partial line-of-sight (LOS) errors. Extensive experiments involving three obstacles (humans, walls, and glass) and two sites show an average NLOS identification accuracy of 95.05%, with LOS/NLOS recall rates of 95.72%/94.15%. The mitigated LOS errors are reduced by 50.4%, while the average improvement in the accuracy of the three types of NLOS ranging errors is 61.8%, reaching up to 76.84%. Overall, this method achieves a reduction in LOS and NLOS ranging errors of 25.19% and 69.85%, respectively, resulting in a 54.46% enhancement in positioning accuracy. This performance surpasses that of state-of-the-art techniques, such as the convolutional neural network (CNN), long short-term memory-extended Kalman filter (LSTM-EKF), least-squares-support vector machine (LS-SVM), and k-nearest neighbor (K-NN) algorithms.

Sulphate resistance prediction of RAC considering the coupling effect of RCA and water-binder ratio based on experimental analysis and LSSVM

ABSTRACT Waste concrete removed from old pavement can be crushed and sieved as recycled aggregate to prepare concrete, and the application of recycled aggregate concrete (RAC) can reduce the occupation of land resources by the waste concrete and save natural aggregate. In this paper, the influence of proportion of recycled coarse aggregate (RCA) replacing natural coarse aggregate (NCA) and water-binder ratio on the sulphate resistance of RAC was studied, and the compressive strength degradation rate (Df ) was introduced to evaluate the sulphate resistance of RAC. The experimental results showed that the compressive strength of RAC before and after sulphate attacks linearly decreased with the growth of the replacement rate of RCA, water-binder ratio, and the number of sulphate wet-dry cycles, respectively; while Df of RAC increased with the rise of the replacement rate of RCA, water-binder ratio, and sulphate wet-dry cycles, respectively. The results also indicated that Df of all groups of RAC was less than 25%, which can meet the engineering requirements. In addition, the Df model of RAC under a sulphate attack based on experimental data and that based on Least-Squares Support Vector Machine (LSSVM) was built and compared, and both models predict the sulphate resistance of RAC well. The prediction model based on LSSVM has higher prediction accuracy.

Improved prediction of groundwater depth in urban areas using the EWT-S–G-GRU and LSSVM model: a case study of Xinxiang City

ABSTRACT The improvement in predicting groundwater depth accuracy has significant implications for the management, ecological environment protection, and economic and social development of regional water resources. Employing empirical wavelet transform (EWT) for nonlinear processing, Savitzky–Golay (S–G) filtering to reduce high-frequency noise, gate recurrent unit (GRU) neural network for linear feature signal processing, and least squares support vector machine (LSSVM) for nonlinear signal handling, we established a comprehensive model combining EWT-S–G-GRU and LSSVM. The results demonstrate that the proposed model exhibits superior accuracy in groundwater depth prediction, with an average relative error of only 2.14% and Nash–Sutcliffe efficiency (NSE) of 0.93. This outperforms the other four models, with average relative errors of 12.88%, 11.90%, 7.07%, and 11.10%, and NSE values of 0.58, 0.56, 0.71, and 0.73, respectively. The superiority of the model established in this study is attributed to its effective handling of both nonlinear and linear features of groundwater, thereby enhancing predictive accuracy. The EWT-S–G-GRU and LSSVM model proves to be more reliable in revealing the spatial distribution and dynamic changes of future groundwater, providing a robust reference for the rational development and utilization of groundwater in the urban area of Xinxiang City.

Method for Fault Diagnosis of Track Circuits Based on a Time–Frequency Intelligent Network

In response to the limitations posed by noise interference in complex environments and the narrow focus of existing diagnosis methods for jointless track circuit faults, an innovative approach is put forward in this study. It involves the application of the continuous wavelet transform (CWT) for signal preprocessing, along with the integration of a deep belief network (DBN) and a genetic algorithm (GA) to improve the least-squares support vector machine (LSSVM) model for intelligent time–frequency fault diagnosis. Initially, the raw induced voltage signals are transformed using continuous wavelet transformation resulting in wavelet time–frequency representations that combine temporal and spectral information. Subsequently, these time–frequency representations are fed into the deep belief networks, which perform semi-supervised dimensionality reduction and feature extraction, thereby uncovering distinct fault characteristics in the track circuit. Finally, the genetic algorithms are employed to improve the kernel function and penalty factor parameters of the least-squares support vector machine, thus establishing an optimal DBN-GA-LSSVM diagnostic model. Experimental validation demonstrates the effectiveness of the proposed time–frequency intelligent network model by leveraging the advantages of deep belief networks in hierarchical feature extraction and the superior performance of the least-squares support vector machine in addressing high-dimensional pattern recognition problems with limited samples. The achieved accuracy rate on the testing dataset reaches an impressive 99.6%. Consequently, this comprehensive approach provides a viable solution for data-driven track circuit fault diagnosis.

A novel robust intelligent fault diagnosis method for rolling bearings based on SPAVMD and WOA-LSSVM under noisy conditions

Aiming at the problem that the fault information of rolling bearings under harsh operation environment is easily submerged by strong noise interference, which causes the traditional method to be difficult to identify bearing faults effectively, this paper proposes a novel robust intelligent fault diagnosis method for rolling bearings based on sparsity-assisted parameter adjustable variational mode decomposition (VMD) and whale optimization algorithm-based optimized least-squares support vector machine (WOA-LSSVM). Firstly, a sparsity measurement named the improved Gini index is introduced as the fitness function of grid search algorithm to adaptively adjust and search for the optimal decomposed mode number K and penalty factor α of VMD. Additionally, VMD containing the optimal parameters is adopted to decompose the original bearing vibration signal into several intrinsic mode function (IMF), and the effective signal reconstruction is performed by screening the sensitive IMF components according to the effective weighted kurtosis Gini index criterion. Subsequently, the refine composite multi-scale dispersion entropy of the reconstructed signal is further calculated to establish a multi-dimensional feature vector set. Finally, the constructed feature vector set is fed into the WOA-LSSVM to achieve automatic fault identification of rolling bearings. The effectiveness of the proposed method is verified by two experimental examples. Experimental results show that the proposed method has higher fault recognition accuracy and better robustness against noise than other homologous methods in noisy conditions. This study provides a new perspective for the developing of robust diagnosis methods.

Crashworthiness Analysis and Optimization of Thin-Walled Structures Based on the TSSA-LSSVM Surrogate Model

Thin-Walled Structures (TWS) play a crucial role in enhancing both the collision safety and lightweight characteristics of vehicles. Structural design and optimization have demonstrated significant potential in improving their crashworthiness and overall weight reduction. Several interconnected factors, including the cross-sectional shape and parameters of distinctive structural characteristics (known as induction grooves), impact the TWS’ resistance to collisions. Another critical aspect in the optimization of TWS design is striking the delicate balance between the surrogate model’s effectiveness and precision. Taking all these factors into account, this paper conducts an in-depth investigation into the cross-sectional shape, grooves, and other geometric attributes of various TWSs through a comparative analysis. It reveals that the thin-walled structure with a Double-Ribbed Rectangle cross-section (DR-TWS) proves to be the optimal choice, particularly in low-speed axial impact scenarios. Furthermore, we present an enhanced Least Squares Support Vector Machine (LSSVM) method, augmented with an adaptive [Formula: see text]-distribution Sparrow Search Algorithm (TSSA), referred to as the TSSA-LSSVM surrogate model. The optimized design of the DR-TWS is realized through the integration of the TSSA-LSSVM surrogate model with the adaptive hybrid multi-objective particle swarm optimization (AHMOPSO) algorithm. This paper not only provides substantial theoretical support but also offers valuable guidelines for the application of TWSs in the automotive industry.

Feasibility of using colorimetric devices for whole and ground coffee roasting degrees prediction.

Coffee roasting is one of the crucial steps in obtaining a high-quality product as it forms the product's color and flavor characteristics. Roast control is made by visual inspection or traditional instruments such as the Agtron spectrophotometer, which can have high implementation costs. Therefore, the present study evaluated colorimetric approaches (a bench colorimeter, smartphone digital images, and a colorimetric sensor) to predict the Agtron roasting degrees of whole and ground coffee. Two calibration approaches were assessed, that is, multiple linear regression and least-squares support vector machine. For that, 70 samples of whole and ground roasted coffees comprising the Agtron roasting range were prepared. The results showed that all three colorimetric acquisition types were efficient for the model building, but the bench colorimeter and the smartphone digital images generally performed with good determination coefficients and low errors as measured by external validation. For the whole bean coffee, the best model presented a determination coefficient (R2) of 0.99 and a root-mean-squared error (RMSE) of 1.91%, while R2 of 0.99 and RMSE of 0.87% was obtained for ground coffee, both using the colorimeter. The obtained models presented good prediction capability, as assessed by external validation and randomization tests. The obtained findings point to an alternative for coffee roasting monitoring that can lead to higher digitalization and local control of the process, even for smaller producers, due to its lower costs. © 2024 Society of Chemical Industry.

Prediction of Energy Consumption in a Coal-Fired Boiler Based on MIV-ISAO-LSSVM

Aiming at the problem that the energy consumption of the boiler system varies greatly under the flexible peaking requirements of coal-fired units, an energy consumption prediction model for the boiler system is established based on a Least-Squares Support Vector Machine (LSSVM). First, the Mean Impact Value (MIV) algorithm is used to simplify the input characteristics of the model and determine the key operating parameters that affect energy consumption. Secondly, the Snow Ablation Optimizer (SAO) with tent map, adaptive t-distribution, and the opposites learning mechanism is introduced to determine the parameters in the prediction model. On this basis, based on the operation data of an ultra-supercritical coal-fired unit in Xinjiang, China, the boiler energy consumption dataset under variable load is established based on the theory of fuel specific consumption. The proposed prediction model is used to predict and analyze the boiler energy consumption, and a comparison is made with other common prediction methods. The results show that compared with the LSSVM, BP, and ELM prediction models, the average Relative Root Mean Squared Errors (aRRMSE) of the LSSVM model using ISAO are reduced by 2.13%, 18.12%, and 40.3%, respectively. The prediction model established in this paper has good accuracy. It can predict the energy consumption distribution of the boiler system of the ultra-supercritical coal-fired unit under variable load more accurately.

Power Transformer Classification through Dissolved Gas Analysis Utilizing Least-Square Support Vector Machine

Power Transformer Classification through Dissolved Gas Analysis Utilizing Least-Square Support Vector Machine

Rolling bearing fault diagnosis based on improved whale-optimization- algorithm–variational-mode-decomposition method

Rolling bearings are a key component of rotating machinery and their health directly affects the safe operation of mechanical equipment. Therefore, fault diagnose for rolling bearings is very important. The fault diagnosis process of rolling bearings includes three stages: signal decomposition, feature extraction, and pattern recognition. Variational mode decomposition (VMD) can suppress end effects, but improper parameter settings will cause information losses or excessive decomposition. In this work, an improved whale optimization algorithm (IWOA) is applied to parameter settings of VMD. Correspondingly, an IWOA-VMD signal decomposition method is proposed. The decomposed signal is combined with a Laplace score method and classifier to remove the redundancy and noise in the feature set and obtain a low-dimensional sensitive feature subset. Then, aiming at the problem of the parameter settings of a least squares support vector machine (LSSVM) affecting the recognition performance and accuracy, a salp swarm algorithm (SSA) is used to globally optimize the penalty parameter and kernel width in the LSSVM to establish an SSA-LSSVM fault recognition model. This model is applied to the fault diagnosis of rolling bearings. In particular, rolling bearing fault samples at Case Western Reserve University are used to verify the method. The results indicate that the proposed method is effective and improves the speed and accuracy of fault diagnosis.

Modeling and estimation of water activity for the ionic-liquid-based aqueous ternary systems by smart paradigms

In this study, different artificial intelligence (AI) techniques are employed to determine the water activity of ionic liquid-based ternary systems as a function of pressure, temperature, and the type and concentration of the components. The proposed model has been selected among five artificial neural networks (ANNs), bagged trees (BT), decision trees (DT), gaussian process regression (GPR), adaptive neuro-fuzzy inference systems (ANFIS), and least square support vector machines (LS-SVM) by trial-and-error process and ranking analysis of statistical and computational indicators. The most accurate model is the cascade feed-forward neural network with fifteen hidden neurons which are equipped with logarithm and hyperbolic tangent sigmoid transfer functions. This model estimates 1829 empirical water activity for ionic liquid-based ternary systems with AARD=0.25 %, MSE=1.73 × 10−5, RMSE=4.16 × 10−3, and R2=0.9919.

Synchronous Retrieval of Wheat Cab and LAI from UAV Remote Sensing: Application of the Optimized Estimation Inversion Framework

Chlorophyll a and b content (Cab) and leaf area index (LAI) are two key parameters of crops, and their quantitative inversions are important for growth monitoring and the field management of wheat. However, due to the close correlation between the spectral signals of these two parameters and the effects of soil and atmospheric conditions, as well as modeling errors, synchronous retrieval of LAI and Cab from remote sensing data is still a challenging task. In a previous study, we introduced the optimal estimation theory and established the inversion framework by coupling the PROSAIL (PROSPECT + SAIL) model with the unified linearized vector radiative transfer model (UNL-VRTM). The framework fully utilizes the simulated radiance spectra for synchronous retrieval of Cab and LAI at the UAV observation scale and has good convergence and self-consistency. In this study, based on this inversion framework, synchronized retrieval of Cab and LAI was carried out by real wheat UAV observation data and validated with the ground-measured data. By comparing with the empirical statistical model constructed by the PROSAIL model and coupled model, least squares support vector machine (LSSVM), and random forest (RF), the proposed method has the highest accuracy of Cab and LAI estimated from UAV multispectral data (for Cab, R2 = 0.835, RMSE = 14.357; for LAI, R2 = 0.892, RMSE = 0.564). Our proposed method enables the fast and efficient estimation of Cab and LAI in multispectral data without prior measurements and training.

Photovoltaic output power estimation model based on multi-meteorological variable dimension reduction and improved variational mode decomposition algorithm

ABSTRACT The instability and intermittence of photovoltaic (PV) power generation challenge the large-scale integration of PV power generation systems into the power grid and the development of energy storage systems. To accurately predict the output power of PV power generation, a power prediction model based on the optimized variational modal decomposition (OVMD)-adaptive t-distribution sparrow search algorithm (tSSA)-least squares support vector machine (LSSVM) algorithm was built. The performance of the model was then verified by three-year meteorological data and real-time PV output power data of two different climatic regions in China. The mean absolute percentage error (MAPE) and root mean square error (RMSE) of this model were less than 3.1% and 0.40, respectively, and the determination coefficient (R2) of this model was greater than 95%. Moreover, the prediction accuracy was greatly improved compared to models based on the support vector machine (SVM), LSSVM, variational modal decomposition (VMD)-LSSVM, and VMD-SSA-LSSVM algorithms. Finally, by performing dimension reduction of meteorological variables related to PV output power, the performance of the proposed OVMD-tSSA-LSSVM model was further improved.

Classification of wheat grain varieties using terahertz spectroscopy and convolutional neural network

Wheat quality and quantity differ in diverse climates. Therefore, it is essential to identify the variety before purchasing and warehousing. In this study, a study on variety discrimination for 12 wheat varieties (strong-gluten wheat, medium-gluten wheat, weak-gluten wheat) using the Terahertz time-domain spectroscopy (THz-TDS) technology in combination with a Convolutional neural network (CNN). Firstly, the original Time-domain spectra (TDS) of wheat in the range of 0.1–2.0 THz were acquired, and the Frequency domain spectra (FDS), the absorption coefficient spectra in the range of 0.2–1.0 THz were obtained through Fourier Transform. Then, Competitive adaptive reweighted sampling (CARS) algorithms were applied to screen the feature spectrum. Finally, the Support vector machine (SVM), the Least square support vector machine (LS-SVM), the Back-propagation neural networks (BPNN) and CNN models were constructed using feature spectral data. By comparing the four models, it was found that the calibration set accuracy and prediction set accuracy of the CNN model reached 98.7% and 97.8% respectively, with an error recognition rate of only 2.2%. The research results show that combining THz-TDS technology with CNN has the advantages of accurate recognition and high efficiency. It can identify different wheat varieties and can be used for seed classification and quality detection.