Squares-support Vector Machine Research Articles

Prediction of Grease Performance and Optimal Additive Ratio Based on the SSA-GDA-LSSVM Model

In this paper, to address the issue of compounding three additives in PTFE grease, we propose a machine learning model based on SSA-GDA-LSSVM to predict both the tribological performance and the optimal ratio of additives in PTFE grease. Gaussian data augmentation expanded the experimental data, and the Sparrow Algorithm optimized hyperparameters of the Least Squares Support Vector Machine. SHAP analysis clarified model predictions, and a Non-Dominated Sorting Genetic Algorithm identified optimal additive ratios, which were experimentally validated. The results showed that the model predicted friction coefficients and wear scar widths with R² values exceeding 0.97, and the experimental error for optimal ratios was less than 1%.

A Hybrid Prediction Model for Short-Term Load Forecasting in Power Systems

Short-term load forecasting (STLF) plays a vital role in effective power system management by assisting power dispatch centers in developing generation plans and ensuring smooth system operation. This study introduces a novel hybrid prediction model called iSSA-LSSVM to tackle the STLF challenge. By integrating the Salp Swarm Algorithm (SSA) with Least Squares Support Vector Machines (LSSVM), the iSSA-LSSVM model significantly improves LSSVM's prediction accuracy. One of the key contributions is the model's ability to autonomously ne-tune LSSVM hyperparameters, eliminating the need for manual adjustments and optimizing performance. Modifying the SSA within iSSA-LSSVM enhances the original algorithm's exploration and exploitation capabilities, ensuring better search efficiency and precision. Using a dataset with four independent variables as input and electrical power output as the target variable, the model demonstrates superior predictive performance. Comparative analysis with three other models shows that iSSA-LSSVM achieves a lower Mean Square Error (MSE) and faster convergence. This improvement in accuracy and efficiency enhances STLF, allowing power dispatch centers to develop more precise generation plans and ensure more reliable power system operation.

Quantification of soluble protein content and characterization of protein secondary structure by Raman spectroscopy combined with chemometrics

In the large-scale industrialized production of soymilk, the accurate monitoring of protein content and its secondary structure is crucial to guarantee product quality and enhance nutritional value, while traditional protein detection methods have significant limitations. In this study, Raman spectroscopy was used to characterize the changes in the relative content of protein secondary structures during boiling, in which the α-helix structure was converted to other unstable secondary structures, indicating that the protein was thermally denatured. The study also accurately predicted the soluble protein content of soymilk for the first time by combining Raman spectroscopy and chemometrics based on algorithms such as Partial Least Squares (PLS), Least Squares Support Vector Machines (LSSVM), and Artificial Neural Networks (ANN). The results showed that the relative percentage deviation (RPD) values of all models were greater than 2, indicating that the models all possessed a certain degree of robustness; The PLS-based regression prediction model (with an average RPD value of 3.5592 and an average R2p of 0.9597) had the best overall prediction. The study's results can guide the soymilk production industry into online monitoring of soymilk quality.

Discriminating Between Biotic and Abiotic Stress in Poplar Forests Using Hyperspectral and LiDAR Data

Sustainable forest management faces challenges from various biotic and abiotic stresses. The Asian longhorned beetle (ALB) and drought stress both induce water shortages in poplar trees, but require different management strategies. In northwestern China, ALB and drought stress caused massive mortality in poplar shelterbelts, which seriously affected the ecological functions of poplars. Developing a large-scale detection method for discriminating them is crucial for applying targeted management. This study integrated UAV-hyperspectral and LiDAR data to distinguish between ALB and drought stress in poplars of China’s Three-North Shelterbelt. These data were analyzed using a Partial Least Squares-Support Vector Machine (PLS-SVM). The results showed that the LiDAR metric (elev_sqrt_mean_sq) was key in detecting drought, while the hyperspectral band (R970) was key in ALB detection, underscoring the necessity of integrating both sensors. Detection of ALB in poplars improved when the poplars were well watered. The classification accuracy was 94.85% for distinguishing well-watered from water-deficient trees, and 80.81% for detecting ALB damage. Overall classification accuracy was 78.79% when classifying four stress types: healthy, only ALB affected, only drought affected, and combined stress of ALB and drought. The results demonstrate the effectiveness of UAV-hyperspectral and LiDAR data in distinguishing ALB and drought stress in poplar forests, which contribute to apply targeted treatments based on the specific stress in poplars in northwest China.

Rapid and non-destructive determination of oil content in fuzzy cottonseeds via near-infrared reflectance spectroscopy

In China, enterprises specializing in cottonseed processing predominantly acquire fuzzy cottonseeds. However, the expeditious determination of oil content within these fuzzy cottonseeds remains a major challenge. This particular challenge considerably influences the economic efficiency of enterprises engaged in cottonseed processing. To address this concern, this research acquired near-infrared spectral (NIRS) information from fuzzy cottonseeds and applied the Standard Normal Variate (SNV) pretreatment to eliminate data noise. Subsequently, the Uninformative Variable Elimination (UVE) was employed to select feature wavelength points reflecting fuzzy cottonseed oil content. To further accurately extract the feature wavelengths of fuzzy cottonseeds, Competitive Adaptive Reweighted Sampling (CARS) and Random Frog (RF) were utilized for a second round of feature wavelength extraction. Finally, prediction models for fuzzy cottonseed oil content were established using partial least squares regression (PLSR), Support Vector Regression (SVR), and Least Squares Support Vector Machines (LSSVM). Results revealed that the LSSVM model constructed using the feature selected by the UVE-CARS had the best performance in predicting the oil content of fuzzy cottonseeds. The values of Coefficient of Determination (R2), Root Mean Square Error (RMSE), Residual Error Ratio (RER), and Residual Predictive Deviation (RPD) were 0.8605, 0.0075, 12.8116, and 3.8154, respectively. The LSSVM model built on the feature wavelength points selected by the UVE-RF demonstrated the second-best prediction performance with an R2 of 0.8329, an RMSE of 0.0091, an RER of 10.4861, and an RPD of 3.1228. The validation experiments corroborated these findings, it was found that the R2 was 0.8122, the RMSE was 0.0063, The RER was 7.6190, and the RPD was 2.9036. This underscores the effectiveness of the method employed in this study for detecting the oil content in fuzzy cottonseeds. The approach presented in this study serves as a crucial reference for the swift and accurate assessment of oil content in fuzzy cottonseeds, providing an impactful technical solution for cottonseed processing enterprises to enhance their economic outcomes.

A Mechanical Fault Diagnosis Method for UCG-Type On-Load Tap Changers in Converter Transformers Based on Multi-Feature Fusion

The On-Load Tap Changer (OLTC) is the only movable mechanical component in a converter transformer. To ensure the reliable operation of the OLTC and to promptly detect mechanical faults in OLTCs to prevent them from developing into electrical faults, this paper proposes a fault diagnosis method for OLTCs based on a combination of Particle Swarm Optimization (PSO) algorithm and Least Squares Support Vector Machine (LSSVM) with multi-feature fusion. Firstly, a multi-feature extraction method based on time/frequency domain statistics, synchrosqueezed wavelet transform, singular value decomposition, and multi-scale modal decomposition is proposed. Meanwhile, the random forest algorithm is used to screen features to eliminate the influence of redundant features on the accuracy of fault diagnosis. Secondly, the PSO algorithm is introduced to optimize the hyperparameters of LSSVM to obtain optimal parameters, thereby constructing an optimal LSSVM fault diagnosis model. Finally, different types of feature combinations are utilized for fault diagnosis, and the impact of these feature combinations on the fault diagnosis results is compared. Experimental results indicate that features of different types can complement each other, making the OLTC state information carried by multi-dimensional features more comprehensive, which helps to improve the accuracy of fault diagnosis. Compared with four traditional fault diagnosis methods, the proposed method performs better in fault diagnosis accuracy, achieving the highest accuracy of 98.58%, which can help to detect mechanical faults in the OLTC early and reduce the system’s downtime.

Short-term Traffic Flow Prediction Based on KELM Optimized By Improved Slime Mould Algorithm

Traffic flow prediction plays a crucial role in improving transportation efficiency and enhancing Intelligent Transportation Systems (ITS). However, the temporal, spatial, and nonlinear nature of traffic flow data presents challenges for accurate short-term prediction. We propose a short-term traffic flow prediction model based on Kernel Extreme Learning Machine (KELM) optimized by the improved Slime Mould Algorithm (SMA). KELM is an improved version of Extreme Learning Machine (ELM) that incorporates kernel functions for improved generalization and stability. SMA is a meta-heuristic algorithm inspired by the behavior of slime mould in foraging, known for its strong global searching ability. For better performance, three strategies are introduced: the Good Point Set method for optimizing the initial population, the combination of Opposition Based Learning (OBL) and Differential Evolution (DE) to improve the slime mould generation mechanism, and the use of adaptive t distribution mutation to enhance convergence speed. After comparing the performance of these improved SMAs on twelve test functions, the ISMA improved by integrating three strategies as mentioned above is best. Then the ISMA is applied to search for the optimal parameters of KELM model. Finally, the optimized KELM with optimal parameters is applied to predict the short-term traffic flow on given traffic data set. Experimental results demonstrate that the proposed model, KELM optimized by ISMA namely ISMA-KELM, outperforms existing models such as Random Forest (RF), Least Squares Support Vector Machine (LSSVM), KELM optimized by Tuna Swarm Optimization Algorithm (TSO-KELM), and KELM optimized by SMA (SMA-KELM) in terms of traffic flow prediction accuracy. The proposed model ISMA-KELM provides a promising approach for addressing the challenges of traffic flow prediction, offering improved accuracy and efficiency in real-time traffic management systems.

Detection of hypoglycaemia in type 1 diabetes through breath volatile organic compound profiling using gas chromatography-ion mobility spectrometry.

To evaluate the relationship between breath volatile organic compounds (VOCs) and glycaemic states in individuals with type 1 diabetes (T1D), focusing on identifying specific VOCs as biomarkers for hypoglycaemia to offer a non-invasive diabetes-monitoring method. Ten individuals with T1D underwent induced hypoglycaemia in a clinical setting. Breath samples, collected every 10-15 minutes, were analysed using gas chromatography-ion mobility spectrometry (GC-IMS). Correlation analysis and machine learning models, including Partial Least Squares Discriminant Analysis (PLS-DA) and Support Vector Machine classifiers, were used to classify glycaemic states based on VOC profiles. Statistical analysis revealed moderate correlations between specific VOCs (e.g. isoprene, acetone) and venous blood glucose levels. Machine learning models showed high accuracy in classifying glycaemic states, with the best performance achieved by a two-class PLS-DA model showing an accuracy of 93%, sensitivity of 92% and specificity of 94%. Key biomarkers identified included isoprene, acetone, 2-butanone, methanol, ethanol, 2-propanol and 2-pentanone. This study shows the potential of breath VOCs to accurately classify glycaemic states in individuals with T1D. While key biomarkers such as isoprene, acetone and 2-butanone were identified, the analysis emphasizes the importance of using overall VOC patterns rather than individual compounds, which can be markers for multiple conditions. Machine learning models leveraging these patterns achieved high accuracy, sensitivity and specificity. These findings suggest that breath analysis using GC-IMS could be a viable non-invasive method for monitoring glycaemic states and managing diabetes.

Discrete ripplet-II transform feature extraction and metaheuristic-optimized feature selection for enhanced glaucoma detection in fundus images using least square-support vector machine

Discrete ripplet-II transform feature extraction and metaheuristic-optimized feature selection for enhanced glaucoma detection in fundus images using least square-support vector machine

A novel infrared thermography image analysis for transformer condition monitoring

Electrical systems are deeply ingrained in most industrial facilities, their maintenance is increasingly becoming a critical and significant component of economic policy. Condition monitoring of electrical transformers is essential for improving their dependability and availability, averting costly maintenance and additional significant breakdowns. This research adopts the approach based on infrared thermography techniques (IRT) to keep eye on electrical transformers and detect their defects. Thermal images of the transformer were captured at two distinct operating states with an infrared camera. These images were then compiled into a dataset for further analysis. This method uses infrared imaging (IRT), along with feature analysis and machine learning, to identify issues in electrical transformers in a new way. To find the best performing machine learning model, different techniques are compared here in terms of their accuracy and stability. Two approaches are investigated for identifying features in thermography images. Approach-1 employed five common machine learning algorithms, such as Support Vector Machine (SVM), K-Nearest Neighbours (KNN), Decision Tree (DT), Logistic Regression (LR), and Least Squares Support Vector Machine (LS-SVM). Approach-2 utilized four deep learning techniques, such as MobileNetV2 (MNV2), InceptionV3 (InV3), DenseNet121(DN121), and our proposed modified VGG-16. Among all evaluated methods, the modified VGG-16 architecture achieved the highest level of dependability, demonstrating exceptional efficiency and accuracy in transformer condition monitoring and fault diagnosis.

Nutritional Quality Analysis and Classification Detection of Buckwheat in Different Harvest Periods.

For buckwheat, the optimal harvest period is difficult to determine-too early or too late a harvest affects the nutritional quality of buckwheat. In this paper, physical and chemical tests are combined with a method using near-infrared spectroscopy nondestructive testing technology to study buckwheat harvest and determine the optimal harvest period. Physical and chemical tests to determine the growth cycle were performed at 83 days, 90 days, 93 days, 96 days, 99 days, and 102 days, in which the buckwheat grain starch, fat, protein, total flavonoid, and total phenol contents were assessed. Spectral images of buckwheat in six different harvest periods were collected using a near-infrared spectral imaging system. Four preprocessing methods (SNV, S-G, DWT, and the normaliz function) and three dimensionality reduction algorithms (IVSO, VCPA, VISSA) were used to process the raw buckwheat spectral data, and the full and eigen spectra were established as a random forest (RF). Random forest (RF) and Least Squares Support Vector Machine (LS-SVM) classification models were used to determine the full and eigen spectra, respectively, and the optimal model for the buckwheat single harvest period was determined and validated. Through physical and chemical tests, it was concluded that the 90-day harvest buckwheat grain protein, fat, and starch contents were the highest, and that the total flavonoid and total phenolic contents were also high. The SNV preprocessing method was the most effective, and the feature bands extracted using the IVSO algorithm were more representative. The IVSO-RF model was the best discriminative model for the classification of buckwheat in different harvest periods, with the correct rates of the training and prediction sets reaching 100% and 96.67%, respectively. When applying the IVSO-RF model to the buckwheat single harvest period to verify the classification, the correct rate of the training set for each harvest period reached 96%, and that of the prediction set reached 100%. Near-infrared spectroscopy combined with the IVSO-RF modeling method for buckwheat harvest period detection is a rapid, nondestructive classification method. When this was combined with physical and chemical analyses, it was determined that a growth cycle of 90 days is the best harvest period for buckwheat. The results of this study can not only improve the quality of buckwheat crops but also be applied to other crops to determine their optimal harvest period.

Optimized Random Forest Method for 3D Evaluation of Coalbed Methane Content Using Geophysical Logging Data.

Accurate evaluation of coalbed methane (CBM) content is crucial for effective exploration and development. Traditional gas content measurement methods based on laboratory analysis of drill core samples are costly, whereas geophysical logging methods offer a cost-effective alternative by providing continuous high-resolution profiles of rock layer physical properties. However, the relationship between CBM content and geophysical logging data is complex and nonlinear, necessitating an advanced prediction method. This study focuses on the No. 3 coal seam in the Shizhuang South Block of the Qinshui Basin, utilizing geophysical logging data and 148 sets of laboratory core samples. We employed the Random Forest (RF) method optimized with a simulated annealing-genetic algorithm (SA-GA) to develop the SA-GA-RF model for evaluating CBM content. The model's performance was validated using test data and new CBM well data, and it was applied to calculate the vertical gas content profiles of No. 3 coal seam across 128 wells. The SA-GA-RF model demonstrated an average relative error of 13.13% in the test data set, outperforming Backpropagation Neural Network (BPNN), Least Squares Support Vector Machine (LSSVM), Extreme Learning Machine (ELM), and multivariate regression (MR) methods. The model also exhibited strong generalizability in new wells and improved model-building efficiency compared to traditional cross-validation grid search methods. The construction of a three-dimensional CBM content model, incorporating well coordinates and elevation data, allowed for detailed identification of high gas content areas and layers. This three-dimensional model offers a more precise characterization than traditional two-dimensional isopleth maps, providing valuable insights for CBM exploration, reserve evaluation, and production optimization.

Hybrid intelligent and numerical methods to estimate the transmission coefficients of rectangular floating breakwaters

ABSTRACT Breakwaters are used to reduce incoming wave energy at harbors and shorelines. This paper presents a comparison of novel two-dimensional hybrid intelligent models for the idealization of the effects of waves on the performance of moored rectangular floating breakwaters (FBs). Fluid structure interactions (FSIs) were idealized by airy-type monochromatic regular waves generated in a numerical wave tank. The coupled Volume of Fluid-Fast Fictitious Domain (VOF-FFD) interpolation method was used to evaluate FB motions. Different forms of Least Squares Support Vector Machine Methods (LSSVMs) that utilized 183 data streams were used to model FB performance for different wave height-to-water depth ratios, dimensional aspect ratios, and specific length-to-water depth ratios. Of those, 80% were used to train the model and 20% to test it. Parametric studies have shown that during training a Least Squares Support Vector Machine Method-Bat Algorithm (LSSVM-BA) with R2 = 0.8725, MAE = 0.0276, and RMSE = 0.0488 presents the most appropriate model for the evaluation of FB performance. Notwithstanding this, during testing a Least Squares Support Vector Machine Method-Cuckoo Search (LSSVM-CS) Algorithm with corresponding values of 0.6841, 0.0519, and 0.0708 performs better.

Compressive strength detection of tunnel lining using hyperspectral images and machine learning

Traditional tunnel lining strength detection techniques are mostly contact-based, with relatively low detection efficiency. This study innovatively proposes hyperspectral imaging method to rapidly detect tunnel concrete lining strength from a machine vision perspective. Hyperspectral cameras were employed in indoor experiments to capture hyperspectral images of concrete specimens with different compressive strength levels. The differences of concrete strength based on hyperspectral reflectance characteristics were analysed using hyperspectral images and machine learning algorithms. Firstly, the K-Nearest Neighbors (KNN) classification algorithm was used to predict the classification of the concrete hyperspectral dataset with accuracy mostly exceeding 90 %. The results indicate distinctive differences in hyperspectral reflectance characteristics among concrete specimens. Furthermore, compressive strength prediction of different concrete specimens was carried out using Principal component regression (PCR), Partial least squares regression (PLSR), and Least Squares Support Vector Machine (LSSVM) machine learning models on both original and Savitzky-Golay(S-G) processed spectral data. LSSVM and PLSR models performed excellently in the visible light spectrums(400–1000 nm), with LSSVM excelling in the near-infrared spectrums(900–1700 nm). Finally, the feasibility of using Hyperspectral imaging(HSI) technology to detect tunnel lining strength was demonstrated at the shield tunnel model site. The predicted results were validated by combining with strength values measured by the rebound hammer, and presented promoising research path for automated non-destructive detection of concrete tunnel lining strength.

Machine learning models applied to altimetry era tide gauge and grid altimetry data for comparative long-term trend estimation: A study from Shikoku Island, Japan

Estimation of sea level trends is essential for understanding sea level rise dynamics. In this study, the performance of traditional Ordinary Least Squares (OLS) linear trend forecasting is compared with modern machine learning techniques, specifically Random Forests (RF) and Least Squares Support Vector Machines (LS-SVM).These methods are applied to 50 years of long-term tide gauge (TG) data from six tide gauge stations off the coast of Shikoku Island, Japan, and CMEMS Grid Altimetry data from 1993 to the present. The analysis uses OLS, RF, and LS-SVM to estimate trends from both data sets and compares the results. The objective is to determine the consistency and accuracy of RF and LS-SVM methods compared to the OLS method. The results indicate that machine learning algorithms (LS-SVM) effectively estimate sea level trends, offering potential improvements in precision for both long-term and medium-term analyses. Shikoku Island's coastal sea level trend is determined as 2.91±1.44 mm/yr using TG data and 3.00±1.52 mm/yr using CMEMS Grid Altimeter data with the OLS approach. Using the LS-SVM approach, the trend is found as 2.96±1.58 mm/yr with TG data and 3.02±1.60 mm/yr with CMEMS Grid Altimetry data. The novelty of this study lies in its thorough comparison of traditional and machine learning approaches for sea level trend estimation, providing valuable insights for future predictions of the sea level rise.

A novel optimization rainfall coupling model based on stepwise decomposition technique

Traditional decomposition integration models decompose the original sequence into subsequences, which are then proportionally divided into training and testing periods for modeling. Decomposition may cause data aliasing, then the decomposed training period may contain part of the test period data. A more effective method of sample construction is sought in order to accurately validate the model prediction accuracy. Semi-stepwise decomposition (SSD), full stepwise decomposition (FSD), single model semi-stepwise decomposition (SMSSD), and single model full stepwise decomposition (SMFSD) techniques were used to create the samples. This study integrates Variational Mode Decomposition (VMD), African Vulture Optimization Algorithm (AVOA), and Least Squares Support Vector Machine (LSSVM) to construct a coupled rainfall prediction model. The influence of different VMD parameters α is examined, and the most suitable stepwise decomposition machine learning coupled model algorithm for various stations in the North China Plain is selected. The results reveal that SMFSD is relatively the most suitable tool for monthly precipitation forecasting in the North China Plain. Among the predictions for the five stations, the best overall performance is observed at Huairou Station (RMSE of 18.37 mm, NSE of 0.86, MRE of 107.2%) and Jingxian Station (RMSE of 24.74 mm, NSE of 0.86, MRE of 51.71%), while Hekou Station exhibits the poorest performance (RMSE of 25.11 mm, NSE of 0.75, MRE of 173.75%).

Lithium-Ion Battery Health State Estimation Based on Feature Reconstruction and Optimized Least Squares Support Vector Machine

Abstract The state of health (SOH) of a battery is the main indicator of battery life. In order to improve the SOH estimation accuracy, a model framework for lithium-ion battery health state estimation with feature reconstruction and improved least squares support vector machine is proposed. First, the indirect health features (HF) are obtained by processing multiple health features extracted from the charging and discharging phases through principal component analysis to remove the information redundancy among multiple features. Subsequently, multiple smooth component subsequences of different frequencies are obtained by using variational modal decomposition to efficiently capture the overall downtrend and regeneration fluctuations of the data. Then, use the sparrow search algorithm to optimize the least squares support vector machine to build an estimation model, predict and superimpose the reconstructed fusion features of multiple feature subsequences. Finally, use the mapping relationship between the reconstructed HF and the SOH for the estimation. The NASA battery dataset and the University of Maryland battery dataset (CACLE) are used to perform validation tests on multiple batteries with different cycle intervals. The results show that the mean absolute error and root mean square error are less than 1% and the method has high-estimation accuracy and robustness.

Feature extraction method for ultrasonic pipeline defects based on fractional-order VMD

ABSTRACT Ensuring the safety of pipeline transportation is vital for societal well-being. Traditional methods for identifying defects in steel pipelines through ultrasonic echo signal pattern recognition often fail to extract comprehensive and effective features crucial for accurate defect detection. This study introduces an innovative feature extraction method employing a fractional Fourier transform variational modal decomposition (FRFT-VMD), hereafter referred to as the fractional-order VMD algorithm. This method utilises the fourth-order central moment and envelope entropy to optimise several key parameters: the fractional order in the Fourier transform, the number of decomposition layers, and the penalty factor in variational modal decomposition. To evaluate the effectiveness of the proposed method, ultrasonic echo signals from both finite element simulations and experimental platforms were analysed using the FRFT-VMD technique. The features extracted were then classified using a Least Squares Support Vector Machine (LSSVM) to determine defect depths. The results show a recognition accuracy of 95.2% in simulated signals and 89.1% in experimentally measured signals across various defect depths, indicating a significant improvement over existing feature extraction methodologies. The fractional-order VMD algorithm proves to be superior in extracting features that enhance the accuracy of defect identification in steel pipelines.

A generation of synthetic samples and artificial outliers via principal component analysis and evaluation of predictive capability in binary classification models

Unbalanced sample groups tend to yield models with a higher prevalence of predominant classes. A sample group with balanced classes contributes to the development of more robust models with improved predictive capability to classify classes equally. In the literature, two methodologies for sample balancing can be found: elimination (undersampling) and synthetic sample generation (oversampling). Undersampling methodologies result in the loss of real samples, while oversampling methods may introduce issues related to adding non-real signals to the original spectra. To overcome these challenges, this paper aimed to utilize Principal Component Analysis (PCA) for the generation of virtual samples (synthetic samples and artificial outliers) to balance data in multivariate classification models. The proposed methodology was applied to data from mid-infrared spectroscopy (MIR) and high-resolution mass spectrometry (HRMS) with Partial Least Squares Discriminant Analysis (PLS-DA) and Support Vector Machine (SVM) models. The constructed models demonstrate that the addition of virtual samples enhances performance parameters (e.g., false negative rate, false positive rate, accuracy, sensitivity, specificity, among others) compared to unbalanced models, while also mitigating overfitting (a problem found in unbalanced models). Performance parameters exhibited a more significant improvement percentage using the non-linear model (SVM) compared to the linear model (PLS-DA). Furthermore, the created virtual spectra do not introduce new signals, i.e., original, and virtual spectra exhibit a similar spectral profile, differing only in the intensity levels. Finally, all models demonstrated good predictive capability according to permutation testing for the binary model developed in this work, limiting the rate of class permutation retention (between 40 % and 60 % of the y-vector remained in the original class). All created models exhibited accuracy values higher than the accuracy distribution of models with permuted classes for the test group.

Application of four intelligent models for the estimation of formation pressure of natural gas hydrate

Natural gas hydrate is a clean and efficient resource that contains lots of heat energy. The production, processing and transportation of natural gas can be influenced dramatically by natural gas hydrate. Natural gas hydrate will accumulate at the bottom of the pipeline, even block the pipeline and reduce the flow rate of the natural gas. This article presents four intelligent models named GWO-SVM (Grey Wolf Optimizer connected with Support Vector Machine), GWO-LSSVM (Grey Wolf Optimizer connected with Least Squares Support Vector Machine), GA-LSSVM (Genetic Algorithm connected with Least Squares Support Vector Machine), and GWO-RBF (Grey Wolf Optimizer connected with Radial Basis Function) to predict the formation pressure of natural gas hydrate in different systems. In order to determine the degree of influence of each input variable on the formation pressure of natural gas hydrate, this paper uses Spearman correlation analysis method to calculate the correlation coefficients between the formation pressure of natural gas hydrate and various influencing factors. The contradistinction among different intelligent models shows that the GWO-LSSVM model has the exceptional performance of the smallest average absolute relative deviation (AARD=2.89%), the minimum standard deviation (SD=0.05) with all 244 data of natural gas hydrate formation pressure in different systems. When compared to six commonly used empirical models for predicting formation pressure in common natural gas hydrate, the GWO-LSSVM model has the highest prediction accuracy, with an AARD of only 1.82%. Furthermore, when compared to three thermodynamic models employed for predicting formation pressure in sour natural gas hydrate, the GWO-LSSVM model has the highest prediction accuracy, with an AARD of 2.38%. The leverage method demonstrates that four intelligent models have excellent robustness in reflecting the inner relationships among natural gas hydrate formation pressure and influential factors. The four intelligent models for predicting hydrate formation pressure will be helpful to find hydrate formation pressure in different systems.