Support Vector Regression Method Research Articles

Optimization studies on batch extraction of phenolic compounds from Azadirachta indica using genetic algorithm and machine learning techniques

Phenolic compounds play a crucial role as secondary metabolites due to their substantial biological activity and medicinal value. These compounds are present in various parts of plant species. This study focused on solid-liquid batch extraction to recover total phenolic compounds from Azadirachta indica leaves. The experimental design was based on the Taguchi L16 array, considering four independent factors: extraction time, temperature, particle size, and solid-to-solvent ratio. Among these factors, the particle size exerted the maximum influence. Particle size inversely affects the yield of total phenolic content (TPC), while temperature, time, and solid-to-liquid ratio have a direct impact. The process factors concerned were investigated both experimentally and through machine learning techniques. Support vector regression (SVR) and random forest method (RFM) algorithms were utilized for predicting TPC, while a genetic algorithm (GA) was employed to derive optimal process parameters. The GA predicts the optimal extraction factors, yielding the maximum TPC. During this study, these factors were the following: particle size of 0.15 mm, extraction time of 40 min, solid-to-liquid ratio of 1:25 g/mL, and a temperature of 55 °C, with a predicted value of 23.039 mg GAE/g of plant material. Notably, in this study, the SVR values of TPC yield closely matched the experimental values for the training and test data set when compared with the random forest method values.

A machine learning approach for the estimation of photocatalytic activity of ALD ZnO thin films on fabric substrates

Research in the field of photocatalytic wastewater treatment is striving to enhance catalyst materials to achieve high-performance systems. A promising approach to this goal has been immobilizing photocatalytic materials on fibrous substrates via atomic layer deposition (ALD). Nevertheless, both the ALD process and the assessment of photocatalytic performance involve a multitude of parameters necessitating thorough investigation. In this study, we employ popular machine-learning algorithms, including Support Vector Regression (SVR) and Artificial Neural Networks (ANN), to predict the photocatalytic activity of ALD-coated textiles. The photocatalytic activity is evaluated through methylene blue and methyl orange degradation tests. Machine learning algorithms are tested and trained using the k-fold cross-validation technique. The findings demonstrate that the ANN and SVR methods utilized in this research can predict catalytic activity with mean absolute percentage errors (MAPE) of 2.35 and 3.25, respectively. This study illuminates that, within the defined range of process parameters, the photocatalytic activity of ALD-coated textiles can be precisely estimated with suitable machine-learning algorithms.

Using a hybrid approach for wind power forecasting in Northwestern Mexico

Wind energy is an important renewable source that has been considerably developed recently. In order to obtain successful 24-h lead-time wind power forecasts for operational and commercial uses, a combination of physical and statistical models is desirable. In this paper, a hybrid methodology that employs a numerical weather prediction model (Weather Research and Forecasting) and a neural network (NN) algorithm is proposed and assessed. The methodology is applied to a wind farm in northwestern Mexico, a region with high wind potential where complex geography adds large uncertainty to wind energy forecasts. The energy forecasts are then evaluated against actual on-site power generation over one year and compared with two reference models: decision trees (DT) and support vector regression (SVR). The proposed method exhibits a better performance with respect to the reference methods, showing an hourly normalized mean absolute percentage error of 6.97%, which represents 6 and 13 percentage points less error in wind power forecasts than with DT and SVR methods, respectively. Under strong synoptic forcing, the NN wind power forecast is not very accurate, and novel approaches such as hierarchical algorithms should be employed instead. Overall, the proposed model is capable of producing high-quality wind power forecasts for most weather conditions prevailing in this region and demonstrates a good performance with respect to similar models for medium-term wind power forecasts.

Online Dynamic Modelling for Digital Twin Enabled Sintering Systems: An Iterative Update Data-Driven Method

The sintering process is a crucial thermochemical process in the blast furnace iron-making system. Tumble strength (TS), as a vital performance to assess sinter quality, is difficult to monitor due to the lack of timely measurement. Constructing a data-driven model for TS is an alternative for monitoring TS. However, the time-varying dynamic sintering process makes the task of modelling challenging. And the data are incomplete and insufficient in practice for modelling since there are unknown time delays in the system and lack actual TS value. The digital twin (DT) technique is a powerful tool to simulate the system dynamics with the real-time interaction between physical processes and virtual agents in cyberspace. This paper introduces a DT-enabled equivalent of the sintering system and proposes online data-driven modelling for TS monitoring. The time delay in the system is estimated for variable sequence alignment based on a modified maximum information coefficient method. The data used for modelling is enriched based on a multi-source information fusion technique. An adaptive update method is proposed to deal with the time-varying dynamics. The iterative forgetting factor-based algorithm is designed for the support vector regression method and guarantees a fast computational speed. Implementation and validation of the model on a DT-enabled sintering system show the efficiency of the proposed method. The accuracy of TS monitoring reaches 99.6% by analysis of 3 months’ data.

A hybrid approach combining data envelopment analysis and recurrent neural network for predicting the efficiency of research institutions

The traditional data envelopment analysis (DEA) method relies solely on current and historical performance and is unable to predict future efficiency. To overcome this limitation, we propose a novel prediction model called the input–output memory network (IOMN), which extends the Recurrent Neural Network (e.g., Long Short-Term Memory (LSTM) and Gate Recurrent Unit (GRU) to forecast the input and output variables in future periods. We then combined the IOMN with the DEA-Malmquist index to evaluate the efficiency of Decision-Making Units (DMUs) from the past to the future. To demonstrate the effectiveness of our proposed model, we conducted a case study of 15 research institutions in the Chinese Academy of Sciences (CAS) and compared its performance with those of linear regression (LR) and support vector regression (SVR) methods. The comparative results show that the IOMN significantly outperformed both the LR and SVR methods. Finally, we fed the past and predicted future data into the DEA-Malmquist index to examine the productivity efficiency trends in research institutions over time. Our results indicate that the annual mean of total factor productivity (TFP) continued to fluctuate at approximately 1 during the observation period. In the future, the geometric mean of TFP will decline slightly by 2.3%, the geometric mean of efficiency change (EC) will decline by 5%, and the geometric mean of technical change (TC) will increase by 2.9%.

Remaining useful life prediction of lithium-ion batteries using EM-PF-SSA-SVR with gamma stochastic process

Due to the complex changes in physicochemical properties of lithium-ion batteries during the process from degradation to failure, it is difficult for methods based on physical or data-driven models to fully characterize this nonlinear process, and existing methods that hybridize physical and data-driven models suffer from ambiguous hybridization, which results in the vast majority of existing methods for predicting the remaining useful life (RUL) of lithium-ion batteries suffering from a lack of accuracy and robustness. In this study, a novel hybrid approach based on empirical modeling and data-driven techniques is proposed for predicting the RUL of lithium-ion batteries. To better capture its complexity, stochasticity, and state transition, and improve the modeling accuracy and RUL prediction precision, Gamma stochasticity and state-space modeling are used to empirically model the complex Li-ion battery degradation process. Moreover, the expectation maximization (EM) method of particle filtering (PF) was used to estimate the hidden parameters of the empirical model, and the estimated parameters were corrected using an optimized support vector regression (SVR) method to enhance the generalization performance and robustness of the data-driven model. The results show that the gamma state-space model is effective in capturing the inherent stochastic properties of the battery degradation and the proposed hybrid method outperforms the existing prediction methods in RUL prediction. The experiments show that the sparrow search algorithm (SSA) optimized SVR is considered to be the most effective correction method for the estimated parameters, while the new EM-PF-SSA-SVR hybrid method provides better performance for state assessment and RUL prediction of lithium-ion batteries. It is indicated that the proposed EM-PF-SSA-SVR method with Gamma stochastic process has hybrid validity and superior performance with equal performance and less parameter computation relative to the existing state-of-the-art deep learning RUL prediction methods.

Seepage Prediction Model for Roller-Compacted Concrete Dam Using Support Vector Regression and Hybrid Parameter Optimization

In this study, a seepage prediction model was established for roller-compacted concrete dams using support vector regression (SVR) with hybrid parameter optimization (HPO). The model includes data processing via HPO and machine learning through SVR. HPO benefits from the correlation extraction capability of grey relational analysis and the dimensionality reduction technique of principal component analysis. The proposed model was trained, validated, and tested using 22 years of monitoring data regarding the Shuidong Dam in China. We compared the performance of HPO with other popular methods, while the SVR method was compared with the traditional time-series prediction method of long short-term memory (LSTM). Our findings reveal that the HPO method proves valuable real-time dam safety monitoring during data processing. Meanwhile, the SVR method demonstrates superior robustness in predicting seepage flowrate post-dam reinforcement, compared with LSTM. Thus, the developed model effectively identifies the factors related to seepage and exhibits high accuracy in predicting fluctuation trends regarding the Shuidong Dam, achieving a determination coefficient R2 > 0.9. Further, the model can provide valuable guidance for dam safety monitoring, including diagnosing the efficacy of monitoring parameters or equipment, evaluating equipment monitoring frequency, identifying locations sensitive to dam seepage, and predicting seepage.

Evaluation of Deep Learning Algorithms for KASE Index Prediction

Security indexes are the primary instruments that are used to analyze the circumstances of the financial market. A considerable proportion of a country’s total wealth is held in the form of investments in its stock market. If future stock market patterns can be accurately predicted, it may be possible for investors to optimize the returns they receive on their investments. Yet, as a result of nonlinearity and nonstationarity, it is difficult to accurately forecast financial data. The purpose of this investigation is to determine how well predictive machine learning models perform in relation to the stock market. This analysis makes use of daily close price data for the KASE index on the Kazakhstan Stock Exchange from March 2018 all the way through March 2023. In order to make accurate predictions, we use a total of four different machine learning algorithms. According to the findings, the deep learning model fares far better than the competition in terms of the accuracy of its predictions. In addition, the neural network approach and the random forest method are considered to be the best, while the support vector regression method is placed second. Criteria for evaluation include mean absolute error (also known as MAE), mean square error (also known as MSE), and root mean square error (RMSE).

Application of XGB-based metaheuristic techniques for prediction time-to-failure of mining machinery

Mining equipment is a critical component in the success of mining operations, and unplanned downtime can be costly. Efficient and intelligent maintenance planning is therefore essential to minimize downtime and maximize productivity. Consecutive time-to-failure (TTF) is an important indicator of machine reliability, and accurate TTF predictions enable effective preventive maintenance planning. Therefore, this study proposes hybrid models of extreme gradient boosting (XGB), optimized by particle swarm optimization (PSO) and gray wolf optimization (GWO), for the prediction of TTF in mining machinery. Additionally, validation of the hybrid model was conducted using support vector regression (SVR) method. Historical data on mining machine failures were collected, and a case study was conducted to investigate shovels in an open-pit mine. The PSO-XGB method was found to be the most accurate predictor of failure time with R2 values of 0.99, RMSE values of 50.66 and 51.77, MAE values of 4.52 and 10.81, and AARE values of 1.15 and 1.24 in the training and testing phases. This research highlights the importance of efficient and intelligent maintenance planning to minimize downtime and optimize productivity in mining operations.

Machine learning based pavement performance prediction for data-driven decision of asphalt pavement overlay

This study is to develop pavement performance models using support vector regression and ensemble machine learning methods for selection of pavement overlay strategy. Predictive models of pavement distresses were developed based on the Long-Term Pavement Performance (LTPP) data and compared using support vector machine, random forest regression, gradient boosting machine, and stacking ensemble. Gradient boosting machine was found to be a more effective method to establish predictive models of rut depth and International Roughness Index. Stacking ensemble and random forest regression would provide reliable prediction of alligator cracking. The models developed based on the clusters of climate and traffic parameters were found to be more effective. Based on the developed performance models, the effects of asphalt overlays on pavement distresses and service lives were investigated. When the overlay with recycled asphalt concrete (AC) was applied, the propagation of alligator cracking was faster compared to the overlay with virgin asphalt mixture. Milling before overlay tended to slow the increase of IRI but fasten the development of rut depth.

Conflation of modal strain energy, spectral finite element, and machine learning techniques for damage detection in multi-span steel girder bridges with variable sections

A new approach is proposed to detect micro and small damages in multi-span bridges constructed with variable sections by combining modal strain energy and spectral finite element through the use of machine learning techniques. The traditional finite element methods are inefficient in detecting and estimating structures' micro-damages. Therefore, wave propagation-based methods must be applied. Micro-damages can be modeled through the spectral finite element method, widely adopted in wave propagation simulation. Among the frequency-based methods used in damage identification, the modal strain energy is more sensitive, especially in complicated structures. However, this method is difficult to apply to steel girders with variable sections and different support conditions. By combining the modal strain energy and the spectral finite element and applying the support vector regression method, the location and severity of damage in steel girders could be accurately identified to overcome the difficulties of detection in complicated elements. To improve the ability of the support vector regression method, in the damage severity estimation, it was combined with feature selection. To evaluate the proposed method's ability, first, an element of the beam- type, with the characteristics of a spectral finite element, is defined in the OpenSees software. To assess the accuracy of the defined element in detecting micro-damages, a tested numerical model was adopted to measure the accuracy of the provided element. Next, to detect small damages, the model of the steel girder bridge with the defined element was prepared, and the modal properties of the structure were extracted. Due to the change in cross-section along each girder, the modified modal strain energy-based damage index (β) equation was used in the subject model. By applying the damages of different severities in steel girders in single and multiple damage scenarios, the modified damage index β was calculated. With the damage indexes calculated in the different scenarios and applying the machine learning method, the severity of damage was estimated. The obtained results indicate that this proposed method is accurate and satisfactory for detecting and estimating micro and small damages with high accuracy in bridges with variable section girders.

A Novel Support Vector Regression Algorithm for Processing Experimental Data with Perturbation

Processing experimental data with uncertainty is a huge challenge for robust regression modeling and high-accuracy data forecasting. The objective of the present study is the development of a robust support vector regression model for processing observed data of independent variables containing polyhedral perturbation and achievement of high forecasting precision. Firstly, the conception of the data collection of polyhedral perturbation is given and analyzed. Secondly, a novel robust support vector regression model is constructed by replacing the original independent variables in a linear support vector regression model with the independent variables containing polyhedral perturbation. Thirdly, the novel robust linear support vector regression method is also expanded to the non-linear regression model. Both the two models are validated by the linear and nonlinear numerical regression experiments. Comparison of the experimental results show that the proposed method provide more accurate predictions than the traditional regression method for processing data with polyhedral perturbation.

Short-term passenger flow prediction during station closures in subway systems

Passenger flow prediction is critical for subway managers to efficiently organize passenger flow and assign capacity resources. Station closures caused by economic conferences and political events alter the topology of subway networks, resulting in large-scale passenger flows with irregular trends. Thus, predicting subway passenger flow during station closures is a challenging task. In this study, we propose a hybrid model for predicting short-term passenger flow during subway station closures. First, a spatiotemporal ensemble prediction model is used to capture the temporal characteristics of passenger flow at each station and the spatial characteristics of passenger flow among distinct groups of stations. Then, to address the effects of station closures, a support vector regression method is devised to investigate the residual time series from the ensemble model. Finally, a case study of Beijing subway is examined to validate the performance of the proposed model. The results indicate that the proposed model is superior to other state-of-the-art models in terms of accuracy and reliability in estimating the effects of station closures. Furthermore, the residuals from station closures are thoroughly converted into white noise and the proposed model is more interpretable than those in previous studies.

A comparative study of creep-fatigue life prediction for complex geometrical specimens using supervised machine learning

This study proposes a supervised machine learning approach to predict the creep-fatigue life of complex geometrical specimens. Seven different specimens were tested under creep-fatigue loading, and finite element analysis and test results showed that the stress distribution and life of the specimens were significantly influenced by the diameters and arrangement of holes. Characteristic parameters were proposed to describe the specimens' features, and support vector regression (SVR) and artificial neural network (ANN) methods were utilized to predict their life. The results indicate that both methods are effective in predicting the life of the specimens, with the ANN showing better performance when input data is limited. This study offers valuable insights into the leading factors behind the failure of complex geometrical specimens under creep-fatigue loading.

The Support Vector Regression Method Performance Analysis in Predicting National Staple Commodity Prices

Support Vector Regression (SVR) is a supervised learning algorithm to predict continuous variable values. The basic goal of the SVR algorithm is to find the most suitable decision line. SVR has been successfully applied to several issues in time series prediction. In this research, SVR is used to predict the price of staple commodity, which are constantly changing in price at any time due to several factors making it difficult for the public to get groceries that are easy to reach. National staple commodity data consisting of 17 commodities, including shallots, honan garlic, kating garlic, medium rice, premium rice, red cayenne peppers, curly red chilies, red chili peppers, meat of broiler chicken, beef hamstrings, granulated sugar, imported soybeans, bulk cooking oil, premium packaged cooking oil, simple packaged cooking oil, broiler chicken eggs, and wheat flour. With a data set for the last 3 years, including from January 1, 2020, to December 31, 2022. There are 3 variables in the data set, namely commodity, date, and price. This research divides the entire dataset into 80% training and 20% testing data. The results of this research show that SVR using the RBF kernel produces good forecasting accuracy for all datasets with an average Mean Square Error (MSE) training data of 6,005 while data testing is 6,062, Mean Absolute Deviation (MAD) of training data is 6,730 while data testing is 6.6831, Mean Absolute Percentage Error (MAPE) training data is 0.0148 while data testing is 0.0147, and Root Mean Squared Error (RMSE) training data is 7.772 while data testing is 7.746.

A novel quantitative measurement method for formation element via iterative support vector regression

Pulsed neutron sources provide notable advantages in well logging tools. They emit high-energy neutrons and can be controlled effectively. These emitted neutrons induce gamma rays, which can be detected and utilized to estimate diverse formation properties. One such application is the determination of elemental concentrations using gamma spectroscopy, enabling accurate characterization of mineralogy and total organic carbon in complex reservoirs. However, in field tests, the extraction of net inelastic energy spectra becomes challenging due to the diverse environment. Additionally, the measurement of actual gamma spectra is susceptible to the effects of formation conditions and detector performance, resulting in statistical errors and unavoidable noise within the measured gamma spectra. Consequently, the accuracy of elemental concentration estimation is reduced. To address these challenges, this paper proposes a novel method for accurately estimating formation elemental concentrations. Initially, the capture deduction coefficient is calculated by fitting the gamma time spectra iteratively, enabling the extraction of net inelastic gamma spectra in various formation environments. Subsequently, an iterative support vector regression (ISVR) method is proposed to obtain the stable and accurate non-negative elemental yields, even in the presence of gamma spectrum noise. Finally, the obtained elemental yields are converted to accurate elemental concentrations using the oxide closure model and pseudo-capture spectroscopy. The accuracy and robustness of proposed method are validated on simulated mixed spectra with high noise. Field test results are compared with core analysis results to assess the accuracy of the determined elemental concentrations.

Predicting minimum miscible pressure in pure CO2 flooding using machine learning: Method comparison and sensitivity analysis

CO2 injection for enhanced oil recovery (EOR) is widely recognized as an efficient technique for carbon capture, utilization, and storage (CCUS). This operation has a significant impact on various technical parameters, emphasizing the need to carefully consider and select the optimum approach. Among these factors, the minimum miscible pressure (MMP) plays a crucial role in determining the effectiveness and performance of CO2 injection. Therefore, this study aims to assess the reliability of machine learning (ML) in predicting the MMP of pure CO2 and examine the influence of different independent parameters. To achieve this, five ML methods were employed to predict the pure CO2 MMP, and the results were compared to statistical evaluations based on empirical correlations. In addition, three types of data with different functional input parameters were used in this research. Two types of data were obtained from existing literature, while the third category was collected from the thesis and PVT reports for specific Iraqi oil fields. The ML models were constructed by splitting the dataset into 20% for testing and 80% for training using Python programming. The significance of this study lies in its ability to identify the most efficient approach for forecasting MMP. The results of this work revealed that the K-nearest neighbors (KNN) model indicated the best statistical evaluation among the ML learning algorithms for two types of data (2) and (3) in predicting the MMP for pure CO2 flooding. This was evidenced by the lowest mean square error and the highest coefficient of determination. Additionally, the findings indicated that the support vector regression (SVR) method is an effective technique for smaller datasets. Moreover, the sensitivity analysis and assessment of the relative impacts of various input parameters revealed that the prediction of MMP is most sensitive to the composition of the injected gas and temperature, accounting for 46% and 28.5% of the variation, respectively. Finally, the presented ML models indicate exceptional accuracy, speed, adaptability in handling diverse conditions, and cost-effectiveness when compared to conventional approaches. These results verify the ability of ML models to provide high-quality predictions.

IMPLEMENTASI GRIDSEARCHCV PADA SUPPORT VECTOR REGRESSION (SVR) UNTUK PERAMALAN HARGA SAHAM

Stock is a sign of the capital participation of a person or authority in a company (PT). PT Anabatic Technologies Tbk (ATIC) is one of the service providers and IT consultants that is included in the technology sector, which is a new sector in the IDX-IC classification. ATIC stock trading was temporarily suspended due to a significant increase in cumulative prices. This indicates that stock prices tend to be volatile and non-linear. The Support Vector Regression (SVR) method can be used to predict stock prices. SVR is able to solve non-linear data problems by using kernel functions so it can overcome overfitting problems and will give good performance. The SVR problem is difficult to determine the optimal hyperparameters, so this research uses grid search cross validation (GridSearchCV). In this research, ATIC’s daily closing price data was used with 1007 training data and 252 testing data. The results show that the best model is SVR with a linear kernel and the hyperparameters used are Cost and epsilon . The linear kernel SVR model produces a MSE of 0,001237173; SMAPE of 0,1167301; and = 0,9206643

Optimal design of thermal cycles for experimental processing of advanced TRIP-assisted galvanized steels using support vector regression and kernel-based gradient evolution method

Optimal design of thermal cycles for experimental processing of advanced TRIP-assisted galvanized steels using support vector regression and kernel-based gradient evolution method

Support Vector Regression for Bluetooth Ranging in Multipath Environments

Ranging solutions for Internet of Things (IoT) localization applications seek to provide high accuracy with low cost of implementation. Among candidate IoT technologies that may fit this criterion, Bluetooth is a desirable choice as Bluetooth Low Energy (BLE) support is ubiquitous in modern smartphones, providing low implementation costs and low power consumption. Recent advancements in BLE ranging technology employ the Multi-Carrier Phase Difference technique which takes two-way channel frequency response (CFR) measurements. However, accurate ranging with these measurements is challenging due to many closely spaced multipath components from squaring the one-way CFR, a single or low number of snapshots, and model imperfections that arise in practical scenarios. To overcome these challenges, we propose a data-driven support vector regression (SVR) approach. Using real-world BLE measurements, our proposed SVR method demonstrates decimeter-level accuracy with single antenna devices, whereas Multiple Signal Classification (MUSIC), a popular model-based method, requires multiple antennas to obtain comparable performance. Moreover, we show robustness in different multipath environments including indoor, outdoor, and non-line-of-sight conditions, we determine generalization capabilities with training size, and we analytically establish the reduction in computational complexity compared to MUSIC.