Proposed Support Vector Regression Model Research Articles

Short-term photovoltaic power production forecasting based on novel hybrid data-driven models

The uncertainty associated with photovoltaic (PV) systems is one of the core obstacles that hinder their seamless integration into power systems. The fluctuation, which is influenced by the weather conditions, poses significant challenges to local energy management systems. Hence, the accuracy of PV power forecasting is very important, particularly in regions with high PV penetrations. This study addresses this issue by presenting a framework of novel forecasting methodologies based on hybrid data-driven models. The proposed forecasting models hybridize Support Vector Regression (SVR) and Artificial Neural Network (ANN) with different Metaheuristic Optimization Algorithms, namely Social Spider Optimization, Particle Swarm Optimization, Cuckoo Search Optimization, and Neural Network Algorithm. These optimization algorithms are utilized to improve the predictive efficacy of SVR and ANN, where the optimal selection of their hyperparameters and architectures plays a significant role in yielding precise forecasting outcomes. In addition, the proposed methodology aims to reduce the burden of random or manual estimation of such paraments and improve the robustness of the models that are subject to under and overfitting without proper tuning. The results of this study exhibit the superiority of the proposed models. The proposed SVR models show improvements compared to the default SVR models, with Root Mean Square Error between 12.001 and 50.079%. Therefore, the outcomes of this research work can uphold and support the ongoing efforts in developing accurate data-driven models for PV forecasting.

Evaluation of Weather Information for Short-Term Wind Power Forecasting with Various Types of Models

The rising share of renewable energy in the energy mix brings with it new challenges such as power curtailment and lack of reliable large-scale energy grid. The forecasting of wind power generation for provision of flexibility, defined as the ability to absorb and manage fluctuations in the demand and supply by storing energy at times of surplus and releasing it when needed, is important. In this study, short-term forecasting models of wind power generation were developed using the conventional time-series method and hybrid models using support vector regression (SVR) based on rolling origin recalibration. For the application of the methodology, the meteorological database from Korea Meteorological Administration and actual operating data of a wind power turbine (2.3 MW) from 1 January to 31 December 2015 were used. The results showed that the proposed SVR model has higher forecasting accuracy than the existing time-series methods. In addition, the conventional time-series model has high accuracy under proper curation of wind turbine operation data. Therefore, the analysis results reveal that data curation and weather information are as important as the model for wind power forecasting.

Impact of Cloud Reduction on MODIS Thermal Infrared Sea Surface Temperature Retrieval

Considering the significant influence of clouds on thermal infrared (TIR) data sea surface temperature (SST) retrieval, this study focuses on the reduction of cloud influence on Moderate Resolution Imaging Spectroradiometer (MODIS) longwave SST retrieval. First, the quality level (QL) 3 MODIS SSTs are classified into cloudy or clear-sky pixels based on MODIS cloud mask products. The cloudy SST pixels, flagged as confident cloudy or probably cloudy in the cloud masks, are further classified into three groups according to their associated daytime cloud-top and optical properties. Taking the SSTs measured by 11 buoys over one year as the reference data, those three groups of cloudy SSTs are significantly underestimated, with biases of −33.45 °C, −6.35 °C, and −4.72 °C. The QL 3 clear-sky SSTs are identified as probably and confident clear in the cloud masks and are not influenced by clouds, with a bias of nearly 0 °C. Then, three support vector regression (SVR) models are individually proposed for the three groups of cloudy SSTs. Taking the cloud-top and optical parameters as inputs of the proposed SVR models, we can obtain SSTs with a bias of 0 °C and a root-mean-square error (RMSE) of less than 1.6 °C for the three groups of cloudy SSTs. The method proposed in this study shows the potential for TIR SST estimation under some cloudy conditions using satellite remote sensing cloud products. Considering the RMSE of 0.4 °C in operational sea surface temperature and sea ice analysis, further study is needed before actual applying the method proposed in this study.

Demulsification of crude oil emulsions using ionic liquids: A computational intelligence approach

Ionic liquids have enormous applications in various areas of technology. Among their usage is for the destabilization of crude oil emulsions produced in oilfields. Herein, we propose a support vector regression (SVR) based model for smart screening and prediction water/oil separation driven by ionic liquids demulsifiers. The proposed SVR model applies attributes such as crude oil-water volumes, asphaltenes/resins content, emulsification speed/time, demulsification temperature, ionic liquid concentration, ionic liquid molecular weight, and demulsification time to predict the extent of demulsification performance of representative ionic liquids. Accordingly, the predicted demulsification efficiencies of the assessed ionic liquids exhibited significant matches with the experimental results. The developed approach demonstrated excellent accuracy as indicated by the root mean square error (RSME) values: 4.0123 and 19.6478 for the training and testing datasets, respectively. Additionally, the model demonstrated a considerable correlation coefficient (R2): 97.86 % and 75.97 % for the demulsification efficiency of tested ionic liquids in the training and testing datasets, respectively, thereby consolidating an appreciable agreement between the measured and the predicted results. It is envisaged that the SVR model employed in this study would greatly enhance the smart screening of ionic liquids used for demulsification activities in the petroleum and other related industries.

Continuous Blood Pressure Estimation From Non-Invasive Measurements Using Support Vector Regression.

Blood pressure (BP) is one of the most crucial vital signs of the human body that can be assessed as a critical risk factor for severe health conditions such as cardiovascular diseases (CVD) and hypertension. An accurate, continuous, and cuff-less BP monitoring technique could help clinicians improve the prevention, detection, and diagnosis of hypertension and manage related treatment plans. Notably, the complex and dynamic nature of the cardiovascular system necessitates that any BP monitoring system could benefit from an intelligent technology that can extract and analyze compelling BP features. In this study, a support vector regression (SVR) model was developed to estimate systolic blood pressure (SBP) and diastolic blood pressure (DBP) continuously. We selected a set of features commonly used in previous studies to train the proposed SVR model. A total of 120 patients with available ECG, PPG, DBP and SBP data were chosen from the Medical Information Mart for Intensive Care (MIMIC III) dataset to validate the proposed model. The results showed that the average root mean square error (RMSE) of 2.37 mmHg and 4.18 mmHg were achieved for SBP and DBP, respectively.

Predicting Health Spending in Turkey Using the GPR, SVR, and DT Models

Rising healthcare costs for countries and the long-term maintainability of this situation are at the center of the political agenda. The steady increase in health spending puts pressure on government budgets, healthcare, and personal patient financing. Policymakers would like to plan reforms to reduce these costs to adapt to problems that may arise. This has led planners to decision support systems and forecasting models. In this paper, three machine learnings algoritms, namely Support Vector Regression (SVR), Decision Tree Regression (DT), and Gaussian Process Regression (GPR) are employed to design a forecasting model for Health Spendings (HS) of Turkey considering various determinants. Gross domestic product per capita, urban population rate, unemployment rate, population ages 65 and above, the life expectancy, the physicians’ rate, and the total number of hospital beds are used as inputs. The data set consists of 30 years between 1990- 2019, which splits as training and test sets. Developed models were compared considering performance metrics, and the most accurate model was identified. The coefficient of determinations (R2 ) for SVR, GPR, and DT models are 0.9929, 0.9989, and 0.9611 in the training phase, 0.9536, 0.8944, and 0.1166 in the testing stage, respectively. Therefore, the SVR model has accurate prediction results with the highest R2 and the least root mean square error values in the testing phase. The study showed that the proposed SVR model reduced RMSE value by 32.02% and 39.66% compared to the GPR and DT models, respectively. Consequently, the Health Spendings of Turkey can be predicted by employing SVR with high accuracy. 

Reduction of computational error by optimizing SVR kernel coefficients to simulate concrete compressive strength through the use of a human learning optimization algorithm

This research presents a new model for finding optimal conditions in the concrete technology area. To do that, results of a series of laboratory investigations on concrete samples were considered and used to design several artificial intelligence (AI) models. The data samples include 8 parameters i.e., silica fume replacement ratio, fly ash replacement ratio, fine aggregate, water content, high rate water reducing agent, coarse aggregate, total cementitious material, and age of samples, were used to predict and optimize the compressive strength of concrete samples. For optimization purposes, this study used a human learning optimization (HLO) algorithm to find the optimal results as well as optimizing the kernel coefficients of the support vector regression (SVR) models. Initially, to form the core of this research, various models were constructed and proposed to design the required relationship between the data using SVR. Since different SVR kernels have their own coefficients, using optimization theory, the probability of error in the models was reduced and the models were identified and executed with the highest accuracy. Finally, the polynomial model was selected as the model with the lowest computational error and the highest accuracy for evaluating the compressive strength of the concrete samples. The accuracy of the proposed SVR model for training and testing data was obtained as the coefficient of determination (R2) = 0.9694 and R2 = 0.9470, respectively. This function was considered as a relation, to be developed by the HLO algorithm to find optimal options under different conditions. The results for 14 samples, which are the most important examples of this research, showed that the optimal states are obtained with a high level of accuracy. This confirms the proper use/develop of the SVR-HLO algorithm in designing the predictive model as well as finding optimal conditions in the concrete technology area.

Performance Evaluation of Hybrid WOA-SVR and HHO-SVR Models with Various Kernels to Predict Factor of Safety for Circular Failure Slope

To detect areas with the potential for landslides, slopes are routinely subjected to stability analyses. To this end, there is a need to adopt appropriate mitigation techniques. In general, the stability of slopes with circular failure mode is defined as the factor of safety (FOS). The literature includes a variety of numerical/analytical models proposed in different studies to compute the FOS values of slopes. However, the main challenge is to propose a model for solving a non-linear relationship between independent parameters (which have a great impact on slope stability) and FOS values of slopes. This creates a problem with a high level of complexity and with multiple variables. To resolve the problem, this study proposes a new hybrid intelligent model for FOS evaluation and analysis of slopes in two different phases: simulation and optimization. In the simulation phase, different support vector regression (SVR) kernels were built to predict FOS values. The results showed that the radius basis function (RBF) kernel produces more accurate performance prediction compared with the other applied kernels. The prediction accuracy of this kernel was obtained as coefficient of determination = 0.94, which indicates a high prediction capacity during the simulation phase. Then, in the optimization phase, the proposed SVR model was optimized through the use of two well-known techniques, namely, the whale optimization algorithm (WOA) and Harris hawks optimization (HHO), and the optimum input parameters were obtained. The optimal results confirmed that both optimization techniques are able to achieve a high value for FOS of slopes; however, the HHO shows a more powerful process in FOS maximization compared with the WOA technique. In addition, the developed model was also successfully validated using new data with nine data samples.

Energy parameter and electronic descriptor for carbon based catalyst predicted using QM/ML

Descriptor based model can be efficient in identifying an optimal carbon-based catalyst for oxygen evolution reaction (OER). Here, we correlate the O-atom adsorption strength with the OER activity of graphene nanoribbon systems and define the energy parameters (ΔGO-ΔGOH) to identify the overpotential (ɳ). The π electron based descriptor can predict the catalytic activity of the graphene surfaces. Machine learning algorithms like Multiple Linear Regression, Random Forest Regression and Support Vector Regression (SVR) are trained on the data generated by density functional theory to predict the overpotential. An optimal active site for OER using proposed SVR model is identified with overpotential (0.29 V) and then validate through DFT calculations. To generalize the study, we used SVR model on N doped GNR to predict the site-specific activity towards OER. Such a combined approach can be extended to estimate the site-specific OER activity of different carbon catalysts at a dramatically reduced computational cost.

Downscaling and improving the wind forecasts from NWP for wind energy applications using support vector regression

The stochastic nature of wind poses challenges in the large scale integration of wind energy with the grid. Wind characteristics at a site may significantly vary with time. which will be reflected on the wind power production. Understanding and managing such variations could be challenging for wind farm owners. energy traders and grid operators. In this work. we propose models based on support vector regression (SVR) to downscale the speed and direction of wind at a specific site using both historical observed measurements and numerical weather predictions (NWP). Several meteorological variables. considered to have potential influence on the wind. were used in the feature selection for the models. The models are then optimally developed and used to predict the wind speed and direction at the site considered. In view of the two of Nord pool’s energy markets namely the intraday and day ahead markets. approaches for short-term forecasts (t + 1 hours) and medium-term recursive forecasts (t + 36 hours) were developed. The proposed SVR models are found to be accurate and efficient in correcting the NWP information and predicting the wind speed and direction for the short-term forecasts. For medium-term forecasts. the developed models could outperform the NWP. especially for the wind speed predictions.

A class of new Support Vector Regression models

We propose a novel convex loss function termed as ‘ϵ-penalty loss function’, to be used in Support Vector Regression (SVR) model. The proposed ϵ-penalty loss function is shown to be optimal for a more general noise distribution. The popular ϵ-insensitive loss function and the Laplace loss function are particular cases of the proposed loss function. Making the use of the proposed loss function, we have proposed two new Support Vector Regression models in this paper. The first model which we have termed with ‘ϵ-Penalty Support Vector Regression’ (ϵ-PSVR) model minimizes the proposed loss function with L2-norm regularization. The second model minimizes the proposed loss function with L1-Norm regularization and has been termed as ‘L1-Norm Penalty Support Vector Regression’ (L1- Norm PSVR) model. The proposed loss function can offer different rates of penalization inside and outside of the ϵ-tube. This strategy enables the proposed SVR models to use the full information of the training set which make them to generalize well. Further, the numerical results obtained from the experiments carried out on various artificial, benchmark datasets and financial time series datasets show that the proposed SVR models own better generalization ability than existing SVR models.

A machine learning-based model to estimate the density of nanofluids of nitrides in ethylene glycol

The density of nanofluids is an important thermophysical property whose value is required to evaluate various heat-transfer parameters such as the Reynolds number, the Nusselt number, pressure loss, and the Darcy friction factor. The determination of these parameters is central to the design of many heat-transfer applications. Notably, the density of nanofluids has received relatively little research attention compared with other thermophysical properties. The present study thus focuses on the development of a support vector regression model to estimate the densities of aluminum nitride, titanium nitride, and silicon nitride nanoparticles dispersed in ethylene glycol solution. As inputs, the proposed model uses the mass fraction, temperature, nanoparticle size, and the molecular weight of the nanoparticles. The proposed model predicts the nanofluid densities with high accuracy, as determined by a correlation coefficient of 99.87% and an absolute average relative deviation of 0.0701. To further highlight the accuracy of the proposed model, we compare its results with those of the model of Pak and Cho. The Pak and Cho results deviate considerably from the experimental data except at 298 K. Overall, the proposed support vector regression model is much more accurate than the Pak and Cho model. We thus conclude that the machine learning approach is more reliable for obtaining rapid estimates of the density of nanofluids.

Support Vector Regression for the Relationships between Ground Motion Parameters and Macroseismic Intensity in the Sichuan–Yunnan Region

In this paper, a nonlinear regression method called a support vector regression (SVR) is presented to establish the relationship between engineering ground motion parameters and macroseismic intensity (MSI). Sixteen ground motion parameters, including peak ground acceleration (PGA), peak ground velocity (PGV), Arias intensity, Housner intensity, acceleration spectrum intensity, velocity spectrum intensity, and others, are considered as candidates for feature selection to generate optimal SVR models. The datasets with both useable strong ground motion records and corresponding investigated MSIs in the Sichuan–Yunnan region, China, are all collected, and these 125 pairs of datasets are used for selecting features and comparing regression results. Nine ground motion parameters are selected as the most relevant features: PGA is the first fundamental one and PGV is the fifth relevant feature. Based on performance measures on the testing dataset, the best SVR model is given when the number of features is one all the way up to nine. According to predicted accuracy, SVR models with Gaussian kernel give much better MSI prediction than linear kernel SVR models and linear regression models. In particular, the Gaussian kernel SVR of PGA gives much higher MSI prediction accuracy than the linear regression model of PGV and PGA. The proposed SVR models are valid for MSI values from VI to IX, and they can be used for rapid mapping damage potential and reporting seismic intensity for this high-seismic-activity region.

An Estimation Model on Electricity Consumption of New Metro Stations

Electricity consumption of metro stations increases sharply with expansion of a metro network and this has been a growing cause for concern. Based on relevant historical data from existing metro stations, this paper proposes a support vector regression (SVR) model to estimate daily electricity consumption of a newly constructed metro station. The model considers some major factors influencing the electricity consumption of metro station in terms of both the interior design scheme of a station (e.g., layout of the station and allocation of facilities) and external factors (e.g., passenger volume, air temperature and relative humidity). A genetic algorithm with five-fold cross-validation is used to optimize the hyper-parameters of the SVR model in order to improve its accuracy in estimating the electricity consumption of a metro station (ECMS). With the optimized hyper-parameters, results from case studies on the Beijing Subway showed that the estimating accuracy of the proposed SVR model could reach up to 95% and the correlation coefficient was 0.89. It was demonstrated that the proposed model could outperform the traditional methods which use a back-propagation neural network or multivariate linear regression. The method presented in this paper can be an adequate tool for estimating the ECMS and should further assist in the delivery of new, energy-efficient metro stations.

Performance Comparison of Generalized Regression Network, Radial Basis Function Network and Support Vector Regression for Wind Power Forecasting

With the increasing global warming and enormous pollution, it is obvious to generate power from renewable energy sources. Wind power generation is volatile and intermittent in nature. Stable and reliable power supply may not be feasible with wind power. Power supply reliability is important as much as its availability. To deal with variability of wind power generation, power sector is highly dependent on forecasting methods and techniques. In addition to the implementation of improved neural network models as generalized regression neural network (GRNN), radial basis function neural network (RBFN), support vector regression (SVR) model is used in short term wind power forecasting (WPF).The performance of these models in WPF is compared in terms of the mean absolute percentage error (MAPE). To carry WPF, the data of meteorological parameters like wind speed, temperature and historical wind power data of Indian Kolkata region are used for forecasting. Except the cases where the wind power generation is very less, SVR model has performed better than GRNN and RBFN in short term WPF. Short term WPF using GRNN is also considered reliable, but gives higher errors in terms of MAPE than SVR model. It is found that the proposed SVR model gives most accurate short term wind power forecasting.

Dynamic Behavioral Modeling of RF Power Amplifier Based on Time-Delay Support Vector Regression

A new, dynamic behavioral modeling technique, based on a time-delay support vector regression (SVR) method, is presented in this paper. As an advanced machine learning algorithm, the SVR method provides an effective option for behavioral modeling of radio frequency (RF) power amplifiers (PAs), taking into account the effects of both device nonlinearity and memory. The basic theory of the proposed modeling technique is given, along with a detailed model extraction procedure. Unlike traditional artificial neural network (ANN) techniques, which take time to determine the best configuration of the model, the SVR method can obtain the optimal model in short time, using the grid-search technique. An example of an optimal SVR model selection applied to an RF PA is also given; the performance of the selected model presents a big improvement when compared with the default SVR model. Experimental validation is performed using an LDMOS PA, a single device gallium nitride (GaN) PA, and a Doherty GaN PA, revealing that the new modeling methodology provides very efficient and extremely accurate prediction. Compared with traditional Volterra models, canonical piecewise linear models, and ANN-based models, the proposed SVR model gives improved performance with reasonable complexity. In addition, it is shown that the model can predict accurately the behavior of the PA under input power levels that are different from those under which it is extracted.

Robust Support Vector Regression in Primal with Asymmetric Huber Loss

As real world data sets in general contain noise, construction of robust regression learning models to fit data with noise is an important and challenging research problem. It is all the more difficult to learn regression function with good generalization performance for input samples corrupted by asymmetric noise and outliers. In this work, we propose novel robust regularized support vector regression models with asymmetric Huber and e-insensitive Huber loss functions leading to strongly convex minimization problems in simpler form whose solutions are obtained by simple functional iterative method. Numerical experiments performed on (1) synthetic data sets with different noise models and having outliers; (2) real world data sets, clearly show the effectiveness and applicability of the proposed support vector regression models with asymmetric Huber loss.

Prediction of Resilient Modulus from Post-compaction Moisture Content and Physical Properties Using Support Vector Regression

The present study assesses the use of support vector machine regression to predict the variation of resilient modulus with post-compaction moisture content of soils commonly encountered in Oklahoma, Pennsylvania and Wisconsin. Results show the prediction model using the support vector regression (SVR) approach is a function of degree of saturation, moisture content and plasticity index. The developed model is compared to current models in the literature. Results indicate the proposed SVR model gives more accurate values than current regression models. This model will better predict changes in the bearing capacity of pavements due to seasonal variations of moisture content.

Developing a dengue forecast model using machine learning: A case study in China.

BackgroundIn China, dengue remains an important public health issue with expanded areas and increased incidence recently. Accurate and timely forecasts of dengue incidence in China are still lacking. We aimed to use the state-of-the-art machine learning algorithms to develop an accurate predictive model of dengue.Methodology/Principal findingsWeekly dengue cases, Baidu search queries and climate factors (mean temperature, relative humidity and rainfall) during 2011–2014 in Guangdong were gathered. A dengue search index was constructed for developing the predictive models in combination with climate factors. The observed year and week were also included in the models to control for the long-term trend and seasonality. Several machine learning algorithms, including the support vector regression (SVR) algorithm, step-down linear regression model, gradient boosted regression tree algorithm (GBM), negative binomial regression model (NBM), least absolute shrinkage and selection operator (LASSO) linear regression model and generalized additive model (GAM), were used as candidate models to predict dengue incidence. Performance and goodness of fit of the models were assessed using the root-mean-square error (RMSE) and R-squared measures. The residuals of the models were examined using the autocorrelation and partial autocorrelation function analyses to check the validity of the models. The models were further validated using dengue surveillance data from five other provinces. The epidemics during the last 12 weeks and the peak of the 2014 large outbreak were accurately forecasted by the SVR model selected by a cross-validation technique. Moreover, the SVR model had the consistently smallest prediction error rates for tracking the dynamics of dengue and forecasting the outbreaks in other areas in China.Conclusion and significanceThe proposed SVR model achieved a superior performance in comparison with other forecasting techniques assessed in this study. The findings can help the government and community respond early to dengue epidemics.

Support Vector Regression Prediction and Analysis of the Copper (II) Biosorption Efficiency

Support vector regression (SVR) has been used for the prediction of biosorption efficiency of Cu(II) using litter of natural trembling poplar (Populus tremula) forest (LNTPF) as a low-cost biosorbent. The proposed SVR model has been compared with the most widely used multiple linear regression (MLR) model based on statistical parameters for the unseen test data set in terms of coefficient of determination (R2), average absolute relative error, root mean square error and standard deviation. The SVR-based model is found to be superior to the MLR model. The effect of various parameters (adsorbent concentration, pH, particle size, initial Cu(II) concentration, agitating speed and temperature) on the Cu(II) biosorption efficiency using the SVR-based model has also been studied. The simulation results of the SVR-based model agree appreciably with the available experimental results. This novel application of SVR can be successfully applied to model other such systems with good accuracy and high generalization ability.