Poly Kernel Research Articles

Which Machine Learning Algorithm Is Best Suited for Estimating Reference Evapotranspiration in Humid Subtropical Climate?

ABSTRACTTimely and reliable estimates of reference evapotranspiration (ET0) are imperative for robust water resources planning and management. Applying machine learning (ML) algorithms for estimating ET0 has been evolving, and their applicability in different sectors is still a compelling field of research. In this study, four Gaussian process regression (GPR) algorithms—polynomial kernel (PK), polynomial universal function kernel (PUK), normalized poly kernel (NPK), and radial basis function (RBF)—were compared against widely used random forest (RF) and a simpler locally weighted linear regression (LWLR) algorithm at a humid subtropical region in India. The sensitivity analysis of the input variables was followed by application of the best combination of variables in algorithm testing and training for generating ET0. The results were then compared against the Penman–Monteith method at both daily and monthly time steps. The results indicated that ET0 is least sensitive to wind speed at 2 m height. Additionally, at a daily time step, RF, followed by PUK, generated the best results during both training and testing phases. In contrast, at a monthly time step, using multiple model evaluation matrices, PUK followed by RF performed best. These results demonstrate the application of the ML algorithms is subjected to user‐required time steps. Although this study focused on Northwest India, the findings are relevant to all humid subtropical regions across the world.

MSTD: a framework for rolling bearing fault diagnosis based on multi-scale and soft-threshold denoising

ABSTRACT In real industrial environments, the data collected is often subject to noise interference, which leads to challenges for traditional deep learning fault diagnosis models, including insufficient autonomous learning capabilities, complex feature extraction processes, and sensitivity to noise. To address these issues, this paper proposes a bearing fault diagnosis convolutional neural network based on multi-scale and soft-threshold denoising (MSTD). In this model, Poly Kernel Feature Fusion (PKFF) is utilised to extract high-quality features from raw vibration data, enhancing the model’s ability to recognise complex fault patterns; Residual anchor attention (RAA) reduces redundancy and irrelevant information between channels, optimising inter-channel information to ensure the model’s effectiveness when handling multi-channel data; Soft threshold CG (ST-CG) utilises a soft-threshold denoising method to suppress noise after feature extraction, thereby more clearly revealing fault characteristics. Soft-threshold processing not only enhances the model’s noise resistance but also preserves critical fault information, enabling the model to maintain high diagnostic accuracy even under low signal-to-noise ratio conditions. The proposed model demonstrates outstanding performance on the CWRU and MFS-MG datasets, exhibiting superior feature extraction capabilities and robustness in high-noise environments compared to other models, significantly improving diagnostic accuracy and noise resistance.

Early detection of pine wilt disease based on UAV reconstructed hyperspectral image.

Pine wilt disease (PWD) is a highly destructive infectious disease that affects pine forests. Therefore, an accurate and effective method to monitor PWD infection is crucial. However, the majority of existing technologies can detect PWD only in the later stages. To curb the spread of PWD, it is imperative to develop an efficient method for early detection. We presented an early stage detection method for PWD utilizing UAV remote sensing, hyperspectral image reconstruction, and SVM classification. Initially, employ UAV to capture RGB remote sensing images of pine forests, followed by labeling infected plants using these images. Hyperspectral reconstruction networks, including HSCNN+, HRNet, MST++, and a self-built DW3D network, were employed to reconstruct the RGB images obtained from remote sensing. This resulted in hyperspectral images in the 400-700nm range, which were used as the dataset of early PWD detection in pine forests. Spectral reflectance curves of infected and uninfected plants were extracted. SVM algorithms with various kernel functions were then employed to detect early pine wilt disease. The results showed that using SVM for early detection of PWD infection based on reconstructed hyperspectral images achieved the highest accuracy, enabling the detection of PWD in its early stage. Among the experiments, MST++, DW3D, HRNet, and HSCNN+ were combined with Poly kernel SVM performed the best in terms of cross-validation accuracy, achieving 0.77, 0.74, 0.71, and 0.70, respectively. Regarding the reconstruction network parameters, the DW3D network had only 0.61M parameters, significantly lower than the MST++ network, which had the highest reconstruction accuracy with 1.6M parameters. The accuracy was improved by 27% compared to the detection results obtained using RGB images. This paper demonstrated that the hyperspectral reconstruction-poly SVM model could effectively detect the Early stage of PWD. In comparison to UAV hyperspectral remote sensing methods, the proposed method in this article offers a same precision, but a higher operational efficiency and cost-effectiveness. It also enables the detection of PWD at an earlier stage compared to RGB remote sensing, yielding more accurate and reliable results.

Field Obstacle Detection and Location Method Based on Binocular Vision

When uncrewed agricultural machinery performs autonomous operations in the field, it inevitably encounters obstacles such as persons, livestock, poles, and stones. Therefore, accurate recognition of obstacles in the field environment is an essential function. To ensure the safety and enhance the operational efficiency of autonomous farming equipment, this study proposes an improved YOLOv8-based field obstacle detection model, leveraging depth information obtained from binocular cameras for precise obstacle localization. The improved model incorporates the Large Separable Kernel Attention (LSKA) module to enhance the extraction of field obstacle features. Additionally, the use of a Poly Kernel Inception (PKI) Block reduces model size while improving obstacle detection across various scales. An auxiliary detection head is also added to improve accuracy. Combining the improved model with binocular cameras allows for the detection of obstacles and their three-dimensional coordinates. Experimental results demonstrate that the improved model achieves a mean average precision (mAP) of 91.8%, representing a 3.4% improvement over the original model, while reducing floating-point operations to 7.9 G (Giga). The improved model exhibits significant advantages compared to other algorithms. In localization accuracy tests, the maximum average error and relative error in the 2–10 m range for the distance between the camera and five types of obstacles were 0.16 m and 2.26%. These findings confirm that the designed model meets the requirements for obstacle detection and localization in field environments.

Ship Detection in Synthetic Aperture Radar Images Based on BiLevel Spatial Attention and Deep Poly Kernel Network

Synthetic aperture radar (SAR) is a technique widely used in the field of ship detection. However, due to the high ship density, fore-ground-background imbalance, and varying target sizes, achieving lightweight and high-precision multiscale ship object detection remains a significant challenge. In response to these challenges, this research presents YOLO-MSD, a multiscale SAR ship detection method. Firstly, we propose a Deep Poly Kernel Backbone Network (DPK-Net) that utilizes the Optimized Convolution (OC) Module to reduce data redundancy and the Poly Kernel (PK) Module to improve the feature extraction capability and scale adaptability. Secondly, we design a BiLevel Spatial Attention Module (BSAM), which consists of the BiLevel Routing Attention (BRA) and the Spatial Attention Module. The BRA is first utilized to capture global information. Then, the Spatial Attention Module is used to improve the network’s ability to localize the target and capture high-quality detailed information. Finally, we adopt a Powerful-IoU (P-IoU) loss function, which can adjust to the ship size adaptively, effectively guiding the anchor box to achieve faster and more accurate detection. Using HRSID and SSDD as experimental datasets, mAP of 90.2% and 98.8% are achieved, respectively, outperforming the baseline by 5.9% and 6.2% with a model size of 12.3 M. Furthermore, the network exhibits excellent performance across various ship scales.

Application of Naive Bayes Classifier Method to Analyze Social Media User Sentiment Towards the Presidential Election Phase

This research aims to analyze the sentiment of social media users towards the election. The author collected data in this research through a literature study and observation. The author uses a classification method with the Naïve Bayes Classifier Algorithm and Support Vector Machine to analyze sentiment results. Next, this research extracts word assessment features using TextBlob, which changes text into positive or negative classes. Based on the research results, after going through the text preprocessing stage of more than 15,000 tweets, 11,000 clean tweets were obtained, which were then labelled using the text blob library in Python. The labelling results show that 4,000 tweets are positive, and the rest are harmful, indicating that most social media users' sentiment towards the election is positive. Words that often appear in the positive class express support and confidence in implementing elections that are considered honest and fair. On the other hand, words in the negative class reflect negative sentiment towards implementing elections, which are considered unsuccessful and time-consuming. The Naïve Bayes method provides accuracy, precision, and recall values of 85%, 80%, and 75%. In the Support Vector Machine method, testing is carried out with three kernels (linear, RBF, and poly), where the poly kernel with the best parameter values C is ten and degree is 1 produces the highest accuracy, precision, and recall of 90%, 90%, and 85%, respectively.

Social Media Based Film Recommender System (Twitter) on Disney+ with Hybrid Filtering Using Support Vector Machine

In the current era, the culture of watching TV shows and movies has been made easier by the presence of the internet. Now, watching movies on platforms can be done from anywhere, one of which is Disney+. At times, people find it challenging to decide which film to watch given the multitude of genres and film titles available on these platforms. One solution to this issue is a recommendation system that can suggest films based on ratings. The recommendation system to be utilized involves Collaborative Filtering, Content-Based Filtering, and Hybrid Filtering. This is because Collaborative Filtering and Content-Based Filtering encounter issues like cold start, sparsity, and overspecialization. Thus, the objective of this study is to develop a recommendation system using Hybrid Filtering combined with Support Vector Machine (SVM). In this research, classification will be carried out using poly, linear, and RBF kernels with varying parameters. Techniques such as TF-IDF, RMSE, tuning, and data balancing with SMOTEN will be implemented to enhance accuracy during the classification process. The evaluation employed in this study utilizes the confusion matrix. Support Vector Machine, when tuned and combined with SMOTEN, achieves noteworthy results, particularly with the RBF kernel which attains a Precision score of 0.94. Recall produces a value of 0.93 with the Poly kernel, while the highest Accuracy, at 0.93, is achieved with the RBF kernel. Furthermore, the RBF kernel also demonstrates the highest F1-Score of 0.93. These findings illustrate elevated precision, recall, accuracy, and F1-Score within the context of hybrid filtering, achieved through the application of Support Vector Machine for classification and the implementation of the SMOTEN technique.

Carbon Trading Price Prediction of Three Carbon Trading Markets in China Based on a Hybrid Model Combining CEEMDAN, SE, ISSA, and MKELM

Carbon trading has been deemed as the most effective mechanism to mitigate carbon emissions. However, during carbon trading market operation, competition among market participants will inevitably occur; hence, the precise forecasting of the carbon trading price (CTP) has become a significant element in the formulation of competition strategies. This investigation has established a hybrid CTP forecasting framework combining complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), sample entropy (SE) method, improved salp swarm algorithm (ISSA), and multi-kernel extreme learning machine (MKELM) methods to improve forecasting accuracy. Firstly, the initial CTP data sequence is disintegrated into several intrinsic mode functions (IMFs) and a residual sequence by a CEEMDAN method. Secondly, to save calculation time, SE method has been utilized to reconstruct the IMFs and the residual sequence into new IMFs. Thirdly, the new IMFs are fed into the MKELM model, combing RBF and the poly kernel functions to utilize their superior learning and generalization abilities. The parameters of the MKELM model are optimized by ISSA, combining dynamic inertia weight and chaotic local searching method into the SSA to enhance the searching speed, convergence precision, as well as the global searching ability. CTP data in Guangdong, Shanghai, and Hubei are selected to prove the validity of the established CEEMDAN-SE-ISSA-MKELM model. Through a comparison analysis, the established CEEMDAN-SE-ISSA-MKELM model performs the best with the smallest MAPE and RMSE values and the highest R2 value, which are 0.76%, 0.53, and 0.99, respectively, for Guangdong,. Thus, the presented model would be extensively applied in CTP forecasting in the future.

Indirect prediction of remaining useful life for lithium-ion batteries based on improved multiple kernel extreme learning machine

Lithium-ion batteries (LIBs) will undergo aging after a certain period of operation. Predicting RUL in advance is critical to ensure a reliable energy supply from the battery system. To address the problem of poor robustness of the single-core extreme learning machine model in predicting the RUL of LIBs, a multiple kernel extreme learning machine (MKELM) integrating radial basis function (RBF) and Poly kernel function is proposed. Firstly, introduce the sparrow search algorithm improved by logistic chaotic mapping (ISSA) to optimize the parameter selection of the MKELM, which overcomes the inefficiency of parameter selection, and improves the efficiency of the global optimal parameter selection. Secondly, from the charging/discharging processes, extract the health indicators (HIs) and examine the correlation capacity via methods of Pearson, Spearman, and Kendall. Finally, validate the ISSA-MKELM with the NASA battery dataset, and the prediction results are compared with traditional methods. Results show that, with the lowest prediction error (less than 2.38 %), ISSA-MKELM has good state tracking fit and excellent computational efficiency performance but also has strong generalization and robustness.

Comparative Assessment of Improved SVM Method under Different Kernel Functions for Predicting Multi-scale Drought Index

This paper focus on the drought monitoring and forecasting for semi-arid region based on the various machine learning models and SPI index. Drought phenomena are crucial role in the agriculture and drinking purposes in the area. In this study, Standardized Precipitation Index (SPI) was used to predicted the future drought in the upper Godavari River basin, India. We have selected the ten input combinations of ML model were used to prediction of drought for three SPI timescales (i.e., SPI -3, SPI-6, and SPI-12). The historical data of SPI from 2000 to 2019 was used for creation of ML models SPI prediction, these datasets was divided into training (75% of the data) and testing (25% of the data) models. The best subset regression method and sensitivity analysis were applied to estimate the most effective input variables for estimation of SPI 3, 6, and 12. The improved support vector machine model using sequential minimal optimization (SVM-SMO) with various kernel functions i.e., SMO-SVM poly kernel, SMO-SVM Normalized poly kernel, SMO-SVM PUK (Pearson Universal Kernel) and SMO-SVM RBF (radial basis function) kernel was developed to forecasting of the SPI-3,6 and 12 months. The ML models accuracy were compared with various statistical indicators i.e., root mean square error (RMSE), mean absolute error (MAE), relative absolute error (RAE), root relative squared error (RRSE), and correlation coefficient (r). The results of study area have been showed that the SMO-SVM poly kernel model precisely predicted the SPI-3 (R2 = 0.819) and SPI-12 (R2 = 0.968) values at Paithan station; the SPI-3 (R2 = 0.736) and SPI-6 (R2 = 0.841) values at Silload station, respectively. The SMO-SVM PUK kernel is found that the best ML model for the prediction of SPI-6 (R2 = 0.846) at Paithan station and SPI-12 (R2 = 0.975) at the Silload station. The compared with SVM-SMO poly kernel and SVM-SMO PUK kernel was observed, these models are best forecasting of drought (i.e. SPI-6 and SPI-12), while SVM-SMO poly kernel is good for SPI-3 prediction at both stations. The results have been showed the ability of the SVM-SMO algorithm with various kernel functions successfully applied for the forecasting of multiscale SPI under the climate changes. It can be helpful for decision making in water resource management and tackle droughts in the semi-arid region of central India.

Predicting acceptance of the bank loan offers by using support vector machines

Loans are one of the main profit sources in banking system. Banks try to select reliable customers and offer them personal loans, but customers can sometimes reject bank loan offers. Prediction of this problem is an extra work for banks, but if they can predict which customers will accept personal loan offers, they can make a better profit. Therefore, at this point, the aim of this study is to predict acceptance of the bank loan offers using the Support Vector Machine (SVM) algorithm. In this context, SVM was used to predict results with four kernels of SVM, with a grid search algorithm for better prediction and cross validation for much more reliable results. Research findings show that the best results were obtained with a poly kernel as 97.2% accuracy and the lowest success rate with a sigmoid kernel as 83.3% accuracy. Some precision and recall values are lower than normal ones, like 0.108 and 0.008 due to unbalanced dataset, like for 1 true value, there are 9 negative values (9.6% true value). This study recommends the use of SVC in banking system while predicting acceptance of bank loan offers.

How do multiple kernel functions in machine learning algorithms improve precision in flood probability mapping?

With climate change, hydro-climatic hazards, such as floods in the Himalayas regions, are expected to worsen, thus likely to overwhelm humans and socioeconomic system. Precisely, the Koshi River basin (KRB) is often impacted by floods over time. However, studies on estimating and predicting floods are still scarce in this basin. This study aims at developing a flood probability map using machine learning algorithms (MLAs): Gaussian process regression (GPR) and support vector machine (SVM) with multiple kernel functions including Pearson VII function kernel (PUK), polynomial, normalized poly kernel, and radial basis kernel function (RBF). Historical flood locations from available (topography, hydrogeology, and environmental) datasets were further considered to build a flood probability model. Two datasets were carefully chosen to measure the feasibility and robustness of MLAs: the training dataset (flood locations between 2010 and 2019) and the testing dataset (flood locations of 2020) with thirteen flood influencing factors. Validation of the MLAs was performed with statistical indices such as the coefficient of determination (r2: 0.546 –.995), mean absolute error (MAE: 0.009 –373), root mean square error (RMSE: 0.051–0.466), relative absolute error (RAE: 1.81–8.55%), and root-relative square error (RRSE: 10.19–91.00%). Results showed that the SVM-Pearson VII kernel (PUK) yielded better prediction than other algorithms. The resultant map from SVM-PUK revealed that 27.99% area with low, 39.91% area with medium, 31.00% with high, and 1.10% area with very high probabilities of flooding in the study area. The flood probability map, derived in this study, could add great value to the effort of flood risk mitigation and planning processes in KRB.

Application of soft computing approaches to predict gabion weir oxygen aeration efficiency

ABSTRACT Aeration is a physical process through which air is brought into contact with water, and thereby both times of contact and contact area between water and air are enhanced. Gabion weirs are eco-friendly alternative structures having the least negative environmental impact. The paper examines the performance of the neuro-fuzzy, neural network, and Gaussian process regression (GPR) models in estimating the gabion weir oxygen aeration efficiency using the test data sets with regards to the evaluation indicators. The model utilizes non-dimensional inputs: Froude number, Reynolds number, the ratio of mean size to weir length, and porosity of gabion weir particles. The GPR with normalized poly kernel model having the highest value of correlation coefficient (CC), 0.9132, and value of root mean square error (RMSE), 0.392, outperforms all applied models and is followed by the neural network model with CC 0.8952 and RMSE 0.398. The neuro-fuzzy triangular shape with CC 0.8865 and RMSE 0.07576 is an equally performing model; nevertheless, all applied AI-based models are mostly performing well, but all mathematical models are exceptionally performing poorly except the nonlinear model with CC 0.8612 and RMSE 0.0029. To assess the relative relevance of input parameters on the output results, sensitivity and parametric analysis are also performed.

Uncertain Interval Forecasting for Combined Electricity-Heat-Cooling-Gas Loads in the Integrated Energy System Based on Multi-Task Learning and Multi-Kernel Extreme Learning Machine

The accurate prediction of electricity-heat-cooling-gas loads on the demand side in the integrated energy system (IES) can provide significant reference for multiple energy planning and stable operation of the IES. This paper combines the multi-task learning (MTL) method, the Bootstrap method, the improved Salp Swarm Algorithm (ISSA) and the multi-kernel extreme learning machine (MKELM) method to establish the uncertain interval prediction model of electricity-heat-cooling-gas loads. The ISSA introduces the dynamic inertia weight and chaotic local searching mechanism into the basic SSA to improve the searching speed and avoid falling into local optimum. The MKELM model is established by combining the RBF kernel function and the Poly kernel function to integrate the superior learning ability and generalization ability of the two functions. Based on the established model, weather, calendar information, social–economic factors, and historical load are selected as the input variables. Through empirical analysis and comparison discussion, we can obtain: (1) the prediction results of workday are better than those on holiday. (2) The Bootstrap-ISSA-MKELM based on the MTL method has superior performance than that based on the STL method. (3) Through comparing discussion, we discover the established uncertain interval prediction model has the superior performance in combined electricity-heat-cooling-gas loads prediction.

Applications of Gaussian process regression for predicting blue water footprint: Case study in Ad Daqahliyah, Egypt

Timely and reliable water footprint prediction is imperative and prerequisite to mitigate climate risk and ensure water and food security and enhance the water-use efficiency. This study aims to model the Water Footprint (WF) by using the four kernels of Gaussian processes models (Polynomial, Normalized Poly, Radial Basis Function RBF, and Pearson Universal Function PUK) and select the best kernel with best climate scenario. This study investigates the predicting WF of maize based on meteorological variables including maximum temperature (Tmax), minimum temperature (Tmin), mean temperature (Tmean), solar radiation (SR), wind speed (WS), and vapor pressure deficit (VPD), Extraterrestrial radiation (Ra) relative humidity (RH) and crop coefficient (Kc) by applying ten scenarios of climate variables in the Egyptian Nile Delta, Ad Daqahliyah Governorate for predicting blue WF of maize during 2000–2019. The main findings are following as, firstly; based on developing four kernels, the performance of the PUK kernel in predicting blue WF is far better than the other three kernels followed by the Poly kernel. Secondly; for PUK kernel, model 7 (Tmax, Tmin, Tmean, WS, Sunshine Hours (SH), VPD and SR) has good performance which is close to models 8 (model 7 +Ra), model 9 (model 7 +Ra and RH) and model 10 (all inputs). Thirdly; in all four kernels, the error rate in small blue WF values is higher than the other values, moreover, the error value decreases at the medium blue WF values, while, it increases again at large WF values. Therefore, the developed models in this study can help and promote the decision makers to manage and secure the water resources management under the extreme climate events.

Comparative Analysis of Identifying Accuracy of Online Misinformation of Covid-19 Using SVM Algorithm with Decision Tree Classification

Aim: To improve the accuracy percentage of predicting misinformation about COVID-19 using SVM algorithm. Materials and methods: Support Vector Machine (SVM) with sample size = 20 and Decision Tree classification with sample size = 20 was iterated at different times for predicting the accuracy percentage of misinformation about COVID19. The Novel Poly kernel function used in SVM maps the dataset into higher dimensional space which helps to improve accuracy percentage. Results and Discussion: SVM has significantly better accuracy (94.48%) compared to Decision Tree accuracy (93%). There was a statistical significance between SVM and the Decision Tree (p=0.000) (p<0.05 Independent Sample T-test). Conclusion: SVM with Novel Poly kernel helps in predicting with more accuracy the percentage of misinformation about COVID-19.

Daily Suspended Sediment Prediction Using Seasonal Time Series and Artificial Intelligence Techniques

Estimating the amount of suspended sediment in rivers correctly is important due to the adverse impacts encountered during the design and maintenance of hydraulic structures such as dams, regulators, water channels and bridges. The sediment concentration and discharge currents have usually complex relationship, especially on long term scales, which can lead to high uncertainties in load estimates for certain components. In this paper, with several data-driven methods, including two types of perceptron support vector machines with radial basis function kernel (SVM-RBF), and poly kernel learning algorithms (SVM-PK), Library SVM (LibSVM), adaptive neuro-fuzzy (NF) and statistical approaches such as sediment rating curves (SRC), multi linear regression (MLR) are used for forecasting daily suspended sediment concentration from daily temperature of water and streamflow in the river. Daily data are measured at Augusta station by the US Geological Survey. 15 different input combinations (1 to 15) were used for SVM-PK, SVM-RBF, LibSVM, NF and MLR model studies. All approaches are compared to each other according to three statistical criteria; mean absolute errors (MAE), root mean square errors (RMSE) and correlation coefficient (R). Of the applied linear and nonlinear methods, LibSVM and NF have good results, but LibSVM generates a slightly better fit under whole daily sediment values.

Tweets Responding to the Indonesian Government’s Handling of COVID-19: Sentiment Analysis Using SVM with Normalized Poly Kernel

Background: Handling COVID-19 (Corona Virus Disease-2019) in Indonesia was once trending on Twitter. The Indonesian government's handling evoked pros and cons in the community. Public opinions on Twitter can be used as a decision support system in making appropriate policies to evaluate government performance. A sentiment analysis method can be used to analyse public opinion on Twitter.Objective: This study aims to understand public opinion trends on COVID-19 in Indonesia both from a general perspective and an economic perspective.Methods: We used tweets from Twitterscraper library. Because they did not have a label, we provided labels using sentistrength_id and experts to be classified into positive, negative, and neutral sentiments. Then, we carried out a pre-processing to eliminate duplicate and irrelevant data. Next, we employed machine learning to predict the sentiments for new data. After that, the machine learning algorithms were evaluated using confusion matrix and K-fold cross-validation.Results: The SVM analysis on the sentiments on general aspects using two-classes dataset achieved the highest performance in average accuracy, precision, recall, and f-measure with the value of 82.00%, 82.24%, 82.01%, and 81.84%, respectively.Conclusion: From the economic perspective, people seemed to agree with the government’s policies in dealing with COVID-19; but people were not satisfied with the government performance in general. The SVM algorithm with the Normalized Poly Kernel can be used as an intelligent algorithm to predict sentiment on Twitter for new data quickly and accurately.

Estimation of sodium adsorption ratio in a river with kernel-based and decision-tree models.

The control of surface water quality plays an important role in the management of water resources. In this context, the estimation and assessment of sodium adsorption ratio (SAR) are required which is one of the significant water quality parameters in the agricultural production sector. Chemical analysis might not, however, be feasible for a longer period of time in all the country-scale rivers. Therefore in this study, a support vector regression (SVR) model with different kernel functions; K nearest neighbour algorithm; and four decision-tree models, namely, Hoeffding tree, random forest, random tree, and REPTree, were used to estimate the SAR value with minimal parameters in the Aladag River in Turkey. In alternative scenarios, a correlation matrix and sensitivity analysis were used to ascertain the model inputs from among the 15 distinct parameters. All 15 parameters were utilized as model inputs in the first scenario, and only the sodium (Na) parameter was utilized as the model input in the final scenario. The accuracy of the aforesaid models was then assessed making use of correlation coefficient, Nash-Sutcliffe model efficiency coefficient, root mean square error, mean absolute error, and Willmott index of agreement. The results indicate that the SVR model with the poly kernel function provides the best estimates of SAR among the considered models. According to the findings, there is no considerable difference between the results acquired in the first and last scenarios, and one can determine the SAR value while making use of machine learning approaches taking into account only Na parameter.

Support vector machine based prediction of photovoltaic module and power station parameters

ABSTRACTThe uncertainty in the output power of the photovoltaic (PV) power generation station due to variation in meteorological parameters is of serious concern. An accurate output power prediction of a PV system helps in better design and planning. The present study is carried out for the prediction of output power of PV generating station by using Support Vector Machines. Two cases are considered in the present study for prediction. Case-I deals with the prediction of PV module parameters such as Voc, Ish, Rs, Rsh, Imax, Vmax, Pmax, and case-II deals with the prediction of power generation parameters such as PDC, PAC, and system efficiency. Historical data of PV power station with an installed capacity of 10 MW and weather information are used as input to develop four different seasons-based SVM models for all parameters. The performance results of the models are presented in terms of Mean Relative Error (MRE) and Root Mean Square Error (RMSE). Additionally, the performance results obtained with polynomial and Radial Based Function kernel are also compared to show that which kernel has better prediction accuracy, and practicability. The result shows that the minimum average RMSE and MRE for case-I with Radial Based Function kernel are 0.034%, 0.055%, 0.002%, 1.726%, 0.044%, 0.047%, 2.342%, and 0.005%, 0.014%, 0.079%, 0.885%, 0.005%, 0.007%, 0.013%, and for case-II with poly kernel are 0.014%, 0.016%, 0.149% and 0.011%, 0.0175, 1.03%, respectively. The present study will be helpful to provide technical guidance to the prediction of the PV power System.