scholarly journals Multi-Kernel Learning based Sugar Industry Load Forecasting

2021 ◽  
Vol 9 (5) ◽  
pp. 275-278
Author(s):  
Yamanappa. N. Doddamani
Keyword(s):  
Support Vector Regression ◽  
Sugar Industry ◽  
Load Forecasting ◽  
Absolute Error ◽  
Power Delivery ◽  
Percentage Error ◽  
Support Vector ◽  
Energy Audit ◽  
Load Demand ◽  
Audit Data

Sugar industry which plans for power usage from Bagasse also needs the load forecasting carried out using the energy audit data. The stochastic nature of the load demand of the sugar industry needs to be forecasted in advance for the assuring uninterrupted power delivery to the industry. The manual energy audit data obtained from the sugar industry for a period of time is obtained and trained on a regression based on MultiKernel Learning (MKL). The Support Vector Regression (SVR) formulation is applied with the MultiKernel topology and the performance parameters including the Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) is observed in the implementation. The algorithm is the Multi Kernel Support Vector Regression algorithm using the Python based toolbox.

scholarly journals Multi-State Load Demand Forecasting Using Hybridized Support Vector Regression Integrated with Optimal Design of Off-Grid Energy Systems—A Metaheuristic Approach

Processes ◽  
2021 ◽  
Vol 9 (7) ◽  
pp. 1166
Author(s):  
Bashir Musa ◽  
Nasser Yimen ◽  
Sani Isah Abba ◽  
Humphrey Hugh Adun ◽  
Mustafa Dagbasi
Keyword(s):  
Support Vector Regression ◽  
Mean Square Error ◽  
Prediction Accuracy ◽  
Demand Forecasting ◽  
Load Forecasting ◽  
Coefficient Of Determination ◽  
Percentage Error ◽  
Support Vector ◽  
Mean Square ◽  
Hybrid Renewable Energy System

The prediction accuracy of support vector regression (SVR) is highly influenced by a kernel function. However, its performance suffers on large datasets, and this could be attributed to the computational limitations of kernel learning. To tackle this problem, this paper combines SVR with the emerging Harris hawks optimization (HHO) and particle swarm optimization (PSO) algorithms to form two hybrid SVR algorithms, SVR-HHO and SVR-PSO. Both the two proposed algorithms and traditional SVR were applied to load forecasting in four different states of Nigeria. The correlation coefficient (R), coefficient of determination (R2), mean square error (MSE), root mean square error (RMSE), and mean absolute percentage error (MAPE) were used as indicators to evaluate the prediction accuracy of the algorithms. The results reveal that there is an increase in performance for both SVR-HHO and SVR-PSO over traditional SVR. SVR-HHO has the highest R2 values of 0.9951, 0.8963, 0.9951, and 0.9313, the lowest MSE values of 0.0002, 0.0070, 0.0002, and 0.0080, and the lowest MAPE values of 0.1311, 0.1452, 0.0599, and 0.1817, respectively, for Kano, Abuja, Niger, and Lagos State. The results of SVR-HHO also prove more advantageous over SVR-PSO in all the states concerning load forecasting skills. This paper also designed a hybrid renewable energy system (HRES) that consists of solar photovoltaic (PV) panels, wind turbines, and batteries. As inputs, the system used solar radiation, temperature, wind speed, and the predicted load demands by SVR-HHO in all the states. The system was optimized by using the PSO algorithm to obtain the optimal configuration of the HRES that will satisfy all constraints at the minimum cost.

scholarly journals Electrical Load Forecasting using SVM Algorithm

2020 ◽  
Vol 8 (6) ◽  
pp. 4811-4816
Keyword(s):  
Support Vector Machine ◽  
Load Forecasting ◽  
Percentage Error ◽  
Support Vector ◽  
Live Load ◽  
Electric Load ◽  
Mean Square ◽  
Electrical Load ◽  
Load Demand ◽  
Svm Algorithm

Electrical load demand is variable in nature. Also, with the increase in technological development and automation, electric load demand tends to rise with time. For this, our generation facilities should be adequate 24x7 to meet the consumer’s load demand effectively. Therefore, load demand needs to be predicted or forecasted to avoid the energy crisis. In this paper, support vector machine (SVM) algorithm is explored for electric load forecasting. The live load data for the period of three months i.e., January to March, 2015, from a typical 66kV sub-station of the Punjab State Power Corporation Limited (PSPCL) for a selected site at Bhai Roopa sub-station, Bathinda, situated in the Punjab state of India, is acquired for the presented simulation study. The collected live data is divided into three categories, i.e., validation, training, and testing for the simulation study considering a SVM approach. Then, based on the environmental data input for the next 50 hours, the electric load is predicted. The obtained results from simulation were validated with the live load data of the selected site and found to be within the permissible limits. The mean square error (MSE), root-mean-square error (RMSE), mean absolute error (MAE), absolute percentage error (APE), mean absolute percentage error (MAPE) and sum of squares error (SSE) were calculated to show the effectiveness of the proposed support vector machine (SVM) algorithm based STLF. SVM is one of the effective machine learning algorithms. The errors so obtained clearly suggest that the proposed SVM algorithm gives reasonably accurate results, and is reliable for electric load forecasting.

scholarly journals Very Short-Term Load Forecasting Using Hybrid Algebraic Prediction and Support Vector Regression

2017 ◽  
Vol 2017 ◽  
pp. 1-9 ◽  
Author(s):  
Marlon Capuno ◽  
Jung-Su Kim ◽  
Hwachang Song
Keyword(s):  
Prediction Model ◽  
Support Vector Regression ◽  
Peak Load ◽  
Load Forecasting ◽  
Weather Forecast ◽  
Warning Signal ◽  
Support Vector ◽  
Short Term ◽  
Load Demand ◽  
Short Term Load Forecasting

This paper presents a model for very short-term load forecasting (VSTLF) based on algebraic prediction (AP) using a modified concept of the Hankel rank of a sequence. Moreover, AP is coupled with support vector regression (SVR) to accommodate weather forecast parameters for improved accuracy of a longer prediction horizon; thus, a hybrid model is also proposed. To increase system reliability during peak hours, this prediction model also aims to provide more accurate peak-loading conditions when considerable changes in temperature and humidity happen. The objective of going hybrid is to estimate an increase or decrease on the expected peak load demand by presenting the total MW per Celsius degree change (MW/C°) as criterion for providing a warning signal to system operators to prepare necessary storage facilities and sufficient reserve capacities if urgently needed by the system. The prediction model is applied using actual 2014 load demand of mainland South Korea during the summer months of July to September to demonstrate the performance of the proposed prediction model.

Prediction of Glass Transition Temperature of Polymer by Support Vector Regression

2012 ◽  
Vol 455-456 ◽  
pp. 436-442
Author(s):  
J.F. Pei ◽  
C.Z. Cai ◽  
X.J. Zhu ◽  
G.L. Wang ◽  
B. Yan
Keyword(s):  
Glass Transition ◽  
Transition Temperature ◽  
Glass Transition Temperature ◽  
Support Vector Regression ◽  
Density Functional ◽  
Absolute Error ◽  
Percentage Error ◽  
Support Vector ◽  
Quantum Chemical Descriptors ◽  
Leave One Out

. Based on two quantum chemical descriptors (the thermal energy Ethermal and the total energy of the whole system EHF) calculated from the structures of the repeat units of polyacrylamides by density functional theory (DFT), the support vector regression (SVR) approach combined with particle swarm optimization (PSO), is proposed to establish a model for prediction of the glass transition temperature (Tg) of polyacrylamides. The prediction performance of SVR was compared with that of multivariate linear regression (MLR). The results show that the mean absolute error (MAE=4.65K), mean absolute percentage error (MAPE=1.28%) and correlation coefficient (R2=0.9818) calculated by leave-one–out cross validation (LOOCV) via SVR models are superior to those achieved by QSPR (MAE=14.25K, MAPE=4.39% and R2=0.9211) and QSPR-LOO (MAE=17.01K, MAPE=5.66% and R2=0.8823) models for the identical samples, respectively. The prediction results strongly demonstrate that the modeling and generalization abilities of SVR model consistently surpass those of QSPR and QSPR-LOO models. It is revealed that the established SVR model is more suitable to be used for prediction of the Tg values for unknown polymers possessing similar structure than the conventional MLR approach. These suggest that SVR is a promising and practical methodology to predict the glass transition temperature of polyacrylamides.

scholarly journals An Accurate Medium-Term Load Forecasting based on Hybrid Technique

2018 ◽  
Vol 12 (1) ◽  
pp. 161
Author(s):  
Z.M. Yasin ◽  
N.F.A. Aziz ◽  
N.A. Salim ◽  
N.A. Wahab ◽  
N.A. Rahmat
Keyword(s):  
Support Vector Machine ◽  
Load Forecasting ◽  
Least Square ◽  
Percentage Error ◽  
Support Vector ◽  
Hybrid Technique ◽  
Medium Term ◽  
Load Demand ◽  
Ant Lion Optimizer ◽  
Ant Lion

An accurate medium term load forecasting is significant for power generation scheduling, economic and reliable operation in power system. Most of classical approach for medium term load forecasting only consider total daily load demand. This approach may not provide accurate results since the load demand is fluctuated in a day. In this paper, a hybrid Ant-Lion Optimizer Least-square Support Vector Machine (ALO-LSSVM) is proposed to forecast 24-hour load demand for the next year. Ant-Lion Optimizer (ALO) is utilized to optimize the RBF Kernel parameters in Least-Square Support Vector Machine (LS-SVM). The objective of the optimization is to minimize the Mean Absolute Percentage Error (MAPE). The performance of ALO-LSSVM technique was compared with those obtained from LS-SVM technique through a 10-fold cross-validation procedure. The historical hourly load data are analyzed and appropriate features are selected for the model. There are 24 inputs and 24 outputs vectors for this model which represents 24-hour load demand for whole year. The results revealed that the high accuracy of prediction could be achieved using ALO-LSSVM.

scholarly journals Path Smoothing With Support Vector Regression

2020 ◽  
Vol 4 (1) ◽  
pp. 142-150
Author(s):  
Donni Richasdy ◽  
Saiful Akbar
Keyword(s):  
Support Vector Regression ◽  
Learning Algorithm ◽  
Absolute Error ◽  
Moving Object ◽  
Percentage Error ◽  
Support Vector ◽  
Average Value ◽  
Path Smoothing ◽  
Aircraft Tracking ◽  
Smoothing Process

One of moving object problems is the incomplete data that acquired by Geo-tracking technology. This phenomenon can be found in aircraft ground-based tracking with data loss come near to 5 minutes. It needs path smoothing process to complete the data. One solution of path smoothing is using physics of motion, while this research performs path smoothing process using machine learning algorithm that is Support Vector Regression (SVR). This study will optimize the SVR configuration parameters such as kernel, common, gamma, epsilon and degree. Support Vector Regression will predict value of the data lost from aircraft tracking data. We use combination of mean absolute error (MAE) and mean absolute percentage error (MAPE) to get more accuracy. MAE will explain the average value of error that occurs, while MAPE will explain the error percentage to the data. In the experiment, the best error value MAE 0.52 and MAPE 2.07, which means error data ± 0.52, this is equal to 2.07% of the overall data value.Keywords: Moving Object, Path Smoothing, Support Vector Regression, MAE

scholarly journals Medium-Term Regional Electricity Load Forecasting through Machine Learning and Deep Learning

Designs ◽  
2021 ◽  
Vol 5 (2) ◽  
pp. 27
Author(s):  
Navid Shirzadi ◽  
Ameer Nizami ◽  
Mohammadali Khazen ◽  
Mazdak Nik-Bakht
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Smart Grids ◽  
Load Forecasting ◽  
Percentage Error ◽  
Support Vector ◽  
Medium Term ◽  
Load Demand ◽  
Electricity Load ◽  
Electricity Load Forecasting

Due to severe climate change impact on electricity consumption, as well as new trends in smart grids (such as the use of renewable resources and the advent of prosumers and energy commons), medium-term and long-term electricity load forecasting has become a crucial need. Such forecasts are necessary to support the plans and decisions related to the capacity evaluation of centralized and decentralized power generation systems, demand response strategies, and controlling the operation. To address this problem, the main objective of this study is to develop and compare precise district level models for predicting the electrical load demand based on machine learning techniques including support vector machine (SVM) and Random Forest (RF), and deep learning methods such as non-linear auto-regressive exogenous (NARX) neural network and recurrent neural networks (Long Short-Term Memory—LSTM). A dataset including nine years of historical load demand for Bruce County, Ontario, Canada, fused with the climatic information (temperature and wind speed) are used to train the models after completing the preprocessing and cleaning stages. The results show that by employing deep learning, the model could predict the load demand more accurately than SVM and RF, with an R-Squared of about 0.93–0.96 and Mean Absolute Percentage Error (MAPE) of about 4–10%. The model can be used not only by the municipalities as well as utility companies and power distributors in the management and expansion of electricity grids; but also by the households to make decisions on the adoption of home- and district-scale renewable energy technologies.

scholarly journals Long-term load forecasting using grey wolf optimizer -least-squares support vector machine

2020 ◽  
Vol 9 (3) ◽  
pp. 417
Author(s):  
Z. M. Yasin ◽  
N. A. Salim ◽  
N.F.A. Aziz ◽  
Y.M. Ali ◽  
H. Mohamad
Keyword(s):  
Support Vector Machine ◽  
Peak Load ◽  
Load Forecasting ◽  
Least Square ◽  
Percentage Error ◽  
Support Vector ◽  
Grey Wolf Optimizer ◽  
Grey Wolf ◽  
Load Demand

<p><span lang="EN-US">Long term load forecasting data is important for grid expansion and power system operation. Besides, it also important to ensure the generation capacity meet electricity demand at all times. In this paper, Least-Square Support Vector Machine (LSSVM) is used to predict the long-term load demand. Four inputs are considered which are peak load demand, ambient temperature, humidity and wind speed. Total load demand is set as the output of prediction in LSSVM. In order to improve the accuracy of the LSSVM, Grey Wolf Optimizer (GWO) is hybridized to obtain the optimal parameters of LSSVM namely GWO-LSSVM. Mean Absolute Percentage Error (MAPE) is used as the quantify measurement of the prediction model. The objective of the optimization is to minimize the value of MAPE. The performance of GWO-LSSVM is compared with other methods such as LSSVM and Ant Lion Optimizer – Least-Square Support Vector Machine (ALO-LSSVM). From the results obtained, it can be concluded that GWO-LSSVM provide lower MAPE value which is 0.13% as compared to other methods.</span></p>

Sign in / Sign up

Export Citation Format

Share Document