Integrated architecture for short term load forecasting using support vector machines

This research paper investigates the application of artificial neural networks (ANNs) and fuzzy logic methods for short-term load forecasting in the Western Libyan Electric Network. Historical hourly load data collected during the year 2023 by the General Electricity Company of Libya (GECOL). The study aims to develop accurate and reliable load forecasting models to support efficient grid operation and resource planning. Historical load data and meteorological variables are utilized to train and evaluate the forecasting models. The performance of the ANNs and fuzzy logic methods is compared, and the suitability of each approach for load forecasting in the Western Libyan Electric Network is assessed. The results demonstrate the effectiveness of both ANNs and fuzzy logic methods in short-term load forecasting, providing valuable insights for electricity providers and policymakers in the region. Keywords: neural' networks, fuzzy logic, load forecasting

Smart distribution network gathers massive load data with SCADA system. Bus load forecasting needs to increase the process speed while assuring the forecasting accuracy, to meet the real-time requirements. This paper proposed a fast bus load forecasting method for smart distribution system. This method utilizes the historical load and temperature data to establish a forecasting model, which derived the reference load data of the forecasting day. With temperature data of the forecasting day and the historical data, it is deduced the load correction of the forecasting day to realize online load forecasting, which increases the forecasting speed. By the instance analysis, the speediness and accuracy of the proposed method is verified.

Support Vector Machine (SVM) is a precise and fast method for the prediction of short-term electrical load and the similar day method is a simple and direct method for load forecasting. This paper tries to combine SVM model and similar day method for next day load forecasting. The proposed method forecasts the load of next day using SVM. Then, the load curve of a similar day is selected to correct the curve forecasted by SVM, which can avoid the appearance of large forecasting error effectively. Corresponding software was developed and used to forecast the next day load in a practical power system, and the final forecasting result is accurate and reliable.

Network traffic prediction performs a main function in characterizing network community performance. An approach which could appropriately seize the salient characteristics of the network visitors could be very useful for network analysis and simulation. Network traffic prediction methods could be divided into two classes: one is the single models and the opposite is the hybrid fashions. The hybrid models integrate the merits of several single models and consequently can enhance the network traffic prediction accuracy. In this paper, a new hybrid network traffic prediction method (EPSVM) primarily based on Empirical Mode Decomposition (EMD), Particle Swarm Optimization (PSO), and Support Vector Machines (SVM) is presented. The EPSVM first utilizes EMD to eliminate the impact of noise signals. Then, SVM is applied to model training and fitting, and the parameters of SVM are optimized by PSO. The effectiveness of the presented method is examined by evaluating it with different methods, including basic SVM (BSVM), Empirical Mode Decomposition processed by SVM (ESVM), and SVM optimized by Particle Swarm Optimization (PSVM). Case studies have demonstrated that EPSVM performed better than the other three network traffic prediction models.

For the planning, operation, and control of power system, load forecasting plays a vital role. It is necessary to forecast the electrical load accurately for the efficient working of distribution systems. Besides these, it is also essential for security analysis, unit commitment, fuel purchase scheduling and maintenance schedule. To minimize the errors in forecasting, the paper uses Support Vector Machines (SVM) and Artificial Neural Networks (ANN) as machine learning models for the load forecasting. Real measurements collected in Indian Institute of Technology Kanpur (IITK) substation are used to validate the study. The data used in the modeling of Feed Forward Neural Network (FFNN) are hourly electricity load and historical temperature data. The hourly load data of IITK substation from April-2017 to June-2017 is used for training of FFNN model and SVM models, and testing was done on test data for next week. It can be seen from the simulation results that hourly load forecast using proposed SVM models are more accurate than FFNN model.

A variety of academic theories are used to research the prediction algorithm including support vector machine. In order to improve instantaneity and accuracy of prediction, wavelet support vector machine prediction algorithm is studied, the structure and process of the algorithm is described also. Mackey-Glass <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">17</sub> time series is used as samples for the prediction result simulation, that is compared with the prediction effect used by basic support vector machine. The result shows that Root Mean Square Error of wavelet support vector machine prediction algorithm is 0.0075, that is superior to the basic support vector machine prediction algorithm.

The evolution of conventional power systems into smart grids demands intelligent and efficient energy management solutions. Accurate load forecasting is essential for optimal generation scheduling, demand-side management, and system reliability. However, traditional methods often fail to capture the nonlinear and time-varying nature of electrical loads. This work proposes a machine learning–based approach for local load forecasting and grid stability enhancement. Techniques such as Artificial Neural Networks (ANN) and Support Vector Machines (SVM) are utilized to analyze historical load and weather data for improved prediction accuracy. In addition, a predictive control strategy is implemented to maintain system stability by mitigating voltage fluctuations, frequency deviations, and load imbalances. The proposed model dynamically adjusts control actions such as demand response and load shedding. Simulation results demonstrate reduced forecasting errors (MAE, RMSE) and improved voltage and frequency regulation, ensuring enhanced reliability, efficiency, and stable operation of modern smart grid systems. Keywords: Smart Grid; Machine Learning; Load Forecasting; Grid Stability; Predictive Control.

Short-term load forecasting plays a vital role for the power system’s safe operation and production arrangements. This paper regards electric load forecasting as a nonlinear time series prediction problem, and establishes the electric load autoregressive prediction model (ARX Model) based on historical load data, approaching ARX model parameters with RBF neural network, estimating model parameters with using a Structured Nonlinear Parameters Optimization Method (SNPOM), proposing cycle prediction method for short-term electric load forecasting. Experimental results show that the method has higher prediction accuracy for short-term electric load forecasting. Keywords: Short-term load forecasting; time series; RBF-ARX model; cycle forecasting; structured nonlinear parameter optimization method

Small Modular Reactors (SMRs) are gaining increasing attention as flexible and low-emission energy sources capable of supporting the transition to a sustainable power grid. Their ability to perform load-following operations is critical for ensuring grid reliability in systems with high shares of variable energy resources, such as wind and solar. This paper presents a seasonal assessment of SMR load-following capabilities using forecasted load demand profiles obtained from hybrid deep learning models. Multiple forecasting models were evaluated to first identify the most accurate short-term electricity load demand forecast model. A hybrid deep learning model based on the combination of Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory (BiLSTM), i.e., hybrid CNN+BiLSTM, achieved the best performance by generating more accurate week-ahead load forecasts based on historical temperature and load data. These forecasts were used to simulate the response of a dynamic SMR model developed in Simulink, incorporating realistic ramp-rate and control constraints. Performance of the models is evaluated across four seasonal scenarios and test results show that the SMR is able to effectively track forecasted load profiles year-round, with Mean Absolute Percentage Error (MAPE) below 1.5%. These findings demonstrate the potential of forecast-informed SMR operation as a reliable and dispatchable clean energy solution for future grid architectures.

Aiming at the shortcomings of traditional load forecasting strategy, this paper considers the comprehensive application of various meteorological factors in load forecasting, and it proposes a power system short term load forecasting method based on the least squares support vector machine (LSSVM) model. Firstly, it provides a theoretical basis for selecting modeling variables by correlation analysis, and deals with the load data to optimize the sample quality by similarity principle. Secondly, it establishes a weather LSSVM load forecasting model, and the model is trained and tested by historical load data of a certain region. Finally, the accuracy and validity of the prediction model are verified by simulation.

The accurate prediction of power system load is extremely important for the operation of the power market and the safe operation of the power grid. In order to improve the accuracy of short-term load forecasting of power systems, a combination model based on long short term memory network (LSTM) and light gradient boosting machine (LightGBM) is proposed. The experiment first decomposes historical load data by EMD, uses historical weather data and load data decomposed by EMD to establish LSTM prediction model and LightGBM prediction model respectively, and then these two predicted values are linearly combined to obtain the final predicted value. The electrical load data of the 2016 Electrician Mathematical Contest in Modeling is used as an example to verify. The experimental results show that the LSTM-LightGBM combined model has higher forecasting accuracy and application prospects for power load forecasting than traditional load forecasting methods and standard LSTM and LightGBM load forecasting methods.

This paper aims to develop a load forecasting method for short-term load forecasting, based on an adaptive two-stage hybrid network with self-organized map (SOM) and support vector machine (SVM). In the first stage, a SOM network is applied to cluster the input data set into several subsets in an unsupervised manner. Then, groups of 24 SVMs for the next day's load profile are used to fit the training data of each subset in the second stage in a supervised way. The proposed structure is robust with different data types and can deal well with the nonstationarity of load series. In particular, our method has the ability to adapt to different models automatically for the regular days and anomalous days at the same time. With the trained network, we can straightforwardly predict the next-day hourly electricity load. To confirm the effectiveness, the proposed model has been trained and tested on the data of the historical energy load from New York Independent System Operator.

The present study proposes a short-term load forecasting method based on weighted grey relational degree and improved support vector machines with the artificial bee colony algorithm (ABC-SVM). The entropy weight method was employed to obtain the weight of load-related physical information, and the historical and forecast load data selected based on the weighted grey relational degree were input into the support vector machine (SVM) to build a forecasting model. Meanwhile, the SVM parameters were optimized by the improved artificial bee colony algorithm before the model was used to perform load forecasting. The experimental results show that the proposed method could effectively improve the accuracy of the forecasting model and simplify the calculation, thus having research and practical value.

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.

This paper presents an improved support vector machine method based on the Hadoop, which can solve the problem of short-term load forecasting in the big data environment. In the control design, the fuzzy C clustering model of quantum behaved particle swarm optimization is used to analyze the historical load data. Then, the load data and the influencing factors data are stored and matched in Cloud platform, through this way which can improve the accuracy of the input data and the speed of the response. At last, the model is established. According to the certain area load big data, the improved support vector machine forecast model and the normal support vector machine forecast model are compared in forecasting accuracy and forecasting time. The results indicate that Hadoop-SVM has better predictive ability and faster computing time when processing the big data.