Accurate Load Prediction Research Articles

Intelligent grid load forecasting based on BERT network model in low-carbon economy

In recent years, the reduction of high carbon emissions has become a paramount objective for industries worldwide. In response, enterprises and industries are actively pursuing low-carbon transformations. Within this context, power systems have a pivotal role, as they are the primary drivers of national development. Efficient energy scheduling and utilization have therefore become critical concerns. The convergence of smart grid technology and artificial intelligence has propelled transformer load forecasting to the forefront of enterprise power demand management. Traditional forecasting methods relying on regression analysis and support vector machines are ill-equipped to handle the growing complexity and diversity of load forecasting requirements. This paper presents a BERT-based power load forecasting method that leverages natural language processing and image processing techniques to enhance the accuracy and efficiency of transformer load forecasting in smart grids. The proposed approach involves using BERT for data preprocessing, analysis, and feature extraction on long-term historical load data from power grid transformers. Multiple rounds of training and fine-tuning are then conducted on the BERT architecture using the preprocessed training datasets. Finally, the trained BERT model is used to predict the transformer load, and the predicted results are compared with those obtained based on long short-term memory (LSTM) and actual composite values. The experimental results show that compared with LSTM method, the BERT-based model has higher short-term power load prediction accuracy and feature extraction capability. Moreover, the proposed scheme enables high levels of accuracy, thereby providing valuable support for resource management in power dispatching departments and offering theoretical guidance for carbon reduction initiatives.

Research on short-term load forecasting of new-type power system based on GCN-LSTM considering multiple influencing factors

With the construction of new-type power system under the ”double carbon” target and the increasing diversification of the energy demand of the user side, the short-term load forecasting of power system is facing new challenges. In order to fully exploit the massive information contained in big data, this paper proposed a new short-term load forecasting method for new-type power system considering multiple factors, which based on Graph Convolutional Network (GCN) and Long Short-Term Memory network (LSTM). Spearman rank correlation coefficient was used to analyze the correlation between load and meteorological factors, and a quantitative model including meteorological factors, date factors and regional factors was established. Thus, GCN and LSTM were jointly used to extract the spatial and temporal characteristics of massive data respectively, and finally the short-term power load forecasting was achieved. The public data sets were used for performance verification compared with three comparison models, LSTM, CNN-LSTM and TCN-LSTM. The results show that the proposed method can make full use of the influence of multi-dimensional data, meanwhile improve the load prediction accuracy and training efficiency effectively.

Numerical study of trimaran wave load based on time-domain Rankine method

Due to the complex fluid interference between outrigger and main hull, it is difficult to predict the trimaran load fluctuation accurately in different marine environments. Therefore, a time-domain Rankine method is developed to improve the accuracy of trimaran load prediction. This method successfully adds the nonlinear load components in time-domain load simulation by taking into account the factors such as instantaneous hull wetted surface, steady ship waves, green wave and slamming. Additionally, the nonlinear growth of green wave and slamming is also observed at outrigger and wet deck, with the increasing of speeds and wave heights. Finally, the relatively smaller errors in the Rankine method are confirmed by comparison with the values from a trimaran model test and linear time-domain load prediction method. Furthermore, the weakly nonlinear Rankine method is considered to be more suitable for trimaran load prediction under harsh marine environment.

A Short-term Electricity Load Forecasting Method Based on WOA-BiLSTM-Attention

Under the “double carbon” policy, it is time to build a new green and safe power system. Special attention should be paid to the low accuracy in short-term power load prediction. Therefore, this paper presents a technique for anticipating short-term power demand based on a bidirectional long short-term memory network with a whale-optimized attention mechanism (WOA-BiLSTM-Attention), which is used to forecast and analyze the measured power values in a certain area. The experimental findings reveal that the suggested technique has much greater prediction accuracy and convergence speed, as well as superior stability, when compared to LSTM, BiLSTM, and BiLSTM based on attention mechanism, making it a good reference for power system planning and stability.

A Short-term Power Load Forecasting Method Based on An Improved VMD-LSTM Algorithm

Accurate power load prediction is a crucial basis to ensure the balance of power supply and demand. This paper provides a modified variational modal decomposition (VMD) - long short-term memory (LSTM) algorithm, using genetic algorithm (GA) to iteratively optimize the number of decomposition layers and penalty factor parameters, and using VMD to obtain the modal components, fuse them with the filtered features. Then the reconstruction data is obtained through normalization according to the parameters of the original load characteristic data set. Finally, LSTM is constructed to excavate temporal characteristics and obtain power load prediction results. Simulation fully demonstrates the superior performance of this modified model in terms of prediction effect.

Enhancing smart home energy efficiency through accurate load prediction using deep convolutional neural networks

Enhancing smart home energy efficiency through accurate load prediction using deep convolutional neural networks

Real-time load forecasting model for the smart grid using bayesian optimized CNN-BiLSTM

A smart grid is a new type of power system based on modern information technology, which utilises advanced communication, computing and control technologies and employs advanced sensors, measurement, communication and control devices that can monitor the status and operation of various devices in the power system in real-time and optimise the dispatch of the power system through intelligent algorithms to achieve efficient operation of the power system. However, due to its complexity and uncertainty, how to effectively perform real-time prediction is an important challenge. This paper proposes a smart grid real-time prediction model based on the attention mechanism of convolutional neural network (CNN) combined with bi-directional long and short-term memory BiLSTM.The model has stronger spatiotemporal feature extraction capability, more accurate prediction capability and better adaptability than ARMA and decision trees. The traditional prediction models ARMA and decision tree can often only use simple statistical methods for prediction, which cannot meet the requirements of high accuracy and efficiency of real-time load prediction, so the CNN-BiLSTM model based on Bayesian optimisation has the following advantages and is more suitable for smart grid real-time load prediction compared with ARMA and decision tree. CNN is a hierarchical neural network structure containing several layers such as a convolutional layer, pooling layer and fully connected layer. The convolutional layer is mainly used for extracting features from data such as images, the pooling layer is used for the dimensionality reduction of features, and the fully connected layer is used for classification and recognition. The core of CNN is the convolutional operation, a locally weighted summation operation on the input data that can effectively extract features from the data. In the convolution operation, different features can be extracted by setting different convolution kernels to achieve feature extraction and classification of data. BiLSTM can capture semantic dependencies in both directions. The BiLSTM structure consists of two LSTM layers that process the input sequence in the forward and backward directions to combine the information in both directions to obtain more comprehensive contextual information. BiLSTM can access both the front and back inputs at each time step to obtain more accurate prediction results. It effectively prevents gradient explosion and gradient disappearance while better capturing longer-distance dependencies. The CNN-BiLSTM extracts features of the data and then optimises them by Bayes. By collecting real-time data from the power system, including power, load, weather and other factors, our model uses the features of CNN-BiLSTM to deeply learn real-time load data from smart grids and extract key features to achieve future load prediction. Meanwhile, the Bayesian optimisation algorithm based on the model can optimise the model’s hyperparameters, thus improving the model’s prediction performance. The model can achieve accurate prediction of a real-time power system load, provide an important reference for the dispatch and operation of the power system, and help optimise the operation efficiency and energy utilisation efficiency of the power system.

Deep learning predictions of unsteady aerodynamic loads on an airfoil model pitched over the entire operating range

For the design and certification of wind turbines, it is essential to provide fast and accurate unsteady aerodynamic load prediction models for the whole operational range of angle of attack, up to 180° for vertical-axis and 90° for horizontal-axis wind turbines. This work describes a computationally efficient unsteady forces prediction model based on a deep learning approach, namely the bidirectional long short-term memory (BiLSTM) algorithm, for an airfoil pitched over the full operational range of angles of attack up to 180°. No model has been developed to capture the unsteady forces at high angles of attack. Novel features based on operating conditions and the steady polars of the airfoil are used as inputs for the BiLSTM model. Direct measurements of steady and unsteady forces on a NACA 0021 airfoil model were conducted at reduced frequencies up to 0.075 and a Reynolds number of 120 000 in an open-jet wind tunnel for model learning and testing. The unsteady forces vary significantly from the steady values at high pitching amplitudes and post-stall angles, which, if not accounted for when simulating wind turbine performance, would result in inaccurate predictions. Furthermore, measurements revealed the effect of unsteady vorticity development and shedding on aerodynamic forces under forward and reverse flow conditions. The BiLSTM model is capable of capturing the underlying physics of unsteady aerodynamic forces under extreme operating conditions.

Optimized Seq2Seq model based on multiple methods for short-term power load forecasting

Accurate power load prediction plays a key role in reducing resource waste and ensuring stable and safe operations of power systems. To address the problems of poor stability and unsatisfactory prediction accuracy of existing prediction methods, in this paper, we propose a novel approach for short-term power load prediction by improving the sequence to sequence (Seq2Seq) model based on bidirectional long-short term memory (Bi-LSTM) network. Different from existing prediction models, we apply convolutional neural network, attention mechanism, and Bayesian optimization for the improvement of the Seq2Seq model. Moreover, in the data processing stage, we use the random forest algorithm for feature selection, and also adopt the weighted grey relational projection algorithm for holiday load processing to process the data and thereby overcome the difficulty of holiday load prediction. To validate our model, we choose the power load dataset in Singapore and Switzerland as experimental data and compare our prediction results with those by other models to show that our method can generate a higher prediction accuracy.

Pre-Attention Mechanism and Convolutional Neural Network Based Multivariate Load Prediction for Demand Response

The construction of smart grids has greatly changed the power grid pattern and power supply structure. For the power system, reasonable power planning and demand response is necessary to ensure the stable operation of a society. Accurate load prediction is the basis for realizing demand response for the power system. This paper proposes a Pre-Attention-CNN-GRU model (PreAttCG) which combines a convolutional neural network (CNN) and gate recurrent unit (GRU) and applies the attention mechanism in front of the whole model. The PreAttCG model accepts historical load data and more than nine other factors (including temperature, wind speed, humidity, etc.) as input. The attention layer and CNN layer effectively extract the features and weights of each factor. Load forecasting is then performed by the prediction layer, which consists of a stacked GRU. The model is verified by industrial load data from a German dataset and a Chinese dataset from the real world. The results show that the PreAttCG model has better performance (3~5% improvement in MAPE) than both LSTM with only load input and LSTM with all factors. Additionally, the experiments also show that the attention mechanism can effectively extract the weights of relevant factors affecting the load data.

Machine-learning-based models versus design-oriented models for predicting the axial compressive load of FRP-confined rectangular RC columns

To improve the prediction accuracy of axial compressive load of FRP-confined concrete columns, machine-learning techniques have been used recently. However, few studies have used machine learning to estimate the axial compressive load of FRP-confined rectangular RC columns. Therefore, this study introduces a state-of-art review of externally strengthened rectangular concrete columns with FRP composites: the influential parameters were introduced and their effects on the strength, ductility, and failure mode of FRP-strengthened columns were discussed. Hence, the critical design parameters were identified and used as input features in machine-learning modeling. From a practical point of view, special attention was paid to collecting any dataset related to steel reinforced rectangular concrete columns and externally confined with different types of FRP composites. These collected datasets were used to generate machine-learning models to predict the axial compressive load of rectangular RC columns confined by external FRP sheets, and were compared with existing design-oriented models. The proposed machine-learning models are found to be in good agreement with the test results in the datasets. In the comparison between the existing design-oriented models and the developed machine-learning models, the gradient boosting (GB) and random forest (RF) regressors are more accurate, and both methods achieved the lowest deviation value.

Data driven machine learning models for short‐term load forecasting considering electrical vehicle load

AbstractElectric vehicles (EVs) are gaining popularity due to their fuel efficiency and ability to reduce greenhouse gas emissions. Significant penetration of EVs with unregulated charging, which can have a substantial impact on power networks. Accurate load predictions, including the charging of EVs are crucial for ensuring the cost‐effective and dependable operation of power systems. In order to estimate short‐term load in the presence of EV load, the XG (extreme gradient) boost algorithm is proposed and the effectiveness of the performances are checked against other models. A variety of distinct meteorological parameters and the electrical load pattern for the years 2017 and 2018 for Northeast India are used to train the machine learning classifier models. R‐squared value analyses are also performed to identify the most correlated input parameters that influence the results of various models. Analysis shows that temperature, cloud cover, heat index, dew point, wind chill, and perceived temperature are substantially connected with electricity consumption. The performance of XG boost is outperformed by comparing prediction results with decision tree, random forest, K nearest neighbors, and logistic regression. Three separate case studies were employed for verification using the precision, F1 score, sensitivity, specificity, and accuracy metrics. Our findings demonstrate that, in comparison to other forecasting models, XG Boost exhibits higher accuracy (83.84%, 81.51%, and 85.97%) and robust forecasts.

Seasonal peak load prediction of underground gas storage using a novel two-stage model combining improved complete ensemble empirical mode decomposition and long short-term memory with a sparrow search algorithm

Accurate seasonal peak load (SPL) prediction of underground gas storage (UGS) is of great significance for enterprises to formulate scheduling plans. In this work, a novel two-stage model is proposed to predict it. First, improved complete ensemble empirical mode decomposition (ICEEMDAN) is used to decompose the seasonal peak load into several intrinsic mode function. Second, the highest frequency components are smoothed by using Gaussian smoothing (GS), and second highest frequency component is decomposed into several components by applying wavelet transform (WT). Third, long short-term memory (LSTM) optimized by improved sparrow search algorithm (ISSA) is utilized to predict them, and the prediction results of these components are recombined to obtain the final prediction results. Finally, the historical data of Wen 23 in the Zhongyuan gas storage group is taken as a case, and three key issues are discussed. The results: (1) Utilizing ICEEMDAN, GS, WT, and ISSA can effectively improve the prediction performance of the LSTM; (2) The prediction performance of one-step is better than that of multi-step; (3) The comprehensive error and stability evaluation index is reasonable. The following conclusion is reached: the established prediction model can be used as a reference for the development of the UGS scheduling platform.

A Flight Parameter-Based Aircraft Structural Load Monitoring Method Using a Genetic Algorithm Enhanced Extreme Learning Machine

High-precision operational flight loads are essential for monitoring fatigue of individual aircraft and are usually determined by flight parameters. To tackle the nonlinear relationship between flight loads and flight parameters for more accurate prediction of flight loads, artificial neural networks have been widely studied. However, there are still two major problems, namely the training strategy and sensitivity analysis of the flight parameters. For the first problem, the gradient descent method is usually used, which is time-consuming and can easily converge to a local solution. To solve this problem, an extreme learning machine is proposed to determine the weights based on a Moore–Penrose generalized inverse. Moreover, a genetic algorithm method is proposed to optimize the weights between the input and hidden layers. For the second problem, a mean impact value (MIV) method is proposed to measure the sensitivity of the flight parameters, and the neuron number in the hidden layer is also optimized. Finally, based on the measured dataset of an aircraft, the proposed flight load prediction method is verified to be effective and efficient. In addition, a comparison is made with some well-known neural networks to demonstrate the advantages of the proposed method.

A novel combined model for heat load prediction in district heating systems

Accurate heat load prediction is essential for improving the operational efficiency of district heating systems (DHSs). Numerous heat load prediction models have been proposed to improve the accuracy and stability. However, previous prediction models are limited by training data and hyperparameters and cannot obtain accurate and stable prediction results. This study proposed a combined prediction model to overcome the limitations of a single model. In the proposed model, four classical single models were selected as individual models for prediction. Subsequently, the minimum sum of squares of the combined prediction errors algorithm was used to calculate the weighting coefficient of each individual model. The final prediction results were obtained by combining the prediction results of the four models. The 1-h historical operation data from three heat substations in a DHS in northeast China were adopted to evaluate the performance of the proposed model. The multi-step prediction experiment results of the three datasets revealed that the average mean absolute percentage error values of the proposed model for the one-step, two-step, and three-step predictions reached 2.6%, 4.9%, and 7.0%, respectively; these percentages were significantly lower than those of the four individual models and three representative combined models. Therefore, the proposed model has better prediction performance than previous models in terms of accuracy and stability and can provide a theoretical basis for the precise regulation of DHSs.

Metro Electric Traction Load Forecasting Based on Prophet-GRU Hybrid Model

In order to further improve the metro electric traction load forecasting and provide support for energy conservation and sustainable development of urban rail transit. In this paper, a Prophet-GRU hybrid model based on weight selection is proposed. This model combines the advantages of Prophet and GRU, takes account of timing characteristics and influence factors synthetically, and improves the accuracy of electric traction load prediction. This paper first introduces the background of electric metro traction load and the model used in this paper and then proposes electric traction load forecasting based on the Prophet-GRU hybrid model. Taking a subway line as an example, firstly, the Prophet and GRU are used for prediction respectively, and then the error reciprocal method is used to combine the two models to obtain the final forecasting result. Compared with the single Prophet and GRU, it is verified that the model has a higher precision forecasting ability.

Short-term Power Load Forecasting Based on Orthogonal PCA-LPP Dimension Reduction and IGWO-BiLSTM

Background: Accurate power load forecasting is of great significance in ensuring power load planning, reliability and economic operation. The traditional power load is easy to be affected by climate, environment, date type and other factors, resulting in the problem of poor forecasting accuracy. Therefore, it is necessary to study power load forecasting. Objective: Through machine learning, dimension reduction method and intelligent optimization algorithm, the accuracy of load forecasting is improved Methods: In order to fully extract load information and improve the accuracy of short-term load forecasting for campus electricity, an improved combination of orthogonal dimensionality reduction and Bilstm is proposed to optimize the hyperparameters in BiLSTM using an improved gray wolf algorithm. Firstly, using the advantages of principal component analysis (PCA) and Locality Preserving Projection (LPP) to maintain the global and local structure of the data, respectively, the Orthogonal PCA-LPP（OPCA-LPP） dimensionality reduction method is proposed to reduce the dimensionality of the original multidimensional data. Finally, the dimensionality-reduced data is used as the input of BiLSTM and optimized by the improved Gray Wolf algorithm, which can enhance the prediction capability of the model and thus achieve accurate prediction of short-term electric load. Results: The Mae and RMSE of this paper are 1.6585 and 1.7602 respectively. The results show that the method proposed in this paper is reasonable Conclusion: This method is applied to power load forecasting. The comparative experimental results show that this method reduces the dimension of data input, simplifies the complexity of network input data, and improves the accuracy of load forecasting. Compared with other methods, it can effectively improve the accuracy of load forecasting, and provide a basis for formulating reasonable power grid operation mode and balanced dispatching of power grid.

Fractional multivariate grey Bernoulli model combined with improved grey wolf algorithm: Application in short-term power load forecasting

The accurate prediction of power load is helpful to make reasonable power generation plans and scientific dispatching schemes and achieve the goal of energy saving and emission reduction.There are many factors influencing the power load, and there may be a nonlinear relationship between the power load and these influencing factors. A new fractional multivariate grey Bernoulli model, referred to as MFGBM (q, r, N), is developed in this article for the short-term prediction of power load. The fractional differential equation and fractional accumulation generation are integrated into MFGBM (q, r, N). Second, this paper improves the grey wolf algorithm to better optimize many parameters in the model. The algorithm is improved by using a chaotic Tent map to optimize the initial population composition, adding inertia weights to change the position vector of the grey wolf, and introducing a nonlinear function and Lévy flight to enhance local exploitation and global exploration ability. Finally, this paper selects the daytime and nighttime power loads in Australia and takes the electricity price, humidity, and temperature as the influencing factors to validate the prediction capability of MFGBM (q, r, N). Our findings indicate that MFGBM (q, r, N) is highly applicable to short-term power system prediction.

Machine Learning Approach to Predict Building Thermal Load Considering Feature Variable Dimensions: An Office Building Case Study

An accurate and fast building load prediction model is critically important for guiding building energy system design, optimizing operational parameters, and balancing a power grid between energy supply and demand. A physics-based simulation tool is traditionally used to provide the building load demand; however, it is constrained by its complex model development process and requirement for engineering judgments. Machine learning algorithms (i.e., data-driven models) based on big data can bridge this gap. In this study, we used the massive energy data generated by a physics-based tool (EnergyPlus) to develop three data-driven models (i.e., LightGBM, random forest (RF), and long-short term memory (LSTM)) and compared their prediction performances. The physics-based models were developed using office prototype building models as baselines, and ranges were provided for selected key input parameters. Three different input feature dimensions (i.e., six-, nine-, and fifteen-input feature selections) were investigated, aiming to meet different demands for practical applications. We found that LightGBM significantly outperforms the RF and LSTM algorithms, not only with respect to prediction accuracy but also in regard to computation cost. The best prediction results show that the coefficient of variation of the root mean squared error (CVRMSE), squared correction coefficient (R2), and computation time are 5.25%, 0.9959, and 7.0 s for LightGBM, respectively, evidently better than the values for the algorithms based on RF (18.54%, 0.9482, and 44.6 s) and LSTM (22.06%, 0.9267, and 758.8 s). The findings demonstrate that a data-driven model is able to avoid the process of establishing a complicated physics-based model for predicting a building’s thermal load, with similar accuracy to that of a physics-based simulation tool.

Detection of Outliers in Time Series Power Data Based on Prediction Errors

The primary focus of smart grid power analysis is on power load forecasting and data anomaly detection. Efficient and accurate power load prediction and data anomaly detection enable energy companies to develop reasonable production and scheduling plans and reduce waste. Since traditional anomaly detection algorithms are typically for symmetrically distributed time series data, the distribution of energy consumption data features uncertainty. To this end, a time series outlier detection approach based on prediction errors is proposed in this paper, which starts by using an attention mechanism-based convolutional neural network (CNN)-gated recursive unit (GRU) method to obtain the residual between the measured value and its predicted value, and the residual data generally conform to a symmetric distribution. Subsequently, for these residual data, a random forest classification algorithm based on grid search optimization is used to identify outliers in the power consumption data. The model proposed in this paper is applied to both classical and real energy consumption datasets, and the performance is evaluated using different metrics. As shown in the results, the average accuracy of the model is improved by 25.2% and the average precision is improved by 17.2%, with an average recall improvement of 16.4% and an average F1 score improvement of 26.8% compared to the mainstream algorithms.