Historical Load Data Research Articles

Construction and Application of Short-Term and Mid-Term Power System Load Forecasting Model Based on Hybrid Deep Learning

Power system load forecasting model plays an important role in all aspects of power system planning, operation and control. Therefore, accurate power load forecasting provides an important guarantee for the stable operation of the power grid system. This paper first analyzes the current status of power system load forecasting, and finds that there are still some deficiencies in the existing forecasting models. In order to make up for these shortcomings, this paper proposes construction of short-term and mid-term power system load forecasting models based on hybrid deep learning. In the data preprocessing part, this paper proposes to use the exponential weight moving average method to process missing values. The detection method of abnormal value is the GeneralizedESDTestAD(GESD). This paper analyzes the historical load data of a regional power grid and four industries, and proposes a short-term power system load forecasting model based on Bi-directional Long Short-Term Memory(BiLSTM); For mid-term load forecasting, this paper first uses random forest and Pearson correlation coefficient to select features. Then a hybrid deep learning model is constructed based on BiLSTM and random forest. After optimizing the parameters of the model, a mid-term power system load forecasting model based on hybrid deep learning is constructed. Finally, the benchmark models are selected for comparative experiments. The experimental results show that the MAPE of the proposed model is 2.36%, which is better than the benchmark models. This proves that the proposed hybrid model can effectively improve the accuracy of power system load forecasting.

An investigation of the customer baseline load (CBL) calculation for industrial demand response participants –A regional case study from Iran

In today's power systems, performing demand response (DR) programs is considered a solution for sustainable operation, carbon reduction, and facilitating renewable energy integration. Currently in Iran, several DR programs are available to industrial customers. Moreover, the design of a market-based DR has recently been considered by the power system operator. In DR programs, customer baseline load (CBL) is used to determine the level of customers’ demand reduction and is key to the billing process; thus, the CBL calculation methodology can directly affect the efficiency of these programs. This paper investigates the current CBL calculation method in Iran's power system and, as its main contribution, proposes some effective methods to reduce the probable incentives for CBL manipulation. In order to show the effectiveness of the methods, a real study case is considered based on the historical load data of the industries in the Khorasan region.

Performance analysis and comparison of various techniques for short-term load forecasting

Rapidly varying load demand is one of the greatest problems that distribution system operators are now experiencing. Many researchers have been implemented the load demand requirement using traditional methods and machine learning methods. Both the methods have their own pros and cons according to dataset available for a particular site. This paper aims to predict load demand using different stochastic and deterministic approaches for short term load forecasting (STLF). Utilizing a variety of dependent characteristics, historical load data for two years is collected from IEEE dataport. The methods used in this study are exponential smoothing of traditional method and machine learning methods such as support vector machine (SVM), ensemble, artificial neural network (ANN), convolution neural network (CNN), long-short term memory (LSTM) and CNN-LSTM hybrid Model. All of these methods are compared and the best option is chosen of coefficient of correlation (R). The best method is found to be CNN-LSTM hybrid with R of 95.05%. The algorithm is implemented in PYTHON platform which is more compatible for machine learning applications.

Machine Learning Techniques for Decarbonizing and Managing Renewable Energy Grids

Given the vitality of the renewable-energy grid market, the optimal allocation of clean energy is crucial. An optimal dispatching method for source–load coordination of renewable-energy grid is proposed. An improved K-means clustering algorithm is used to preprocess the source data and historical load data. A support vector machine is used to predict the cluster of renewable-energy grid resources and load data, and typical scenarios are selected from the prediction results. Taking typical scenarios as a representative, the probability distribution of wind power output is accurately obtained. An optimization model of the total operation cost of the renewable-energy grid is established. The experimental results show that the algorithm reduces the error between the predicted value and the actual value. Our method can improve the real-time prediction accuracy of the renewable-energy grid system and increase the economic benefits of the renewable energy grid.

Implementation of EDGE Computing Platform in Feeder Terminal Unit for Smart Applications in Distribution Networks with Distributed Renewable Energies

Under the plan of net-zero carbon emissions in 2050, the high penetration of distributed renewable energies in distribution networks will cause the operation of more complicated distribution networks. The development of edge computing platforms will help the operator to monitor and compute the system status timely and locally, and it can ensure the security operation of the system. In this paper, a novel EDGE computing platform that is implemented by a graphics processing unit in the existing feeder terminal unit (FTU) is proposed for smart applications in distribution networks with distributed renewable energies and loads. This platform makes timely forecasts of the feeder status for the next seven days in accordance with historical weather, sun, and loading data. The forecast solver uses the machine learning long short-term memory (LSTM) method. Thereafter, the power calculation analyzers transform feeder topology into the circuit model for transient-state, steady-state, and symmetrical component analyses. An important-factor explainer parses the LSTM model into the concise value of each historical datum. All information transports to remote devices via the internet for the real-time monitor feature. The software stack of the EDGE platform consists of the database archive file system, time-series forecast solver, power flow analyzers, important-factor explainer, and message queuing telemetry transport (MQTT) protocol communication. All open-source software packages, such as SQLite, LSTM, ngspyce, Shapley Additive Explanations, and Paho-MQTT, form the aforementioned function. The developed EDGE forecast and power flow computing platform are helpful for achieving FTU becoming an Internet of Things component for smart operation in active distribution networks.

Short-term residential load forecasting using Graph Convolutional Recurrent Neural Networks

The abundance of energy consumption data collected by smart meters has inspired researchers to employ deep neural networks to solve the existing problems in the power industry, such as Short-Term Load Forecasting (STLF). Most studies addressing the STLF problem, focus on historical load data and to achieve higher performance, they supplement costly accessible environmental and calendar variables with data. This approach ignores the existing spatial information among the consumers which subsequently might lead into the emergence of similar consumption patterns. In this paper, we present a Graph Convolutional Recurrent Neural Network, a novel neural architecture, for STLF problem that combines Graph Convolutional Networks and Long Short-Term Memory networks to simultaneously extract spatial and temporal information from users with similar consumption patterns. Our model captures spatial information from users without prior knowledge of their geographic location and does not rely on additional environmental variables. We compared our model to traditional baseline models for STLF using two real-world electricity consumption datasets. The empirical results demonstrate a significant improvement in prediction compared with the baseline models, exhibiting a 9.5% and an 8% improvement in terms of Mean Absolute Percentage Error, in the Customer Behavior Trials and Low Carbon London datasets, respectively.

Co-Optimizing the Siting and Sizing of Batteries and the Siting of Isolation Devices in Distribution Systems

This paper presents an optimization model to help distribution system planners to site and size battery energy storage systems (BESSs) and to site isolation devices. The purpose is twofold: (i) to enhance system reliability and (ii) to generate revenue by taking advantage of energy arbitrage. Our work fills a research gap regarding the optimal coordination of the siting of BESSs and isolation devices. Input data for the model includes utility historical load and fault data. This data is used to generate multiple scenarios pertaining to normal and faulted operating conditions. The proposed optimization model is recast as a mixed-integer linear programming (MILP) problem and solved using a commercial solver. It is successfully applied in a real-world case study.

A robust support vector regression model for electric load forecasting

Electric load forecasting is a crucial part of business operations in the energy industry. Various load forecasting methods and techniques have been proposed and tested. With growing concerns about cybersecurity and malicious data manipulations, an emerging topic is to develop robust load forecasting models. In this paper, we propose a robust support vector regression (SVR) model to forecast the electricity demand under data integrity attacks. We first introduce a weight function to calculate the relative importance of each observation in the load history. We then construct a weighted quadratic surface SVR model. Some theoretical properties of the proposed model are derived. Extensive computational experiments are based on the publicly available data from Global Energy Forecasting Competition 2012 and ISO New England. To imitate data integrity attacks, we have deliberately increased or decreased the historical load data. Finally, the computational results demonstrate better accuracy of the proposed robust model over other recently proposed robust models in the load forecasting literature.

Self-Attention-Based Short-Term Load Forecasting Considering Demand-Side Management

Accurate and rapid forecasting of short-term loads facilitates demand-side management by electricity retailers. The complexity of customer demand makes traditional forecasting methods incapable of meeting the accuracy requirements, so a self-attention based short-term load forecasting (STLF) considering demand-side management is proposed. In the data preprocessing stage, non-parametric kernel density estimation is used to construct customer electricity consumption feature curves, and then historical load data are used to delineate the feasible domain range for outlier detection. In the feature selection stage, the feature data are selected using variational modal decomposition and a maximum information coefficient to enhance the model prediction accuracy. In the model prediction stage, the decomposed intrinsic mode function components are independently predicted and reconstructed using an Informer based on improved self-attention. Additionally, the novel AdaBlief optimizer is used to optimize the model parameters. Cross-sectional and longitudinal experiments are conducted on a regional-level load dataset set in Spain. The experimental results prove that the proposed method is superior to other methods in STLF.

Boruta-XGBoost Electricity Theft Detection Based on Features of Electric Energy Parameters

Electricity theft detection is critical for the safe and effective development of the electric power system. The existing methods that are used to detect electricity theft rely on historical load data and are considered to have poor timeliness. Their detection results have limited reference to power supply enterprises’ investigation on electricity theft. Therefore, this paper proposes the Boruta-XGBoost power theft detection model based on multiple features of electric energy parameters. The model converts electricity theft detection problem into a multiclass classification problem. First, the features of various electric energy parameters that are collected by the electric energy metering device are used to construct the original dataset. Then, the Boruta algorithm is used to select features and reconstruct the dataset based on the selection results. Finally, the reconstructed dataset is used to train an XGBoost model that can detect the type of electricity theft based on the features of real-time electric energy parameters. In order to verify the effectiveness of the model, a comparative experiment is conducted in this paper. The reconstructed dataset is used to train the following algorithms: LightGBM, CART, SVM, and logistic regression models. The test results show that the Boruta-XGBoost model in this paper has the best effect on electricity theft detection.

Application of LSTM-LightGBM Nonlinear Combined Model to Power Load Forecasting

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.

Optimal placement of remote-controlled switches in distribution networks considering load forecasting

The application of long-term load forecasting (LTLF) in solving the problem of optimal placement of remote-controlled switches (RCSs) in distribution networks is considered in this paper. The mixed-integer linear programming (MILP) formulation for the RCSs placement problem is modified to include LTLF results obtained using the proposed reference historical load data model. Different methods are selected and validated in terms of forecasting accuracy. The modified MILP model for RCSs placement problem has been tested on distribution network RBTS-bus4 which has been used extensively for optimal switch placement studies. It is shown that LTLF results could significantly improve the accuracy of total cost estimation when compared to approximated constant annual load increase rates regularly used in previous research papers. Load profiling data have been included in LTLF analysis. The forecasting of annual U.S. dollar values has been performed to additionally improve the optimal solution accuracy. It is highly recommended for distribution companies to consider LTLF before making the final decision on RCSs investment.

Spatio-Temporal Two-Dimensions Data Based Customer Baseline Load Estimation Approach Using LASSO Regression

Customer baseline load (CBL) estimation plays a crucial role in financial settlement for incentive-based demand response (DR). Most current CBL estimation methods utilize temporal auto-correlation or spatial cross-correlation to estimate the CBL, which would produce large errors when the load pattern suddenly changes on the DR event day or the number of CONTROL customers is insufficient. To this end, a spatio-temporal CBL estimation approach is proposed to improve the estimation accuracy in this paper. First, all CONTROL customers are grouped into several non-overlapping clusters by the K-means algorithm. Each DR customer is then matched to the most similar cluster according to the similarity between the typical load pattern and the cluster center. Second, spatio-temporal features are extracted from the historical load data of the DR customer itself and the load data of the matched CONTROL customers on the DR event day. Third, the least absolute shrinkage and selection operator (LASSO) regression model is constructed to estimate the CBL of each DR customer. The effectiveness and superiority of the proposed approach have been verified on a real-world load dataset.

Capacity subscription grid tariff efficiency and the impact of uncertainty on the subscribed level

While volume-based grid tariffs have been the norm for residential consumers, capacity-based tariffs will become more relevant with the increasing electrification of society. A further development is capacity subscription, where consumers are financially penalised for exceeding their subscribed capacity, or alternatively their demand is limited to the subscribed level. The penalty or limitation can either be static (always active) or dynamic, meaning that it is only activated when there are active grid constraints. We investigate the cost impact for static and dynamic capacity subscription tariffs, for 84 consumers based on six years of historical load data. We use several approaches for finding the optimal subscription level ex ante. The results show that annual costs remain both stable and similar for most consumers, with a few exceptions for those that have high peak demand. In the case of a physical limitation, it is important to use a stochastic approach for the optimal subscription level to avoid excessive demand limitations. Facing increased peak loads due to electrification, regulators should consider a move to capacity-based tariffs in order to reduce cross-subsidisation between consumers and increase cost reflectivity without impacting the DSO cost recovery.

Fuzzy-Swarm Intelligence-Based Short-Term Load Forecasting Model as a Solution to Power Quality Issues Existing in Microgrid System

Load demand is highly stochastic and uncertain. This is because it was highly influenced by a number of variables like load type, weather conditions, time of day, the seasonality factor, economic constraints, and other randomness effects. The loads are categorized as holiday loads (national and religious), weekdays, and weekend days. The nonlinearity and uncertain characteristics of electrical load in a microgrid are one of the major sources of power quality problems in a microgrid system, and they can be handled using an accurate load forecast model. The fuzzy load prediction model can effectively handle these nonlinearity and uncertainty characteristics to have an accurate load forecast, but the main challenge with this model is its inability to accommodate a large volume of historical load and weather information when the membership function of the input and output fuzzy variables and the number of the fuzzy rules are tremendous. The swarm intelligence load forecast model based on particle swarm optimization algorithms can improve the limitations of the fuzzy system and increase its forecasting performance. The parameters of time, temperature, historical load, and error correction factors are considered as the Fuzzy and Fuzzy-PSO model input variables, while the forecasted industrial load is the only output variable. The Gaussian membership function is considered for both the input and output fuzzy variables. A 3-year historical hourly load data of an Ethiopian industrial system is used to train and validate both prediction models. The mean absolute percentage error (MAPE) is used to evaluate the performance of these prediction models. The Fuzzy-PSO load prediction model shows results that have superior performance to the fuzzy-alone load prediction results.

Short-Term Combined Forecasting Method of Park Load Based on CEEMD-MLR-LSSVR-SBO

To improve the accuracy of park load forecasting, a combined forecasting method for short-term park load is proposed based on complementary ensemble empirical mode decomposition (CEEMD), sample entropy, the satin bower bird optimization algorithm (SBO), and the least squares support vector regression (LSSVR) model. Firstly, aiming at the random fluctuation of park load series, the modes with different characteristic scales are divided into low-frequency and high-frequency according to the calculation of sample entropy, which is based on the decomposition of historical park load data modes by CEEMD. The low-frequency is forecast by multiple linear regression (MLR), and the high-frequency component is the training input of the LSSVR forecasting model. Secondly, the SBO algorithm is adopted to optimize the regularization parameters and the kernel function width of LSSVR. Then, the park load forecasting model of each sequence component is built. The forecast output of each sequence component is superimposed to get the final park load forecast value. Finally, a case study of a park in Liaoning Province has been performed with the results proving that the proposed method significantly outperforms the state-of-art in reducing the difficulty and complexity of forecasting effectively, also eliminating the defect of large reconstruction error greatly through the decomposed original sequence by the ensemble empirical model.

Short-term microgrid load probability density forecasting method based on k-means-deep learning quantile regression

Traditional short-term load forecasting (STLF) methods for large utility grid systems usually provide the forecasted load with deterministic points. However, deterministic load forecasting cannot reveal the load pattern and uncertainty of controllable load in a microgrid, where the prediction errors may exceed the expected range due to the high volatility and strong randomness. In order to deal with this matter, a probability density forecasting method is proposed to predict the microgrid load with uncertainty for robust power scheduling in this paper. The proposed probability forecasting method effectively combines several data-driven and statistical algorithms, including the k-means algorithm, quantile regression long short-term memory neural network (QRLSTM), and kernel density estimation (KDE). Firstly, similar days related to the prediction day are selected through the k-means algorithm, and the historical load data of these selected days are divided into two subsets including the training dataset and the testing dataset. Secondly, a QRLSTM-based model is established and used to predict the microgrid load for different quantiles. Finally, the probability density function of the predicted points is obtained by KDE on the target day. The prediction accuracy is evaluated roundly and the results demonstrate that the proposed method can effectively reproduce the probability density distribution of the load and provide noticeably better performance than some benchmark methods.

Short-Term Power Load Forecasting of Integrated Energy System Based on Attention-CNN-DBILSTM

In view of the fact that the potential high-dimensional features in the historical sequence are difficult to be effectively extracted by traditional power load forecasting methods and the coupling factors of electricity, heat, and gas have not been considered, the correlation of electric heating and gas load is considered in this paper, and a short-term power load forecasting method for integrated energy systems based on Attention-CNN- (Convolutional Neural Network-) DBILSTM (Deep Bidirectional Long-Short-Term Memory) is proposed. First, the correlation between the multiple load influencing factors is considered, and the Pearson coefficient is used to quantitatively calculate the correlation between the multiple loads. Second, a CNN network consisting of a one-dimensional convolutional layer and a pooling layer is established. High-dimensional features reflecting the dynamic changes of the load are extracted, and the proposed feature vector is constructed in the form of time series as the input of the DBILSTM network; the dynamic change law of time series data is modeled and learned. Then, the Attention mechanism is introduced to assign different weights to the hidden state of DBILSTM through the mapping weight and the learning parameter matrix, to reduce the loss of historical information and strengthen the impact of key information, and the Dense layer is used to output the load prediction results. Finally, the influence of the correlation of multiple loads and its influencing factors on the power load forecasting results is analyzed, based on the historical load data of the integrated energy system in a certain area of Northeast China. The simulation results of the calculation example show that the prediction accuracy of the method reaches 97.99%, and the integrated energy system electric, heat, and gas load correlation coefficients as the input parameters of the Attention-CNN-DBILSTM network can reduce the average prediction error by 0.37%∼1.93%. The proposed method has been verified to effectively improve the prediction accuracy by comparison with the prediction model results of CNN-LSTM network, CNN-BILSTM network, and CNN-DBILSTM network.

Forecasting of Electric Load Using a Hybrid LSTM-Neural Prophet Model

Load forecasting (LF) is an essential factor in power system management. LF helps the utility maximize the utilization of power-generating plants and schedule them both reliably and economically. In this paper, a novel and hybrid forecasting method is proposed, combining a long short-term memory network (LSTM) and neural prophet (NP) through an artificial neural network. The paper aims to predict electric load for different time horizons with improved accuracy as well as consistency. The proposed model uses historical load data, weather data, and statistical features obtained from the historical data. Multiple case studies have been conducted with two different real-time data sets on three different types of load forecasting. The hybrid model is later compared with a few established methods of load forecasting found in the literature with different performance metrics: mean average percentage error (MAPE), root mean square error (RMSE), sum of square error (SSE), and regression coefficient (R). Moreover, a guideline with various attributes is provided for different types of load forecasting considering the applications of the proposed model. The results and comparisons from our test cases showed that the proposed hybrid model improved the forecasting accuracy for three different types of load forecasting over other forecasting techniques.

A Forecast Model of City Natural Gas Daily Load Based on Data Mining

Data mining technology is more and more widely used in the daily load forecasting of natural gas systems. It is still difficult to carry out high-precision, timely intraday load forecasting and intraday load dynamic characteristics clustering for natural gas systems. Based on data mining technology, this paper proposes a stable intraday load forecasting method for the natural gas flow state-space model. The load sensitivity under the current operating conditions of the system is obtained by calculation; the sample space of the state space is established through data processing; the partitions under different clustering radii are calculated; and the best intraday load flow is obtained through the state space effectiveness evaluation method. The experimental results show that the model load forecasting accuracy and relative error reached 98.5% and 0.026, respectively, which solved the problem of processing the long-term accumulated historical data of gas intra-day load. At the same time, the amount of data calculation was reduced by 33.6%, which effectively promoted the quantification of intraday load influencing factors and qualitative analysis.