Figure 3 - uploaded by Svetlana Chemetova
Content may be subject to copyright.
![Supervision and control of the distribution network, InovGrid Project (taken from [7]).](profile/Svetlana-Chemetova/publication/309217981/figure/fig2/AS:418542454362119@1476799705960/Supervision-and-control-of-the-distribution-network-InovGrid-Project-taken-from-7.png)
Source publication
The importance of forecasting has become more evident with the appearance of the open electricity market and the restructuring of the national energy sector. This paper presents a new approach to load forecasting in the medium voltage distribution network in Portugal. The forecast horizon is short term, from 24 h up to a week. The forecast method i...
Context in source publication
Context 1
... proposed change was implemented by a technological renovation and organizational adequacy of the distribution system operation and relationship with other stakeholders, based on a infrastructure that aims to respond to the needs arising from energy efficiency, remote management, distributed generation and microgeneration, and assume active control of the intelligent network ( figure 3). ...
Similar publications
Direct current (DC) power grids based on flexible high-voltage DC technology have become a common solution of facilitating the large-scale integration of distributed energy resources (DERs) and the construction of advanced urban power grids. In this study, a typical topology analysis is performed for an advanced urban medium-voltage DC (MVDC) distr...
Citations
... At the DN level, short-to-medium range load forecasts are studied. For example, the Portuguese medium voltage DN load for 24 hours to a week ahead was assessed using both Artificial Neural Networks (ANN) and a regression model in [14]. On the household level, the STLF is usually analysed using the high temporal resolution data collected from smart meters. ...
This paper proposes a generalised and robust multi-factor Gated Recurrent Unit (GRU) based Deep Learning (DL) model to forecast electricity load in distribution networks during wildfire seasons. The flexible modelling methods consider data input structure, calendar effects and correlation-based leading temperature conditions. Compared to the regular use of instantaneous temperature, the Mean Absolute Percentage Error (MAPE) is decreased by 30.73% by using the proposed input feature selection and leading temperature relationships. Our model is generalised and applied to eight real distribution networks in Victoria, Australia, during the wildfire seasons of 2015-2020. We demonstrate that the GRU-based model consistently outperforms another DL model, Long Short-Term Memory (LSTM), at every step, giving average improvements in Mean Squared Error (MSE) and MAPE of 10.06% and 12.86%, respectively. The sensitivity to large-scale climate variability in training data sets, e.g. El Ni\~no or La Ni\~na years, is considered to understand the possible consequences for load forecasting performance stability, showing minimal impact. Other factors such as regional poverty rate and large-scale off-peak electricity use are potential factors to further improve forecast performance. The proposed method achieves an average forecast MAPE of around 3%, giving a potential annual energy saving of AU\$80.46 million for the state of Victoria.
... Based on neural networks, [101] presented an extensive comparative model of forecasting methods in a radially configured power system. Different approaches, such as combined regression models and artificial neural networks (ANN) [71][72][73][74], adoptive load forecasting (ALF) [75], multilayer bidirectional recurrent neural network (BRNN) [76], deep learning (DL) and image encoding (IE) [77], and gray system theory (GST) [78], have been reported in the literature for load forecasting to enhance the reliability and efficacy of radial configuration-based distribution networks. An extensive and comprehensive review was carried out [102,103] to investigate the different load forecasting models and techniques for distribution network. ...
To enhance the reliability and resilience of power systems and achieve reliable delivery of power to end users, smart distribution networks (SDNs) play a vital role. The conventional distribution network is transforming into an active one by incorporating a higher degree of automation. Replacing the traditional absence of manual actions, energy delivery is becoming increasingly dependent on intelligent active system management. As an emerging grid modernization concept, the smart grid addresses a wide range of economic and environmental concerns, especially by integrating a wide range of active technologies at distribution level. At the same time, these active technologies are causing a slew of technological problems in terms of power quality and stability. The development of such strategies and approaches that can improve SDN infrastructure in terms of planning, operation, and control has always been essential. As a result, a substantial number of studies have been conducted in these areas over the last 10–15 years. The current literature lacks a combined systematic analysis of the planning, operation, and control of SDN technologies. This paper conducts a systematic survey of the state-of-the-art advancements in SDN planning, operation, and control over the last 10 years. The reviewed literature is structured so that each SDN technology is discussed sequentially from the viewpoints of planning, operation, and then control. A comprehensive analysis of practical SND concepts across the globe is also presented in later sections. The key constraints and future research opportunities in the existing literature are discussed in the final part. This review specifically assists readers in comprehending current trends in SDN planning, operation, and control, as well as identifying the need for further research to contribute to the field.
... Electric power load forecasting on a medium voltage level based on regression models and ANN was proposed in [19] using time series DSO telemetry data and weather records from the Portuguese Institute of Sea and Atmosphere, and applied to the urban area of Évora, one of Portugal's first smart cities. A new top-down algorithm based on a similar day-type method to compute an accurate short-term distribution loads forecast, using only SCADA data from transmission grid substations was proposed in [20]. ...
Short-term electric power load forecasting is a critical and essential task for utilities in the
electric power industry for proper energy trading, which enables the independent system operator to operate the network without any technical and economical issues. From an electric power distribution system point of view, accurate load forecasting is essential for proper planning and operation. In order to build most robust machine learning model to forecast the load with a good accuracy irrespective of weather condition and type of day, features such as the season, temperature, humidity and day-status are incorporated into the data. In this paper, a machine learning model, namely a regression tree, is used to forecast the active power load an hour and one day ahead. Real-time active power load data to train and test the machine learning models are collected from a 33/11 kV substation located in Telangana State, India. Based on the simulation results, it is observed that the regression tree model is able to forecast the load with less error.
... The maximum limit could be avoided with a higher level of automation in terms of load modeling and feeder management. Forecasting algorithms will be important assistants in the management of the distribution networks [14][15]. ...
... NNs have been used for MV load forecasting with their approximation capability of nonlinear mapping [19], [20]. To cope with uncertainties in load forecasting, researchers propose NN-based PLF methods [21], where a basic load forecasting model is applied to historical data set repeatedly to obtain multiple load predictions. ...
... If MV substation load data are available, go to step (2). If there are only outgoing line load data, go to step (3); 2) Based on section II.C, MV substation load patterns are clustered with K-means algorithm, and the sum of load patterns in each cluster is calculated; 3) Based on PLF method introduced in section III, deterministic MV load predictions y l,t and corresponding prediction variances σ 2 l,t are obtained according to (8)-(11); 4) Based on the method proposed in section IV.A, the deterministic prediction of add-up error y εa,t and corresponding prediction variance σ 2 εa,t are obtained; 5) The deterministic predictions and prediction variances of MV loads and add-up error are aggregated to calculate the deterministic load prediction y hv,t and prediction variance σ 2 hv,t of HV substation according to (16) and (17), and then PIs are obtained according to (18)- (20). The flowchart of the proposed method is shown in Fig. 9. ...
Load forecasting has always been an essential part of power system planning and operation. In recent decades, the competition of the market and the requirements of renewable integration lead more attention to probabilistic load forecasting methods, which can reflect forecasting uncertainties through prediction intervals and hence benefit decision-making activities in system operation. Moreover, with the development of smart grid and power metering techniques, power companies have collected enormous load data about electricity customers and substations. The abundant load data allow us to utilize medium voltage measurement data to achieve better accuracy in high voltage transmission substation load forecasting. In this paper, a bottom-up probabilistic forecasting method is proposed for high voltage transmission substation short-term load forecasting, in which the probability distributions of medium voltage day-ahead load forecasting values are estimated and added up to form high voltage load predictions. Two bottom-up frameworks based on load patterns collected from medium voltage outgoing lines and substations are proposed respectively, in which mismatches between load data at different levels are estimated for correcting high voltage predictions. The comparison of predictions obtained by traditional and bottom-up methods demonstrates that the proposed method obtains high voltage load forecasting more accurately and give narrower prediction intervals.
... As a classical time series forecasting problem, short-term load forecasting (STLF) has attracted much attention from both academia and industry since the 1950s, and various methods have been proposed ever since [3][4][5][6][7][8][9][10][11][12][13]. The existing algorithms can be roughly classified into two categories: statistical methods and machine learning methods. ...
... Machine learning methods mainly include shallow networks and deep learning methods. Shallow neural network models are data-driven adaptive method, which simulates the relationship between input and output through nonlinear mapping to learn the patterns hidden in the electrical load [10][11][12]. Among numerous shallow neural network models, the ensemble learning ones usually have excellent performance, which combine multiple prediction algorithms and achieve enhanced prediction accuracy with mild increase of complexity [13,14]. ...
Short term load forecasting (STLF) is one of the basic techniques for economic operation of the power grid. Electrical load consumption can be affected by both internal and external factors so that it is hard to forecast accurately due to the random influencing factors such as weather. Besides complicated and numerous internal patterns, electrical load shows obvious yearly, seasonal, and weekly quasi-periodicity. Traditional regression-based models and shallow neural network models cannot accurately learn the complicated inner patterns of the electrical load. Long short-term memory (LSTM) model features a strong learning capacity to capture the time dependence of the time series and presents the state-of-the-art performance. However, as the time span increases, LSTM becomes much harder to train because it cannot completely avoid the vanishing gradient problem in recurrent neural networks. Then, LSTM models cannot capture the dependence over large time span which is of potency to enhance STLF. Moreover, electrical loads feature data imbalance where some load patterns in high/low temperature zones are more complicated but occur much less often than those in mild temperature zones, which severely degrades the LSTM-based STLF algorithms. To fully exploit the information beneath the high correlation of load segments over large time spans and combat the data imbalance, a deep ensemble learning model within active learning framework is proposed, which consists of a selector and a predictor. The selector actively selects several key load segments with the most similar pattern as the current one to train the predictor, and the predictor is an ensemble learning-based deep learning machine integrating LSTM and multi-layer preceptor (MLP). The LSTM is capable of capturing the short-term dependence of the electrical load, and the MLP integrates both the key history load segments and the outcome of LSTM for better forecasting. The proposed model was evaluated over an open dataset, and the results verify its advantage over the existing STLF models.
... Багатофакторні регресійні моделі та моделі часових рядів також отримали поширення у прогнозуванні споживання для вугільних шахт [9]. При порівнянні якості прогнозування за допомогою різних підходів [10,11] та їх комбінацій [12] можна зробити висновок про те, що використання комбінацій різних методів дозволяє зменшити відносну похибку прогнозування, а найбільшу точність прогнозування показують комбінації різних методів з нейромережевим підходом [13,14]. Але комбінованій системі на основі нечіткої логіки і генетичного алгоритму не приділено значної уваги. ...
In the paper, a two-stage methodology has been proposed for the real-time load estimation in the electric distribution substations (EDSs) from the electric distribution networks (EDNs) using the data provided by pilot EDSs selected to be integrated into the SCADA system considering a clustering-based correlation criterion and multiple linear regression (MLR) models. Testing the proposed methodology has been done in an EDN from an urban area with 39 substations belonging to a Romanian Distribution Network Operator (DNO). The obtained results confirmed the accuracy of the proposed approach, recording a value of the mean average percentage error (MAPE) by 2.60% compared with the simple linear regression method, 12.82%, and the profiling method, 9.96%.
Currently, distribution system operators (DSOs) are asked to operate distribution grids, managing the rise of the distributed generators (DGs), the rise of the load correlated to heat pump and e-mobility, etc. Nevertheless, they are asked to minimize investments in new sensors and telecommunication links and, consequently, several nodes of the grid are still not monitored and tele-controlled. At the same time, DSOs are asked to improve the network’s resilience, looking for a reduction in the frequency and impact of power outages caused by extreme weather events. The paper presents a machine learning GIS-based approach to estimate a secondary substation’s load profiles, even in those cases where monitoring sensors are not deployed. For this purpose, a large amount of data from different sources has been collected and integrated to describe secondary substation load profiles adequately. Based on real measurements of some secondary substations (medium-voltage to low-voltage interface) given by Unareti, the DSO of Milan, and georeferenced data gathered from open-source databases, unknown secondary substations load profiles are estimated. Three types of machine learning algorithms, regression tree, boosting, and random forest, as well as geographic information system (GIS) information, such as secondary substation locations, building area, types of occupants, etc., are considered to find the most effective approach.
