energies-logo

Journal Browser

Journal Browser

Computational Intelligence and Load Forecasting in Power Systems

A special issue of Energies (ISSN 1996-1073). This special issue belongs to the section "F5: Artificial Intelligence and Smart Energy".

Deadline for manuscript submissions: closed (30 January 2022) | Viewed by 14672

Special Issue Editor

Department of Electrical and Computer Engineering, University of Thessaly, 37 Glavani–28th October Str., Deligiorgi Building, 382 21 Volos, Greece
Interests: demand side management; forecasting; load profiling

Special Issue Information

Dear Colleagues,

Load forecasting is a key tool in the design of the electricity system. It refers to the process of predicting the amount of demand for electricity in an area and/or a transmission network over a period of time. The forecasting aims to determine the amount of electricity in a future time horizon. Load forecasting may correspond to a prediction of total energy, hourly load, peak load and load duration curve. Load forecasting explores issues such as the demand for installed capacity to meet potential demand growth, the type of energy resources to be used, the development of the transmission and distribution systems, the demand by type of consumer and by geographical area in order to implement demand side management measures and others.

Since power systems are gradually transforming to smart grids, new issues arise that can be addressed with robust forecasting algorithms. Also, the deregulation of electricity markets provide new opportunities for many market participants. Forecasting can aid to the strategic actions of market players to minimize risks and increase profits.

In the context of these challenges, the main scope of this Special Issue is to develop new algorithms for load forecasting. State-of-the-art papers together with innovative case studies are invited. Multi-disciplinary research and cutting-edge approaches are welcomed in order to address the challenges that are raised by power systems and electricity markets.

Asst. Prof. Dr. Ioannis Panapakidis
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Energies is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • deep learning
  • shallow learning
  • short-term forecasting
  • medium-term forecasting
  • long-term forecasting
  • auto-regressive models
  • neural networks

Published Papers (6 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

14 pages, 2248 KiB  
Article
Clustering Informed MLP Models for Fast and Accurate Short-Term Load Forecasting
by Athanasios Ioannis Arvanitidis, Dimitrios Bargiotas, Aspassia Daskalopulu, Dimitrios Kontogiannis, Ioannis P. Panapakidis and Lefteri H. Tsoukalas
Energies 2022, 15(4), 1295; https://0-doi-org.brum.beds.ac.uk/10.3390/en15041295 - 10 Feb 2022
Cited by 9 | Viewed by 1929
Abstract
The stable and efficient operation of power systems requires them to be optimized, which, given the growing availability of load data, relies on load forecasting methods. Fast and highly accurate Short-Term Load Forecasting (STLF) is critical for the daily operation of power plants, [...] Read more.
The stable and efficient operation of power systems requires them to be optimized, which, given the growing availability of load data, relies on load forecasting methods. Fast and highly accurate Short-Term Load Forecasting (STLF) is critical for the daily operation of power plants, and state-of-the-art approaches for it involve hybrid models that deploy regressive deep learning algorithms, such as neural networks, in conjunction with clustering techniques for the pre-processing of load data before they are fed to the neural network. This paper develops and evaluates four robust STLF models based on Multi-Layer Perceptrons (MLPs) coupled with the K-Means and Fuzzy C-Means clustering algorithms. The first set of two models cluster the data before feeding it to the MLPs, and are directly comparable to similar existing approaches, yielding, however, better forecasting accuracy. They also serve as a common reference point for the evaluation of the second set of two models, which further enhance the input to the MLP by informing it explicitly with clustering information, which is a novel feature. All four models are designed, tested and evaluated using data from the Greek power system, although their development is generic and they could, in principle, be applied to any power system. The results obtained by the four models are compared to those of other STLF methods, using objective metrics, and the accuracy obtained, as well as convergence time, is in most cases improved. Full article
(This article belongs to the Special Issue Computational Intelligence and Load Forecasting in Power Systems)
Show Figures

Figure 1

15 pages, 8910 KiB  
Article
High Precision LSTM Model for Short-Time Load Forecasting in Power Systems
by Tomasz Ciechulski and Stanisław Osowski
Energies 2021, 14(11), 2983; https://0-doi-org.brum.beds.ac.uk/10.3390/en14112983 - 21 May 2021
Cited by 36 | Viewed by 2422
Abstract
The paper presents the application of recurrent LSTM neural networks for short-time load forecasting in the Polish Power System (PPS) and a small region of a power system in Central Poland. The objective of the present work was to develop an efficient and [...] Read more.
The paper presents the application of recurrent LSTM neural networks for short-time load forecasting in the Polish Power System (PPS) and a small region of a power system in Central Poland. The objective of the present work was to develop an efficient and accurate method of forecasting the 24-h pattern of power load with a 1-h and 24-h horizon. LSTM showed effectiveness in predicting the irregular trends in time series. The final forecast is estimated using an ensemble consisted of five independent predictions. Numerical experiments proved the superiority of the ensemble above single predictor resulting in a reduction of the MAPE the RMSE error by more than 6% in both forecasting tasks. Full article
(This article belongs to the Special Issue Computational Intelligence and Load Forecasting in Power Systems)
Show Figures

Figure 1

14 pages, 2928 KiB  
Article
Multiple-Load Forecasting for Integrated Energy System Based on Copula-DBiLSTM
by Jieyun Zheng, Linyao Zhang, Jinpeng Chen, Guilian Wu, Shiyuan Ni, Zhijian Hu, Changhong Weng and Zhi Chen
Energies 2021, 14(8), 2188; https://0-doi-org.brum.beds.ac.uk/10.3390/en14082188 - 14 Apr 2021
Cited by 15 | Viewed by 1718
Abstract
With the tight coupling of multi-energy systems, accurate multiple-load forecasting will be the primary premise for the optimal operation of integrated energy systems. Therefore, this paper proposes a Copula correlation analysis combined with deep bidirectional long and short-term memory neural network forecasting model. [...] Read more.
With the tight coupling of multi-energy systems, accurate multiple-load forecasting will be the primary premise for the optimal operation of integrated energy systems. Therefore, this paper proposes a Copula correlation analysis combined with deep bidirectional long and short-term memory neural network forecasting model. First, Copula correlation analysis is used to conduct correlation analysis on multiple loads and various influencing factors. The influencing factors that have a great correlation with multiple loads were screened out as the input feature set of the model to eliminate the influence of interfering factors. Then, a deep bidirectional long and short-term memory neural network was constructed. Combined with the input feature set screened by the Copula correlation analysis method, the useful information contained in the historical data was more comprehensively learned from the forward and backward directions for training and forecasting. Through the actual calculation example analysis and comparison with other models, the forecasting accuracy of the method presented in this paper was improved to a certain extent. Full article
(This article belongs to the Special Issue Computational Intelligence and Load Forecasting in Power Systems)
Show Figures

Figure 1

31 pages, 3909 KiB  
Article
Techno-Economic Analysis of a Stand-Alone Hybrid System: Application in Donoussa Island, Greece
by Michail Katsivelakis, Dimitrios Bargiotas, Aspassia Daskalopulu, Ioannis P. Panapakidis and Lefteri Tsoukalas
Energies 2021, 14(7), 1868; https://0-doi-org.brum.beds.ac.uk/10.3390/en14071868 - 28 Mar 2021
Cited by 23 | Viewed by 3702
Abstract
Hybrid Renewable Energy Systems (HRES) are an attractive solution for the supply of electricity in remote areas like islands and communities where grid extension is difficult. Hybrid systems combine renewable energy sources with conventional units and battery storage in order to provide energy [...] Read more.
Hybrid Renewable Energy Systems (HRES) are an attractive solution for the supply of electricity in remote areas like islands and communities where grid extension is difficult. Hybrid systems combine renewable energy sources with conventional units and battery storage in order to provide energy in an off-grid or on-grid system. The purpose of this study is to examine the techno-economical feasibility and viability of a hybrid system in Donoussa island, Greece, in different scenarios. A techno-economic analysis was conducted for a hybrid renewable energy system in three scenarios with different percentages of adoption rate (20%, 50% and 100%)and with different system configurations. Using HOMER Pro software the optimal system configuration between the feasible configurations of each scenario was selected, based on lowest Net Present Cost (NPC), minimum Excess Electricity percentage, and Levelized Cost of Energy (LCoE). The results obtained by the simulation could offer some operational references for a practical hybrid system in Donoussa island. The simulation results confirm the application of a hybrid system with 0% of Excess Electricity, reasonable NPC and LCoE and a decent amount of renewable integration. Full article
(This article belongs to the Special Issue Computational Intelligence and Load Forecasting in Power Systems)
Show Figures

Figure 1

13 pages, 1405 KiB  
Article
Deep Learning Approach to Power Demand Forecasting in Polish Power System
by Tomasz Ciechulski and Stanisław Osowski
Energies 2020, 13(22), 6154; https://0-doi-org.brum.beds.ac.uk/10.3390/en13226154 - 23 Nov 2020
Cited by 8 | Viewed by 1830
Abstract
The paper presents a new approach to predicting the 24-h electricity power demand in the Polish Power System (PPS, or Krajowy System Elektroenergetyczny—KSE) using the deep learning approach. The prediction system uses a deep multilayer autoencoder to generate diagnostic features and an ensemble [...] Read more.
The paper presents a new approach to predicting the 24-h electricity power demand in the Polish Power System (PPS, or Krajowy System Elektroenergetyczny—KSE) using the deep learning approach. The prediction system uses a deep multilayer autoencoder to generate diagnostic features and an ensemble of two neural networks: multilayer perceptron and radial basis function network and support vector machine in regression model, for final 24-h forecast one-week advance. The period of the data that is the subject of the experiments is 2014–2019, which has been divided into two parts: Learning data (2014–2018), and test data (2019). The numerical experiments have shown the advantage of deep learning over classical approaches of neural networks for the problem of power demand prediction. Full article
(This article belongs to the Special Issue Computational Intelligence and Load Forecasting in Power Systems)
Show Figures

Figure 1

14 pages, 1459 KiB  
Article
An Ultra-Short-Term Electrical Load Forecasting Method Based on Temperature-Factor-Weight and LSTM Model
by Dengyong Zhang, Haixin Tong, Feng Li, Lingyun Xiang and Xiangling Ding
Energies 2020, 13(18), 4875; https://0-doi-org.brum.beds.ac.uk/10.3390/en13184875 - 17 Sep 2020
Cited by 7 | Viewed by 1752
Abstract
Ultra-short-term electrical load forecasting is an important guarantee for the safety and efficiency of energy system operation. Temperature is also an important factor affecting the changes in electric load. However, in different cases, the impact of temperature on load forecasting will vary greatly, [...] Read more.
Ultra-short-term electrical load forecasting is an important guarantee for the safety and efficiency of energy system operation. Temperature is also an important factor affecting the changes in electric load. However, in different cases, the impact of temperature on load forecasting will vary greatly, and sometimes even lead to the decrease of forecasting accuracy. This often brings great difficulties to researchers’ work. In order to make more scientific use of temperature factor for ultra-short-term electrical load forecasting, especially to avoid the negative influence of temperature on load forecasting, in this paper we propose an ultra-short-term electrical load forecasting method based on temperature factor weight and long short-term memory model. The proposed method evaluates the importance of the current prediction task’s temperature based on the change magnitude of the recent load and the correlation between temperature and load, and therefore the negative impacts of the temperature model can be avoided. The mean absolute percentage error of proposed method is decreased by 1.24%, 1.86%, and 6.21% compared with traditional long short-term memory model, back-propagation neural network, and gray model on average, respectively. The experimental results demonstrate that this method has obvious advantages in prediction accuracy and generalization ability. Full article
(This article belongs to the Special Issue Computational Intelligence and Load Forecasting in Power Systems)
Show Figures

Graphical abstract

Back to TopTop