Dear Colleagues,

Power system operation and planning rely on forecasting of demand variables such as hourly load, peak load, and total energy. Based on the time horizon, forecasting is categorized as short-term, medium-term, and long-term. Short-term load forecasting (STLF) is the foundation where power system operation is built upon on an intraday and day-ahead basis. Paradigms of applications are unit commitment, hydrothermal coordination, optimal load flow, demand response, and others. Medium-term load forecasting (MTLF) is mostly concerned with fuel import decisions and power unit maintenance scheduling. Long-term load forecasting (LTLF) is exploitable in power system planning.

Until recently, the system operator was responsible for providing the official predictions for the national system level. However, due to the deregulation and increase of competition of modern-day power markets, the strategic actions of various entities such as generation companies, retailers, aggregators, and others rely on accurate load predictions. Moreover, a robust forecasting model for a prosumer would lead to the optimal management of the resources, i.e., energy management, generation, and storage.

Further, the power system gradually transforms into a smart grid where the focus is mostly in small scale instead of system wide level loads such as loads of residences, buildings, distribution buses, and others. Another aspect of smart grid is smart metering, where the data are recorded in low time resolution, resulting in the collection of large amounts of data.

Load forecasting models can be, in general, classified into time series models and computational intelligence models. Time series models such as ARMA and ARIMA demand a priori the definition of the structure of the model. On the other hand, computational intelligence models such as neural networks, support vector machines, neurofuzzy systems, bio-inspired algorithms, and others are trained by the data, and the structure is determined after training. To overcome the drawbacks of each model, hybrid models have been proposed that combine models of a different type or a model and a data preprocessing technique. In addition, deep learning is a new and promising trend in machine learning that has not sufficiently been tested in load forecasting studies.

In the context of these challenges, the main scope of this Special Issue is to develop new methods applicable in short-, medium-, and long-term load forecasting. State-of-the-art papers together with innovative case studies are invited. Multidisciplinary research and cutting-edge approaches are welcomed in order to address the challenges that are raised by modern power systems, smart grids, and competitive power 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. Forecasting is an international peer-reviewed open access quarterly 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 1800 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.

• Time series and computational intelligence models for short-, medium-, and long-term forecasting
• Application of time series processing techniques for load forecasting: Wavelets, empirical mode decomposition, principal component analysis, and others
• Deep learning methods
• Short-term load forecasting exploitation in power systems operations
• Medium- and long-term forecasting exploitation in power systems planning
• Load forecasting and deregulated power markets
• Load forecasting for smart homes and smart buildings
• Load forecasting for distribution systems and buses
• Big data analytics application in load forecasting
• Spatial and temporal load forecasting