Advanced Energy Forecasting in Smart Grids

The basepoint of most of the studies inside the power and energy industry is the forecasted energy signals. By energy signals, here, we refer to the conventional ones such as load, price, PV and wind power generations as well as newly emerged ones like demand-response, EV charging demand, and also reactive power forecasting. The forecast values of the aforementioned signals play an important role in energy systems planning and designing, operation and control, maintenance and many other analysis and studies. Recent technological evolution in all parts of the energy industry caused a changing situation with a rapid dynamic which caused these predictions to be more complex than before.

This Research Topic will target frontier research in the forecasting of energy signals. Both Review and Original Research articles are welcome. Real-world use cases discussing new application areas and resulting new developments are especially welcome. Considering the recent COVID-19 pandemic, research and experiences who fit this topic are particularly welcome. This special section is an extension of the previous special topic called "Frontier Research in Energy Forecasting". We welcome papers and case reports from different energy-linked disciplines and utilities.

Topics that are of interest within the scope of this Research Topic include, but are not limited to:

1. The energy signals to be forecasted:
a) Electric load forecasting;
  b) Electricity price forecasting;
c) Renewable generation (e.g. PV (Solar) and wind power generations) forecasting;
d) Demand-response forecasting;
e) EV charging demand forecasting;
f) Reactive power forecasting;
g) Dynamic thermal rating forecasting.

2. Energy forecasting according to the power system pillars:
a) Energy forecasting in smart grids;
  b) Energy forecasting in microgrids;
c) Energy forecasting in the active buildings (nano-grids).

3. Energy forecasting of the different time horizon:
a) Very-short-term energy forecasting;
b) Short-term energy forecasting;
c) Mid-term energy forecasting;
d) Long-term energy forecasting.

4. Methods and approaches of the energy forecasting:
a) Machine learning, artificial neural networks, and Statistical-based forecasting;
b) Evolutionary and advanced computational approaches;
c) Hybrid forecast engines;
d) Hierarchical forecasters;
e) Probabilistic forecasting;
f) Bayesian approaches to forecasting;
g) Real-time forecasting using big data;
h) Quantum artificial intelligence application in forecasting;
i) Deep learning application in forecasting;
j) Reinforcement learning application in forecasting;
k) Beyond point forecasting (e.g. interval forecasting, density forecasting);
l) Forecasting based on the smart meters and IoT data.

5. Data processing and input data validation for the energy forecasting purpose:
a) Data imputation;
b) Data reduction;
c) Data clustering;
d) Data classification;
e) Correlation analysis;
f) Pattern recognition;
g) Machine learning.

6. Energy forecasting based on exogenous data/parameters:
a) Price-based load forecasting;
b) Whether based energy forecasting;
c) Solar forecasting considering the cloud movement;
d) Solar forecasting based on the image processing
e) Application of Lyapunov vectors for forecasting in coupled systems;
f) Weather based dynamic thermal rating forecasting.

7. Data attacks and manipulation management in forecasting:
a) Forecasting in the presence of data attacks
b) Resilient data forecasting algorithms
c) Data privacy management in the forecasting

8. Forecasting data market
a) Pricing of data for different forecastings
b) Profit and cost distribution of forecasting among different players