Short-Term Wind Energy Forecasting with Independent daytime/Nighttime machine Learning Models

Al-Hajj, Rami; Fouad, Mohamad M.; Assi, Ali; Mabrouk, Emad

doi:10.1109/icrera55966.2022.9922820

Cited by 12 publications

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to the research background, a majority of the studies focused on the optimal power flow problem when renewables [17][18][19][20][21][22][23] and flexible sources are widely allocated across the power system. Considering the voluminous data acquired from all these units, controlling and managing these power sources becomes complicated and challenging [5].…”

Section: Contributionsmentioning

confidence: 99%

Network‐constrained unit commitment‐based virtual power plant model in the day‐ahead market according to energy management strategy

Pirouzi

2023

IET Generation Trans & Dist

113

View full text Add to dashboard Cite

Energy management of a virtual power plant (VPP) that consists of wind farm (WF), energy storage systems and a demand response program is discussed in the present study. The introduced strategy is realized at the electrical power transmission level and takes into account the collaboration between VPPs in day‐ahead energy and reserve markets. One notable feature of the proposed strategy is attempting to make the revenue of VPPs close to the operating cost of generating units as much as possible. The objective function is subjected to the network‐constrained unit commitment model, up and down reserve requirements and the proposed VPP constraints. This method is taking into account the uncertainty in system and VPP loads, day‐ahead market energy and reserve price and WF power generation. This strategy is applied as hybrid stochastic‐robust scheduling to the VPP at the electrical power transmission level, where the scenario‐based stochastic programming models the uncertainty of day‐ahead market prices, and the bounded uncertainty‐based robust optimization has been adopted to model the uncertainties related to the load and WF power. Scheme has been tested on IEEE systems. According to the obtained results, the proposed coordination of VPPs in the mentioned markets demonstrates capability of suggested strategy.

show abstract

Section: Contributionsmentioning

confidence: 99%

Network‐constrained unit commitment‐based virtual power plant model in the day‐ahead market according to energy management strategy

Pirouzi

2023

IET Generation Trans & Dist

113

View full text Add to dashboard Cite

show abstract

“…This paper presents the utilization of the inverter-based virtual power plant (IBVPP) in smart distribution networks (SDNs) by taking into account some indicators, including VSI and opera-tion goals of the DSO. The VPPs, in the current study, contain RESs [24][25][26][27][28]. Also, flexibility of the VPPs is enhanced using flexible sources (e.g.…”

Section: Contributionsmentioning

confidence: 99%

Energy management of distribution network with inverter‐based renewable virtual power plant considering voltage security index

Azarhooshang,

Rezazadeh

2023

IET Renewable Power Gen

View full text Add to dashboard Cite

Virtual power plants (VPP) with resources and storages are able to control the active power of the network. They are also connected to the network through an inverter, which is capable of controlling reactive power. Therefore, it is expected that the optimal use of inverter‐based VPP can play an effective role in improving the economic and technical status of the distribution network. So, the operation of a smart distribution system is presented in this paper by considering inverter‐based VPPs constrained to the operator's measures. The weighted sum of expected energy loss (EEL) and voltage security index (VSI) is minimized while considering AC optimal power flow equations, restrictions of network's security, and operating model of the inverter‐based VPPs. Uncertainties with an origin of the amount of demand, renewable energy, and parameters of mobile energy storage are also discussed. The uncertainties are modelled using a stochastic optimization approach relying on the unscented transformation (UT). Evaluating inverter‐based VPP performance, providing models of flexible resources such as responsive loads and mobile storages, checking network voltage security status, and modelling uncertainties using the UT method are among the innovations of this study. According to the results, it is demonstrated that the technical situation of the distribution system is improved with the help of optimal management of the VPP. With energy management of the inverter‐based VPP, the suggested design has succeeded to enhance the operating status (voltage security) of the system by approximately 33–73% (12%) in comparison to power flow studies.

show abstract

“…Neural networks, which simulate the structure and function of the human brain by the use of a computer system, are a new technique for estimating electrical load and energy, primarily for predicting very short-term and short-term loads [ 7 ]. Artificial neural networks are a popular model that has been extensively studied in the field of load prediction using AI methods [ 8 ].…”

Section: Introductionmentioning

confidence: 99%

Improving the Method of Short-term Forecasting of Electric Load in Distribution Networks using Wavelet transform combined with Ridgelet Neural Network Optimized by Self-adapted Kho-Kho Optimization Algorithm

Wang,

Sun,

Fathi

et al. 2024

Heliyon

View full text Add to dashboard Cite

Short-Term Wind Energy Forecasting with Independent daytime/Nighttime machine Learning Models

Cited by 12 publications

References 16 publications

Network‐constrained unit commitment‐based virtual power plant model in the day‐ahead market according to energy management strategy

Network‐constrained unit commitment‐based virtual power plant model in the day‐ahead market according to energy management strategy

Energy management of distribution network with inverter‐based renewable virtual power plant considering voltage security index

Improving the Method of Short-term Forecasting of Electric Load in Distribution Networks using Wavelet transform combined with Ridgelet Neural Network Optimized by Self-adapted Kho-Kho Optimization Algorithm

Contact Info

Product

Resources

About