Applying reinforcement learning and tree search to the unit commitment problem

Mars, Patrick de; O’Sullivan, Aidan

doi:10.1016/j.apenergy.2021.117519

Cited by 32 publications

(14 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, if rand P3 is less than P 3 , an aggressive siege fight is chosen, as depicted in ( 14) and ( 15) [34]. (15) Where BestVulture 1 (i) and BestVulture 2 (i) represent the firstand second-best vultures in the current iteration, P(i) is the current position of the vultures, and R(i), d(t), and F are variables defined previously. The levy flight equation is used to enhance the efficiency of the AVOA algorithm and is computed according to Equation (16).…”

Section: Phase 3: Explorationmentioning

confidence: 99%

See 1 more Smart Citation

Solving wind‐integrated unit commitment problem by a modified African vultures optimization algorithm

Abuelrub,

Awwad,

Al‐Masri

2023

IET Generation Trans & Dist

View full text Add to dashboard Cite

Unit commitment (UC) stands out as a significant challenge in electrical power systems. With the rapid growth in power demand and the pressing issues of fossil fuel scarcity and global warming, it has become crucial to enhance the utilization of renewable energy sources. This study focuses on addressing the UC problem by incorporating a wind farm and proposes a modified version of the metaheuristic African vultures optimization algorithm (AVOA) in binary form, utilizing the sigmoid transfer function. The modified AVOA employs multiple phase‐shift tactics to overcome premature local optima. By determining the on/off status of generating units, the modified AVOA improves the algorithm's effectiveness. Additionally, the paper develops an auto‐regressive moving average model (ARMA) to forecast wind speeds, with the AVOA assisting in selecting the optimal orders (q and p) of the ARMA model. This is done using historical wind speed data to capture uncertainty in the wind speed. The wind power is then calculated using various models and integrated into the UC problem. The effectiveness of the modified AVOA is examined on the IEEE 30‐bus system. The binary AVOA (BAVOA) outperforms several algorithms presented in the case study, demonstrating its superiority. Furthermore, the results indicate that BAVOA delivers superior outcomes within the discrete search space when compared to the continuous search space.

show abstract

Section: Phase 3: Explorationmentioning

confidence: 99%

“…The algorithm can self-learn, and it was tested on the IEEE 118-bus system and the State Grid Hunan Electric Power Company. The authors in [15] used reinforcement learning with tree search to solve unit commitment under RES and load uncertainties. The effectiveness of this algorithm was tested on a 10-unit system.…”

Section: Introductionmentioning

confidence: 99%

Solving wind‐integrated unit commitment problem by a modified African vultures optimization algorithm

Abuelrub,

Awwad,

Al‐Masri

2023

IET Generation Trans & Dist

View full text Add to dashboard Cite

show abstract

“…In recent years, many researchers have adopted data-driven approaches like reinforcement learning (RL) to overcome the limitations of traditional approaches in PSD [7]. In [8], a distributed Q-learning algorithm is adopted to improve robustness through local information communication; In [9], the policy gradient algorithm is adopted to achieve a fast and efficient search of the action space through a bootstrap tree search approach; In [10], multi-scene parallel optimal scheduling is implemented; In [11], better convergence and economy are achieved through multi-intelligence reinforcement learning.…”

Section: Introductionmentioning

confidence: 99%

A Deep Reinforcement Learning Method for Unit Commitment with Renewable Energy Generators

Rao,

Zheng

2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

To cope with the challenges of complex modeling, difficult solutions, and low efficiency brought by a significant amount of renewable energy units to the power system dispatch (PSD), a deep reinforcement learning method for unit commitment with renewable energy generators is proposed. Firstly, a PPO-based reinforcement learning method is utilized to promptly and effectively resolve the unit commitment. A mathematical optimization technique is employed to solve the security-constrained economic dispatch, considering the “N-1” security check. By integrating these two methods, the decision-making efficiency is enhanced while ensuring the security of the result. The proposed method’s efficiency is affirmed by case simulation.

show abstract

“…This makes UC heavily rely on accurate forecasts while the day-ahead forecasts of renewable generation are presently unreliable. Alternative methods, including reinforcement learning [7][8][9], however, have still been proposed as offline control, nor considered complex network constraints such as transmission line capacity, which is infeasible in actual scheduling. On the other hand, ED is usually modeled as a convex optimization problem and can be solved by the interior point and dual methods [10][11][12], but these methods still encounter problems of convergence and computational speed in the face of AC power flow models, in which ED turns into a non-convex problem.…”

mentioning

confidence: 99%

Real-time scheduling of renewable power systems through planning-based reinforcement learning

Liu¹,

Wang²,

Ye³

et al. 2023

Preprint

View full text Add to dashboard Cite

The growing renewable energy sources have posed significant challenges to traditional power scheduling. It is difficult for operators to obtain accurate day-ahead forecasts of renewable generation, thereby requiring the future scheduling system to make real-time scheduling decisions aligning with ultra-short-term forecasts. Restricted by the computation speed, traditional optimization-based methods can not solve this problem. Recent developments in reinforcement learning (RL) have demonstrated the potential to solve this challenge. However, the existing RL methods are inadequate in terms of constraint complexity, algorithm performance, and environment fidelity. We are the first to propose a systematic solution

show abstract

Applying reinforcement learning and tree search to the unit commitment problem

Cited by 32 publications

References 21 publications

Solving wind‐integrated unit commitment problem by a modified African vultures optimization algorithm

Solving wind‐integrated unit commitment problem by a modified African vultures optimization algorithm

A Deep Reinforcement Learning Method for Unit Commitment with Renewable Energy Generators

Real-time scheduling of renewable power systems through planning-based reinforcement learning

Contact Info

Product

Resources

About