Sequential Reconfiguration of Unbalanced Distribution Network with Soft Open Points Based on Deep Reinforcement Learning

نویسندگان

چکیده

With the large-scale distributed generations (DGs) being connected to distribution network (DN), traditional day-ahead reconfiguration methods based on physical models are challenged maintain robustness and avoid voltage off-limits. To address these problems, this paper develops a deep re-inforcement learning method for sequential with soft open points (SOPs) real-time data. A state-based decision model is first proposed by constructing Marko process-based SOP joint optimization so that decisions can be achieved in milliseconds. Then, reinforcement framework including branching double $Q$ (BDDQN) multi-policy actor-critic (MPSAC) proposed, which has significantly improved efficiency of multi-dimensional mixed-integer action space. And influence DG load uncertainty control results been minimized using status DN make decisions. The numerical simulations IEEE 34-bus 123-bus systems demonstrate effectively reduce operation cost solve overvoltage problem caused high ratio photovoltaic (PV) integration.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Operation Scheduling of MGs Based on Deep Reinforcement Learning Algorithm

: In this paper, the operation scheduling of Microgrids (MGs), including Distributed Energy Resources (DERs) and Energy Storage Systems (ESSs), is proposed using a Deep Reinforcement Learning (DRL) based approach. Due to the dynamic characteristic of the problem, it firstly is formulated as a Markov Decision Process (MDP). Next, Deep Deterministic Policy Gradient (DDPG) algorithm is presented t...

متن کامل

Reinforcement Learning with Soft

It is widely accepted that the use of more compact representations than lookup tables is crucial to scaling reinforcement learning (RL) algorithms to real-world problems. Unfortunately almost all of the theory of reinforcement learning assumes lookup table representations. In this paper we address the pressing issue of combining function approximation and RL, and present 1) a function approx-im...

متن کامل

Reinforcement Learning with Deep Energy-Based Policies

We propose a method for learning expressive energy-based policies for continuous states and actions, which has been feasible only in tabular domains before. We apply our method to learning maximum entropy policies, resulting into a new algorithm, called soft Q-learning, that expresses the optimal policy via a Boltzmann distribution. We use the recently proposed amortized Stein variational gradi...

متن کامل

Vision-based Deep Reinforcement Learning

Recently, Google Deepmind showcased how Deep learning can be used in conjunction with existing Reinforcement Learning (RL) techniques to play Atari games[11], beat a world-class player [14] in the game of Go and solve complicated riddles [3]. Deep learning has been shown to be successful in extracting useful, nonlinear features from high-dimensional media such as images, text, video and audio [...

متن کامل

Dueling Network Architectures for Deep Reinforcement Learning

In recent years there have been many successes of using deep representations in reinforcement learning. Still, many of these applications use conventional architectures, such as convolutional networks, LSTMs, or auto-encoders. In this paper, we present a new neural network architecture for model-free reinforcement learning inspired by advantage learning. Our dueling architecture represents two ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Journal of modern power systems and clean energy

سال: 2023

ISSN: ['2196-5420', '2196-5625']

DOI: https://doi.org/10.35833/mpce.2022.000271