Enhancing deep reinforcement learning with integral action to control tokamak safety factor
نویسندگان
چکیده
Recent advances in the use of Artificial Intelligence to control complex systems make it suitable for profile plasma control. In this work, we propose an algorithm based on Deep Reinforcement Learning safety factor with a feedback design. For purpose, first derive device-specific control-oriented model fast simulation time. Then, order enhance robustness respect external disturbances and errors, include error time integrator into controller. A cascade kinetic magnetic models is used learning procedure Finally, illustrate efficiency proposed design procedure, obtained controller tested reference simulator, Raptor simulator.
منابع مشابه
Learning to Factor Policies and Action-Value Functions: Factored Action Space Representations for Deep Reinforcement learning
Deep Reinforcement Learning (DRL) methods have performed well in an increasing numbering of high-dimensional visual decision making domains. Among all such visual decision making problems, those with discrete action spaces often tend to have underlying compositional structure in the said action space. Such action spaces often contain actions such as go left, go up as well as go diagonally up an...
متن کاملMulti-Task Deep Reinforcement Learning for Continuous Action Control
In this paper, we propose a deep reinforcement learning algorithm to learn multiple tasks concurrently. A new network architecture is proposed in the algorithm which reduces the number of parameters needed by more than 75% per task compared to typical single-task deep reinforcement learning algorithms. The proposed algorithm and network fuse images with sensor data and were tested with up to 12...
متن کاملContinuous control with deep reinforcement learning
We adapt the ideas underlying the success of Deep Q-Learning to the continuous action domain. We present an actor-critic, model-free algorithm based on the deterministic policy gradient that can operate over continuous action spaces. Using the same learning algorithm, network architecture and hyper-parameters, our algorithm robustly solves more than 20 simulated physics tasks, including classic...
متن کاملSafety Factor Profile Control in a Tokamak
Give us 5 minutes and we will show you the best book to read today. This is it, the safety factor profile control in a tokamak that will be your best choice for better reading book. Your five times will not spend wasted by reading this website. You can take the book as a source to make better concept. Referring the books that can be situated with your needs is sometime difficult. But here, this...
متن کاملDeep Reinforcement Learning in Parameterized Action Space
Recent work has shown that deep neural networks are capable of approximating both value functions and policies in reinforcement learning domains featuring continuous state and action spaces. However, to the best of our knowledge no previous work has succeeded at using deep neural networks in structured (parameterized) continuous action spaces. To fill this gap, this paper focuses on learning wi...
متن کاملذخیره 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: Fusion Engineering and Design
سال: 2023
ISSN: ['1873-7196', '0920-3796']
DOI: https://doi.org/10.1016/j.fusengdes.2023.114008