Wide-Range Variable Cycle Engine Control Based on Deep Reinforcement Learning
In this paper, a controller design method based on deep reinforcement learning is proposed for a wide-range variable cycle engine with a turbine interstage mixed architecture. The PID controller is subject to limitations, including single-input single-output limitations, low regulation efficiency, a...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-05-01
|
| Series: | Aerospace |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2226-4310/12/5/424 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849711027070959616 |
|---|---|
| author | Yaoyao Ding Fengming Wang Yuanwei Mu Hongfei Sun |
| author_facet | Yaoyao Ding Fengming Wang Yuanwei Mu Hongfei Sun |
| author_sort | Yaoyao Ding |
| collection | DOAJ |
| description | In this paper, a controller design method based on deep reinforcement learning is proposed for a wide-range variable cycle engine with a turbine interstage mixed architecture. The PID controller is subject to limitations, including single-input single-output limitations, low regulation efficiency, and poor adaptability when confronted with contemporary variable cycle engines that exhibit complex and multi-variable operating conditions. To solve this problem, this paper adopts a deep reinforcement learning method based on a deep deterministic policy gradient algorithm, and it applies an action space pruning technique to optimize the controller, which significantly improves the convergence speed of network training. In order to verify the control performance, two typical flight conditions are selected for simulation experiments as follows: in the first scenario, H = 0 km and Ma = 0, while in the second scenario, H = 10 km and Ma = 0.9. A comparison of the simulation results shows that the proposed deep reinforcement learning controller effectively addresses the engine’s multi-variable coupling control problem. In addition, it reduces response time by an average of 44.5%, while maintaining a similar overshoot level to that of the PID controller. |
| format | Article |
| id | doaj-art-21211f96f321434d8d100a916c7d5c4e |
| institution | DOAJ |
| issn | 2226-4310 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Aerospace |
| spelling | doaj-art-21211f96f321434d8d100a916c7d5c4e2025-08-20T03:14:43ZengMDPI AGAerospace2226-43102025-05-0112542410.3390/aerospace12050424Wide-Range Variable Cycle Engine Control Based on Deep Reinforcement LearningYaoyao Ding0Fengming Wang1Yuanwei Mu2Hongfei Sun3School of Aerospace Engineering, Xiamen University, Xiamen 361102, ChinaAero Engine Academy of China, Beijing 101304, ChinaAero Engine Academy of China, Beijing 101304, ChinaSchool of Aerospace Engineering, Xiamen University, Xiamen 361102, ChinaIn this paper, a controller design method based on deep reinforcement learning is proposed for a wide-range variable cycle engine with a turbine interstage mixed architecture. The PID controller is subject to limitations, including single-input single-output limitations, low regulation efficiency, and poor adaptability when confronted with contemporary variable cycle engines that exhibit complex and multi-variable operating conditions. To solve this problem, this paper adopts a deep reinforcement learning method based on a deep deterministic policy gradient algorithm, and it applies an action space pruning technique to optimize the controller, which significantly improves the convergence speed of network training. In order to verify the control performance, two typical flight conditions are selected for simulation experiments as follows: in the first scenario, H = 0 km and Ma = 0, while in the second scenario, H = 10 km and Ma = 0.9. A comparison of the simulation results shows that the proposed deep reinforcement learning controller effectively addresses the engine’s multi-variable coupling control problem. In addition, it reduces response time by an average of 44.5%, while maintaining a similar overshoot level to that of the PID controller.https://www.mdpi.com/2226-4310/12/5/424wide-rangevariable cycle enginePID controldeep reinforcement learningaction space pruning |
| spellingShingle | Yaoyao Ding Fengming Wang Yuanwei Mu Hongfei Sun Wide-Range Variable Cycle Engine Control Based on Deep Reinforcement Learning Aerospace wide-range variable cycle engine PID control deep reinforcement learning action space pruning |
| title | Wide-Range Variable Cycle Engine Control Based on Deep Reinforcement Learning |
| title_full | Wide-Range Variable Cycle Engine Control Based on Deep Reinforcement Learning |
| title_fullStr | Wide-Range Variable Cycle Engine Control Based on Deep Reinforcement Learning |
| title_full_unstemmed | Wide-Range Variable Cycle Engine Control Based on Deep Reinforcement Learning |
| title_short | Wide-Range Variable Cycle Engine Control Based on Deep Reinforcement Learning |
| title_sort | wide range variable cycle engine control based on deep reinforcement learning |
| topic | wide-range variable cycle engine PID control deep reinforcement learning action space pruning |
| url | https://www.mdpi.com/2226-4310/12/5/424 |
| work_keys_str_mv | AT yaoyaoding widerangevariablecycleenginecontrolbasedondeepreinforcementlearning AT fengmingwang widerangevariablecycleenginecontrolbasedondeepreinforcementlearning AT yuanweimu widerangevariablecycleenginecontrolbasedondeepreinforcementlearning AT hongfeisun widerangevariablecycleenginecontrolbasedondeepreinforcementlearning |