A Reinforcement Learning Approach for Transient Control of Liquid Rocket Engines

Waxenegger-Wilfing, Günther; Dresia, Kai; Deeken, Jan Christian; Oschwald, Michael

doi:10.48550/arxiv.2006.11108

Cited by 2 publications

(2 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…But this does not work for higher state dimensions. A deep reinforcement learning approach is investigated for optimal control of a generic gas-generator engine's continuous start-up phase [16]. The modeling and simulation tool EcosimPro is used as an engine simulator to train the NN controller.…”

Section: Start-up Control Of Gas-generator Engines Using Deep Reinfor...mentioning

confidence: 99%

Machine Learning Methods for the Design and Operation of Liquid Rocket Engines -- Research Activities at the DLR Institute of Space Propulsion

Waxenegger-Wilfing,

Dresia,

Deeken

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

The last years have witnessed an enormous interest in the use of artificial intelligence methods, especially machine learning algorithms. This also has a major impact on aerospace engineering in general, and the design and operation of liquid rocket engines in particular, and research in this area is growing rapidly. The paper describes current machine learning applications at the DLR Institute of Space Propulsion. Not only applications in the field of modeling are presented, but also convincing results that prove the capabilities of machine learning methods for control and condition monitoring are described in detail. Furthermore, the advantages and disadvantages of the presented methods as well as current and future research directions are discussed.Recent research and development activities at the DLR Institute of Space Propulsion prove the feasibility of such methods for supporting the design and operation of liquid rocket engines. So far, NNs are used for the prediction of heat transfer in rocket engine cooling channels and fatigue life estimation. Other applications include the automatic discovery of suitable precursors to combustion instabilities and optimal control of the engines.

show abstract

Section: Start-up Control Of Gas-generator Engines Using Deep Reinfor...mentioning

confidence: 99%

Machine Learning Methods for the Design and Operation of Liquid Rocket Engines -- Research Activities at the DLR Institute of Space Propulsion

Waxenegger-Wilfing,

Dresia,

Deeken

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Related work by McCartney et al 20 uses the detrended fluctuation analysis (DFA) spectrum of the pressure signal as input to a random forest and finds that this approach compares favorably to precursors from the literature. Recent works have investigated machine learning methods for the design and operation of cryogenic rocket engines [21][22][23][24] .…”

Section: Introductionmentioning

confidence: 99%

Early Detection of Thermoacoustic Instabilities in a Cryogenic Rocket Thrust Chamber using Combustion Noise Features and Machine Learning

Waxenegger-Wilfing,

Sengupta,

Martin

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Combustion instabilities are particularly problematic for rocket thrust chambers because of their high energy release rates and their operation close to the structural limits. In the last decades, progress has been made in predicting high amplitude combustion instabilities but still, no reliable prediction ability is given. Reliable early warning signals are the main requirement for active combustion control systems. In this paper, we present a data-driven method for the early detection of thermoacoustic instabilities. Recurrence quantification analysis is used to calculate characteristic combustion features from short-length time series of dynamic pressure sensor data. Features like the recurrence rate are used to train support vector machines to detect the onset of an instability a few hundred milliseconds in advance. The performance of the proposed method is investigated on experimental data from a representative LOX/H 2 research thrust chamber. In most cases, the method is able to timely predict two types of thermoacoustic instabilities on test data not used for training. The results are compared with state-of-the-art early warning indicators.High-amplitude pressure oscillations known as combustion instability are an issue for all chemical propulsion systems. Combustion instabilities are particularly problematic for rocket thrust chambers because of their high energy release rates. Especially thermoacoustic oscillations are a major hazard, but difficult to predict. An important question is whether features of combustion noise can be used to construct reliable early warning signals for representative rocket thrust chambers. Among other things, instability precursors are needed for active combustion control systems. In this study, we use recurrence quantification analysis and well-known machine learning models to construct an early warning signal that should be able to predict high-amplitude instabilities a few hundred milliseconds in advance. The performance of the proposed method is investigated on experimental data from a representative LOX/H 2 research thrust chamber. We describe the test case in detail and also evaluate other measures like the Hurst exponent.

show abstract

A Reinforcement Learning Approach for Transient Control of Liquid Rocket Engines

Cited by 2 publications

References 32 publications

Machine Learning Methods for the Design and Operation of Liquid Rocket Engines -- Research Activities at the DLR Institute of Space Propulsion

Machine Learning Methods for the Design and Operation of Liquid Rocket Engines -- Research Activities at the DLR Institute of Space Propulsion

Early Detection of Thermoacoustic Instabilities in a Cryogenic Rocket Thrust Chamber using Combustion Noise Features and Machine Learning

Contact Info

Product

Resources

About