Forecasting Thermoacoustic Instabilities in Liquid Propellant Rocket Engines Using Multimodal Bayesian Deep Learning

TL;DR

This paper develops a Bayesian neural network model to forecast thermoacoustic instabilities in liquid rocket engines using multimodal sensor data, achieving accurate predictions and uncertainty quantification with limited experimental data.

Contribution

It introduces an autoregressive Bayesian deep learning approach for predicting instabilities in rocket combustors, incorporating multiple sensor inputs and interpretability techniques.

Findings

Accurately forecast pressure amplitudes 500 ms in advance

High-frequency dynamic pressure signals are highly informative

Predictive uncertainties are well-characterized and respond to sensor failures

Abstract

The 100 MW cryogenic liquid oxygen/hydrogen multi-injector combustor BKD operated by the DLR Institute of Space Propulsion is a research platform that allows the study of thermoacoustic instabilities under realistic conditions, representative of small upper stage rocket engines. We use data from BKD experimental campaigns in which the static chamber pressure and fuel-oxidizer ratio are varied such that the first tangential mode of the combustor is excited under some conditions. We train an autoregressive Bayesian neural network model to forecast the amplitude of the dynamic pressure time series, inputting multiple sensor measurements (injector pressure/ temperature measurements, static chamber pressure, high-frequency dynamic pressure measurements, high-frequency OH* chemiluminescence measurements) and future flow rate control signals. The Bayesian nature of our algorithms allows us to work with a dataset whose size is restricted by the expense of each experimental run, without making overconfident extrapolations. We find that the networks are able to accurately forecast the evolution of the pressure amplitude and anticipate instability events on unseen experimental runs 500 milliseconds in advance. We compare the predictive accuracy of multiple models using different combinations of sensor inputs. We find that the high-frequency dynamic pressure signal is particularly informative. We also use the technique of integrated gradients to interpret the influence of different sensor inputs on the model prediction. The negative log-likelihood of data points in the test dataset indicates that predictive uncertainties are well-characterized by our Bayesian model and simulating a sensor failure event results as expected in a dramatic increase in the epistemic component of the uncertainty.