We have for the first time achieved cavity ignition and sustainable combustion of liquid kerosene in supersonic flow of Mach number 2.52 using a laser-induced plasma (LIP) on a model supersonic combustor equipped with dual cavities in tandem as flameholders. The liquid kerosene of ambient temperature is injected from the front wall of the upstream cavity, while the ignitions have been conducted in both cavities. High-speed chemiluminescence imaging shows that the flame kernel initiated in the downstream cavity can propagate contraflow into upstream cavity and establish full sustainable combustion. Based on the qualitative distribution of the kerosene vapor in the cavity, obtained using the kerosene planar laser-induced fluorescence technique, we find that the fuel atomization and evaporation, local hydrodynamic and mixing conditions in the vicinity of the ignition position and in the leading edge area of the cavity have combined effects on the flame kernel evolution and the eventual ignition results.

To optimize a laser ignition scheme, absorption rate measurements and Schlieren visualizations are performed on dual-pulse laser-induced breakdowns (LIBs) at incident energies from 50 mJ to 200 mJ and pulse intervals that range from 20 ns to 250 μs in quiescent air at atmospheric pressure. For comparison, experiments on single-pulse LIBs are also conducted. The shock loss is determined using a semi-empirical model (Jones' model), and quantitative information on the spatial distribution of the hot plume is extracted from Schlieren images using in-house code. The results reveal that multi-location laser ignition can be achieved without reducing the energy absorption or strengthening the shock loss only when the energy of each laser pulse exceeds 200 mJ. This requirement is because the absorption rate of single-pulse LIB decreases significantly when the laser energy is lower than 200 mJ, and the shock loss of single-pulse LIB invariably accounts for approximately 80% of the absorbed laser energy at various incident energies. Compared with single-pulse LIB, dual-pulse LIB with a pulse interval of less than 200 ns is slightly inferior in terms of energy absorption and shock loss; however, the advantages of a larger initial plasma volume and lower energy dissipation can compensate for this deficiency. Therefore, dual-pulse laser ignition is a promising alternative to single-pulse laser ignition. Moreover, ignition reliability can be enhanced by initially releasing the laser pulse with higher energy when the energies of the successive pulses are not the same because of higher energy absorption and lower shock loss. In addition, the spatial distribution of the resulting hot plume is relatively centralized, which helps to reduce energy and radical dissipation. However, a pulse interval longer than 200 ns should be avoided for dual-pulse LIB because the laser energy cannot be utilized efficiently.

To reduce the energy of an individual laser pulse, dual-pulse laser ignitions (LIs) at various pulse intervals were investigated in a Mach 2.92 scramjet engine fueled with ethylene. For comparison, experiments on a single-pulse LI were also performed. Schlieren visualization and high-speed photography were employed to observe the ignition processes simultaneously. The results indicate that the energy of an individual laser pulse can be reduced by half via a dual-pulse LI method as compared with a single-pulse LI with the same total energy. The reduction of the individual laser pulse energy degrades the requirements on the laser source and the beam delivery system, which facilitates the practical application of LI in hypersonic vehicles. A pulse interval shorter than 40 μs is suggested for dual-pulse LI in the present study. Because of the intense heat loss and radical dissipation in high-speed flows, the pulse interval for dual-pulse LI should be short enough to narrow the spatial distribution of the initial flame kernel.

Combustion mode analysis is essential to a rocket-based combined cycle (RBCC) combustor because it may experience multiple combustion modes during the operation. In this study, a method based on an autoencoder and a K-means algorithm was proposed for combustion mode analysis. Flame chemiluminescence images and schlieren images of three combustion modes observed in an RBCC combustor were used to evaluate this method. Two autoencoders that followed the same encoder–decoder architecture were developed separately to generate the latent space representations of flame chemiluminescence images and schlieren images. In the latent space, the centroids and boundaries of different combustion modes were determined using the K-means algorithm. Each autoencoder was trained using 750 images and tested using another 3000 images. The method achieved an accuracy up to 99% on both flame chemiluminescence images and schlieren images. The images generated by the decoder suggested that the autoencoder captured the important features (e.g., primary reaction zone and shock wave) of the reacting flow field. The autoencoder developed for flame chemiluminescence images also successfully detected the combustion mode transition during an ignition process, which suggested that it had the potential to monitor the combustion mode in a real time manner. However, the autoencoder failed on monitoring combustion mode transition when it came to the schlieren images because the optical access of the training data was not exactly the same. Therefore, it is essential to ensure that the optical accesses of different combustion modes are exactly the same when schlieren images are employed for combustion mode analysis.