A flame’s structural feature is a crucial parameter required to comprehensively understand the interaction between turbulence and flames. The generation and evolution processes of the structure feature have rarely been investigated in lean blowout (LBO) flame instability states. Hence, to understand the precursor features of the LBO flame, this work employed high-speed OH-PLIF measurements to acquire time-series LBO flame images and developed a novel feature extraction method based on a deep neural network to quantify the LBO features in real time. Meanwhile, we proposed a deep neural network segmentation method based on a tri-map called the Fire-MatteFormer, and conducted a statistical analysis on flame surface features, primarily holes. The statistical analysis results determined the relationship between the life cycle of holes (from generation to disappearance) and their area, perimeter, and total number. The trained Fire-MatteFormer model was found to represent a viable method for determining flame features in the detection of incipient LBO instability conditions. Overall, the model shows significant promise in ascertaining local flame structure features.

A temporal pre-shaped burst-mode hybrid fiber-bulk laser system was illustrated at a 10 kHz rate with a narrow spectral linewidth. A theoretical model was proposed to counteract the temporal profile distortion and compensate for the desired one, based on reverse process of amplification. For uniformly modulated injection, amplified shapes were recorded and investigated in series for their varied pulse duration, envelope width and amplification delay, respectively. The pre-shaped output effectively realized a uniform distribution on a time scale for both the burst envelope and pulse shape under the action of the established theoretical method. Compared with previous amplification delay methods, this model possesses the capacity to extend itself for applications in burst-mode shaping with variable parameters and characteristics. The maximum pulse energy was enlarged up to 9.68 mJ, 8.94 mJ and 6.57 mJ with a 300 ns pulse duration over envelope widths of 2 ms to 4 ms. Moreover, the time-averaged spectral bandwidths were measured and characterized with Lonrentz fits of 68.3 MHz, 67.2 MHz and 67.7 MHz when the pulse duration varied from 100 ns to 300 ns.

This paper presents the flow field reconstruction and prediction of powder fuel transport systems based on representative feature extraction from scattering images using deep learning techniques. A laboratory-built powder fuel supply system was used to conduct scattering spectroscopy experiments on boron-based fuel under various flow rate conditions. Based on the acquired scattering images, a prediction and reconstruction method was developed using a deep network framework composed of a Stacked Autoencoder (SAE), a Backpropagation Neural Network (BP), and a Long Short-Term Memory (LSTM) model. The proposed framework enables accurate classification and prediction of the dynamic evolution of flow structures based on learned representations from scattering images. Experimental results show that the feature vectors extracted by the SAE form clearly separable clusters in the latent space, leading to high classification accuracy under varying flow conditions. In the prediction task, the feature vectors predicted by the LSTM exhibit strong agreement with ground truth, with average mean square error, mean absolute error, and r-square values of 0.0027, 0.0398, and 0.9897, respectively. Furthermore, the reconstructed images offer a visual representation of the changing flow field, validating the model’s effectiveness in structure-level recovery. These results suggest that the proposed method provides reliable support for future real-time prediction of powder fuel mass flow rates based on optical sensing and imaging techniques.