Search

The rapid and accurate reconstruction of the supersonic combustor flow field is of great significance for sensing and predicting the combustion state. Existing deep learning methods pay less attention to the convergence speed of flow field reconstruction, which results in longer training and prediction times for the models. This study proposes a method for reconstructing the flow field in supersonic combustor by combining a reduced-order model based on proper orthogonal decomposition (POD) with a multi-branch convolutional neural network. This method first analyzes the effectiveness of POD reconstruction. Then, based on the wall pressure data of the supersonic engine combustor, it performs flow field image reconstruction. Finally, through error calculation and gradient updating with low-resolution principal component flow field shadow images obtained from the POD algorithm, the high-precision and efficient prediction of flow field images is achieved. Different equivalence ratio hydrogen fuel combustion experiments were conducted in a pulsed combustion wind tunnel with an incoming flow Mach number of 2.5. The learning model was trained and tested using the dataset obtained from these experiments. Numerous experiments demonstrated that the model can effectively reconstruct the wave structures of complex flow fields. Multiple evaluation indicators indicated that the reconstructed flow field of the combustor shows good agreement with that obtained from ground wind tunnel testing. Furthermore, after introducing the POD dimensionality reduction model, the training time was reduced by 32.03%, effectively improving the training time complexity of the model. [ABSTRACT FROM AUTHOR]

This paper proposed the physical information residual spatial pyramid pooling (PIResSpp) convolutional neural network that is highly robust and introduces a residual neural network architecture that can satisfactorily fit high-dimensional functions by using jumping connections to reduce the risk of overfitting. Key features of the flow field were extracted by using pooling kernels of different sizes and were then stitched together to fuse its local and global features. The axisymmetric inlet of the scramjet generated by the Bezier curve was established through highly precise numerical simulations, and datasets of flow fields under different geometric configurations were constructed according to the parametric design. The PIResSpp model was trained on a sample dataset, and mapping relationships were established between the parameters of incoming flow/those of the geometry of the inlet, and the velocity, pressure, and density fields in it. Finally, the results of reconstruction of the flow field at the inlet with different design parameters were tested and compared with the outcomes of various deep learning models. The results show that the average peak signal-to-noise ratio of the flow field reconstructed by the proposed model was 36.427, with a correlation coefficient higher than 97%. [ABSTRACT FROM AUTHOR]

Using various types of data to extract different features and reconstruct flame spontaneous image, the rapid acquisition of flame propagation data and automatic identification of the combustion state in a scramjet combustion chamber in confined space can be promoted. A reconstruction model based on multi-source information fusion is proposed that can effectively reconstruct a flame self-luminous image based on the combustor wall pressure and schlieren image data. A dense convolutional neural network is used to construct and reconstruct the model and train the network based on the experimental datasets. The ground pulse combustion wind tunnel test is conducted under the condition of hydrogen fuel self-ignition using the variation rules of different injection pressures. The evolution process of the combustor wall pressure, schlieren image, and flame self-luminescence image is obtained over time and a dataset is constructed. The reconstruction accuracy of the flame spontaneous luminescence image using the schlieren image and the fusion of the schlieren image and wall pressure under different injection pressure is comprehensively analyzed. Moreover, both qualitative and quantitative results show that the prediction accuracy of flame spontaneous illumination based on multi-source information fusion well matches the experimental result, and the flame boundary, flame head region, flame area, and burning intensity are all well reconstructed. The average peak signal to noise ratio reached 31.57 dB, average structural similarity was 0.939, average mean square error was 0.979, correlation coefficient was 0.0013, and intersection over union reached 0.876. • A dense convolution neural network model was proposed to reconstruct flame. • Multi-source information fusion for flame reconstruction is proposed firstly. • Flame can be reconstructed accurately by wall pressure and schlieren images. • Results are helpful to optimize the optical measurement equipment of scramjet. [ABSTRACT FROM AUTHOR]

Two models based on the deep learning-based convolutional neural network (CNN) and the re-parameterized convolutional neural network (RepCNN) were designed to reconstruct the flame in the combustor. Experiments were performed on a ground-pulse combustion wind tunnel at a fixed inlet Mach number of 2.5 and different pressures to inject hydrogen to obtain the relevant datasets. The results showed that both models could reconstruct the image of the flame in the combustor based on pressures of the upper and lower walls as well as the pressure at which hydrogen was injected. The average structural similarity index between the reconstructed image of the flame and its actual/original image was 0.9553, the average peak signal-to-noise ratio was 34.201, and the average correlation coefficient was 0.9819. The speed of reconstruction of the image using the RepCNN model improved by 40.7% at the cost of a slightly lower accuracy compared with the CNN model, and it took only 2.85 ms to reconstruct the image of a single flame. The lightweight feature of the RepCNN provides an important foundation for monitoring the model to reconstruct the image of the flame in real time. The work here simplifies requirements on the hardware for ground wind tunnel tests and provides a new idea for examining the characteristics of the flame in small combustors. [ABSTRACT FROM AUTHOR]

The flame development prediction of a scramjet combustor forecasts the combustion state and provides valuable information for active flow control. Experiments were performed on a hydrogen-fueled scramjet at different equivalence ratios in a ground pulse combustion wind tunnel with a Mach-2.5 incoming flow. Five image datasets of the flame evolution process were constructed at different predicted periods. The memory fusion cascade network (MFCN) was developed to predict flame images after a certain span using flame image sequences of the previous periods. A complete evaluation system was constructed to compare and analyze the performances of MFCN, Kongs, and ResNet16 models in multi- and long-span conditions. Experimental results show that MFCN achieves a maximum increase of 46.16% of the peak signal-to-noise ratio index, 69.14% of the structural correlation coefficient index, and 5.72% of the correlation coefficient index in the independent test set. Moreover, the volume of the model only reaches the KB level, which has the characteristics of being lightweight. MFCN outperforms other methods in terms of the prediction accuracy and maintains stable prediction results during multi- and long-span tasks. [ABSTRACT FROM AUTHOR]

Scramjet is the main power device of hypersonic vehicles. With the gradual expansion of wide velocity domain, shock wave/shock wave and shock wave/boundary layer are the main phenomena in scramjet isolator. When the leading edge of the shock train is pushed out from the inlet of the isolator, the engine will not start. Therefore, it is very important to detect the flow field structure in the isolator and the leading edge position of the shock train. The traditional shock train detection methods have low detection accuracy and slow detection speed. This paper describes a method based on deep learning to reconstruct the flow field in the isolator and detect the leading edge of the shock train. Under various back pressure conditions, the flow field images of computational fluid dynamics (CFD) data and the corresponding upper and lower wall pressure data were obtained, and a data set corresponding to pressure and flow field was constructed. By constructing and training convolutional neural networks, a mapping model with pressure information as input and flow field image as output is obtained, and then the leading edge position of shock train is detected on the output flow field image. The experimental results show that the average structure similarity (SSIM) between the reconstructed flow field image and the CFD flow field image is 0.902, the average peak signal-to-noise ratio (PSNR) is 25.289, the average correlation coefficient (CORR) is 0.956, and the root mean square error of shock train leading edge detection is 3.28 mm. Moreover, if the total pressure input is appropriately reduced, the accuracy of flow field reconstruction does not decline significantly, which means that the model has a certain robustness. Finally, in order to improve the detection accuracy of the leading edge position, we fine tuned the model and obtained another detection method, which reduced the root mean square error of the detection results to 1.87 mm. [ABSTRACT FROM AUTHOR]

This paper reports experiments on a hydrogen-fueled scramjet performed at different equivalence ratios in a ground pulse combustion wind tunnel with a Mach-2.5 incoming flow. In the non-chemical reaction flow before the fuel was ignited, the flow field was oscillatory, and from the pressure monitor, the oscillation period was 0.07 s and the oscillation amplitude was 0.035 MPa. Schlieren and flame self-luminescence images of the combustor flow were obtained synchronously, and the flow-field structure was stable with the flame concentrated in the shear layer. Deep learning was used to subject the low-resolution combustion flow field to super-resolution analysis to reconstruct a high-resolution flow field. To improve the spatial resolution of the flow field during self-ignition of the hydrogen-fueled scramjet and study the flow mechanism and coupling rule of turbulent fluctuations in the ignition process, a multipath asymmetric residual network (MARN) is proposed based on a single-path super-resolution convolutional neural network (SRCNN) and a residual network model (ResNet_16). The experimental results show that compared with SRCNN and ResNet_16, MARN has the best precision and performance regarding the super-resolution flow field in the self-ignition of hydrogen fuel in terms of the mean peak signal-to-noise ratio, mean structural similarity, and average correlation coefficient as well as being the least complicated. The proposed method offers the possibility of developing lightweight super-resolution models for the flow fields in supersonic combustors; it shows enormous potential for revealing the physical flow of the fuel and air mixture, and it offers accurate forecasts of self-ignition times. [ABSTRACT FROM AUTHOR]

The data-driven intelligent reconstruction of a flow field in a supersonic combustor aids the real-time monitoring of wave system evolution in a scramjet flow field structure, allowing the determination of the combustion state for active flow control. In this paper, a deep learning architecture based on a multi-branch fusion convolutional neural network (MBFCNN) is proposed to reconstruct the flow field in a supersonic combustor. Experiments on hydrogen-fueled scramjets with different equivalence ratios were carried out in a direct-connected supersonic pulse combustion wind tunnel with an inflow Mach number of 2.5 to establish a dataset for MBFCNN network training and testing. The trained model successfully reconstructed the flow field structure from measured wall pressure data. The flow field reconstruction model provided a rich information source for the evolution of the wave system structure under the self-ignition conditions of the hydrogen-fueled scramjet, greatly improving the detection accuracy. The proposed deep learning architecture method was compared with basic convolutional neural network and symmetric convolutional neural network methods. The three methods all accurately reconstructed the flow field of the supersonic combustor. However, the proposed MBFCNN provided the best reconstruction results, and its average linear correlation coefficient in the test set was 0.952. The proposed MBFCNN had a lower mean square error and higher peak signal-to-noise ratio than the other two methods, which verified that the proposed model is eminently able to reconstruct and predict the flow field of a supersonic combustor. [ABSTRACT FROM AUTHOR]

The detection of supersonic combustion flow field provides valuable reference information for the efficient combustion of hypersonic vehicles. From the perspectives of scramjet operation during flight as well as experiment where the measured information is limited, flow field prediction using deep learning technology is a promising method that can provide future flow field evolution in scramjet combustors. This study proposes a convolutional neural network model of multi-temporal pressure information fusion to predict combustion flow field based on wall pressure. In addition, it analyzes the effect of various wall pressures on the flow field prediction. An experimental dataset of supersonic ground-pulse wind tunnel tests with different injection pressures is built through the combustion chamber wall under self-ignition conditions of a hydrogen-fueled scramjet. The results showed that the predicted flow field of the combustor agrees with the ground test, average correlation coefficient reached more than 90% and when the wall pressure number is 32, the average peak signal to noise ratio reaches about 23. and the main wave structure and turbulent boundary layer region are restored even if there is a shock boundary layer interaction. [ABSTRACT FROM AUTHOR]