The main challenge and limitation of combustion diagnosis technique lies in the volumetric/spatial resolution. Frequently reported technique widely use laser bean scanning to get inner knowledge of combustion, however due to the line-of-sight nature, spatial resolution along the “path” is lacked. With the ability of 3-D tomography detection, CTC technique (computed tomography of chemiluminescence) shows advantage over other techniques; but not yet widely used in combustion science study. In CTC measurement, “dense projection” is required to make satisfying tomography reconstruction, but this is practically hard to meet for high cost and spatial limitation. At sparse projection conditions, tomography reconstruction is distorted with the appearance of geometric blur. The reason of occurrence of blur is analyzed in this paper, and an improved method for tomography reconstruction is proposed. Test results show that blur distortion could be effectively eliminated using the improved method at sparse projection conditions. Other factors and constraints that have influence on flame tomography measurement are discussed in this paper, and quantitative evaluations were made through tests.
The attitude angle disturbance of helicopter payload platform can significantly reduce the density distribution and imaging accuracy of laser point cloud measured by airborne LiDAR, so an attitude angle stabilization device is designed to compensate the attitude angle disturbance in real time. In order to eliminate strong coupling effect of the control system of the attitude angle stabilization device, decoupling control based on neural network inverse system is designed. Firstly, the dynamic model of the control system is established; secondly, the neural network inverse dynamic model is built, and a compound control strategy with PID feedback controller and neural network inverse system feed-forward controller is adopted to realize real-time decoupling control and improve the static and dynamic control performance; finally, simulation experiment for the decoupling control is carried out, and results show that the control system has fine static and dynamic performance. The designed compound decoupling control strategy can effectively improve the control accuracy and has excellent anti-jamming robustness against random error disturbance.