RBFNN Online Rescue Orbit Decision for Launch Vehicle Thrust Drop
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Launch vehicles under thrust drop fault face challenges in completing missions due to the difficulty in determining optimal rescue orbital elements, which is compounded by the large search space of unknown rescue orbits, affecting calculation efficiency.
Innovation Solution
An optimal rescue orbital elements online decision-making method based on RBFNN is developed, which involves establishing dynamic equations, constructing optimization problems, using the adaptive pseudo-spectral method to solve offline, normalizing data, selecting a radial basis function neural network data center, and training the network to establish a nonlinear mapping from fault states to optimal rescue orbital elements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the coupling optimization method is used to solve rescue trajectory decision and trajectory optimization, then the solution provides good online performance, but the search space of unknown rescue orbits is very large which affects calculation efficiency
Solution Approach 1:
The patent segments the complex trajectory optimization problem into two independent parts: (1) optimal rescue orbital elements decision-making, and (2) trajectory optimization. By decoupling these problems, the search space is reduced from the entire trajectory space to only the orbital elements space, significantly improving calculation efficiency while maintaining online rescue performance
Solution Approach 2:
The patent performs preliminary action by using offline computations (adaptive pseudo-spectral method) to generate training data for the RBFNN model. This preliminary work captures the complex nonlinear relationships between fault states and optimal orbital elements, allowing the online phase to use a simplified neural network model that is much faster to compute
2Manufacturing precision
If the search space for unknown rescue orbits is explored, then the optimal rescue orbital elements can be found, but the calculation efficiency is reduced due to the large search space
Solution Approach 1:
The patent creates a computational model (RBFNN) that copies the complex nonlinear mapping relationship between fault states and optimal orbital elements. Instead of directly searching through the entire orbital space, the model learns from offline training data and provides accurate predictions online, achieving both high precision and efficiency
Solution Approach 2:
The patent changes the approach from direct trajectory optimization to orbital elements optimization. By focusing on optimizing only the orbital elements (semi-major axis, eccentricity, inclination, etc.) rather than the entire trajectory, the search space is dramatically reduced while maintaining the accuracy of the rescue orbit determination
Data Source
AI summary
An optimal rescue orbital elements online decision-making method based on RBFNN for launch vehicles under thrust drop fault includes establishing the flight dynamic equations of launch vehicles in the second-stage ascending phase in the geocentric inertial coordinate system, to construct a series of optimization problems of maximum semi-major axis of circular orbit under the thrust drop fault. The method further includes using the adaptive pseudo-spectrum method to solve the optimization problems of maximum semi-major, and using the maximum and minimum method to normalize the sample data to [−1, 1], using the orthogonal least square method to select the data center of the radial basis function neural network (RBFNN), where the Gaussian function is selected as the radial basis function, and the RBFNN is trained offline to establish a nonlinear mapping relationship from the fault states to the optimal rescue orbital elements.