导航

ACTA AERONAUTICAET ASTRONAUTICA SINICA ›› 2021, Vol. 42 ›› Issue (4): 524208-524208.doi: 10.7527/S1000-6893.2020.24208

• Solid Mechanics and Vehicle Conceptual Design • Previous Articles     Next Articles

Prediction of CMG failure boundary domain based on combined stress test and neural network

HUANG Shouqing1,2, LIU Shouwen1,2, ZHAI Baichen3, ZHOU Yuan1,2, HUANG Xiaokai1,2, QIN Taichun1,2   

  1. 1. Beijing Key Laboratory of Environment & Reliability Test Technology for Aerospace Mechanical & Electrical Products, Beijing 100094, China;
    2. Beijing Institute of Spacecraft Environment Engineering, Beijing 100094, China;
    3. Beijing Institute of Control Engineering, Beijing 100190, China
  • Received:2020-05-11 Revised:2020-06-16 Published:2020-07-10
  • Supported by:
    Civil Aerospace Technology Research Project from State Administration of Science, Technology and Industry for National Defense(A0201); Equipment Research Project of PLA Equipment Development Department (41402010103)

Abstract: This study designs the test equipment for simultaneous simulation of the vacuum thermal environment and angular momentum exchange conditions between a CMG (Control Moment Gyroscope) and a spacecraft, and proposes a combined stress working-state test method to simulate the in orbit vacuum temperature, CMG gimbal rotational speed and spacecraft rotational speed. A quantitative expression method for CMG running status suitable for a neural network model is applied. Based on relatively little amount of test results, the neural network model after training is used to predict the working limit rotational speed matrix, the failure boundary and the failure boundary domain. Furthermore, the influence of experience samples on the prediction results is analyzed, and the coupling effect of each stress on the working domain of other stresses investigated, and the predicted initial value by the neural network proposed to reflect the reliability of the prediction results. The results show that the presented method can not only simulate the real working-state better but also significantly save both test cost and time, in addition to high prediction accuracy and good adaptability under multi-stress working conditions. The prediction accuracy can reach up to 100% and 98.8% based on the two training data sets I and Ⅱ, respectively. The failure boundary of rotational speeds at 55 ℃ is given which cannot be obtained only via the test data, and the engineering experience behind the test data can be internalized.

Key words: control moment gyroscopes, working-state tests, neural networks, failure boundary domains, combined stress

CLC Number: