The invention discloses a rocket control signal correction algorithm based on a BP (Back Propagation) neural network, which is used for amplitude and phase correction of a control system signal, belongs to the field of carrier rocket control, and comprises the following steps of: 1, giving expected correction amplitude and correction phase characteristics within a frequency range considered by a control signal according to rocket body characteristics; 2, generating a BP neural network time domain signal training sample sequence containing a constant value and multiple frequency components according to expected amplitude phase characteristics; 3, designing a BP neural network with a proper scale; 4, training the BP neural network by adopting a sample until a better convergence effect is achieved; and step five, amplitude phase characteristic identification is carried out on the trained BP network, and the effect is verified. According to the method, the principle is simple, the defect that the amplitude-phase characteristics of a linear correction system have a hard relation is overcome, attenuation of high-frequency-band signals is achieved while the low-frequency phase is advanced, a carrier rocket control system can have large margin, and the design difficulty is reduced.

    本发明公开了一种基于BP神经网络的火箭控制信号校正算法,用于控制系统信号的幅值相位校正,属于运载火箭控制领域,包括步骤如下:步骤一、根据箭体特性,在控制信号考虑的频率范围内给出期望的校正幅值和校正相位特性;步骤二、根据期望的幅值相位特性,生成包含常值和多种频率成分的BP神经网络时域信号训练样本序列;步骤三、设计合适规模的BP神经网络;步骤四、采用样本对BP神经网络进行训练直至收敛到较好效果;步骤五、对训练好的BP网络进行幅值相位特性辨识,校验其效果。本方法原理简洁,弥补了线性校正系统幅相特性存在硬性关系的缺陷,在低频相位超前的同时实现高频段信号的衰减,可以使运载火箭控制系统有较大的裕度,降低设计难度。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Rocket control signal correction algorithm based on BP neural network


    Weitere Titelangaben:

    一种基于BP神经网络的火箭控制信号校正算法


    Beteiligte:
    ZHANG ZHIPING (Autor:in) / XIU GUAN (Autor:in) / KUANG DONGZHENG (Autor:in) / JI YANYU (Autor:in) / MAO CHENGYUAN (Autor:in)

    Erscheinungsdatum :

    2023-07-25


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    B64G Raumfahrt , COSMONAUTICS / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen




    An Improved Non-uniformity Correction Algorithm for IRFPA Based on Neural Network

    Wang, Bing-Jian / Wang, Da-Bao / Lai, Rui et al. | IEEE | 2008


    Sounding-Rocket Wind-Response Correction

    Hoult, Charles P | Online Contents | 2016


    Sounding-Rocket Wind-Response Correction

    Hoult, Charles P. | AIAA | 2016


    Solid carrier rocket injection correction method

    AI ZHIHENG / SHU CHANG / MA CHAO | Europäisches Patentamt | 2021

    Freier Zugriff