基于RS-Chaos-LSSVM的光电稳瞄系统稳定精度检测方法

Stabilization Precision Detecting Methods of Photoelectric Aim-stabilized System Based on RS-Chaos-LSSVM

稳定精度是评价光电稳瞄系统性能最为关键的参数之一.针对传统频谱仪测试方法的局限性及测试中干扰因素的不利影响,提出了基于RS-Chaos-LSSVM的光电稳瞄系统稳定精度检测方法.根据光电稳瞄系统测量原理和基本组成,按照RS理论、Chaos和LSSVM模型,设计了RS-Chaos-LSSVM的光电稳瞄系统稳定精度检测系统.LSSVM把位置稳定环上的陀螺仪反馈的实际转台角度,在以DSP为核心的信号检测处理单元中进行稳定精度检测处理.RS理论可以对进入LSSVM模型样本进行属性值约简,以提高实时快速性.Chaos优化LSSVM模型的参数,旨在提高其控制精度.最后对某机载光电稳瞄系统的定频和随机动态稳定精度分别以RS-LSSVM与RS-Chaos-LSSVM的稳定精度检测进行了测试对比.实验仿真数据表明:利用RS-Chaos-LSSVM检测模型在此稳定精度检测时能够有效剔除干扰数据,比Chaos-LSSVM方法的实时性提高了约50%.因此基于RS-Chaos-LSSVM的稳定精度检测方法,具有更高的可靠性和准确性,达到了期望目标.

Stabilization precision is one of the key parameters to evaluate photoelectric aim-stabilization system performance. Aiming at the limitations of the traditional test methods, the adverse effect and the interference factors of test, a new stabilization precision testing method for photoelectric aim-stabilization system based on RS-Chaos-LSSVM is provided. Firstly, the fundamental of the photoelectric aim- stabilization system is explored. And then the stabilization precision testing method for photoelectric aim -stabilization system based on RS-Chaos-LSSVM is achieved using LSSVM and RS theory. The actual turntable angle from the position stable circle is put into the detector, LSSVM model detects and processes stabilization precision in the DSP detector processing unit. The RS theory is to reduce the attributive value of the sample in the LSSVM model, which improves the real-time performance. Then, LSSVM model is trained by using the reduction rules, and its parameters are optimized by using chaos theory for moer accurate control. The random vibration and fixed-frequency dynamic stabilization accuracy of the photoelectric aim-stabilization system are tested. The comparison between the traditional test and RS-Chaos-LSSVM electrical test is made. The tests show that the later way is better than the former. The operation speed and accuracy of RS- Chaos-LSSVM electrical test can meet the requirements of the detection system. It effectively eliminates noise data, improves the testing precision and achieves the expected objectives.