论文部分内容阅读
提出了一种将遗传算法(GA)和量子粒子群(QPSO)算法相结合的新优化算法,该算法通过运用GA中的交叉和变异算子操作来优化QPSO算法,提高QPSO的全局搜索能力,克服其易陷入局部极值的缺点。将其应用到PseudoVoigt型布里渊散射谱特征提取,对不同权重比、不同线宽和不同信噪比下的布里渊散射谱进行了参数估计和分析,通过采集不同温度时的布里渊散射谱实验数据,利用GA-QPSO算法对实验数据进行处理。实验结果表明,利用GA-QPSO算法可以提高布里渊散射谱的频移提取精度,当温度为25℃时,频移拟合误差最大为2.18 MHz,且随着温度的升高,平均拟合误差逐渐减小,在80℃时的频移拟合误差最大为0.065 MHz。因此,将该算法用于布里渊散射温度和应变传感系统,在提高空间分辨率、检测精度等方面具有很好的应用前景。
A new optimization algorithm combining genetic algorithm (GA) with quantum particle swarm optimization (QPSO) is proposed. The algorithm optimizes the QPSO algorithm by using the crossover and mutation operators in GA to improve the global search ability of QPSO. Overcome its easy to fall into the local extreme shortcomings. This method is applied to the feature extraction of Brillouin scattering spectrum of PseudoVoigt type. Parameters of Brillouin scattering spectrum under different weighting ratios, different linewidths and different signal-to-noise ratios are estimated and analyzed. Brillouin Scattering spectral experimental data, the use of GA-QPSO algorithm for experimental data processing. Experimental results show that the GA-QPSO algorithm can improve the frequency shift extraction precision of Brillouin scattering spectrum. When the temperature is 25 ℃, the maximum frequency error fitting error is 2.18 MHz, and with the increase of temperature, the average fitting The error decreases gradually, and the maximum error of frequency-shift fitting at 80 ℃ is 0.065 MHz. Therefore, this algorithm is applied to Brillouin scattering temperature and strain sensing system, which has a good application prospect in improving the spatial resolution and detection accuracy.