基于改进QPSO优化SVR的某电源组合故障预测研究

Research on Fault Prediction of Power Supply Based on SVR Optimized with Improved QPSO

针对电源组合的故障特点,提出了一种基于改进QPSO优化SVR的故障预测方法;文中首先对QPSO算法进行了介绍;然后对支持向量回归机(Support Vect or Regression,SVR)性能影响因素进行了分析,并给出了基于改进QPSO优化SVR参数的算法步骤;最后以制导雷达波束系统中的某电源组合为例进行了仿真分析,预测结果表明,同QPSO算法相比该预测方法误差更小,达到了预期效果。

Aiming at the character of power supply fault prediction, a prediction method based on SVR optimized with improved QPSO is put forward. Firstly, the QPSO algorithm is introduced in the paper; Then the performance impact factors of SVR are analyzed, and the steps of SVR parameters optimized with improved QPSO are given; Finally, taking a power supply of guidance radar beam control system as an example to simulate, the result shows that the error of this method is lower than QPSO algorithm, achieving the expected effect.