改进GWO-RBF神经网络的高频地波雷达海杂波预测模型

Prediction model of high frequency surfacewave radar sea clutter with improved GWO-RBF neural network

高频地波雷达的海上目标探测能力与海杂波的抑制效果息息相关,而海杂波的精确预测又是其有效抑制的重要前提.为实现海杂波的精确预测,提出一种基于改进灰狼算法优化RBF神经网络的预测模型(MGWO-RBF),为解决灰狼优化算法收敛速度慢和易陷入局部最优的缺点,提出一种灰狼狼群分工搜索的策略,使得整个迭代过程中狼群始终兼具大范围搜索和局部探索能力.与经典的粒子群算法、标准灰狼优化算法和动态权重灰狼优化算法对比得出结论:改进的灰狼算法在收敛速度和精度上都有明显提升,MGWO-RBF预测模型对海杂波预测的精度达到96.23%,取得了较好的预测效果.

The sea target detection capability of high frequency surface wave radar is closely related to the suppression effect of sea clutter,and the accurate prediction of sea clutter is an important premise for its effective suppression.In order to realize the accurate prediction of sea clutter,a prediction model based on RBF neural network optimized by improved gray wolf optimization algorithm(MGWO-RBF)is proposed.In order to overcome the two shortcomings of grey wolf optimization algorithm,such as slow convergence speed and tendency to fall into the local optimum,a strategy of gray wolf group division search is proposed,which makes the wolves have the ability of both large-scale search and local exploration in the whole iterative process.Compared with the classical particle swarm optimization algorithm,the standard gray wolf optimization algorithm and the dynamic weight gray wolf optimization algorithm,the improved gray wolf algorithm has a significant improvement in convergence speed and accuracy.The accuracy of MGWO-RBF prediction model for sea clutter prediction reaches 96.23%,achieving a better prediction effect.