基于Chaos-MEA-Elman的电离层临界频率f0F2预测

针对飞机高频通信中通信频率选择不当导致信号衰落等问题,提出一种基于混沌理论和思维进化算法优化的Elman神经网络相结合的临界频率f0F2预测方法。分析和验证f0F2时间序列的混沌特性;采用混沌理论重构技术重建f0F2时间序列的相空间,并根据相空间结构确定Elman神经网络各层节点个数;利用思维进化算法优化Elman神经网络的初始权值和阈值。数值和实验分析表明,Chaos-MEA-Elman算法比Elman神经网络对电离层临界频率f0F2的预测精确度更高,为预测飞机最佳通信频率提供新的方法。

PREDICTION OF IONOSPHERIC CRITICAL FREQUENCY f 0F 2 BASED ON CHAOS-MEA-ELMAN

Aiming at the problem of signal fading caused by improper communication frequency selection in aircraft high-frequency communication,this paper proposes a prediction method of critical frequency f 0F 2 based on chaos theory and Elman neural network optimized by mind evolutionary algorithm.It analyzed and verified the chaotic characteristics of f 0F 2 time series.Then,the chaotic theory reconstruction technique was used to reconstruct the phase space of the f 0F 2 time series,and the number of nodes in each layer of the Elman neural network was determined according to the phase space structure.Finally,the initial weights and thresholds of Elman neural network were optimized by using mind evolutionary algorithm.Numerical and experimental results show that the Chaos-MEA-Elman algorithm has higher prediction accuracy of the ionospheric critical frequency f 0F 2 than the Elman neural network,and it provides a new method for predicting the optimal communication frequency of aircraft.