【摘要】 研究我国南海地区（北纬0°到25°，东经105°到120°）电离层F层临界频率月中值变化特点及利用人工神经网络预报月中值的技术．文章首先给出这个海区上空F2层临界频率月中值随太阳活动水平、地理纬度、季节和昼夜变化规律．结果指出电离层临界频率随纬度分布除出现驼峰现象外，在4和10月份还存在大范围的电离层气泡区域。在这个区域电离层密度出现中间稀周围高特点；电离层临界频率的分形分析得到其相关维为2．7。其次，我们选用5个因子作为网络输入，用电离层观测资料训练网络，建立了我国南海地区电离层模式。利用这个模式预测的电离层临界频率与实际测量的符合较好，可以满足应用要求。更多还原

【Abstract】 In this paper, the ionospheric monthly average characteristics of the F region in the mouth China Sea (including from 0°-25°N, 105°-120°E)are studied. A regional ionospheric model by Artificial Neural Network(ANN) is given. First, variations of the critical frequency with solar active level, latitude, season and diurnal are discussed. Results showed that there is a large hobble area during April and October in this region. From analyses of the ionospheric critical frequency, their correlation dimensions are about 2. 7. Next , five factors are used as the input parameters of this ANN.The comparisons of observed ionosphetic critical frequencies with predicted ones are in very goood agreement.更多还原