碳化硅/碳吸波材料制备及雷达快速精准识别技术研究

Preparation of silicon carbide/carbon absorbers and research on radar fast and accurate identification technology

为预测2 h内的雷暴、大雨、冰雹等高强度、快速变化的恶劣天气现象,减少破坏性天气带来的影响,提出基于多元逻辑回归算法的碳化硅/碳吸波材料雷达恶劣天气快速识别方法。采用最大似然估计方法求解多元逻辑回归模型,利用训练数据的似然函数,同时使用多元逻辑回归算法来过滤雷达杂波,以提高恶劣天气快速识别预警效果。实验结果表明,当样本数量不同时,多元逻辑回归算法模型的准确率和损失率随迭代次数的增加而显著变化。

To predict high-intensity and rapidly changing adverse weather phenomena such as thunderstorms,heavy rain,and hail within 2 hours and reduce the impact of destructive weather,a rapid identification method for adverse weather using silicon carbide/carbon absorbing material radar based on multiple logistic regression algo⁃rithm was proposed.The maximum likelihood estimation method was used to solve the multiple logistic regression model,the likelihood function of training data was utilized,and the multiple logistic regression algorithm was used to filter radar clutter to improve the rapid recognition and warning effect of severe weather.The experimental results showed that when the number of samples was different,the accuracy and loss rate of the multiple logistic regression algorithm model significantly changed with the increase of iteration times.