Filling method for missing data of distributed optical fiber sensor

The invention discloses a method for filling missing data of a distributed optical fiber sensor, and the method comprises the steps: obtaining a data set composed of the data of the distributed optical fiber sensor, carrying out the preprocessing of the data in the data set, carrying out the filling of missing values in the data, and obtaining a preprocessed data set; training a preset autocorrelation neural network model based on the preprocessed data set, dynamically filling missing data to be filled by using the trained neural network model, and outputting the filled data; performing wavelet denoising on the data output from the neural network model to obtain denoised data; and performing local error elimination on the de-noised data to obtain a final missing data filling result. The method has the characteristics of high accuracy and high processing efficiency. 本发明公开了一种用于分布式光纤传感器缺失数据的填补方法，其包括：获取由分布式光纤传感器数据组成的数据集，并对所述数据集中的数据进行预处理，以对其中的缺失值进行填补，得到预处理后的数据集；基于预处理后的数据集对预设的自相关神经网络模型进行训练，利用训练好的神经网络模型对待填充的缺失数据进行动态