The invention discloses an inland ship AIS missing data fitting compensation method, and the method comprises the steps: building an encoder model with hidden layer neurons being mutually orthogonal based on an existing ship AIS training data set, optimizing the weights and deviations of the neurons through a PSO algorithm, and adding the weights and deviations into hidden layers of an encoder network model one by one; according to the method, the feature variables of the AIS historical data can be extracted to the maximum extent, and the AIS missing data can be further predicted according to the feature variables. Compared with the prior art, the method has the characteristics of simple model structure and high prediction precision.
本发明公开了内河船舶AIS缺失数据拟合补偿方法,本发明基于已有船舶AIS训练数据集构建隐含层神经元相互正交的编码器模型,利用PSO算法优化神经元的权重和偏差,并逐个添加到编码器网络模型的隐含层中;本方法可以最大程度上提取AIS历史数据的特征变量,并根据特征变量进一步预测AIS缺失数据。与现有技术相比,本发明具有模型结构简单、预测精度高的特点。
Inland ship AIS missing data fitting compensation method
内河船舶AIS缺失数据拟合补偿方法
2023-07-14
Patent
Electronic Resource
Chinese
IPC: | G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
Engineering Index Backfile | 1961