The invention discloses an urban short-term traffic flow prediction method based on a GCN-IPSO-LSTM combination model, and the method comprises the steps: firstly obtaining urban traffic flow data, carrying out the data preprocessing and data segmentation, and dividing the data into a training set and a test set; then optimizing parameters of the LSTM model by using an improved particle swarm optimization (IPSO) algorithm; then sequentially constructing a graph convolution GCN model and an LSTM model after IPSO optimization, and respectively extracting spatial features and time features in the traffic flow data; and finally, training the constructed CGN-IPSO-LSTM combination model, and carrying out prediction on the test set. According to the invention, the defect of insufficient spatial-temporal feature extraction of the traffic flow by a single prediction model is overcome; according to the method, an LSTM model is optimized, a nonlinear inertia weight, a self-adjusting learning factor and a self-adaptive mutation operation are introduced into a traditional particle swarm algorithm to improve the particle swarm algorithm, the parameters of the LSTM model are optimized by using the improved particle swarm algorithm, an optimal parameter combination can be found more quickly and more accurately, and the generalization ability of the model and the prediction accuracy are further improved.
本发明公开了一种基于GCN‑IPSO‑LSTM组合模型的城市短时交通流预测方法,首先获取城市交通流数据并进行数据预处理与数据分割,将数据划分为训练集和测试集;然后用改进的粒子群优化算法IPSO优化LSTM模型的参数;然后依次构建图卷积GCN模型和IPSO优化后的LSTM模型,分别提取交通流数据中的空间特征和时间特征;最后训练上述所构建的CGN‑IPSO‑LSTM组合模型,并对测试集进行预测。本发明弥补了单一预测模型对交通流的时空特征提取不足的缺点;并在传统的粒子群算法中引入非线性惯性权重、自调整学习因子和自适应变异操作来改进粒子群算法,使用改进后的粒子群算法优化LSTM模型的参数,能更快、更准确地找到最优的参数组合,进一步提高了模型的泛化能力和预测的准确性。
Urban short-term traffic flow prediction method based on GCN-IPSO-LSTM combined model
基于GCN-IPSO-LSTM组合模型的城市短时交通流预测方法
2023-09-08
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G06Q Datenverarbeitungssysteme oder -verfahren, besonders angepasst an verwaltungstechnische, geschäftliche, finanzielle oder betriebswirtschaftliche Zwecke, sowie an geschäftsbezogene Überwachungs- oder Voraussagezwecke , DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES |
Short-term traffic flow prediction method based on Spearman-LSTM model
European Patent Office | 2024
|Short-term traffic flow prediction method based on Conv1D-LSTM model
European Patent Office | 2023
|Short-Term Traffic Flow Prediction: Using LSTM
IEEE | 2020
|Short-term traffic flow prediction method based on improved LSTM
European Patent Office | 2021
|