Short-term traffic flow prediction model construction method and prediction method

• Main Authors: HU YUANJIAO, FENG SHAOWEI, LI WEI, CAO LEI, SUN CHAOYUN, PEI LILI, HAO XUELI
• Format: Patent
• Language: Chinese; English
• Published: 03.11.2020
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS; SIGNALLING; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR; TRAFFIC CONTROL SYSTEMS

The invention discloses a short-time traffic flow prediction model construction method and prediction method, and the method comprises the steps: carrying out the clustering of traffic flow data in aperiod of time, fully extracting data information, inputting the data information to a GRU neural network, taking a short-time traffic flow mode type as an output, and carrying out the training of a GRU neural network model; obtaining a short-time traffic flow prediction model after the training is completed, and achieving the prediction of the short-time traffic flow. According to the method, thehistorical short-time traffic flow data are clustered by adopting the KMeans clustering algorithm aiming at the significant influence of training set selection of the neural network on the predictionresult of the short-time traffic flow, so that the prediction is carried out in a targeted manner, and the accuracy of the prediction result is improved. 本发明公开了一种短时交通流量预测模型构建方法及预测方法，针对一段时间内的交通流量数据进行聚类，充分提取数据信息输入到GRU神经网络，以短时交通流