A Short-term Traffic Flow Prediction Method Based on Combinatorial Logic

Abstract

The present invention relates to a short-term traffic flow prediction method based on combinational logic, comprising: collecting traffic flow data to obtain a first data set; preprocessing the first data set to obtain a second data set; using the second data The set trains the random forest regression prediction model and the BP neural network prediction model respectively, and obtains the prediction result of the random forest regression prediction model and the prediction result of the BP neural network prediction model; the prediction result of the random forest regression prediction model and the described The prediction results of the BP neural network prediction model are fused to form a final model; the second data set is input into the final model to obtain the final prediction result. In the embodiment of the present invention, the final model is obtained by secondary modeling of the prediction results of the random forest regression prediction model and the prediction results of the BP neural network prediction model, and more accurate and more efficient short-term traffic flow prediction is realized with less data .

Description

technical field [0001] The invention belongs to the field of ITS (Intelligent Transportation System), and in particular relates to a short-term traffic flow prediction method based on combinational logic. Background technique [0002] Under the existing traffic conditions, traffic congestion and inefficiency have become a worldwide problem. The importance of traffic issues is not only reflected in people's actions, but also has a decisive impact on urban construction and planning management at the macro level. However, the traffic congestion problem can be effectively alleviated through traffic flow forecasting and intelligent traffic dispatching system. However, short-term traffic flow is highly uncertain and non-linear, and may be affected by weather, traffic control, and emergencies. Simple linear models cannot reflect the changing laws of traffic flow, so more complex models are required. Make predictions. [0003] There is a fuzzy adaptive short-term traffic flow pre...

AI Technical Summary

Problems solved by technology

Deep learning does have great advantages in feature extraction, but this algorithm requires a large amount of data for analysis, but in traffic flow prediction, a large amount of data means data with a large time span, and a lot of data is very important for short-term traffic flow prediction. It is outdated and does not work very well, and the hardware requirements for deep learning training are high, and the time efficiency is not ideal

[0004] At present, the research status of traffic flow at home and abroad lies in the advantages and disadvantages of different models. In different traffic prediction scenarios, combined prediction models are adopted. The existing traffic flow based on the combination of support vector machine (SVM) and BP neural network Forecasting algorithm, this method requires a large number of parameters to be adjusted, and the training period is long, and the prediction accuracy is low

Images