A passenger transport hub area resident number change trend short-time prediction method based on a kNN algorithm

Abstract

The invention discloses a passenger transport hub area resident number change trend short-time prediction method based on a kNN algorithm. The method comprises the following steps of obtaining passenger transport hub area resident situation data in real time through a detection system; according to the date characteristics of the to-be-predicted day, selecting m historical samples similar to the to-be-predicted day as predicted sample spaces; preprocessing abnormal data and noise in the historical sample; determining a characteristic space of the to-be-predicted day and the historical sample corresponding to the to-be-predicted time period, calculating an increment ratio standard deviation of the data of the historical sample and the to-be-predicted day on the characteristic space, and finding out k-day data with the minimum increment ratio standard deviation as k adjacent samples; calculating an increment ratio coefficient of k adjacent samples, and predicting the change trend of thenumber of resident people in the region according to the increment ratio coefficient; and calculating a short-time prediction value of the number of resident people in the region by taking the currentnumber of resident people in the region as a reference. According to the method, the change trend of the number of resident people in the short-time area can be accurately predicted by using the historical data, so that a high-precision short-time area resident number prediction result is obtained based on the calculation of the number of resident people in the current area. The method is suitable for the intelligent traffic field.

Description

technical field [0001] The invention relates to the field of intelligent transportation, and more specifically, relates to a short-term prediction method for the change trend of the number of residents in a passenger transport hub area based on a kNN algorithm. Background technique [0002] Passenger transport hub, as a complex of transportation station facilities, can provide passengers with a variety of services to meet the various needs of passengers, and is committed to making passengers feel convenient and comfortable. The number of resident passengers in the passenger transport hub directly reflects the crowd density and degree of congestion in the hub, and is one of the most important reference indicators for the passenger flow organization plan and the distribution management plan. Accurate short-term prediction of the number of resident passengers is essential for emergency implementation of grading Pre-plan and security management are of great significance. [000...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem that the number of passengers staying in a passenger transport hub is affected by various factors, and the relationship between various factors is intricate, and it is difficult to accurately predict the change of the number of passengers staying in the passenger transport hub, the present invention provides a change in the number of passengers staying in a passenger transport hub area based on the kNN algorithm The trend short-term prediction method, which selects historical samples similar to the feature space change trend of the day to be predicted as adjacent samples, and performs short-term prediction of the change trend of the number of residents in the area on the day to be predicted, which eliminates the The impact of the difference between historical samples and the data level of the day to be predicted ensures the accuracy of short-term predictions, which is of great significance for the emergency implementation of graded plans and security management

Images