Closed parking lot parking demand prediction method based on improved GMDH algorithm

Abstract

The invention discloses a closed parking lot parking demand prediction method based on an improved GMDH algorithm. According to the method, the GMDH algorithm is used to train the entering traffic flow of a closed parking lot, the GMDH algorithm is improved through combining with the idea of Ensemble Learning for solving a problem that the modeling generalization capability of the GMDH algorithm is poor in a training process, and the generalization capability of a GMDH model is improved; and the improved algorithm is applied to the construction of an entrance parking demand prediction model ofthe closed parking lot. According to the method, the dependence on historical data is reduced, the data cost is reduced, the prediction precision is high, and the parking demand of the closed parkinglot can be rapidly and effectively predicted. In the later period, parking lot parking resource sharing and scheduling with different characteristics in a regional mode can be achieved based on the user balance theory in combination with an intelligent traffic guidance system, and a great practical value is achieved for the construction of smart cities.

Description

technical field [0001] The invention belongs to the field of parking planning in traffic planning. Aiming at the problems of unbalanced supply of parking resources and shortage of parking resources caused by changes in parking demand in urban parking lots on working days or non-working days and in different time periods of each day, a method based on Improved GMDH Algorithm for Parking Demand Prediction in Enclosed Parking Lot. Background technique [0002] Under the national "new infrastructure" wave, the new transportation infrastructure has thus entered the "fast lane" of development. Intelligent transportation is an important part of smart cities, and the parking industry belongs to the level of urban transportation. With regard to the difficulty of parking in cities, after several years of development, shared parking and intelligent consumption habits have been formed in many cities in China. Urban traffic big data management drives urban transformation and upgrading. ...

AI Technical Summary

Problems solved by technology

[0004] In order to realize accurate real-time prediction of parking demand in parking lots, the present invention provides a parking demand prediction method for closed parking lots based on the improved GMDH algorithm, which overcomes the deficiencies in the prior art, improves prediction accuracy, reduces dependence on historical data, reduces Data cost, and through improvement to strengthen the generalization ability of the model, it can predict the incoming traffic flow of the closed parking lot in real time, obtain the real-time parking demand of the parking lot, and solve the imbalance between the supply of parking facilities and the parking demand, which leads to Parking contradictions are becoming more and more intense, which leads to problems such as the decline in the operating efficiency of the traffic system and the reduction of road traffic capacity. It is convenient to realize regional parking resource sharing and scheduling of parking lots with different characteristics based on the user equilibrium theory and combined with the intelligent traffic guidance system in the later stage.

Images