A Method for Analysis and Prediction of On-street Parking Demand in Urban Central Area

Abstract

The invention discloses an urban center area roadside parking demand analysis prediction method. The urban center area roadside parking demand analysis prediction method includes the steps: acquiringthe related data of roadside parking facilities in the center area for the last three months, establishing an artificial neural network, performing variation operation on the parameter Beta of the artificial neural network, and obtaining an improved artificial neural network; taking the historical data of the target data to be predicted as the original data, after performing data longitudinal comparison processing and normalization, inputting the processed data as the training sample into the improved artificial neural network; selecting one group of corresponding improved artificial neural network models having the minimum verification sample error as a training model, and inputting the training model into the data to be predicted for prediction, thus obtaining the parking demand Poissondistribution parameter of the day to be predicted; and applying the urban roadside parking demand model which is obtained through fitting to a practical parking lot to predict the roadside parking demand time. The urban center area roadside parking demand analysis prediction method has the advantages of providing a feasible scheme for analysis and prediction of the urban center area roadside parking demand, and having preferable practicality.

Description

technical field [0001] The invention belongs to the technical field of analysis and prediction of urban roadside parking demand, and relates to a method for analysis and prediction of roadside parking demand in a city center area. Background technique [0002] In recent years, with the rapid development of urban economy and the improvement of people's living standards, the number of private cars has also increased dramatically. However, the growth rate of urban parking lot planning and construction lags behind the increase in car ownership, and the supply of parking spaces is seriously lacking. In addition, some cities lack systematic parking facility planning, and these problems have led to increasingly serious problems of parking difficulties and disordered parking on the city's roadside. The problem of on-street parking in the urban central area is particularly serious. How to quickly and accurately grasp the basic characteristics of on-street parking demand in the urban...

AI Technical Summary

Problems solved by technology

These studies show that the solution to the on-street parking problem mainly focuses on guiding parking demand and improving traffic efficiency, but there are the following shortcomings: the temporal and spatial distribution characteristics of urban on-street parking are not explored, and the rules of vehicle arrival and departure are not fully considered. , predict the possible distribution of parking time, and formulate different parking strategies for different time periods and regions, so as to achieve a reasonable diversion of urban parking demand and alleviate urban parking problems

[0004] The traditional Elman artificial neural network uses the momentum gradient descent method to adjust the network weights and weights. When the amount of input data is large, the Elman artificial neural network is easy to enter the local optimal state, and the convergence speed is significantly reduced.

Images