Design And Implementation Of Large-scale Intelligent Parking Management System

The number of parking lots in the nation has been rising recently,but parking lot information management technology is developing slowly.As a result,there is a delay between the time users see parking lot information and the time they actually find parking spaces,which could lead to users arriving at the target parking lot when there are no spaces available.A solution for an intelligent parking lot management system based on the ARIMALSTM algorithm is suggested to address the issues of low parking spot utilization,poor parking lot management,and the time difference between users and information.The paper aims to design and construct an intelligent parking lot management system that is reasonable.The specific work of the paper is as follows:1、The demand analysis for management staff and users was carried out,the system framework design was finished,and the development of a parking management system at home and abroad was examined.The system’s various functions and the related database are then carefully designed and developed in accordance with the requirements analysis.With features like user log-in,parking car evaluation,parking space prediction,parking space reservation,order management,and user management,the system is made up of two components: a We Chat mini program and a web terminal.Users can rent their own parking spaces or reserve other people’s parking spaces in accordance with their own needs to increase the utilization rate of vacant parking spaces.An optimized parking reservation mode is also suggested to address the needs of various users.2 、 The parking lot management system’s parking place prediction feature was investigated.Prior to training the classic prediction models ARIMA and LSTM,the time series of parking spots is first smoothed using data pretreatment technology.The experimental findings reveal that the root means square errors of ARIMA and LSTM are572.74 and 157.02,respectively.The prediction accuracy of LSTM is better than ARIMA,but the predicted value is still offset from the true value.There is a significant difference between the predicted value of ARIMA and the true value.3、It is suggested to use an enhanced algorithm for parking spot number prediction based on the ARIMA-LSTM combination model.First,the LSTM approach is used to train and forecast the residual of ARIMA and nonlinear factors such as holidays and significant events at shopping malls after the linear factors in the data are filtered and the ARIMA method is used to predict the number of parking spots.According to experiments,the model is more reliable and its root mean square errors are 465.46,73.94,and 49.74 less than those of ARIMA,BP,and LSTM,respectively.4、Build test scenarios to assess the performance and functionality of the system.The test results demonstrate that the system can run steadily and generally achieves the desired aims.