New method for predicting remaining parking spaces

• Main Authors: ZHANG KAI, XI GUOLONG, LAI YONGSHOU, ZHANG ZHIQIANG, HOU SHUJIE, WU XICHEN, MA CHANGXI, HE QUANHONG, HUANG XIAOTING, MENG WEI, JIN KECHEN
• Format: Patent
• Language: Chinese; English
• Published: 29.08.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS; SIGNALLING; TRAFFIC CONTROL SYSTEMS

The invention discloses a new method for predicting remaining parking spaces, and belongs to the technical field of intelligent parking. Comprising the following steps: decomposing parking original data into a plurality of IMFs with different frequencies by using a CEEMDAN decomposition algorithm, and decomposing fluctuation characteristics and trends with different scales from the original data step by step to eliminate noise; pCA dimension reduction processing is carried out on the decomposed data, abstract advanced features are extracted, and correlation and redundancy of different time sequences after CEEMDAN decomposition are eliminated; and deep learning prediction is carried out by adopting a GRU neural network. According to the method, thoughts of three disciplines of signal processing, statistics and deep learning are integrated, primary decomposition of original parking data is carried out based on CEEMDAN, noise reduction and feature reconstruction are carried out on a decomposition sequence by uti