Method for predicting parking space occupation of parking lot and related equipment

Abstract

The invention relates to the field of parking lot management, in particular to a method for predicting parking space occupation of a parking lot and related equipment. The method comprises the steps that area type identification information and historical data of a parking lot are acquired, and the historical data comprise corresponding parking space occupation data and weather data in a preset time period; determining a target mode corresponding to the long-term prediction model based on the identification information, the historical data and the date of the day; inputting the historical data and the current day weather into a short-term prediction model and inputting the historical data and the current day weather into a long-term prediction model based on the target mode for prediction processing to obtain short-term prediction data and long-term prediction data respectively; the short-term prediction data and the long-term prediction data are subjected to junction point processing based on the target prediction time, the prediction result of the day is obtained, and the time of the long-term prediction data is larger than or equal to the target prediction time and larger than the time of the short-term prediction data. The parking space occupation of the parking lot can be predicted in real time, and the parking space occupation prediction precision of the parking lot is further improved.

Description

Technical Field

[0001] The present application relates to the field of parking lot management, and in particular to a method for predicting parking lot occupancy and related equipment. Background Art

[0002] The rapid development of parking management systems has greatly improved the efficiency of urban traffic management and resource utilization. With the acceleration of urbanization, the number of vehicles continues to increase, and the problem of parking difficulties has become increasingly prominent. In order to effectively alleviate this problem, many cities have begun to introduce intelligent parking management systems to optimize parking space allocation through data analysis and prediction, improve parking space utilization, and reduce the time car owners spend looking for parking spaces, thereby improving overall traffic flow.

[0003] However, there are some shortcomings in the existing parking space occupancy prediction methods. The first is the low prediction accuracy, especially the inconsistent prediction accuracy over short and long time spans, which leads to low reliability in practical applications. Secondly, most of the existing prediction methods rely on a single data source and fail to make full use of multi-dimensional data for comprehensive analysis, which limits the comprehensiveness and accuracy of the prediction results. These problems make the existing parking space occupancy prediction methods have certain limitations in practical applications and cannot meet the needs of high precision and real-time performance. Summary of the invention

[0004] The purpose of this application is to overcome the above-mentioned technical problems and provide a method and related equipment for predicting parking space occupancy in a parking lot, which can predict the parking space occupancy in a parking lot in real time and further improve the accuracy of parking space occupancy prediction.

[0005] In a first aspect, an embodiment of the present application discloses a method for predicting parking space occupancy in a parking lot, using the following scheme: A method for predicting parking space occupancy in a parking lot, comprising: obtaining area type identification information and historical data of the parking lot, the historical data including corresponding parking space occupancy data and weather data within a preset time period; determining a target mode corresponding to a long-term prediction model based on the identification information area type, the historical data and the date of the day; inputting the historical data and the weather of the day into a short-term prediction model and inputting the target mode into the long-term prediction model for prediction processing, respectively obtaining short-term prediction data and long-term prediction data; performing intersection processing on the short-term prediction data and the long-term prediction data based on a target prediction time, obtaining a prediction result for the day, wherein the time of the long-term prediction data ≥ the target prediction time > the time of the short-term prediction data.

[0006] By adopting the above technical solution, it is possible to achieve accurate prediction of parking space occupancy in parking lots. Specifically: obtaining the identification information and historical data of the parking lot can ensure that subsequent predictions are targeted and improve the accuracy of predictions. Based on the identification information, historical data and the date of the day, the target mode corresponding to the long-term prediction model is determined, so that the prediction model is more adapted to the current specific conditions and the prediction accuracy is further improved. The historical data and the current weather are input into the short-term prediction model and the long-term prediction model for prediction processing, and the short-term prediction data and the long-term prediction data are obtained respectively. Combining the advantages of the two prediction models, the comprehensiveness and reliability of the prediction are improved. The intersection of the short-term prediction data and the long-term prediction data is processed based on the target prediction time, and the final prediction result of the day is obtained. This processing method effectively solves the connection problem between short-term and long-term predictions, avoids the prediction fault phenomenon, and makes the prediction result smoother and more accurate. In this way, the method can effectively predict the parking space occupancy of the parking lot in the future period of time, and provide a scientific basis for management decisions.

[0007] Optionally, the short-term prediction model is used to predict the parking space occupancy data for the next 6 hours; the long-term prediction model is used to predict the parking space occupancy data for the next 24 hours; when the short-term prediction data is , which corresponds to the parking space occupancy data for each hour within 6 hours. The long-term prediction data is , corresponding to the parking space occupancy data for each hour within 24 hours, the prediction result for the day is ; Among them, when the intersection point corresponds to 6-7, the corresponding processing algorithm is , The corresponding parking space occupancy data is 6 o'clock. This corresponds to the parking space occupancy data at 7 o'clock.

[0008] By adopting the above technical solution, it is possible to accurately predict the parking space occupancy in the parking lot within the next 6 hours and 24 hours. Specifically: the short-term prediction model is used to predict the parking space occupancy data within the next 6 hours, which helps to respond to changes in parking demand in a short period of time in a timely manner and improve resource utilization. The long-term prediction model is used to predict the parking space occupancy data within the next 24 hours, which is conducive to planning the allocation of parking spaces in a long period of time in advance and reducing the waste of resources caused by sudden demand. When the short-term prediction data and the long-term prediction data are processed at the intersection between 6 and 7 o'clock, a specific weighted average algorithm is used to ensure the smooth transition of the prediction results between the two time intervals, thereby improving the continuity and accuracy of the prediction. Especially for the two key time nodes of 6 o'clock and 7 o'clock, this processing method can more accurately reflect the changing trend of actual parking space occupancy, thereby improving the reliability of the overall prediction.

[0009] Optionally, the short-term prediction model uses the parking space occupancy data of the previous 24 hours of the current date and the corresponding weather data, as well as the current day's weather, to predict the parking space occupancy in the next 6 hours; The long-term prediction model is configured with four prediction modes corresponding to the regional types, which are as follows: Prediction mode 1, using the parking space occupancy data of the previous 72 hours and the corresponding weather data, as well as the weather of the current day, to predict the parking space occupancy for 24 hours of the current day; Prediction mode 2, using the parking space occupancy data and the corresponding weather data of the previous 24 hours, the parking space occupancy data and the corresponding weather data of the same day last week for 24 hours, and the weather of the current day, to predict the parking space occupancy of the current day for 24 hours; Prediction mode three, using the parking space occupancy data of 48 hours before the working day or holiday day and the corresponding weather data, as well as the weather of the day, to predict the parking space occupancy of 24 hours on the day; Prediction mode 4, using the parking space occupancy data and the corresponding weather data for 24 hours before a working day or a holiday, the parking space occupancy data and the corresponding weather data for 24 hours on the same day last week, and the weather of the day, to predict the parking space occupancy for 24 hours on the day; The target mode is one of prediction mode one, prediction mode two, prediction mode three and prediction mode four.

[0010] By adopting the above technical solution, the accuracy and reliability of parking space occupancy prediction can be improved. Specifically: by using the parking space occupancy data and the corresponding weather conditions of the 24 hours before the current date and the weather of the current day, the recent trend of change can be captured more accurately, thereby improving the prediction accuracy of parking space occupancy in the next 6 hours. Four different prediction modes are set to make long-term predictions based on the parking lot's identification information, historical data, and the corresponding mode matching the current date, thereby effectively improving the overall performance of parking space occupancy prediction.

[0011] Optionally, when the target mode is prediction mode three, determine whether the current date is a weekday or a holiday; if the current date is a weekday, select the parking space occupancy data for the first 48 hours on the weekday from the historical data, and input it together with the current weather into the long-term prediction model to predict the parking space occupancy data for 24 hours on the day; if the current date is a holiday, select the parking space occupancy data for the first 48 hours before the holiday from the historical data, and input it together with the current weather into the long-term prediction model to predict the parking space occupancy data for 24 hours on the day.

[0012] By adopting the above technical solution, the accuracy of the long-term prediction model can be improved. Specifically: when the target mode is prediction mode three, by judging whether the current date is a weekday or a holiday, the corresponding historical data and the current weather are selected for prediction, so that the prediction is closer to the actual car demand. For weekdays, the weekday parking space occupancy data of the previous 48 hours are selected together with the current weather for prediction, which can effectively capture the changing patterns of parking behavior during weekdays and improve the prediction accuracy. For holidays, the holiday parking space occupancy data of the previous 48 hours are selected together with the current weather for prediction, which can also better reflect the parking situation on holidays and reduce errors.

[0013] Optionally, when the target mode is prediction mode four, determine whether the current date is a weekday or a holiday; if the current date is a weekday, select the parking space occupancy data for the first 24 hours on the weekday and the parking space occupancy data for 24 hours on the same day last week from the historical data, and input them into the long-term prediction model together with the current weather to predict the parking space occupancy data for the current 24 hours; if the current date is a holiday, select the parking space occupancy data for the first 24 hours before the holiday and the parking space occupancy data for 24 hours on the same day last week from the historical data, and input them into the long-term prediction model together with the current weather to predict the parking space occupancy data for the current 24 hours.

[0014] By adopting the above technical solution, the most appropriate combination of historical data can be selected for different date types (working days or holidays), thereby improving the accuracy of the long-term prediction model. Specifically: when the target mode is prediction mode 4 and the day is a working day, the data of the last 24 hours and the same day last week are selected from the historical data, and together with the current weather, it can more accurately reflect the parking demand trend of the current working day and reduce the impact of factors such as holidays. If the day is a holiday, the data of the last 24 hours plus the same holiday last week are selected and predicted together with the current weather, which can help capture the parking space occupancy pattern unique to holidays and further improve the prediction accuracy.

[0015] Optionally, it also includes: establishing a mapping table between the identification information, date and multiple modes in the long-term prediction model, wherein the target mode corresponding to the long-term prediction model is determined from the mapping table based on the identification information and the date of the day.

[0016] By adopting the above technical solution, it is possible to quickly and accurately match the target mode of the long-term prediction model that best suits the current parking lot conditions, thereby improving the prediction accuracy and reliability. Specifically: a mapping table is established between identification information, date, and multiple modes in the long-term prediction model, so that the system can automatically select the most appropriate prediction mode based on real-time data. The establishment of the mapping table ensures that the most appropriate historical data can be used for prediction under different conditions (such as weekdays and weekends), thereby improving the accuracy of the prediction results. This method avoids the errors that may be caused by manually selecting the prediction mode, and improves the system's degree of automation and user experience.

[0017] Optionally, before obtaining the historical data and area type of the parking lot, it also includes: collecting historical data of different parking lots; processing the historical data by time period and corresponding parking space occupancy data as a target training set; inputting the target training set into the original short-term prediction model and the original long-term prediction model for training to obtain the short-term prediction model and the long-term prediction model.

[0018] By adopting the above technical solution, it is possible to effectively collect and organize historical data from multiple parking lots and convert them into target training sets that can be used to train short-term prediction models and long-term prediction models. This not only improves the generalization ability of the model, but also makes the short-term prediction data and long-term prediction data generated in the end more accurate and reliable. Specifically: By comprehensively collecting and processing data from different parking lots, the influence of various factors on parking space occupancy can be better captured, thereby improving prediction accuracy. Using multi-source data for model training helps the model learn the characteristics of different types of parking lots, so that it can also perform well when facing new parking scenarios. Based on more accurate prediction results, managers can reasonably arrange parking space resources, reduce vacancy rates, and improve user experience.

[0019] In a second aspect, an embodiment of the present application discloses a system for predicting parking space occupancy in a parking lot, which adopts the following scheme: A method for predicting parking space occupancy in a parking lot, comprising: an acquisition module for acquiring identification information area type and historical data of the parking lot, wherein the historical data includes corresponding parking space occupancy data and weather data within a preset time period; a mode determination module for determining a target mode corresponding to a long-term prediction model based on the identification information, the historical data and the date of the day; a prediction module for inputting the historical data and current weather into a short-term prediction model and inputting the target mode into the long-term prediction model for prediction processing, thereby obtaining short-term prediction data and long-term prediction data respectively; a processing module for performing intersection processing on the short-term prediction data and the long-term prediction data based on a target prediction time, thereby obtaining a prediction result for the day, wherein the time of the long-term prediction data ≥ the target prediction time > the time of the short-term prediction data.

[0020] By adopting the above technical solutions, the acquisition module ensures the comprehensiveness and reliability of the input data by collecting the identification information and historical data of the parking lot. The mode determination module determines the target mode of the long-term prediction model suitable for the current situation based on the identification information, historical data and the date of the day, thereby improving the accuracy of the selection of the prediction model. The prediction module inputs the historical data and the weather of the day into the short-term prediction model and the long-term prediction model for processing, generates short-term prediction data and long-term prediction data, makes full use of the data characteristics of different time scales, and enhances the reliability and coverage of the prediction results. The processing module processes the intersection of the short-term prediction data and the long-term prediction data at the target prediction time, smoothly transitions the two prediction results, reduces the prediction error, and further improves the prediction accuracy. In this way, through multi-dimensional data analysis and intelligent model selection, the accuracy and practicality of parking lot parking space occupancy prediction are significantly improved.

[0021] In a third aspect, an embodiment of the present application discloses an electronic device, which adopts the following solution: An electronic device comprises: a memory and a processor, wherein the memory is used to store a computer program; and the processor is used to implement the steps of any of the above methods when executing the computer program.

[0022] In a fourth aspect, an embodiment of the present application discloses a computer-readable storage medium, which adopts the following scheme: A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0023] In summary, the present application includes at least one of the following beneficial technical effects: 1. Improved the accuracy of parking lot occupancy prediction, especially in short-term and long-term prediction, which improves the reliability in practical applications; 2. Make full use of multi-dimensional data for comprehensive analysis, combining historical parking space occupancy data and weather data and other factors to enhance the comprehensiveness and accuracy of the prediction results; 3. By establishing a mapping table and selecting the appropriate prediction mode, the prediction model is dynamically adjusted according to the specific conditions of different parking lots, further improving the adaptability and accuracy of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flowchart of a method for predicting future parking lot occupancy data disclosed in an embodiment of the present application; Figure 2 This is a schematic diagram of parking space occupancy data of a large shopping mall over a period of time; Figure 3 A schematic diagram of parking space occupancy data for a mixed-function area over a period of time; Figure 4 A schematic diagram of the parking space occupancy data of an office building over a period of time; Figure 5 A schematic diagram of parking space occupancy data in a residential area over a period of time; Figure 6 A schematic diagram of data examples corresponding to four modes of a long-term prediction model in a method for predicting future parking lot occupancy data disclosed in an embodiment of the present application; Figure 7 Schematic diagram of data training sets constructed for short-term prediction models and long-term prediction models; Figure 8 A schematic diagram of the structure of a system for predicting parking space occupancy in a parking lot disclosed in another embodiment of the present application; Fig. 9 This is a schematic diagram of the structure of an electronic device disclosed in yet another embodiment of the present application. DETAILED DESCRIPTION

[0025] The present application is further described in detail below in conjunction with the accompanying drawings.

[0026] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0027] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "an" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0028] It should be understood that although the terms "first", "second", etc. may be used in the present application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0029] The technical solution of the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0030] First embodiment See also Figure 1 , is a method for predicting parking space occupancy in a parking lot disclosed in an embodiment of the present application, comprising the following steps: S10, obtaining identification information and historical data of the parking lot; Among them, the identification information can be understood as the ID information corresponding to the parking lot or the type of shopping mall. The type of shopping mall can be divided into large shopping malls, mixed-function areas, office buildings, residential areas, etc. The historical data includes the parking space occupancy data corresponding to each time period and the corresponding daily weather data within a preset time period such as one year, one month, one week, etc. The weather data can be obtained by obtaining real-time weather data from an external weather station, which is divided into three types: sunny, rainy, and cloudy.

[0031] For example, Figure 2 The corresponding data is the parking space occupancy data of large shopping malls over a period of time. The daily parking space occupancy change trend is basically the same; Figure 3 The corresponding data is the parking space occupancy data of the mixed-function area over a period of time. The daily parking space occupancy change trend is slightly different but consistent overall; Figure 4 The corresponding data is the parking space occupancy data of the office building within a period of time. The parking space occupancy change trends on weekdays and weekends are different. Figure 5 It corresponds to the parking space occupancy data of residential areas over a period of time. Weekdays and weekends have their own changing trends, and the daily parking space occupancy changes are slightly different.

[0032] S20, determining a target mode corresponding to the long-term prediction model based on the identification information, historical data and the current date; Among them, the current date is used to determine whether the current day is a working day or a rest day (including holidays). In this way, the long-term prediction model can be more adapted to the current specific conditions and the prediction accuracy can be further improved.

[0033] It is to be noted that before step S20, the S01. Establish a mapping table of identification information, historical data, dates and various modes in the long-term prediction model; Among them, step S20 corresponds to step S01, including: determining the target mode corresponding to the long-term prediction model from the mapping table based on the identification information, historical data and the date of the day. In this way, the establishment of the mapping table ensures that the most appropriate historical data can be used for prediction under different conditions, thereby improving the accuracy of the prediction results. This method avoids the errors that may be caused by manually selecting the prediction mode, and improves the automation level of the system and user experience.

[0034] See also Figure 6 The long-term prediction model is used to predict the parking space occupancy data for the next 24 hours. There are multiple corresponding modes, as follows: Prediction mode 1 uses the parking space occupancy data of the previous 72 hours and the corresponding weather data of the previous 72 hours, as well as the weather of the day, to predict the parking space occupancy of the day for 24 hours; For example, today's date is Thursday (corresponding to Figure 6 tag data), sunny days, corresponding to the parking space occupancy data of Monday, Tuesday, and Wednesday of this week (corresponding Figure 6 The input data) and the corresponding weather data (sunny / cloudy / rainy), as well as the weather of the day (sunny), are input into the long-term prediction model to predict the parking space occupancy data for 24 hours on that day.

[0035] Prediction mode 2, using the parking space occupancy data and the corresponding weather data of the previous 24 hours, the parking space occupancy data and the corresponding weather data of the same day last week for 24 hours, and the weather of the current day, to predict the parking space occupancy for the current day for 24 hours; For example, if today is Monday and it is cloudy, the parking space occupancy data and corresponding weather data of last Monday and Sunday, as well as the weather of today (cloudy), are selected and input into the long-term prediction model to predict the parking space occupancy data for 24 hours on that day.

[0036] Prediction mode 3 uses the parking space occupancy data and corresponding weather data of the 48 hours before the working day or holiday, as well as the weather of the day, to predict the parking space occupancy of the day for 24 hours; (see the example below) Prediction mode 4 uses the parking space occupancy data and corresponding weather data for the 24 hours before the working day or holiday, the parking space occupancy data and corresponding weather data for the 24 hours on the same day last week, and the weather of the day to predict the parking space occupancy for the 24 hours of the day; (see the example below) The target mode is one of prediction mode one, prediction mode two, prediction mode three and prediction mode four.

[0037] Furthermore, when the target mode is prediction mode 3, it is necessary to first determine whether the current date is a working day or a holiday, and then select one of the following two methods for prediction; 3a. If today is a working day, the parking space occupancy data of the first 48 hours of the working day and the corresponding weather data of the first 48 hours, as well as the weather of the day, are selected from the historical data and input into the long-term prediction model to predict the parking space occupancy data of the 24 hours of the day; For example, if today is Monday and it is rainy, which corresponds to a working day (Monday to Friday), the parking space occupancy data and corresponding weather data of last Thursday and last Friday, as well as the weather of the day (rainy day), are selected and input into the long-term prediction model to predict the parking space occupancy data for the 24 hours of the day, such as Figure 6 shown.

[0038] 3b. If the current date is a holiday, the parking space occupancy data for 48 hours before the holiday is selected from the historical data and input into the long-term prediction model to predict the parking space occupancy data for 24 hours on the current date.

[0039] For example, if today is Sunday and sunny, which corresponds to a rest day (Saturday and Sunday), the parking space occupancy data and corresponding weather data of this Saturday and last Sunday, as well as the weather of the day (sunny), are selected and input into the long-term prediction model to predict the parking space occupancy data for the 24 hours of the day, such as Figure 6 shown.

[0040] When the target mode is prediction mode 4, determine whether the current date is a working day or a holiday; 4a. If today is a working day, the parking space occupancy data and the corresponding weather data for the first 24 hours of the working day, the parking space occupancy data and the corresponding weather data for the same day last week, and the weather for the current day are selected from the historical data and input into the long-term prediction model to predict the parking space occupancy data for the current 24 hours; For example, if today is Monday and sunny, which corresponds to a working day (Monday to Friday), the parking space occupancy data and the corresponding weather data of last Monday and last Friday are selected, and the weather data of the day (sunny) are input into the long-term prediction model to predict the parking space occupancy data for the 24 hours of the day, such as Figure 6 shown.

[0041] 4b. If the current date is a holiday, the parking space occupancy data and the corresponding weather data for 24 hours before the holiday, the parking space occupancy data and the corresponding weather data for 24 hours on the same day last week, and the weather of the current day are selected from the historical data and input into the long-term prediction model to predict the parking space occupancy data for 24 hours on the current day; For example, if today is Sunday and cloudy, which corresponds to a rest day (Saturday and Sunday), the parking space occupancy data of this Saturday and last Sunday and the corresponding weather data are selected, and the weather data of the day (cloudy) are input into the long-term prediction model to predict the parking space occupancy data for the 24 hours of the day, such as Figure 6 shown.

[0042] S30, inputting historical data and the weather of the day into the short-term prediction model and inputting the target mode into the long-term prediction model for prediction processing, and obtaining short-term prediction data and long-term prediction data respectively; In this embodiment, the short-term prediction model can be used to predict the parking space occupancy data for the next 6 hours, and the long-term prediction model can be used to predict the parking space occupancy data for the next 24 hours. Of course, in other embodiments, there is no limitation, and predictions for different time periods can be performed based on actual needs.

[0043] Furthermore, before step S10, the short-term prediction model and the long-term prediction model are also established, specifically including: S02. Collect historical data of different parking lots; Wherein, collecting historical data of different parking lots corresponds to collecting historical data of parking in different types of parking lots in different time periods. This step S02 corresponds to the data set collection stage.

[0044] S03, processing the historical data by time period and corresponding parking space occupancy data as a target training set; The processing described in step S03 can be understood as sorting and grouping the historical data. By normalizing (subtracting the mean and then dividing by the standard deviation), data can be obtained one by one to form a data set, such as Figure 7 As shown (weather data is not filled in). In this way, by comprehensively collecting and processing data from different parking lots, the influence of various factors on parking space occupancy can be better captured, thereby improving the prediction accuracy. This step S03 corresponds to the stage of processing the data in the data set.

[0045] S04. Input the target training set into the original short-term prediction model and the original long-term prediction model for training to obtain a short-term prediction model and a long-term prediction model.

[0046] Among them, step S04 corresponds to inputting the data set into the original model for training, so as to use multi-source data for model training, which helps the model learn the characteristics of different types of parking lots, so that it can also show good performance when facing new parking scenarios. It is noted that the training process of the model is an existing related technology and will not be described in detail here.

[0047] S40, performing handover processing on the short-term prediction data and the long-term prediction data based on the target prediction time to obtain the prediction result for the day; Among them, the time of long-term forecast data ≥ target forecast time > short-term forecast data time. For example, the time of long-term forecast data is 12 hours, the target forecast time is 12 hours, and the time of short-term forecast data is 6 hours. When the short-term forecast data is , which corresponds to the parking space occupancy data for each hour within 6 hours, and the long-term prediction data is , which corresponds to the parking space occupancy data for each hour within 24 hours, the prediction result for the day is ; Among them, when the intersection point corresponds to 6-7, the corresponding processing algorithm is , The corresponding parking space occupancy data is 6 o'clock. The corresponding data is the parking space occupancy data at 7 o'clock. In this way, the smooth transition of the prediction results between the two time intervals can be ensured, which improves the continuity and accuracy of the prediction. Especially for the two key time nodes of 6 o'clock and 7 o'clock, this processing method can more accurately reflect the changing trend of the actual parking space occupancy, thereby improving the reliability of the overall prediction. It should be noted that the final prediction result is mainly based on short-term prediction, supplemented by long-term prediction. At the junction of time, the weighted average algorithm is used to calculate the data predicted by both.

[0048] In summary, the first embodiment of the present invention discloses a method for predicting future parking lot occupancy data. By obtaining the identification information and historical data of the parking lot, subsequent predictions can be ensured to be targeted and the prediction accuracy can be improved. Based on the identification information, historical data and the date of the day, the target mode corresponding to the long-term prediction model is determined, so that the prediction model is more adapted to the current specific conditions and the prediction accuracy is further improved. The historical data and the weather of the day are input into the short-term prediction model and the long-term prediction model for prediction processing, and the short-term prediction data and the long-term prediction data are obtained respectively. Combining the advantages of the two prediction models, the comprehensiveness and reliability of the prediction are improved. The intersection point of the short-term prediction data and the long-term prediction data is processed based on the target prediction time to obtain the final prediction result of the day. This processing method effectively solves the connection problem between the short-term and long-term predictions, avoids the prediction fault phenomenon, and makes the prediction result smoother and more accurate. In this way, the method can effectively predict the parking space occupancy of the parking lot in the future period of time, and provide a scientific basis for management decisions.

[0049] Second embodiment See also Figure 8 In the second embodiment of the present application, a system for predicting parking space occupancy in a parking lot is disclosed. The system includes: an acquisition module 210, a mode determination module 220, a prediction module 230 and a processing module 240.

[0050] Among them, the acquisition module 210 is used to obtain the identification information and historical data of the parking lot, and the historical data includes the corresponding parking space occupancy data and weather data within a preset time period; the mode determination module 220 is used to determine the target mode corresponding to the long-term prediction model based on the identification information, the historical data and the date of the day; the prediction module 230 is used to input the historical data and current weather into the short-term prediction model and input the target mode into the long-term prediction model for prediction processing, and obtain short-term prediction data and long-term prediction data respectively; the processing module 240 is used to perform intersection processing on the short-term prediction data and the long-term prediction data based on the target prediction time to obtain the prediction result of the day, wherein the time of the long-term prediction data ≥ the target prediction time > the time of the short-term prediction data.

[0051] It should be noted that the method based on predicting parking space occupancy of a parking lot disclosed in the second embodiment of the present application is the same as the first embodiment, so it will not be described in detail here. Optionally, each module in this embodiment and the above-mentioned other operations or functions are respectively for implementing the methods in the above-mentioned embodiments.

[0052] Third embodiment See also Fig. 9In the third embodiment of the present application, an electronic device is disclosed, which includes: a memory 310 and a processor 320, the memory 310 is used to store computer programs; the processor 320 is used to implement the steps of a method for predicting parking space occupancy in the above-mentioned first embodiment when executing the computer program. For details, please refer to the above, so it will not be described in detail here.

[0053] The technical effect of an electronic device provided by this embodiment in actual application is the same as the technical effect of a method for predicting parking space occupancy in the first embodiment.

[0054] Fourth embodiment In the fourth embodiment of the present application, a computer-readable storage medium is disclosed. The computer-readable storage medium is, for example, a non-volatile memory, such as: magnetic media (such as hard disks, floppy disks, and tapes), optical media (such as CDROM disks and DVDs), magneto-optical media (such as optical disks), and hardware devices specially constructed to store and execute computer-executable instructions (such as read-only memory (ROM), random access memory (RAM), flash memory, etc.). A computer program is stored on the computer-readable storage medium. The computer-readable storage medium can be used by one or more processors or processing devices to execute the computer program to implement a method for predicting parking space occupancy in the aforementioned embodiment.

[0055] In addition, it can be understood that the aforementioned embodiments are only exemplary descriptions of the present invention. Under the premise that the technical features do not conflict, the structures do not contradict, and the purpose of the present invention is not violated, the technical solutions of the various embodiments can be arbitrarily combined and used in combination.

[0056] In the several embodiments provided by the present invention, it should be understood that the disclosed methods, systems and devices can be implemented in other ways. For example, the modules included in the system described above are only schematic, and the division of modules is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0057] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0058] In addition, each functional unit / module in each embodiment of the present invention may be integrated into one processing unit / module, or each unit / module may exist physically separately, or two or more units / modules may be integrated into one unit / module. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of hardware plus software functional units / modules.

[0059] The above-mentioned integrated unit / module implemented in the form of a software functional unit / module can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for enabling one or more processors of a computer device (which can be a personal computer, a server, or a network device, etc.) to perform some steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store program codes.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting parking space occupancy in a parking lot, characterized in that: include: Obtaining identification information and historical data of the parking lot, wherein the historical data includes corresponding parking space occupancy data and weather data within a preset time period; Determining a target mode corresponding to a long-term prediction model based on the identification information, the historical data, and the current date; Input the historical data and the weather of the day into the short-term prediction model and input the target model into the long-term prediction model for prediction processing to obtain short-term prediction data and long-term prediction data respectively; The short-term prediction data and the long-term prediction data are processed at the intersection point based on the target prediction time to obtain the prediction result for the day, wherein the time of the long-term prediction data ≥ the target prediction time > the time of the short-term prediction data.

2. The method according to claim 1, characterized in that The short-term prediction model is used to predict the parking space occupancy data for the next 6 hours; the long-term prediction model is used to predict the parking space occupancy data for the next 24 hours; When the short-term forecast data is , which corresponds to the parking space occupancy data for each hour within 6 hours. The long-term prediction data is , corresponding to the parking space occupancy data for each hour within 24 hours, the prediction result for the day is ; Among them, when the intersection point corresponds to 6-7, the corresponding processing algorithm is , The corresponding parking space occupancy data is 6 o'clock. This corresponds to the parking space occupancy data at 7 o'clock.

3. The method according to claim 1, characterized in that: The short-term prediction model uses the parking space occupancy data of the previous 24 hours of the current date and the corresponding weather data, as well as the current day's weather, to predict the parking space occupancy in the next 6 hours; The long-term prediction model is configured with four prediction modes corresponding to the regional types, which are as follows: Prediction mode 1, using the parking space occupancy data of the previous 72 hours and the corresponding weather data, as well as the weather of the current day, to predict the parking space occupancy for 24 hours of the current day; Prediction mode 2, using the parking space occupancy data and the corresponding weather data of the previous 24 hours, the parking space occupancy data and the corresponding weather data of the same day last week for 24 hours, and the weather of the current day, to predict the parking space occupancy of the current day for 24 hours; Prediction mode three, using the parking space occupancy data of 48 hours before the working day or holiday and the corresponding weather data, as well as the weather of the day, to predict the parking space occupancy of 24 hours on the day; Prediction mode 4, using the parking space occupancy data and the corresponding weather data for 24 hours before a working day or a holiday, the parking space occupancy data and the corresponding weather data for 24 hours on the same day last week, and the weather of the day, to predict the parking space occupancy for 24 hours on the day; The target mode is one of prediction mode one, prediction mode two, prediction mode three and prediction mode four.

4. The method according to claim 3, characterized in that: When the target mode is prediction mode three, determining whether the current date is a working day or a holiday; If the current date is a working day, the parking space occupancy data for the first 48 hours of the working day is selected from the historical data and input into the long-term prediction model together with the current weather to predict the parking space occupancy data for the 24 hours of the current day; If the current date is a holiday, the parking space occupancy data for 48 hours before the holiday is selected from the historical data and input into the long-term prediction model together with the current weather to predict the parking space occupancy data for 24 hours on the current date.

5. The method according to claim 3, characterized in that: When the target mode is prediction mode 4, determining whether the current date is a working day or a holiday; If the current date is a working day, the parking space occupancy data for the first 24 hours of the working day and the parking space occupancy data for the same day last week are selected from the historical data, and input into the long-term prediction model together with the current weather to predict the parking space occupancy data for the current 24 hours; If the current date is a holiday, the parking space occupancy data for 24 hours before the holiday and the parking space occupancy data for 24 hours on the same day last week are selected from the historical data and input into the long-term prediction model together with the current weather to predict the parking space occupancy data for 24 hours on the current day.

6. The method according to claim 1, characterized in that Also includes: A mapping table of the identification information, historical data, date and multiple modes in the long-term prediction model is established, wherein the target mode corresponding to the long-term prediction model is determined from the mapping table based on the identification information and the current date.

7. The method according to claim 1, characterized in that Before obtaining the historical data and area type of the parking lot, the method further includes: Collect historical data from different parking lots; Processing the historical data by time period and corresponding parking space occupancy data as a target training set; The target training set is input into the original short-term prediction model and the original long-term prediction model for training to obtain the short-term prediction model and the long-term prediction model.

8. A system for predicting parking space occupancy, characterized in that: include: An acquisition module, used to acquire identification information and historical data of the parking lot, wherein the historical data includes corresponding parking space occupancy data and weather data within a preset time period; A mode determination module, used to determine a target mode corresponding to a long-term prediction model based on the identification information, the historical data and the current date; A prediction module, used for inputting the historical data and current weather into a short-term prediction model and inputting the target mode into the long-term prediction model for prediction processing, to obtain short-term prediction data and long-term prediction data respectively; The processing module is used to perform intersection processing on the short-term prediction data and the long-term prediction data based on the target prediction time to obtain the prediction result of the day, wherein the time of the long-term prediction data ≥ the target prediction time > the time of the short-term prediction data.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the memory is used to store a computer program; and the processor is used to implement the steps of the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images