Charging pile idle state prediction system and method based on time sequence model

Abstract

The invention discloses a charging pile idle state prediction system and method based on a time sequence model, and the system comprises a current sensor which is used for collecting the current change data of a charging pile; the data processing unit is connected with the current sensor and is used for carrying out abnormal data processing on the current change data to obtain processed current change data; and the prediction server is connected with the data processing unit and is used for predicting the future idle time period and the idle window duration of the charging pile according to the processed current change data.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of automobile charging, in particular to a charging pile idle state prediction system and method based on a time sequence model. BACKGROUND

[0002] In a traditional electric vehicle charging scenario, users are blind when looking for charging piles, cannot predict future idle periods of charging piles, and thus cause blind queuing or long-distance detours, and the average waiting time is too long during peak hours, increasing the time cost of charging; for operators, the use rates of charging piles in urban cores and suburbs are uneven.

[0003] In related technologies, it is not possible to distinguish which charging piles have long queuing times during charging peak hours, causing uneven use of charging piles, long queuing times for vehicle owners, and other problems, and no effective solutions have been proposed. SUMMARY

[0004] Embodiments of the present application provide a charging pile idle state prediction system and method based on a time sequence model to at least solve the problem in related technologies that it is not possible to distinguish which charging piles have long queuing times during charging peak hours, causing uneven use of charging piles and long queuing times for vehicle owners.

[0005] According to one embodiment of an embodiment of the present application, a charging pile idle state prediction system based on a time sequence model is provided, comprising: a current sensor for collecting current change data of a charging pile; a data processing unit connected to the current sensor, configured to process abnormal data of the current change data to obtain processed current change data; and a prediction server connected to the data processing unit, configured to predict future idle periods and idle window lengths of the charging pile according to the processed current change data.

[0006] In one exemplary embodiment, the current sensor is further configured to: identify the charging pile as a charging state if the current of the charging pile is greater than a preset threshold; collect current data of the charging pile at a preset period, and generate the current change data according to a plurality of the current data, wherein the current change data includes: a plurality of the current data, position information of the charging pile, first time information of a plurality of the current data, and second time information of the charging pile in the charging state; store the current change data to a cache module, and upload the current change data to a backend database through a message queue.

[0007] In an example embodiment, the data processing unit is further configured to: identify abnormal data points in the current change data through a sliding window; and correct the abnormal data points through an abnormal correction algorithm to obtain the processed current change data, when a proportion of the abnormal data points in a single sliding window is less than an abnormal threshold.

[0008] In an example embodiment, the data processing unit is further configured to: extract time features from the current change data to obtain the time features; determine a charging pile density of a region where the charging pile is located according to the location information; and obtain environmental feature data of the region where the charging pile is located, wherein the processed current change data comprises the time features, the charging pile density, and the environmental feature data.

[0009] In an example embodiment, the prediction server is further configured to: adjust feature weights of a prediction model according to the environmental feature data to obtain an adjusted prediction model, wherein the prediction model is set in the prediction server, and the prediction model is a long short-term memory network model; obtain historical current data of the charging pile, and obtain historical time features corresponding to the historical current data; and process the historical current data, the historical time features, and the processed current change data through the adjusted prediction model to obtain the future idle time period and the idle window length.

[0010] In an example embodiment, the prediction server is further configured to: obtain real-time location information of a target object; recommend and sort a plurality of charging piles according to the real-time location information, the future idle time period, and the idle window length of the plurality of charging piles to obtain a recommendation list, wherein the recommendation list comprises the sorted plurality of charging piles; and send the recommendation list to the target object.

[0011] According to another embodiment of the present application, a charging pile idle state prediction method based on a time sequence model is also provided, which is applied to the charging pile idle state prediction system based on the time sequence model, and comprises: collecting current change data of a charging pile; processing abnormal data in the current change data to obtain processed current change data; and predicting a future idle time period and an idle window length of the charging pile according to the processed current change data.

[0012] According to another aspect of the present application, a computer-readable storage medium is also provided, which stores a computer program, wherein the computer program is configured to execute the charging pile idle state prediction method based on the time sequence model when running.

[0013] According to a further aspect of the embodiments of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-mentioned time-series model based charging pile idle state prediction method through the computer program.

[0014] According to a further aspect of the embodiments of the present application, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the steps of the method described in the embodiments of the present application.

[0015] In the embodiments of the present application, a time-series model based charging pile idle state prediction system is provided, comprising a current sensor, a data processing unit and a prediction server. The current sensor is deployed in the output line of the charging pile to capture current fluctuations and obtain current change data. The data processing unit receives the current change data and processes abnormal data to obtain processed current change data. The prediction server accurately predicts when the charging pile will be idle in the future and how long the available window will last according to the processed current change data, thereby guiding the electric vehicle user to the charging pile that will be idle soon, reducing the waiting time and enhancing the charging experience. The above-mentioned system solves the problem that in the related art, it is not possible to distinguish which charging piles have long queuing times during the charging peak period, resulting in uneven charging pile utilization and long queuing times for vehicle owners. BRIEF DESCRIPTION OF DRAWINGS

[0016] The accompanying drawings, which are included to provide a further understanding of the present application, form a part of the present application and illustrate the illustrative embodiments of the present application and the explanation of the present application, and do not constitute improper limitations on the present application. In the drawings:

[0017] Figure 1 is a hardware structure block diagram of an optional time-series model based charging pile idle state prediction system according to an embodiment of the present application;

[0018] Figure 2 is a structure block diagram of an optional time-series model based charging pile idle state prediction system according to an embodiment of the present application;

[0019] Figure 3 is a running logic schematic diagram of an optional time-series model based charging pile idle state prediction system according to an embodiment of the present application;

[0020] Figure 4 is a flowchart of a time-series model based charging pile idle state prediction method according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] In order to better understand the technical scheme of the present application, the technical scheme in the embodiments of the present application will be clearly and completely described in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work should fall within the scope of protection of the present application.

[0022] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0023] The method embodiments provided by the embodiments of the present application can be executed in a charging pile idle state prediction system based on a timing model or a similar operation system. Taking an example of running on a charging pile idle state prediction system based on a timing model, Figure 1 is a hardware structure block diagram of a charging pile idle state prediction system based on a timing model according to an embodiment of the present application. As Figure 1 indicated, the charging pile idle state prediction system based on a timing model can include one or more (only one is shown in Figure 1 ) processor 102 (the processor 102 can include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data. In an exemplary embodiment, the above-mentioned charging pile idle state prediction system based on a timing model can further include a transmission device 106 for communication function and an input and output device 108. Those skilled in the art can understand that Figure 1 The structure shown is only schematic, which does not limit the structure of the above-mentioned charging pile idle state prediction system based on a timing model. For example, the charging pile idle state prediction system based on a timing model can further include more or less components than those shown in Figure 1 , or have a different configuration with the same function as Figure 1 or more functions than Figure 1 .

[0024] The memory 104 can be configured to store computer programs, for example, software programs of application software and modules, such as a computer program corresponding to the charging pile idle state prediction method based on a timing model in the embodiments of the present application. The processor 102 can execute various functional applications and data processing, that is, implement the above method, by running the computer programs stored in the memory 104. The memory 104 can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage systems, flash memories, or other non-volatile solid-state memories. In some examples, the memory 104 can further include a memory remotely arranged with respect to the processor 102, and the remote memory can be connected to the secure text through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0025] The transmission system 106 is configured to receive or send data via a network. The specific example of the above network can include a wireless network provided by a communication provider of the charging pile idle state prediction system based on a timing model. In one example, the transmission system 106 includes a network adapter (Network Interface Controller, NIC) which can be connected to other network devices through a base station so as to communicate with the Internet.

[0026] In the embodiments of the present application, a charging pile idle state prediction system based on a timing model is provided, Figure 2 According to the embodiments of the present application, an optional structure block diagram of the charging pile idle state prediction system based on a timing model is provided, and the system includes:

[0027] The current sensor 22 is configured to collect current change data of the charging pile.

[0028] The data processing unit 24 is connected with the current sensor and is configured to perform abnormal data processing on the current change data to obtain processed current change data.

[0029] The prediction server 26 is connected with the data processing unit and is configured to predict future idle time periods and idle window time lengths of the charging pile according to the processed current change data.

[0030] The system comprises a current sensor, a data processing unit and a prediction server. The current sensor is arranged in the output line of the charging pile to capture current fluctuations and obtain current change data. The data processing unit receives the current change data and processes abnormal data to obtain processed current change data. The prediction server accurately predicts when the charging pile will be idle and how long the available window will last based on the processed current change data, thereby guiding the electric vehicle user to the charging pile that will be idle, reducing the waiting time and enhancing the charging experience. The system solves the problem of uneven use of charging piles and long queuing time of vehicle owners in related technologies during the charging peak period.

[0031] Optionally, the current sensor 22 is further configured to: identify the charging pile as being in a charging state when the current of the charging pile is greater than a preset threshold; collect current data of the charging pile at a preset period, and generate the current change data based on the current data, wherein the current change data comprises the current data, position information of the charging pile, first time information of the current data, and second time information of the charging pile in the charging state; and store the current change data in a cache module and upload the current change data to a backend database through a message queue.

[0032] When the current sensor detects that the current of the charging pile exceeds the preset threshold, the charging pile is marked as entering the charging mode to distinguish between the charging state and the standby state. Then the current sensor continuously captures the current value at a fixed time interval, such as every 500 milliseconds. The accumulated data constructs a detailed current change profile, which not only contains the instantaneous current reading, but also embeds the charging pile positioning information, the accurate time stamp of each measurement, and the specific time when the charging pile starts charging. These comprehensive information constitutes the entire content of the current change data.

[0033] In order to ensure the integrity of the data when the network environment is unstable, the current change data is temporarily saved in the cache module, and then pushed to the backend real-time database in order through a message queue system such as RabbitMQ, realizing the persistent storage and efficient management of the data.

[0034] In this embodiment, through real-time monitoring by the current sensor and effective management of the data, the working state of the charging pile can be accurately captured, the high-quality transmission of the data is ensured, and data support is provided for accurate prediction of the idle state of the charging pile, thereby significantly enhancing the reliability and efficiency of the electric vehicle charging service.

[0035] Optionally, the data processing unit 24 is further configured to: identify abnormal data points in the current change data through a sliding window; and correct the abnormal data points through an abnormal correction algorithm if a proportion of the abnormal data points in a single sliding window is less than an abnormal threshold, to obtain the processed current change data.

[0036] The data processing unit focuses on the quality control of current change data. By implementing a sliding window technique, the data processing unit can finely screen out abnormal points in the current data. This process is similar to rolling a fixed length interval in a continuous data stream to check the consistency of the data in the interval.

[0037] When abnormal data points are found in the sliding window, it is determined whether the proportion of the abnormal data points is less than an abnormal threshold (e.g., 5%). If it is less, an abnormal correction algorithm is started to repair the abnormal values, ensuring the accuracy of the current change data.

[0038] Through this embodiment, the reliability of the current change data is significantly improved, the deviation of the prediction model caused by abnormal data is reduced, the accuracy of the charging pile idle state prediction is improved, more stable and accurate charging information is provided for the electric vehicle user, and the charging experience and resource utilization efficiency are further optimized.

[0039] Optionally, the data processing unit 24 is further configured to: perform time feature extraction on the current change data to obtain time features; determine a charging pile density of an area where the charging pile is located according to the location information; and obtain environmental feature data of the area where the charging pile is located, wherein the processed current change data includes the time features, the charging pile density, and the environmental feature data.

[0040] The data processing unit analyzes the time stamp of the current change data to identify the specific date, hour, and even minute to which the data belongs. At the same time, by analyzing the day of the week or whether it is a weekday, the periodic pattern of charging demand is understood.

[0041] Further, the data processing unit triggers from the location information of the charging pile to evaluate the charging pile density, i.e., the distribution density of the charging pile in a specific area, which reflects the association between charging convenience and potential waiting time, and is crucial for predicting the utilization rate of the charging pile.

[0042] In addition, the data processing unit also needs to integrate environmental data, including non-structural influencing factors such as holiday classification and weather conditions around the charging pile. These data usually need to be imported from third-party services, adding a change dimension of real-world scenarios to the prediction model.

[0043] The aggregated and processed current change data is no longer limited to the current value itself, but is enriched with time characteristics, charging pile density and environmental characteristics, forming important input data required for the prediction model.

[0044] This embodiment greatly enhances the diversity of data inputs and the relevance to reality of the prediction model by deeply mining the time, space and environmental information hidden in the current change data. This improves the accuracy and practicality of predicting the idle status of charging piles, provides users with more refined charging planning suggestions, effectively shortens charging waiting time, and optimizes the electric vehicle charging experience.

[0045] In an exemplary embodiment, the prediction server 26 is further configured to: adjust the feature weights of the prediction model according to the environmental feature data to obtain an adjusted prediction model, wherein the prediction model is set in the prediction server and the prediction model is a long short-term memory network model; acquire historical current data of the charging pile and acquire historical time features corresponding to the historical current data; and process the historical current data, the historical time features, and the processed current change data through the adjusted prediction model to obtain the future idle time period and the idle window duration.

[0046] The prediction server dynamically adjusts the model feature weights based on environmental feature data collected from the charging pile area, such as holidays and weather conditions, to ensure that the prediction model can fully reflect the impact of the external environment on charging demand and improve prediction accuracy.

[0047] The server then automatically retrieves historical current data from the charging stations, along with corresponding historical time features such as date, time, and weekday, as the basis for model training and prediction. By integrating historical data, time features, and processed current change data, the adjusted prediction model can comprehensively analyze and output specific predictions for future charging station idle periods and the expected idle window duration, providing data support for users' charging decisions.

[0048] It should be noted that the Long Short-Term Memory Network model, as a time series forecasting tool in artificial intelligence, can capture long-term dependencies in data and deeply learn the usage patterns of charging piles. Even if there are time intervals in the data sequence, it can accurately predict the state changes of charging piles.

[0049] This embodiment significantly improves the accuracy and timeliness of charging pile idle state prediction by dynamically adjusting the feature weights of the LSTM prediction model through the introduction of environmental feature data and combining it with the comprehensive analysis of historical and real-time data of charging piles. This, in turn, optimizes resource allocation and the user charging experience. In short, it achieves an intelligent upgrade of the prediction model, bringing a revolutionary breakthrough to the prediction of charging pile idle state and effectively alleviating the supply and demand contradiction during peak charging periods.

[0050] In an exemplary embodiment, the prediction server 26 is further configured to: obtain real-time location information of the target object; sort the multiple charging piles according to the real-time location information, the future idle time periods and idle window durations of the multiple charging piles, and obtain a recommendation list, wherein the recommendation list includes the sorted multiple charging piles; and send the recommendation list to the target object.

[0051] This embodiment illustrates how a prediction server efficiently utilizes prediction results to provide personalized charging station recommendation services for electric vehicle users. The server obtains the user's real-time geographical location information and, combined with predictions of future idle periods and idle window durations for multiple charging stations, implements an intelligent sorting algorithm to generate a customized charging station recommendation list. This list not only lists multiple charging stations but, more importantly, sorts them by predicted availability and distance from the user, prioritizing those charging stations that are both close to the user and about to become available, greatly improving the efficiency and experience of users searching for charging stations.

[0052] This embodiment cleverly combines real-time user location information with predictions of future charging station availability through intelligent recommendation and ranking by a prediction server. This provides an efficient and personalized charging station navigation service, significantly improving user decision-making efficiency and charging experience during peak charging periods, and promoting the intelligent development of electric vehicle charging infrastructure. In short, it achieves prediction-based personalized charging station recommendations, making charging more convenient and efficient for electric vehicle users.

[0053] In an optional embodiment, Figure 3 The overall operation logic of the charging pile idle state prediction system based on the time series model is shown above, as follows: Figure 3 As shown, it includes:

[0054] First, the raw data is obtained by collecting current change data from the charging piles. Since the charging pile current is significantly higher during charging than in standby mode, a current sensor (such as a Hall sensor) is connected in series in the charging pile output line to acquire current changes. When the current exceeds a certain indicator (such as 16A), it is identified as a charging state. The charging status is checked and transmitted every 500ms. Data such as charging pile location, charging pile number, current time, and current day of the week are stored in a cache module to avoid data loss due to network instability, and then uploaded to the backend database via a message queue (such as RabbitMQ).

[0055] For the collected charging pile current change data, the system first filters out abnormal data from the collected charging pile status and power data, and uses a sliding window to correct the status. Points that meet the overall window condition (number of abnormal points less than 5%) but whose individual points do not meet the condition are processed as anomalies. After anomaly processing, time features such as time, day of the week, and whether it is a weekday are obtained. Charging pile density is obtained through charging pile location and charging pile number. Finally, a Long Short-Term Memory (LSTM) network combined with weight allocation (an attention enhancement mechanism that automatically adjusts feature weights based on external factors such as holidays and weather) is used to predict how long a charging pile is likely to be idle and the duration of the idle window.

[0056] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0057] Figure 4 This application discloses a charging pile idle state prediction method based on a time-series model, applied to the aforementioned charging pile idle state prediction system based on a time-series model. The method includes the following steps:

[0058] Step S402: Collect current change data of the charging pile;

[0059] Step S404: Perform abnormal data processing on the current change data to obtain processed current change data;

[0060] Step S406: Predict the future idle time period and idle window duration of the charging pile based on the processed current change data.

[0061] The above scheme captures current fluctuations to obtain current change data; it then processes the abnormal data to obtain processed current change data; based on the processed current change data, it accurately predicts when the charging pile will become idle and how long the available window will last, thereby guiding electric vehicle users to the charging piles that are about to become idle, reducing waiting time and enhancing the charging experience. The above scheme solves the problem in related technologies that it is impossible to distinguish which charging piles have long queues during peak charging periods, resulting in uneven utilization of charging piles and long waiting times for car owners.

[0062] In one exemplary embodiment, the method further includes: identifying the charging pile as being in a charging state when the current of the charging pile is detected to be greater than a preset threshold; collecting current data of the charging pile according to a preset period, generating current change data based on multiple current data, wherein the current change data includes: multiple current data, location information of the charging pile, first time information of the multiple current data, and second time information of the charging pile being in the charging state; storing the current change data in a cache module, and uploading the current change data to a backend database through a message queue.

[0063] When the current sensor detects that the charging pile current exceeds a preset threshold, it marks the charging pile as entering charging mode to distinguish between charging and standby states. Then, the current sensor continuously captures current values ​​at fixed time intervals, such as every 500 milliseconds. The accumulated data constructs a detailed overview of current changes, including not only instantaneous current readings but also embedded charging pile location information, precise timestamps of each measurement, and the specific time when the charging pile was activated. This comprehensive information constitutes the entirety of the current change data.

[0064] To ensure data integrity in unstable network environments, current change data is temporarily stored in a cache module and then pushed to the backend real-time database in an orderly manner through a message queue system, such as RabbitMQ, to achieve persistent data storage and efficient management.

[0065] In this embodiment, the real-time monitoring and effective management of data by the current sensor can accurately capture the working status of the charging pile, ensure high-quality data transmission, provide data support for accurate prediction of the charging pile's idle status, and significantly enhance the reliability and efficiency of electric vehicle charging services.

[0066] In one exemplary embodiment, the method further includes: identifying abnormal data points in the current change data through a sliding window; and correcting the abnormal data points by an anomaly correction algorithm when the proportion of the abnormal data points in a single sliding window is less than an anomaly threshold, thereby obtaining the processed current change data.

[0067] The data processing unit focuses on quality control of current variation data. By implementing a sliding window technique, the data processing unit can finely filter out outliers in the current data. This process is similar to rolling a fixed-length interval in a continuous data stream to check the consistency of the data within the interval.

[0068] When abnormal data points are found within the sliding window, it is confirmed whether the proportion of abnormal data points is lower than the abnormal threshold (e.g., 5%). If it is lower, the abnormal correction algorithm is activated to repair these abnormal values ​​and ensure the accuracy of the current change data.

[0069] This embodiment significantly improves the reliability of current change data, reduces the deviation of the prediction model caused by abnormal data, thereby improving the accuracy of charging pile idle state prediction, providing electric vehicle users with more stable and accurate charging information, and further optimizing the charging experience and resource utilization efficiency.

[0070] In an exemplary embodiment, the method further includes: extracting time features from the current change data to obtain time features; determining the charging pile density in the area where the charging pile is located based on the location information; and acquiring environmental feature data of the area where the charging pile is located, wherein the processed current change data includes: the time features, the charging pile density, and the environmental feature data.

[0071] The data processing unit parses the timestamps of the current change data to identify the specific date, hour, and even minute to which the data belongs. At the same time, by analyzing the day of the week or whether it is a weekday, it can gain insight into the periodic patterns of charging demand.

[0072] Furthermore, the data processing unit triggers the assessment of charging pile density based on the location information of the charging piles, that is, the density of the distribution of charging piles in a specific area. This indicator reflects the correlation between charging convenience and potential waiting time, and is crucial for predicting the utilization rate of charging piles.

[0073] In addition, the data processing unit also needs to integrate environmental data, including unstructured influencing factors such as holidays and weather conditions around the charging stations. This data usually needs to be imported from third-party services to add a dimension of real-world scenario changes to the prediction model.

[0074] The aggregated and processed current change data is no longer limited to the current value itself, but is enriched with time characteristics, charging pile density and environmental characteristics, forming important input data required for the prediction model.

[0075] This embodiment greatly enhances the diversity of data inputs and the relevance to reality of the prediction model by deeply mining the time, space and environmental information hidden in the current change data. This improves the accuracy and practicality of predicting the idle status of charging piles, provides users with more refined charging planning suggestions, effectively shortens charging waiting time, and optimizes the electric vehicle charging experience.

[0076] In an exemplary embodiment, the method further includes: adjusting the feature weights of the prediction model based on the environmental feature data to obtain an adjusted prediction model, wherein the prediction model is set in the prediction server and the prediction model is a long short-term memory network model; acquiring historical current data of the charging pile and acquiring historical time features corresponding to the historical current data; and processing the historical current data, the historical time features, and the processed current change data through the adjusted prediction model to obtain the future idle time period and the idle window duration.

[0077] The prediction server dynamically adjusts the model feature weights based on environmental feature data collected from the charging pile area, such as holidays and weather conditions, to ensure that the prediction model can fully reflect the impact of the external environment on charging demand and improve prediction accuracy.

[0078] The server then automatically retrieves historical current data from the charging stations, along with corresponding historical time features such as date, time, and weekday, as the basis for model training and prediction. By integrating historical data, time features, and processed current change data, the adjusted prediction model can comprehensively analyze and output specific predictions for future charging station idle periods and the expected idle window duration, providing data support for users' charging decisions.

[0079] It should be noted that the Long Short-Term Memory Network model, as a time series forecasting tool in artificial intelligence, can capture long-term dependencies in data and deeply learn the usage patterns of charging piles. Even if there are time intervals in the data sequence, it can accurately predict the state changes of charging piles.

[0080] This embodiment significantly improves the accuracy and timeliness of charging pile idle state prediction by dynamically adjusting the feature weights of the LSTM prediction model through the introduction of environmental feature data and combining it with the comprehensive analysis of historical and real-time data of charging piles. This, in turn, optimizes resource allocation and the user charging experience. In short, it achieves an intelligent upgrade of the prediction model, bringing a revolutionary breakthrough to the prediction of charging pile idle state and effectively alleviating the supply and demand contradiction during peak charging periods.

[0081] In one exemplary embodiment, the method further includes: obtaining real-time location information of the target object; sorting the multiple charging piles according to the real-time location information, the future idle time periods of the multiple charging piles, and the idle window duration to obtain a recommendation list, wherein the recommendation list includes the sorted multiple charging piles; and sending the recommendation list to the target object.

[0082] This embodiment illustrates how a prediction server efficiently utilizes prediction results to provide personalized charging station recommendation services for electric vehicle users. The server obtains the user's real-time geographical location information and, combined with predictions of future idle periods and idle window durations for multiple charging stations, implements an intelligent sorting algorithm to generate a customized charging station recommendation list. This list not only lists multiple charging stations but, more importantly, sorts them by predicted availability and distance from the user, prioritizing those charging stations that are both close to the user and about to become available, greatly improving the efficiency and experience of users searching for charging stations.

[0083] This embodiment cleverly combines real-time user location information with predictions of future charging station availability through intelligent recommendation and ranking by a prediction server. This provides an efficient and personalized charging station navigation service, significantly improving user decision-making efficiency and charging experience during peak charging periods, and promoting the intelligent development of electric vehicle charging infrastructure. In short, it achieves prediction-based personalized charging station recommendations, making charging more convenient and efficient for electric vehicle users.

[0084] Embodiments of this application also provide a storage medium including a stored program, wherein the program executes any of the methods described above when it is run.

[0085] Optionally, in this embodiment, the storage medium may be configured to store program code for performing the following steps:

[0086] S1, collect current change data of the charging pile;

[0087] S2, perform abnormal data processing on the current change data to obtain processed current change data;

[0088] S3, based on the processed current change data, predict the future idle time period and idle window duration of the charging pile.

[0089] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0090] Specific examples in this embodiment can be found in the examples described in the above embodiments and exemplary implementations, and will not be repeated here.

[0091] Embodiments of this application also provide an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0092] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0093] S1, collect current change data of the charging pile;

[0094] S2, perform abnormal data processing on the current change data to obtain processed current change data;

[0095] S3, based on the processed current change data, predict the future idle time period and idle window duration of the charging pile.

[0096] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0097] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium storing the computer program product, wherein the computer program, when executed by a processor, implements the steps of the methods described in various embodiments of this application.

[0098] Optionally, in this embodiment, the computer program described above can be configured to perform the following steps when executed by the processor:

[0099] S1, collect current change data of the charging pile;

[0100] S2, perform abnormal data processing on the current change data to obtain processed current change data;

[0101] S3, based on the processed current change data, predict the future idle time period and idle window duration of the charging pile.

[0102] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0103] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing systems. They can be centralized on a single computing system or distributed across a network of multiple computing systems. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage system for execution by the computing system. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0104] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A charging pile idle state prediction system based on a timing model, characterized in that, include: A current sensor is used to collect data on current changes in the charging pile. A data processing unit, connected to the current sensor, is used to perform abnormal data processing on the current change data to obtain processed current change data. A prediction server, connected to the data processing unit, is used to predict the future idle periods and idle window durations of the charging pile based on the processed current change data.

2. The charging station vacancy state prediction system of claim 1, wherein, The current sensor is also used for: If the current of the charging pile is detected to be greater than a preset threshold, the charging pile will be identified as being in a charging state. The charging pile's current data is collected according to a preset cycle, and the current change data is generated based on the multiple current data. The current change data includes: multiple current data, the location information of the charging pile, first time information of the multiple current data, and second time information of the charging pile being in the charging state. The current change data is stored in the cache module and then uploaded to the backend database via a message queue.

3. The charging station vacancy prediction system of claim 1, wherein, The data processing unit is further configured to: Abnormal data points in the current change data are identified by using a sliding window; If the proportion of abnormal data points in a single sliding window is less than the abnormal threshold, the abnormal data points are corrected using an abnormality correction algorithm to obtain the processed current change data.

4. The charging station vacancy prediction system of claim 2, wherein, The data processing unit is further configured to: The time features are extracted from the current change data to obtain the time features; The density of charging piles in the area where the charging piles are located is determined based on the location information. The environmental characteristic data of the area where the charging pile is located is obtained, wherein the processed current change data includes: the time characteristics, the charging pile density, and the environmental characteristic data.

5. The charging station vacancy prediction system of claim 4, wherein, The prediction server is also used for: The feature weights of the prediction model are adjusted based on the environmental feature data to obtain the adjusted prediction model, wherein the prediction model is set in the prediction server and the prediction model is a long short-term memory network model. Obtain the historical current data of the charging pile, and obtain the historical time characteristics corresponding to the historical current data; The adjusted prediction model is used to process the historical current data, the historical time characteristics, and the processed current change data to obtain the future idle time period and the idle window duration.

6. The charging station vacancy prediction system of claim 5, wherein, The prediction server is also used for: Obtain the real-time location information of the target object; Based on the real-time location information, the future idle time periods and idle window durations of the multiple charging piles, the multiple charging piles are recommended and sorted to obtain a recommendation list, wherein the recommendation list includes the sorted multiple charging piles; The recommendation list is sent to the target object.

7. A charging pile idle state prediction method based on a timing model, characterized in that, The charging pile idle state prediction system based on a time-series model, as described in any one of claims 1 to 6, comprises: Collect current variation data of charging piles; The current change data is subjected to anomaly processing to obtain processed current change data; The future idle periods and idle window durations of the charging pile are predicted based on the processed current change data.

8. A computer readable storage medium, characterized in that, The computer readable storage medium comprises a stored program, wherein the program performs the method of claim 7 when executed. 9.An electronic device comprising a memory and a processor, the electronic device characterized by, The memory stores a computer program, and the processor is configured to execute the method of claim 7 by using the computer program.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of claim 7.

Images