A charging station control data management system and method based on artificial intelligence

Abstract

The present invention discloses a charging station control data management system and method based on artificial intelligence, which belongs to the field of charging station control technology. The real-time load data of the charging warehouse and the charging pile are stored, and the load data in each time segment is recorded by dividing the time segment to generate a cyclic digital label; the load fluctuation characteristics of each time segment in different cycle time periods are expressed in the form of a load curve function, and a load curve associated sample pair is established; the time segment is used as a unified scale, and the load curve is fitted in the same time segment, and the cyclic digital label is divided into two categories of normal and abnormal, and the alternating change characteristics of the normal and abnormal cyclic digital label are obtained; the strong and weak correlation between different charging methods is analyzed, and the charging control decision of the charging station is output; thereby helping the charging station to adapt to the changes in the demand for charging methods, while balancing the power demand of the charging station, reducing the load risk of the charging station.

Description

Technical Field

[0001] The present invention relates to the technical field of charging station control, and specifically to an artificial intelligence-based charging station control data management system and method. Background Art

[0002] Charging stations refer to places that provide charging services for electric vehicles, including charging piles and charging warehouse facilities; charging piles are used to directly charge the power batteries on new energy vehicles without the need for replacement. Charging piles have a fast charging speed and can charge electric vehicles in a short time. Charging piles have a large charging power and can meet the needs of fast charging of electric vehicles. However, due to the large charging power of charging piles, if multiple electric vehicles are charged at the same time, it will cause a large load pressure on the power grid; charging warehouses are used to charge power batteries replaced from new energy vehicles. The charging speed of charging warehouses is slow, and it takes a long time to charge electric vehicles. At the same time, the charging power of charging warehouses is small, which can reduce the load pressure on the power grid. However, due to the slow charging speed of charging warehouses, if multiple electric vehicles need to be charged, it will take a long time to complete the charging;

[0003] With the popularization of new energy vehicles and the changing demands for charging methods, increasingly stringent requirements are being placed on the management of charging stations. Future charging stations not only need to adapt to the changes in charging method demands, but also reduce the load risk of charging stations while balancing the power requirements of different charging methods. Summary of the invention

[0004] The purpose of the present invention is to provide a charging station control data management system and method based on artificial intelligence to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] A charging station control data management system based on artificial intelligence, the system includes: a data identification module, a data integration module, a data analysis and processing module and an optimization control module;

[0007] The data identification module is used to cache the real-time load data of the charging warehouse end and the real-time load data of the charging pile end in the charging station in real time; divide the time segment and add a cyclic digital label to each time segment;

[0008] The data integration module is used to establish a load curve sample library, convert the load curve generated in each time segment into a data sample according to the cyclic digital label, and establish a load curve associated sample pair;

[0009] The data analysis and processing module simulates the overall load curve function of the charging station according to the load curve association sample pairs and combines the real-time load data of the charging warehouse and the real-time load data of the charging pile, and divides the cyclic digital tags into abnormal and normal states to obtain a cyclic monitoring sequence set; according to the cyclic monitoring sequence set, analyzes the association value between the two charging modes of the charging warehouse and the charging pile;

[0010] The optimization and control module supervises the charging behavior in the time segment according to the correlation between the charging modes, outputs the correlation between the charging modes in different time segments, and guides the charging station to adopt different charging modes in different time segments based on the output correlation.

[0011] Furthermore, the data identification module also includes a data cache unit and a digital label unit;

[0012] The data cache unit is used to cache the charging warehouse usage status information and the charging pile usage status information of the charging station in real time; the charging warehouse is used to store the power battery replaced from the new energy vehicle and charge the power battery replaced from the new energy vehicle, and the charging pile is used to directly charge the power battery on the new energy vehicle without replacement; the charging warehouse usage status information and the charging pile usage status information respectively record the real-time load data corresponding to each power battery;

[0013] The digital label unit is used to divide the time range within a day into k time segments, and use the day as the cycle time period to cyclically record the real-time load data corresponding to the charging warehouse and the charging pile in each time segment, and record any time segment as , where s represents the serial number of the time segment, then for the wth cycle time period, the time segment Additional cycle digital labels, denoted by ,Right now Represents a time segment The corresponding additional cycle digital label is in the wth cycle time period.

[0014] Furthermore, the data integration module also includes a classification and aggregation unit and a data association unit;

[0015] The classification and aggregation unit is used to establish a load curve data sample library, which stores two load curve data samples corresponding to each cycle digital label, and the two load curves include a charging warehouse load curve and a charging pile load curve; according to the cycle digital label, the time segment All real-time load data generated in the w-th cycle time period are classified and summarized, and converted into the charging warehouse load curve sample and the charging pile load curve data sample according to the data category corresponding to the real-time load data, wherein the data category includes the charging warehouse usage status information and the charging pile usage status information, and the classification and summary is to summarize according to the data category;

[0016] The data association unit establishes a load curve association sample pair for two load curve data samples formed under the same cyclic digital label according to the cyclic digital label, which is recorded as ,in, Indicates that the digital label is in the loop The charging warehouse load curve function formed below is: Indicates that the digital label is in the loop The charging pile load curve function is formed below.

[0017] Furthermore, the data analysis and processing module also includes a data analysis unit and a data processing unit;

[0018] The data analysis unit associates sample pairs according to the load curve to simulate the cycle digital label The overall load curve function of the charging station at the time is recorded as ,and ; Extract the overall load curve function of the charging station The peak data and trough data in the waveform are obtained, and the difference between the connected peak data and trough data is obtained. The number of differences obtained under the loop is obtained in the digital label The average of the differences under the preset curve fluctuation threshold, if in the cycle digital label If the average difference under the curve fluctuation threshold is greater than or equal to the curve fluctuation threshold, the cyclic digital label Mark the exception, otherwise the loop number label Perform normal marking;

[0019] The data processing unit is used to sort the normal and abnormal marking results according to the order of the cyclic digital labels to obtain a set of cyclic monitoring sequences, which is recorded as , where Q represents the serial number of the Qth cycle time period; obtain the number of cycle digital labels marked as abnormal between the xth cycle digital label marked as normal and the x+1th cycle digital label marked as normal in the cycle monitoring sequence set, and record it as the number of xth association groups , and calculate the correlation value between the two charging modes of charging warehouse and charging pile in the same time segment. The specific calculation formula is as follows:

[0020]

[0021] in, Indicates the time segment The correlation value between the two charging methods of charging warehouse and charging pile, represents the number of the x+1th association group, and y represents the total number of association groups in the cyclic monitoring sequence set.

[0022] Furthermore, the optimization and control module also includes an associated behavior optimization unit and a control decision unit;

[0023] The association behavior optimization unit is used to preset the association threshold. If If the correlation value between the two charging modes of the charging warehouse and the charging pile is greater than or equal to the correlation threshold, the two charging modes of the charging warehouse and the charging pile are combined in the time segment. The relationship below is marked as a strong relationship, otherwise it is marked as a weak relationship;

[0024] The control decision unit is used to guide the charging behavior of the charging station. If the two charging modes of the charging warehouse and the charging pile are in the time segment The correlation under this condition is strong, then the charging station pair in the time segment For charging users who come within the day, it is recommended to charge at the charging warehouse first, otherwise charging at the charging pile is recommended.

[0025] A charging station control data management method based on artificial intelligence, the method comprises the following steps:

[0026] Step S100: caching the real-time load data of the charging warehouse and the real-time load data of the charging pile in the charging station in real time; dividing the time segments, and adding a cyclic digital label to each time segment;

[0027] Step S200: Establish a load curve sample library, convert the load curve generated in each time segment into a data sample according to the cyclic digital label, and establish a load curve associated sample pair;

[0028] Step S300: According to the load curve association sample pair, combined with the real-time load data of the charging warehouse and the real-time load data of the charging pile, the overall load curve function of the charging station is simulated, and the cyclic digital tags are divided into abnormal and normal states to obtain a cyclic monitoring sequence set; according to the cyclic monitoring sequence set, the association value between the two charging modes of the charging warehouse and the charging pile is analyzed;

[0029] Step S400: According to the association between charging modes, the charging behavior in the time segment is supervised, the association relationship between the charging modes in different time segments is output, and the charging station is guided to adopt different charging modes in different time segments based on the output association relationship.

[0030] Furthermore, the specific implementation process of step S100 includes:

[0031] Step S101: real-time caching of the charging station's charging warehouse usage status information and the charging station's charging pile usage status information; the charging warehouse is used to store and charge the power batteries replaced from the new energy vehicle, and the charging pile is used to directly charge the power batteries on the new energy vehicle without replacement; the charging warehouse usage status information and the charging pile usage status information respectively record the real-time load data corresponding to each power battery;

[0032] Step S102: Divide the time range within a day into k time segments, and use the day as the cycle time period to cyclically record the real-time load data corresponding to the charging warehouse and the charging pile in each time segment, and record any time segment as , where s represents the serial number of the time segment, then for the wth cycle time period, the time segment Additional cycle digital labels, denoted by ,Right now Represents a time segment The corresponding additional cycle digital label is added during the wth cycle time period.

[0033] According to the above method, charging stations generally include operating only through the charging warehouse battery replacement mode, operating only through the charging pile charging mode, or operating through the integration of the two modes of charging warehouse battery replacement mode and charging pile charging mode; the main feature of the charging warehouse battery replacement mode is that the charging warehouse is used to charge the power battery replaced from the new energy vehicle. The charging speed of the charging warehouse is slow, and it takes a long time to charge the electric vehicle. At the same time, the charging power of the charging warehouse is small, which can reduce the load pressure on the power grid. However, due to the slow charging speed of the charging warehouse, if multiple electric vehicles need to be charged, it will take a long time to complete the charging; the main feature of the charging pile charging mode is that the charging pile is used to directly charge the power battery on the new energy vehicle without the need for replacement. The charging pile has a fast charging speed and can charge the electric vehicle in a short time. When charging an electric vehicle, the charging power of the charging pile is relatively large, which can meet the needs of fast charging of electric vehicles. However, due to the large charging power of the charging pile, if multiple electric vehicles are charging at the same time, it will cause a large load pressure on the power grid; the main feature of the fusion of the two modes introduced in this application is reflected in balancing the power requirements of the charging station for the charging warehouse and the charging pile, so as to take the comprehensive management of the charging station as the first perspective and the user's needs as the second perspective, while meeting the user's charging needs, the risk of load fluctuations in the charging station is optimized and reduced; the invention of this application first stores the real-time load data of the charging warehouse and the charging pile by caching, and records the load data in each time segment by dividing the time segment and taking the day as the cycle time period, so as to facilitate the analysis of load regularity characteristics and the analysis of user charging habits characteristics in this micro-quantified manner.

[0034] Furthermore, the specific implementation process of step S200 includes:

[0035] Step S201: Establish a load curve data sample library, in which two load curve data samples corresponding to each cycle digital label are stored, and the two load curves include a charging warehouse load curve and a charging pile load curve; according to the cycle digital label, the time segment All real-time load data generated in the w-th cycle time period are classified and summarized, and converted into the charging warehouse load curve sample and the charging pile load curve data sample according to the data category corresponding to the real-time load data, wherein the data category includes the charging warehouse usage status information and the charging pile usage status information, and the classification and summary is to summarize according to the data category;

[0036] Step S202: Based on the cyclic digital label, for two load curve data samples formed under the same cyclic digital label, a load curve associated sample pair is established, which is recorded as ,in, Indicates that the digital label is in the loop The charging warehouse load curve function formed below is: Indicates that the digital label is in the loop The charging pile load curve function is formed below.

[0037] According to the above method, the invention of the present application analyzes the load law characteristics for charging stations that operate in a mixed mode. The premise of the analysis is to be carried out under a unified scale, that is, in the form of a load curve function, to show the load fluctuation characteristics of each time segment under different cycle time periods. After dividing different time segments with days as the cycle time period unit, the real-time load data of different charging methods, that is, charging pile charging and charging warehouse charging, in each time segment within each cycle time period are classified and summarized to form a charging warehouse load curve function and a charging pile load curve function; the load curve refers to the load change required when accessing the power grid within a certain period of time, usually drawn with time as the horizontal axis and power as the vertical axis. The load curve function of the present application is fitted by the load change at the charging port of the two charging methods in each time segment within different cycle time periods. The load law in each time segment within different cycle time periods is characterized by the load curve function. For w cycle time periods, the load curves under w time segments are obtained.

[0038] Furthermore, the specific implementation process of step S300 includes:

[0039] Step S301: Associating sample pairs according to load curves to simulate circulating digital labels The overall load curve function of the charging station at the time is recorded as ,and ; Extract the overall load curve function of the charging station The peak data and trough data in the waveform are obtained, and the difference between the connected peak data and trough data is obtained. The number of differences obtained under the loop is obtained in the digital label The average of the differences under the preset curve fluctuation threshold, if in the cycle digital label If the average difference under the curve fluctuation threshold is greater than or equal to the curve fluctuation threshold, the cyclic digital label Mark the exception, otherwise the loop number label Perform normal marking;

[0040] Step S302: Sort the normal and abnormal labeling results according to the order of the cyclic digital labels to obtain a set of cyclic monitoring sequences, recorded as , where Q represents the serial number of the Qth cycle time period; obtain the number of cycle digital labels marked as abnormal between the xth cycle digital label marked as normal and the x+1th cycle digital label marked as normal in the cycle monitoring sequence set, and record it as the number of xth association groups , and calculate the correlation value between the two charging modes of charging warehouse and charging pile in the same time segment. The specific calculation formula is as follows:

[0041]

[0042] in, Indicates the time segment The correlation value between the two charging methods of charging warehouse and charging pile, represents the number of the x+1th association group, and y represents the total number of association groups in the cyclic monitoring sequence set.

[0043] According to the above method, the overall load of the charging station in this application includes the load of the charging warehouse and the charging pile. After dividing the time segments, the time segments are used as a unified scale. In the same time segment, the load curves of the charging warehouse and the charging pile are fitted to obtain the overall load curve of the charging station; combined with the alternating change characteristics of peaks and troughs, the cyclic digital labels are divided into normal and abnormal categories. Under normal circumstances, the overall load curve of the charging station under the cyclic digital label fluctuates less, and the load risk of the charging station under the cyclic digital label is low. However, when an abnormal situation occurs, it means that the overall load curve of the charging station under the cyclic digital label fluctuates more, and the load risk of the charging station under the cyclic digital label is higher; at the same time, under different cyclic digital labels, the characteristics of such abnormal and normal situations may be different. The different reasons are that people's charging needs may change. Changes occur, and then, for the same time segment, normal and abnormal alternations are shown in different cycle time periods. This change reflects the load law of the charging station and the charging habits of users in the same time segment. Then, a set of cyclic monitoring sequences is generated, and the correlation value between the two charging methods of charging warehouse and charging pile is obtained through the correlation value calculation formula. If the fluctuation under the cyclic digital label is normal, the correlation characteristics of the two charging methods are obvious, which is obviously reflected in the fact that the overall load curve of the charging station under the combination of low-power charging of the charging warehouse and high-power charging of the charging pile has a small fluctuation, then the two charging methods are strongly combined, otherwise the combination is weak, and the alternation of strong and weak combination is accompanied by the alternation of normal and abnormal changes, thereby analyzing the strong and weak correlation between the two charging methods. The larger the correlation value, the stronger the correlation between the two charging methods, that is, the strong correlation.

[0044] Furthermore, the specific implementation process of step S400 includes:

[0045] Step S401: preset the correlation threshold. If If the correlation value between the two charging modes of the charging warehouse and the charging pile is greater than or equal to the correlation threshold, the two charging modes of the charging warehouse and the charging pile are combined in the time segment. The relationship below is marked as a strong relationship, otherwise it is marked as a weak relationship;

[0046] Step S402: If the charging methods of the charging station and the charging pile are in the same time segment The correlation relationship under this condition is strong, then the charging station pair in the time segment For charging users who arrive within 24 hours, charging at the charging warehouse is recommended first, otherwise charging at the charging pile is recommended;

[0047] According to the above method, in a strong correlation, since the charging power of the charging pile is relatively large, in order to avoid the load risk of the charging station, charging in the charging warehouse with a smaller charging power is recommended first; on the contrary, in a weak correlation, charging with the charging pile is recommended first; thereby, the power requirements of the charging station for both the charging warehouse and the charging pile are balanced, and while meeting the charging needs of users, the risk of load fluctuations in the charging station is optimized and reduced.

[0048] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: in an artificial intelligence-based charging station control data management system and method provided by the present invention, real-time load data of charging warehouses and charging piles are stored, and the load data in each time segment is recorded by dividing the time segment to generate a cyclic digital label; the load fluctuation characteristics of each time segment in different cycle time periods are expressed in the form of a load curve function, and a load curve associated sample pair is established; with the time segment as a unified scale, the load curve is fitted in the same time segment, and the cyclic digital labels are divided into normal and abnormal categories to obtain the alternating change characteristics of normal and abnormal cyclic digital labels; the strong and weak correlations between different charging methods are analyzed, and the charging control decision of the charging station is output; thereby helping the charging station to adapt to changes in charging method requirements, while balancing the power requirements of the charging station, reducing the load risk of the charging station. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0050] Figure 1 It is a structural schematic diagram of a charging station control data management system based on artificial intelligence of the present invention;

[0051] Figure 2It is a schematic diagram of the steps of a charging station control data management method based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] See also Figure 1-Figure 2 , the present invention provides a technical solution:

[0054] See also Figure 1 In the first embodiment of the present invention, a charging station control data management system based on artificial intelligence is provided, the system comprising: a data identification module, a data integration module, a data analysis and processing module and an optimization control module;

[0055] The data identification module is used to cache the real-time load data of the charging warehouse and the charging pile in the charging station in real time; divide the time segment and add a cyclic digital label to each time segment;

[0056] Wherein, the data identification module also includes a data cache unit and a digital label unit;

[0057] A data cache unit is used to cache the use status information of the charging warehouse and the use status information of the charging pile of the charging station in real time; the charging warehouse is used to store the power battery replaced from the new energy vehicle and charge the power battery replaced from the new energy vehicle, and the charging pile is used to directly charge the power battery on the new energy vehicle without replacement; the use status information of the charging warehouse and the use status information of the charging pile respectively record the real-time load data corresponding to each power battery;

[0058] The digital label unit is used to divide the time range of a day into k time segments, and use the day as the cycle time period to cyclically record the real-time load data corresponding to the charging warehouse and charging pile in each time segment, and record any time segment as , where s represents the serial number of the time segment, then for the wth cycle time period, the time segment Additional cycle digital labels, denoted by ,Right now Represents a time segment The corresponding additional cycle digital label in the wth cycle time period;

[0059] The data integration module is used to establish a load curve sample library, convert the load curve generated in each time segment into a data sample according to the cyclic digital label, and establish a load curve associated sample pair;

[0060] Among them, the data integration module also includes a classification and summary unit and a data association unit;

[0061] The classification and summary unit is used to establish a load curve data sample library. The load curve data sample library stores two load curve data samples corresponding to each cycle digital label. The two load curves include the charging warehouse load curve and the charging pile load curve. All real-time load data generated in the wth cycle time period are classified and summarized, and converted into charging warehouse load curve samples and charging pile load curve data samples according to the data categories corresponding to the real-time load data, wherein the data categories include charging warehouse usage status information and charging pile usage status information, and classified and summarized, that is, summarized according to the data categories;

[0062] The data association unit, based on the cyclic digital label, establishes a load curve association sample pair for the two load curve data samples formed under the same cyclic digital label, which is recorded as ,in, Indicates that the digital label is in the loop The charging warehouse load curve function formed below is: Indicates that the digital label is in the loop The charging pile load curve function formed under

[0063] The data analysis and processing module associates sample pairs according to the load curve, and combines the real-time load data of the charging warehouse and the real-time load data of the charging pile to simulate the overall load curve function of the charging station, and divides the cyclic digital tags into abnormal and normal states to obtain a set of cyclic monitoring sequences; based on the set of cyclic monitoring sequences, the correlation value between the two charging modes of the charging warehouse and the charging pile is analyzed;

[0064] Wherein, the data analysis and processing module also includes a data analysis unit and a data processing unit;

[0065] The data analysis unit associates sample pairs according to the load curve, simulating the cycle of digital labels The overall load curve function of the charging station at the time is recorded as ,and ; Extract the overall load curve function of the charging station The peak data and trough data in the waveform are obtained, and the difference between the connected peak data and trough data is obtained. The number of differences obtained under the loop is obtained in the digital label The average of the differences under the preset curve fluctuation threshold, if in the cycle digital label If the average difference under the curve fluctuation threshold is greater than or equal to the curve fluctuation threshold, the cyclic digital label Mark the exception, otherwise the loop number label Perform normal marking;

[0066] The data processing unit is used to sort the normal and abnormal labeling results according to the order of the cyclic digital labels to obtain a set of cyclic monitoring sequences, which is recorded as , where Q represents the serial number of the Qth cycle time period; obtain the number of cycle digital labels marked as abnormal between the xth cycle digital label marked as normal and the x+1th cycle digital label marked as normal in the cycle monitoring sequence set, and record it as the number of xth association groups , and calculate the correlation value between the two charging modes of charging warehouse and charging pile in the same time segment. The specific calculation formula is as follows:

[0067]

[0068] in, Indicates the time segment The correlation value between the two charging methods of charging warehouse and charging pile, represents the number of the x+1th association group, and y represents the total number of association groups in the cyclic monitoring sequence set;

[0069] The optimization control module supervises the charging behavior in the time segment according to the correlation between the charging modes, outputs the correlation between the charging modes in different time segments, and guides the charging station to adopt different charging modes in different time segments based on the output correlation.

[0070] Among them, the optimization and control module also includes an associated behavior optimization unit and a control decision unit;

[0071] The correlation behavior optimization unit is used to preset the correlation threshold. If the correlation value between the two charging modes of the charging warehouse and the charging pile is greater than or equal to the correlation threshold, the two charging modes of the charging warehouse and the charging pile are combined in the time segment. The relationship below is marked as a strong relationship, otherwise it is marked as a weak relationship;

[0072] The control decision unit is used to guide the charging behavior of the charging station. If the charging warehouse and charging pile are used in the same time segment, The correlation under this condition is strong, then the charging station pair in the time segment For charging users who come within the day, it is recommended to charge at the charging warehouse first, otherwise charging at the charging pile is recommended.

[0073] See also Figure 2 In the second embodiment, a charging station control data management method based on artificial intelligence is provided, and the method includes the following steps:

[0074] Step S100: caching the real-time load data of the charging warehouse and the real-time load data of the charging pile in the charging station in real time; dividing the time segments, and adding a cyclic digital label to each time segment;

[0075] Specifically, the use status information of the charging warehouse and the use status information of the charging pile of the charging station are cached in real time; the charging warehouse is used to store the power batteries replaced from the new energy vehicles and charge the power batteries replaced from the new energy vehicles, and the charging pile is used to directly charge the power batteries on the new energy vehicles without replacement; the use status information of the charging warehouse and the use status information of the charging pile respectively record the real-time load data corresponding to each power battery;

[0076] The time range of a day is divided into k time segments, and the real-time load data of the charging warehouse and charging pile corresponding to each time segment is recorded cyclically with the day as the cycle time period. Any time segment is recorded as , where s represents the serial number of the time segment, then for the wth cycle time period, the time segment Additional cycle digital labels, denoted by ,Right now Represents a time segment The corresponding additional cycle digital label in the wth cycle time period;

[0077] For example, if it takes about 3 hours to fully charge a charging pile, the time range within a day can be divided into segments with 3 hours as the length of the time segment; or, according to people's charging habits, if charging is generally completed after 2 hours, the time range can be divided into segments with 2 hours as the length;

[0078] Step S200: Establish a load curve sample library, convert the load curve generated in each time segment into a data sample according to the cyclic digital label, and establish a load curve associated sample pair;

[0079] Specifically, a load curve data sample library is established, and the load curve data sample library stores two load curve data samples corresponding to each cycle digital label, and the two load curves include a charging warehouse load curve and a charging pile load curve; according to the cycle digital label, the time segment All real-time load data generated in the wth cycle time period are classified and summarized, and converted into charging warehouse load curve samples and charging pile load curve data samples according to the data categories corresponding to the real-time load data, wherein the data categories include charging warehouse usage status information and charging pile usage status information, and classified and summarized, that is, summarized according to the data categories;

[0080] According to the cyclic digital label, for the two load curve data samples formed under the same cyclic digital label, a load curve associated sample pair is established, which is recorded as ,in, Indicates that the digital label is in the loop The charging warehouse load curve function formed below is: Indicates that the digital label is in the loop The charging pile load curve function formed under

[0081] Step S300: According to the load curve association sample pair, combined with the real-time load data of the charging warehouse and the real-time load data of the charging pile, the overall load curve function of the charging station is simulated, and the cyclic digital tags are divided into abnormal and normal states to obtain a cyclic monitoring sequence set; according to the cyclic monitoring sequence set, the association value between the two charging modes of the charging warehouse and the charging pile is analyzed;

[0082] Specifically, according to the load curve, the sample pairs are associated to simulate the cycle digital label The overall load curve function of the charging station at the time is recorded as ,and ; Extract the overall load curve function of the charging station The peak data and trough data in the waveform are obtained, and the difference between the connected peak data and trough data is obtained. The number of differences obtained under the loop is obtained in the digital label The average of the differences under the preset curve fluctuation threshold, if in the cycle digital label If the average difference under the curve fluctuation threshold is greater than or equal to the curve fluctuation threshold, the cyclic digital label Mark the exception, otherwise the loop number label Perform normal marking;

[0083] According to the order of the cyclic digital labels, the normal and abnormal labeling results are sorted to obtain the cyclic monitoring sequence set, which is recorded as , where Q represents the serial number of the Qth cycle time period; obtain the number of cycle digital labels marked as abnormal between the xth cycle digital label marked as normal and the x+1th cycle digital label marked as normal in the cycle monitoring sequence set, and record it as the number of xth association groups , and calculate the correlation value between the two charging modes of charging warehouse and charging pile in the same time segment. The specific calculation formula is as follows:

[0084]

[0085] in, Indicates the time segment The correlation value between the two charging methods of charging warehouse and charging pile, represents the number of the x+1th association group, and y represents the total number of association groups in the cyclic monitoring sequence set;

[0086] Step S400: According to the association between charging modes, the charging behavior in the time segment is supervised, the association relationship between the charging modes in different time segments is output, and the charging station is guided to adopt different charging modes in different time segments based on the output association relationship;

[0087] Specifically, preset correlation thresholds, if in the time segment If the correlation value between the two charging modes of the charging warehouse and the charging pile is greater than or equal to the correlation threshold, the two charging modes of the charging warehouse and the charging pile are combined in the time segment. The relationship below is marked as a strong relationship, otherwise it is marked as a weak relationship;

[0088] If the charging methods of the charging station and the charging pile are in the same time segment The correlation under this condition is strong, then the charging station pair in the time segment For charging users who come within the day, it is recommended to charge at the charging warehouse first, otherwise charging at the charging pile is recommended.

[0089] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0090] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A charging station control data management method based on artificial intelligence, characterized in that: The method comprises the following steps: Step S100: caching the real-time load data of the charging warehouse and the real-time load data of the charging pile in the charging station in real time; dividing the time segments, and adding a cyclic digital label to each time segment; Step S200: Establish a load curve sample library, convert the load curve generated in each time segment into a data sample according to the cyclic digital label, and establish a load curve associated sample pair; Step S300: According to the load curve association sample pair, combined with the real-time load data of the charging warehouse and the real-time load data of the charging pile, the overall load curve function of the charging station is simulated, and the cyclic digital tags are divided into abnormal and normal states to obtain a cyclic monitoring sequence set; according to the cyclic monitoring sequence set, the association value between the two charging modes of the charging warehouse and the charging pile is analyzed; Step S400: According to the association between charging modes, the charging behavior in the time segment is supervised, the association relationship between the charging modes in different time segments is output, and the charging station is guided to adopt different charging modes in different time segments based on the output association relationship; The specific implementation process of step S100 includes: Step S101: real-time caching of the charging station's charging warehouse usage status information and the charging station's charging pile usage status information; the charging warehouse is used to store and charge the power batteries replaced from the new energy vehicle, and the charging pile is used to directly charge the power batteries on the new energy vehicle without replacement; the charging warehouse usage status information and the charging pile usage status information respectively record the real-time load data corresponding to each power battery; Step S102: Divide the time range within a day into k time segments, and use the day as the cycle time period to cyclically record the real-time load data corresponding to the charging warehouse and the charging pile in each time segment, and record any time segment as , where s represents the serial number of the time segment, then for the wth cycle time period, the time segment Additional cycle digital labels, denoted by ,Right now Represents a time segment The corresponding additional cycle digital label in the wth cycle time period; The specific implementation process of step S200 includes: Step S201: Establish a load curve data sample library, in which two load curve data samples corresponding to each cycle digital label are stored, and the two load curves include a charging warehouse load curve and a charging pile load curve; according to the cycle digital label, the time segment All real-time load data generated in the w-th cycle time period are classified and summarized, and converted into the charging warehouse load curve sample and the charging pile load curve data sample according to the data category corresponding to the real-time load data, wherein the data category includes the charging warehouse usage status information and the charging pile usage status information, and the classification and summary is to summarize according to the data category; Step S202: Based on the cyclic digital label, for two load curve data samples formed under the same cyclic digital label, a load curve associated sample pair is established, which is recorded as ,in, Indicates that the digital label is in the loop The charging warehouse load curve function formed below is: Indicates that the digital label is in the loop The charging pile load curve function formed under The specific implementation process of step S300 includes: Step S301: Associating sample pairs according to load curves to simulate circulating digital labels The overall load curve function of the charging station at the time is recorded as ,and ; Extract the overall load curve function of the charging station The peak data and trough data in the waveform are obtained, and the difference between the connected peak data and trough data is obtained. The number of differences obtained under the loop is obtained in the digital label The average of the differences under the preset curve fluctuation threshold, if in the cycle digital label If the average difference under the curve fluctuation threshold is greater than or equal to the curve fluctuation threshold, the cyclic digital label Mark the exception, otherwise the loop number label Perform normal marking; Step S302: Sort the normal and abnormal labeling results according to the order of the cyclic digital labels to obtain a set of cyclic monitoring sequences, recorded as , where Q represents the serial number of the Qth cycle time period; obtain the number of cycle digital labels marked as abnormal between the xth cycle digital label marked as normal and the x+1th cycle digital label marked as normal in the cycle monitoring sequence set, and record it as the number of xth association groups , and calculate the correlation value between the two charging methods of charging warehouse and charging pile in the same time segment. The specific calculation formula is as follows: ； in, Indicates the time segment The correlation value between the two charging methods of charging warehouse and charging pile, represents the number of the x+1th association group, and y represents the total number of association groups in the cyclic monitoring sequence set; The specific implementation process of step S400 includes: Step S401: preset the correlation threshold. If If the correlation value between the two charging modes of the charging warehouse and the charging pile is greater than or equal to the correlation threshold, the two charging modes of the charging warehouse and the charging pile are combined in the time segment. The relationship below is marked as a strong relationship, otherwise it is marked as a weak relationship; Step S402: If the charging methods of the charging station and the charging pile are in the same time segment The correlation under this condition is strong, then the charging station pair in the time segment For charging users who come within the day, it is recommended to charge at the charging warehouse first, otherwise charging at the charging pile is recommended.

2. An artificial intelligence-based charging station control data management system, characterized in that: The system includes: a data identification module, a data integration module, a data analysis and processing module and an optimization and control module; The data identification module is used to cache the real-time load data of the charging warehouse end and the real-time load data of the charging pile end in the charging station in real time; divide the time segment and add a cyclic digital label to each time segment; The data integration module is used to establish a load curve sample library, convert the load curve generated in each time segment into a data sample according to the cyclic digital label, and establish a load curve associated sample pair; The data analysis and processing module simulates the overall load curve function of the charging station according to the load curve association sample pairs and combines the real-time load data of the charging warehouse and the real-time load data of the charging pile, and divides the cyclic digital tags into abnormal and normal states to obtain a cyclic monitoring sequence set; according to the cyclic monitoring sequence set, analyzes the association value between the two charging modes of the charging warehouse and the charging pile; The optimization control module monitors the charging behavior in the time segment according to the correlation between the charging modes, outputs the correlation between the charging modes in different time segments, and guides the charging station to adopt different charging modes in different time segments based on the output correlation. The data identification module also includes a data cache unit and a digital label unit; The data cache unit is used to cache the charging warehouse usage status information and the charging pile usage status information of the charging station in real time; the charging warehouse is used to store the power battery replaced from the new energy vehicle and charge the power battery replaced from the new energy vehicle, and the charging pile is used to directly charge the power battery on the new energy vehicle without replacement; the real-time load data corresponding to each power battery is recorded in the charging warehouse usage status information and the charging pile usage status information respectively; The digital label unit is used to divide the time range within a day into k time segments, and use the day as the cycle time period to cyclically record the real-time load data corresponding to the charging warehouse and the charging pile in each time segment, and record any time segment as , where s represents the serial number of the time segment, then for the wth cycle time period, the time segment Additional cycle digital labels, denoted by ,Right now Represents a time segment The corresponding additional cycle digital label in the wth cycle time period; The data integration module also includes a classification and summary unit and a data association unit; The classification and aggregation unit is used to establish a load curve data sample library, which stores two load curve data samples corresponding to each cycle digital label, and the two load curves include a charging warehouse load curve and a charging pile load curve; according to the cycle digital label, the time segment All real-time load data generated in the w-th cycle time period are classified and summarized, and converted into the charging warehouse load curve sample and the charging pile load curve data sample according to the data category corresponding to the real-time load data, wherein the data category includes the charging warehouse usage status information and the charging pile usage status information, and the classification and summary is to summarize according to the data category; The data association unit establishes a load curve association sample pair for two load curve data samples formed under the same cyclic digital label according to the cyclic digital label, which is recorded as ,in, Indicates that the digital label is in the loop The charging warehouse load curve function formed below is: Indicates that the digital label is in the loop The charging pile load curve function formed under The data analysis and processing module also includes a data analysis unit and a data processing unit; The data analysis unit associates sample pairs according to the load curve to simulate the cycle digital label The overall load curve function of the charging station at the time is recorded as ,and ; Extract the overall load curve function of the charging station The peak data and trough data in the waveform are obtained, and the difference between the connected peak data and trough data is obtained. The number of differences obtained under the loop is obtained in the digital label The average of the differences under the preset curve fluctuation threshold, if in the cycle digital label If the average difference under the curve fluctuation threshold is greater than or equal to the curve fluctuation threshold, the cyclic digital label Mark the exception, otherwise the loop number label Perform normal marking; The data processing unit is used to sort the normal and abnormal marking results according to the order of the cyclic digital labels to obtain a set of cyclic monitoring sequences, which is recorded as , where Q represents the serial number of the Qth cycle time period; obtain the number of cycle digital labels marked as abnormal between the xth cycle digital label marked as normal and the x+1th cycle digital label marked as normal in the cycle monitoring sequence set, and record it as the number of xth association groups , and calculate the correlation value between the two charging modes of charging warehouse and charging pile in the same time segment. The specific calculation formula is as follows: ； in, Indicates the time segment The correlation value between the two charging methods of charging warehouse and charging pile, represents the number of the x+1th association group, and y represents the total number of association groups in the cyclic monitoring sequence set; The optimization and control module also includes an associated behavior optimization unit and a control decision unit; The association behavior optimization unit is used to preset the association threshold. If If the correlation value between the two charging modes of the charging warehouse and the charging pile is greater than or equal to the correlation threshold, the two charging modes of the charging warehouse and the charging pile are combined in the time segment. The relationship below is marked as a strong relationship, otherwise it is marked as a weak relationship; The control decision unit is used to guide the charging behavior of the charging station. If the two charging modes of the charging warehouse and the charging pile are in the time segment The correlation under this condition is strong, then the charging station pair in the time segment For charging users who come within the day, it is recommended to charge at the charging warehouse first, otherwise charging at the charging pile is recommended.

Images