A charging data collection method and system for new energy vehicles
By mining network optimization and mapping output for the key information for charging data acquisition of new energy vehicles, data abnormal characterization parameters are generated, and the problem of low reliability of charging data acquisition is solved, and higher data acquisition accuracy and completeness are achieved.
Patent Information
- Application Number
- CN202311201624.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-09-18
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2043-09-18
AI Technical Summary
The existing charging data acquisition methods for new energy vehicles are relatively low, mainly due to random sampling or equal interval acquisition, which leads to incomplete and inaccurate data.
The pre-built key information mining network is used to optimize the network to form a target key information mining network. Through this network, the key information mining data to be processed is mined, the charging key information feature representation is output, and the data abnormal characterization parameters are generated through the mapping output network. , data collection and confirmation based on these parameters, and the charging data of the target user is determined.
The reliability of charging data acquisition is improved, and the accuracy and completeness of data acquisition is ensured through the high-precision processing capabilities of the neural network, which improves the shortcomings in the prior art.
Smart Images

Figure CN117113033B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and in particular to a charging data collection method and system for a new energy vehicle. Background Art
[0002] As new energy vehicles are increasingly used, the amount of charging data from new energy vehicles is very large, so it has a high application value. However, for a large amount of charging data, it is generally necessary to collect and confirm it. However, in the prior art, the collection of charging data is generally random sampling or equal interval collection, which results in low reliability. Summary of the invention
[0003] In view of this, an object of the present invention is to provide a charging data collection method and system for a new energy vehicle, so as to improve the reliability of charging data collection.
[0004] To achieve the above objectives, the embodiments of the present invention adopt the following technical solutions:
[0005] A charging data collection method for a new energy vehicle, comprising:
[0006] Performing network optimization operations on the pre-built key information mining network to form a target key information mining network corresponding to the key information mining network;
[0007] Through the target key information mining network, the received user charging data to be processed is subjected to key information mining operations to output corresponding charging key information feature representations;
[0008] Through a mapping output network, the key charging information feature representation is mapped and outputted to output data anomaly characterization parameters corresponding to the user charging data to be processed;
[0009] Based on the data anomaly characterization parameter, a data collection confirmation operation is performed on the charging data of the user to be processed, and when it is confirmed that the charging data of the user to be processed is collected, the charging data of the user to be processed is determined as the collected charging data of the target user.
[0010] In some preferred embodiments, in the above-mentioned charging data collection method for new energy vehicles, the step of performing network optimization operation on the pre-built key information mining network to form a target key information mining network corresponding to the key information mining network includes:
[0011] Determine an internal user charging data combination corresponding to the first user charging data, wherein the internal user charging data combination includes at least two internal user charging data of the first user charging data, and the at least two internal user charging data belong to charging data of the same user in different time periods;
[0012] Performing a key information mining operation on the user charging data in the internal user charging data combination through a key information mining network to form a corresponding charging key information cluster, wherein the charging key information cluster includes a charging key information feature representation corresponding to the first user charging data;
[0013] Based on the characteristic representation of the key charging information, an internal matching parameter between the internal user charging data of the first user charging data is analyzed, and an external matching parameter between the internal user charging data and the external user charging data is analyzed, the external user charging data belongs to the second user charging data other than the first user charging data, the first user charging data and the second user charging data belong to charging data of different users, and the first user charging data and the second user charging data belong to the same time period;
[0014] Analyzing a first mining error parameter corresponding to the first user charging data based on the internal matching parameter and the external matching parameter, wherein the first mining error parameter is used to reflect the degree of mining error between the internal matching parameter corresponding to the first user charging data and the external matching parameter corresponding to the first user charging data;
[0015] Based on the charging key information feature representation, a charging preference information identification operation is performed to output a first parameter of the predicted possibility of the internal user charging data in the charging preference information and a second parameter of the predicted possibility of the internal user charging data in the charging preference information;
[0016] Based on the first prediction possibility parameter and the second prediction possibility parameter, a second mining error parameter corresponding to the first user charging data is analyzed, where the second mining error parameter is used to reflect the degree of mining error between the first prediction possibility parameter of the first user charging data and the second prediction possibility parameter of the first user charging data;
[0017] Based on the first mining error parameter and the second mining error parameter, the key information mining network is subjected to a network optimization operation to form a target key information mining network.
[0018] In some preferred embodiments, in the above-mentioned charging data collection method for new energy vehicles, after the step of performing a charging preference information identification operation based on the charging key information feature representation to output a first parameter of the predicted possibility of the internal user charging data in the charging preference information and a second parameter of the predicted possibility of the internal user charging data in the charging preference information, the charging data collection method for new energy vehicles further includes:
[0019] Based on the charging key information feature representation, the internal user charging data is classified to form corresponding classification identification data and a classification possibility degree of the internal user charging data corresponding to the classification identification data;
[0020] Based on the classification possibility degree, determining a third mining error parameter of different first user charging data in the target classification identification data, where the different first user charging data refers to charging data of the same corresponding user but not belonging to the same time period;
[0021] The step of performing a network optimization operation on the key information mining network based on the first mining error parameter and the second mining error parameter to form a target key information mining network includes:
[0022] Based on the first mining error parameter, the second mining error parameter and the third mining error parameter, the key information mining network is subjected to a network optimization operation to form a target key information mining network.
[0023] In some preferred embodiments, in the above-mentioned charging data collection method for new energy vehicles, the step of classifying the internal user charging data based on the charging key information feature representation to form corresponding classification identification data and the classification possibility degree of the internal user charging data corresponding to the classification identification data includes:
[0024] Based on the charging key information feature representation, the internal user charging data is classified to form corresponding classification identification data;
[0025] Performing a comparative analysis operation on the classification identification data and the charging preference information to form a classification possibility degree of the internal user charging data corresponding to the classification identification data;
[0026] The step of classifying the internal user charging data based on the charging key information feature representation to form corresponding classification identification data includes:
[0027] Performing a logo comparison and analysis operation on the key charging information characteristic representation to form a key charging information characteristic representation with identification data;
[0028] Classify the key charging information feature representation with identification data to form a corresponding classification set;
[0029] A marking operation is performed on the identification data with the largest number in the classification set to form classification identification data corresponding to the classification set.
[0030] In some preferred embodiments, in the above-mentioned charging data collection method for new energy vehicles, the internal user charging data combination includes first internal user charging data and second internal user charging data; and the step of performing a charging preference information identification operation based on the charging key information feature representation to output a first parameter of the predicted possibility of the internal user charging data in the charging preference information and a second parameter of the predicted possibility of the internal user charging data in the charging preference information includes:
[0031] Based on the charging key information feature representation, the internal user charging data is classified to form an original classification possibility degree parameter of the internal user charging data in each charging preference information;
[0032] The original classification possibility degree parameter is decomposed to form a first prediction possibility parameter corresponding to each charging preference information of the first internal user charging data and a second prediction possibility parameter corresponding to each charging preference information of the second internal user charging data.
[0033] In some preferred embodiments, in the above-mentioned charging data collection method for new energy vehicles, the step of analyzing the second mining error parameter corresponding to the first user charging data based on the first prediction possibility parameter and the second prediction possibility parameter includes:
[0034] Analyzing first difference data of the predicted possibility of the first user charging data in the target charging preference information based on the first predicted possibility parameter of the target charging preference information and the second predicted possibility parameter of the target charging preference information;
[0035] Analyzing second difference data of the predicted possibility of the first user charging data between the target charging preference information and the other charging preference information based on the first predicted possibility parameter of the target charging preference information, the second predicted possibility parameter of the target charging preference information, the first predicted possibility parameter of other charging preference information, and the second predicted possibility parameter of other charging preference information;
[0036] Based on the first difference data of the predicted possibility and the second difference data of the predicted possibility, analyzing a second mining error parameter to be processed for the charging data of the first user, wherein the second mining error parameter to be processed is used to reflect the degree of mining error between the first parameter of the predicted possibility of the charging data of the first user and the second parameter of the predicted possibility of the charging data of the first user in the target charging preference information;
[0037] A corresponding second mining error parameter is obtained based on the second mining error parameter to be processed.
[0038] In some preferred embodiments, in the above-mentioned charging data collection method for new energy vehicles, the step of analyzing the second mining error parameter corresponding to the first user charging data based on the first prediction possibility parameter and the second prediction possibility parameter includes:
[0039] Based on the first prediction possibility parameter corresponding to each charging preference information of the first internal user charging data and the second prediction possibility parameter corresponding to each charging preference information of the second internal user charging data, analyzing the second mining error parameter to be analyzed corresponding to the first user charging data, the second mining error parameter to be analyzed being used to reflect the mining error degree between the first prediction possibility parameter of the first user charging data and the second prediction possibility parameter of the first user charging data in different charging preference information;
[0040] A corresponding second mining error parameter is obtained based on the second mining error parameter to be analyzed.
[0041] The embodiment of the present invention further provides a charging data acquisition system for a new energy vehicle, comprising:
[0042] A network optimization module, used for performing network optimization operations on the pre-built key information mining network to form a target key information mining network corresponding to the key information mining network;
[0043] A key information mining module, used to perform a key information mining operation on the received user charging data to be processed through the target key information mining network, so as to output a corresponding charging key information feature representation;
[0044] A mapping output module, used to perform a mapping output operation on the charging key information feature representation through a mapping output network, so as to output a data anomaly characterization parameter corresponding to the user charging data to be processed;
[0045] The collection confirmation module is used to perform a data collection confirmation operation on the charging data of the user to be processed based on the data anomaly characterization parameter, and when confirming that the charging data of the user to be processed is collected, determine the charging data of the user to be processed as the collected charging data of the target user.
[0046] In some preferred embodiments, in the charging data collection system for the new energy vehicle, the network optimization module includes:
[0047] A first network optimization submodule is used to determine an internal user charging data combination corresponding to the first user charging data, wherein the internal user charging data combination includes at least two internal user charging data of the first user charging data, and the at least two internal user charging data belong to charging data of the same user in different time periods;
[0048] a second network optimization submodule, configured to perform a key information mining operation on the user charging data in the internal user charging data combination through a key information mining network to form a corresponding charging key information cluster, wherein the charging key information cluster includes a charging key information feature representation corresponding to the first user charging data;
[0049] a third network optimization submodule, configured to analyze, based on the characteristic representation of the key charging information, an internal matching parameter between the internal user charging data of the first user charging data, and an external matching parameter between the internal user charging data and the external user charging data, wherein the external user charging data belongs to a second user charging data other than the first user charging data, the first user charging data and the second user charging data belong to charging data of different users, and the first user charging data and the second user charging data belong to the same time period;
[0050] a fourth network optimization submodule, configured to analyze a first mining error parameter corresponding to the first user charging data based on the internal matching parameter and the external matching parameter, wherein the first mining error parameter is used to reflect a mining error degree between the internal matching parameter corresponding to the first user charging data and the external matching parameter corresponding to the first user charging data;
[0051] a fifth network optimization submodule, configured to perform a charging preference information identification operation based on the charging key information feature representation, so as to output a first parameter of the predicted possibility of the internal user charging data in the charging preference information and a second parameter of the predicted possibility of the internal user charging data in the charging preference information;
[0052] a sixth network optimization submodule, configured to analyze a second mining error parameter corresponding to the first user charging data based on the first prediction possibility parameter and the second prediction possibility parameter, wherein the second mining error parameter is used to reflect a mining error degree between the first prediction possibility parameter of the first user charging data and the second prediction possibility parameter of the first user charging data;
[0053] A seventh network optimization submodule is used to perform a network optimization operation on the key information mining network based on the first mining error parameter and the second mining error parameter to form a target key information mining network.
[0054] In some preferred embodiments, in the above-mentioned charging data collection system for new energy vehicles, the fifth network optimization submodule is specifically used for:
[0055] Based on the charging key information feature representation, the internal user charging data is classified to form an original classification possibility degree parameter of the internal user charging data in each charging preference information;
[0056] The original classification possibility degree parameter is decomposed to form a first prediction possibility parameter corresponding to each charging preference information of the first internal user charging data and a second prediction possibility parameter corresponding to each charging preference information of the second internal user charging data.
[0057] A charging data collection method and system for new energy vehicles provided by an embodiment of the present invention can first perform a network optimization operation on a pre-built key information mining network to form a target key information mining network corresponding to the key information mining network; through the target key information mining network, perform a key information mining operation on the received user charging data to be processed to output the corresponding charging key information feature representation; through the mapping output network, perform a mapping output operation on the charging key information feature representation to output the data anomaly characterization parameters corresponding to the user charging data to be processed; based on the data anomaly characterization parameters, perform a data collection confirmation operation on the user charging data to be processed, and when confirming that the user charging data to be processed is collected, the user charging data to be processed is determined as the collected target user charging data. Based on the above content, the high-precision data processing capability of the neural network can be used to determine a reliable basis for performing a data collection confirmation operation, so that the reliability of charging data collection can be improved and the deficiencies in the prior art can be improved.
[0058] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] Figure 1 This is a structural block diagram of a charging data collection platform for new energy vehicles provided in an embodiment of the present invention.
[0060] Figure 2 A schematic flow chart of the steps included in the method for collecting charging data of a new energy vehicle provided in an embodiment of the present invention.
[0061] Figure 3 A schematic diagram of various modules included in a charging data acquisition system for a new energy vehicle provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0062] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.
[0063] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. 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.
[0064] like Figure 1 As shown, an embodiment of the present invention provides a charging data collection platform for new energy vehicles, wherein the charging data collection platform for new energy vehicles may include a memory and a processor.
[0065] In detail, the memory and the processor are electrically connected directly or indirectly to achieve data transmission or interaction. For example, they can be electrically connected to each other through one or more communication buses or signal lines. The memory can store at least one software function module (computer program) that can exist in the form of software or firmware. The processor can be used to execute the executable computer program stored in the memory, thereby realizing the charging data collection method for new energy vehicles provided by the embodiment of the present invention (as described later).
[0066] Optionally, in one embodiment, the memory may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc.
[0067] It can be selected that in one embodiment, the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a system on chip (SoC), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
[0068] Optionally, in one embodiment, the charging data collection platform for new energy vehicles may also include: Figure 1 More or fewer components as shown, or with Figure 1 The various configurations shown may, for example, include a communication unit for exchanging information with other devices.
[0069] Optionally, in one implementation, the charging data collection platform for the new energy vehicle may be a server with data processing capabilities.
[0070] Combination Figure 2 The embodiment of the present invention also provides a charging data collection method for a new energy vehicle, which can be applied to the charging data collection platform of the new energy vehicle. The method steps defined in the process related to the charging data collection method for the new energy vehicle can be implemented by the charging data collection platform of the new energy vehicle. Figure 2 The specific process shown is explained in detail.
[0071] Step S110 , performing a network optimization operation on the pre-built key information mining network to form a target key information mining network corresponding to the key information mining network.
[0072] In an embodiment of the present invention, the charging data collection platform of the new energy vehicle may perform network optimization operations on the pre-built key information mining network to form a target key information mining network corresponding to the key information mining network.
[0073] Step S120, performing a key information mining operation on the received user charging data to be processed through the target key information mining network to output a corresponding charging key information feature representation.
[0074] In an embodiment of the present invention, the charging data collection platform of the new energy vehicle can perform key information mining operations on the received user charging data to be processed through the target key information mining network to output corresponding charging key information feature representation.
[0075] Step S130, mapping and outputting the key charging information feature representation through a mapping output network to output data anomaly characterization parameters corresponding to the user charging data to be processed.
[0076] In an embodiment of the present invention, the charging data collection platform of the new energy vehicle can map and output the key charging information feature representation through a mapping output network to output data anomaly characterization parameters corresponding to the user charging data to be processed. The mapping output network can be formed by network optimization together with the target key information mining network, and the mapping output network can include a fully connected unit and an activation unit.
[0077] Step S140: performing a data collection confirmation operation on the charging data of the user to be processed based on the data anomaly characterization parameter, and when confirming that the charging data of the user to be processed is collected, determining the charging data of the user to be processed as the collected charging data of the target user.
[0078] In an embodiment of the present invention, the charging data collection platform of the new energy vehicle can perform a data collection confirmation operation on the pending user charging data based on the data anomaly characterization parameter, and when confirming the collection of the pending user charging data, determine the pending user charging data as the collected target user charging data. For example, when the data anomaly characterization parameter is greater than the preset anomaly characterization parameter, it can be determined not to collect the pending user charging data to avoid unreliable user charging data from interfering with subsequent analysis, or, when the data anomaly characterization parameter is greater than the preset anomaly characterization parameter, it can be determined to collect the pending user charging data. At this time, the collected user charging data can be used for abnormal statistical analysis. In this way, based on different application scenarios, the collection confirmation method can be different.
[0079] Based on the above content, namely step S110-step S140, the high-precision data processing capability of the neural network can be used to determine a reliable basis for performing data collection confirmation operations, thereby improving the reliability of charging data collection and improving the deficiencies in the prior art.
[0080] Alternatively, in one implementation, the step of performing a network optimization operation on the pre-built key information mining network to form a target key information mining network corresponding to the key information mining network may further include the following description:
[0081] Determine an internal user charging data combination corresponding to the first user charging data, the internal user charging data combination including at least two internal user charging data of the first user charging data, the at least two internal user charging data belong to charging data of the same user in different time periods, but the at least two internal user charging data belong to the same time period, and in addition, the definition of the time period is not limited and can be configured according to actual application requirements;
[0082] Performing a key information mining operation on the user charging data in the internal user charging data combination through a key information mining network to form a corresponding charging key information cluster, wherein the charging key information cluster includes a charging key information feature representation corresponding to the first user charging data;
[0083] Based on the characteristic representation of the key charging information, an internal matching parameter between the internal user charging data of the first user charging data, such as the cosine similarity of the characteristic representation, is analyzed, and an external matching parameter between the internal user charging data and the external user charging data is analyzed, the external user charging data belongs to the second user charging data other than the first user charging data, the first user charging data and the second user charging data belong to charging data of different users, and the first user charging data and the second user charging data belong to the same time period;
[0084] Analyzing a first mining error parameter corresponding to the first user charging data based on the internal matching parameter and the external matching parameter, wherein the first mining error parameter is used to reflect the degree of mining error between the internal matching parameter corresponding to the first user charging data and the external matching parameter corresponding to the first user charging data;
[0085] Based on the charging key information feature representation, a charging preference information identification operation is performed to output a first parameter of the predicted possibility of the internal user charging data in the charging preference information and a second parameter of the predicted possibility of the internal user charging data in the charging preference information, wherein the charging preference information may refer to information such as the location, time, and frequency of charging;
[0086] Based on the first prediction possibility parameter and the second prediction possibility parameter, a second mining error parameter corresponding to the first user charging data is analyzed, where the second mining error parameter is used to reflect the degree of mining error between the first prediction possibility parameter of the first user charging data and the second prediction possibility parameter of the first user charging data;
[0087] Based on the first mining error parameter and the second mining error parameter, the key information mining network is subjected to a network optimization operation to form a target key information mining network. For example, the first mining error parameter and the second mining error parameter may be weightedly summed to obtain a target mining error parameter. Then, based on the target mining error parameter, the key information mining network may be subjected to a network optimization operation to form a target key information mining network.
[0088] Optionally, in one embodiment, after the step of performing a charging preference information identification operation based on the charging key information feature representation to output a first parameter of the predicted possibility of the internal user charging data in the charging preference information and a second parameter of the predicted possibility of the internal user charging data in the charging preference information, the charging data collection method for the new energy vehicle may further include the following description:
[0089] Based on the charging key information feature representation, the internal user charging data is classified to form corresponding classification identification data and a classification possibility degree of the internal user charging data corresponding to the classification identification data;
[0090] Based on the classification possibility degree, a third mining error parameter of different first user charging data in the target classification identification data is determined, where the different first user charging data refers to charging data of the same corresponding user but not belonging to the same time period, and the third mining error parameter constrains the classification possibility degree of each internal user charging data combination corresponding to the first user charging data in the corresponding classification identification data to be maximized.
[0091] Optionally, in one embodiment, the step of performing a classification operation on the internal user charging data based on the charging key information feature representation to form corresponding classification identification data and a classification possibility degree of the internal user charging data corresponding to the classification identification data may further include the following description:
[0092] Based on the charging key information feature representation, the internal user charging data is classified to form corresponding classification identification data;
[0093] A comparative analysis operation is performed on the classification identification data and the charging preference information to form a classification possibility degree of the internal user charging data corresponding to the classification identification data.
[0094] Optionally, in one implementation, the step of classifying the internal user charging data based on the charging key information feature representation to form corresponding classification identification data may further include the following description:
[0095] The charging key information feature representation is subjected to an identification comparison and analysis operation to form a charging key information feature representation with identification data; and the charging key information feature representation with identification data is subjected to a classification operation to form a corresponding classification set, such as forming multiple classification sets; and a marking operation is performed on the identification data with the largest number in the classification set to form classification identification data corresponding to the classification set, wherein the same classification identification data can be assigned to the charging key information feature representation corresponding to the same first user charging data.
[0096] Optionally, in one embodiment, the internal user charging data combination includes first internal user charging data and second internal user charging data. Based on this, the step of performing a charging preference information identification operation based on the charging key information feature representation to output a first parameter of the predicted possibility of the internal user charging data in the charging preference information and a second parameter of the predicted possibility of the internal user charging data in the charging preference information may further include the following description:
[0097] Based on the charging key information feature representation, the internal user charging data is classified to form an original classification possibility degree parameter of the internal user charging data in each charging preference information;
[0098] The original classification possibility degree parameter is decomposed to form a first prediction possibility parameter corresponding to each charging preference information of the first internal user charging data and a second prediction possibility parameter corresponding to each charging preference information of the second internal user charging data.
[0099] Optionally, in one implementation, the step of analyzing the second mining error parameter corresponding to the first user charging data based on the first prediction possibility parameter and the second prediction possibility parameter may further include the following description:
[0100] Analyzing first difference data of the predicted possibility of the first user charging data in the target charging preference information based on the first predicted possibility parameter of the target charging preference information and the second predicted possibility parameter of the target charging preference information;
[0101] Analyzing second difference data of the predicted possibility of the first user's charging data between the target charging preference information and the other charging preference information based on the first predicted possibility parameter of the target charging preference information, the second predicted possibility parameter of the target charging preference information, the first predicted possibility parameter of other charging preference information, and the second predicted possibility parameter of other charging preference information, wherein the other charging preference information is other charging preference information other than the target charging preference information;
[0102] Based on the first difference data of the predicted possibility and the second difference data of the predicted possibility, analyzing a second mining error parameter to be processed for the charging data of the first user, wherein the second mining error parameter to be processed is used to reflect the degree of mining error between the first parameter of the predicted possibility of the charging data of the first user and the second parameter of the predicted possibility of the charging data of the first user in the target charging preference information;
[0103] The corresponding second mining error parameter is obtained based on the second mining error parameter to be processed. For example, the second mining error parameter to be processed may be used as the corresponding second mining error parameter, or as a part of the second mining error parameter.
[0104] Optionally, in one implementation, the step of analyzing the second mining error parameter corresponding to the first user charging data based on the first prediction possibility parameter and the second prediction possibility parameter may further include the following description:
[0105] Based on the first prediction possibility parameter corresponding to each charging preference information of the first internal user charging data and the second prediction possibility parameter corresponding to each charging preference information of the second internal user charging data, analyzing the second mining error parameter to be analyzed corresponding to the first user charging data, the second mining error parameter to be analyzed being used to reflect the mining error degree between the first prediction possibility parameter of the first user charging data and the second prediction possibility parameter of the first user charging data in different charging preference information;
[0106] The corresponding second mining error parameter is obtained based on the second mining error parameter to be analyzed. For example, the second mining error parameter to be analyzed may be used as the corresponding second mining error parameter, or as a part of the second mining error parameter.
[0107] For the second mining error parameter to be analyzed, the possibility degree of all internal user charging data for different charging preference information can be limited to be kept average. For example, the charging preference information includes the a-th charging preference information and the b-th charging preference information. The difference between the sum of the possibility degree of all internal user charging data for the a-th charging preference information and the sum of the possibility degree of all internal user charging data for the b-th charging preference information can be small, such as less than a preset value.
[0108] Combination Figure 3 The embodiment of the present invention also provides a charging data collection system for new energy vehicles, which can be applied to the charging data collection platform of the new energy vehicles mentioned above. The charging data collection system for new energy vehicles can include the following software modules:
[0109] A network optimization module, used for performing network optimization operations on the pre-built key information mining network to form a target key information mining network corresponding to the key information mining network;
[0110] A key information mining module, used to perform a key information mining operation on the received user charging data to be processed through the target key information mining network, so as to output a corresponding charging key information feature representation;
[0111] A mapping output module, used to perform a mapping output operation on the charging key information feature representation through a mapping output network, so as to output a data anomaly characterization parameter corresponding to the user charging data to be processed;
[0112] The collection confirmation module is used to perform a data collection confirmation operation on the charging data of the user to be processed based on the data anomaly characterization parameter, and when confirming that the charging data of the user to be processed is collected, determine the charging data of the user to be processed as the collected charging data of the target user.
[0113] Optionally, in one implementation, the network optimization module includes:
[0114] A first network optimization submodule is used to determine an internal user charging data combination corresponding to the first user charging data, wherein the internal user charging data combination includes at least two internal user charging data of the first user charging data, and the at least two internal user charging data belong to charging data of the same user in different time periods;
[0115] a second network optimization submodule, configured to perform a key information mining operation on the user charging data in the internal user charging data combination through a key information mining network to form a corresponding charging key information cluster, wherein the charging key information cluster includes a charging key information feature representation corresponding to the first user charging data;
[0116] a third network optimization submodule, configured to analyze, based on the characteristic representation of the key charging information, an internal matching parameter between the internal user charging data of the first user charging data, and an external matching parameter between the internal user charging data and the external user charging data, wherein the external user charging data belongs to a second user charging data other than the first user charging data, the first user charging data and the second user charging data belong to charging data of different users, and the first user charging data and the second user charging data belong to the same time period;
[0117] a fourth network optimization submodule, configured to analyze a first mining error parameter corresponding to the first user charging data based on the internal matching parameter and the external matching parameter, wherein the first mining error parameter is used to reflect a mining error degree between the internal matching parameter corresponding to the first user charging data and the external matching parameter corresponding to the first user charging data;
[0118] a fifth network optimization submodule, configured to perform a charging preference information identification operation based on the charging key information feature representation, so as to output a first parameter of the predicted possibility of the internal user charging data in the charging preference information and a second parameter of the predicted possibility of the internal user charging data in the charging preference information;
[0119] a sixth network optimization submodule, configured to analyze a second mining error parameter corresponding to the first user charging data based on the first prediction possibility parameter and the second prediction possibility parameter, wherein the second mining error parameter is used to reflect a mining error degree between the first prediction possibility parameter of the first user charging data and the second prediction possibility parameter of the first user charging data;
[0120] A seventh network optimization submodule is used to perform a network optimization operation on the key information mining network based on the first mining error parameter and the second mining error parameter to form a target key information mining network.
[0121] Optionally, in one implementation, the fifth network optimization submodule is specifically configured to:
[0122] Based on the charging key information feature representation, the internal user charging data is classified to form an original classification possibility degree parameter of the internal user charging data in each charging preference information;
[0123] The original classification possibility degree parameter is decomposed to form a first prediction possibility parameter corresponding to each charging preference information of the first internal user charging data and a second prediction possibility parameter corresponding to each charging preference information of the second internal user charging data.
[0124] In summary, the charging data collection method and system for new energy vehicles provided by the present invention can first perform network optimization operations on the pre-built key information mining network to form a target key information mining network corresponding to the key information mining network; through the target key information mining network, the received user charging data to be processed is subjected to key information mining operations to output the corresponding charging key information feature representation; through the mapping output network, the charging key information feature representation is subjected to mapping output operations to output data anomaly characterization parameters corresponding to the user charging data to be processed; based on the data anomaly characterization parameters, the user charging data to be processed is subjected to data collection confirmation operations, and when the user charging data to be processed is confirmed to be collected, the user charging data to be processed is determined as the collected target user charging data. Based on the above content, the high-precision data processing capability of the neural network can be used to determine a reliable basis for performing data collection confirmation operations, so that the reliability of charging data collection can be improved and the deficiencies in the prior art can be improved.
[0125] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. 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 data collection method for a new energy vehicle, characterized in that: include: Performing network optimization operations on the pre-built key information mining network to form a target key information mining network corresponding to the key information mining network; Through the target key information mining network, the received user charging data to be processed is subjected to key information mining operations to output corresponding charging key information feature representations; Through a mapping output network, the key charging information feature representation is mapped and outputted to output data anomaly characterization parameters corresponding to the user charging data to be processed; Based on the data anomaly characterization parameter, performing a data collection confirmation operation on the charging data of the user to be processed, and in the case where it is confirmed that the charging data of the user to be processed is collected, determining the charging data of the user to be processed as the collected charging data of the target user; The step of performing a network optimization operation on the pre-built key information mining network to form a target key information mining network corresponding to the key information mining network includes: Determine an internal user charging data combination corresponding to the first user charging data, wherein the internal user charging data combination includes at least two internal user charging data of the first user charging data, and the at least two internal user charging data belong to charging data of the same user in different time periods; Performing a key information mining operation on the user charging data in the internal user charging data combination through a key information mining network to form a corresponding charging key information cluster, wherein the charging key information cluster includes a charging key information feature representation corresponding to the first user charging data; Based on the characteristic representation of the key charging information, an internal matching parameter between the internal user charging data of the first user charging data is analyzed, and an external matching parameter between the internal user charging data and the external user charging data is analyzed, the external user charging data belongs to the second user charging data other than the first user charging data, the first user charging data and the second user charging data belong to charging data of different users, and the first user charging data and the second user charging data belong to the same time period; Analyzing a first mining error parameter corresponding to the first user charging data based on the internal matching parameter and the external matching parameter, wherein the first mining error parameter is used to reflect the degree of mining error between the internal matching parameter corresponding to the first user charging data and the external matching parameter corresponding to the first user charging data; Based on the charging key information feature representation, a charging preference information identification operation is performed to output a first parameter of the predicted possibility of the internal user charging data in the charging preference information and a second parameter of the predicted possibility of the internal user charging data in the charging preference information; Based on the first prediction possibility parameter and the second prediction possibility parameter, a second mining error parameter corresponding to the first user charging data is analyzed, where the second mining error parameter is used to reflect the degree of mining error between the first prediction possibility parameter of the first user charging data and the second prediction possibility parameter of the first user charging data; Based on the first mining error parameter and the second mining error parameter, the key information mining network is subjected to a network optimization operation to form a target key information mining network.
2. The charging data collection method for new energy vehicles according to claim 1, characterized in that: After the step of performing a charging preference information identification operation based on the charging key information feature representation to output a first parameter of the predicted possibility of the internal user charging data in the charging preference information and a second parameter of the predicted possibility of the internal user charging data in the charging preference information, the charging data collection method for the new energy vehicle further includes: Based on the charging key information feature representation, the internal user charging data is classified to form corresponding classification identification data and a classification possibility degree of the internal user charging data corresponding to the classification identification data; Based on the classification possibility degree, determining a third mining error parameter of different first user charging data in the target classification identification data, where the different first user charging data refers to charging data of the same corresponding user but not belonging to the same time period; The step of performing a network optimization operation on the key information mining network based on the first mining error parameter and the second mining error parameter to form a target key information mining network includes: Based on the first mining error parameter, the second mining error parameter and the third mining error parameter, the key information mining network is subjected to a network optimization operation to form a target key information mining network.
3. The charging data collection method for new energy vehicles as claimed in claim 2, characterized in that: The step of classifying the internal user charging data based on the charging key information feature representation to form corresponding classification identification data and the classification possibility degree of the internal user charging data corresponding to the classification identification data includes: Based on the charging key information feature representation, the internal user charging data is classified to form corresponding classification identification data; Performing a comparative analysis operation on the classification identification data and the charging preference information to form a classification possibility degree of the internal user charging data corresponding to the classification identification data; The step of classifying the internal user charging data based on the charging key information feature representation to form corresponding classification identification data includes: Performing a logo comparison and analysis operation on the key charging information characteristic representation to form a key charging information characteristic representation with identification data; Classify the key charging information feature representation with identification data to form a corresponding classification set; A marking operation is performed on the identification data with the largest number in the classification set to form classification identification data corresponding to the classification set.
4. The charging data collection method for new energy vehicles according to claim 1, characterized in that: The internal user charging data combination includes first internal user charging data and second internal user charging data; and the step of performing a charging preference information identification operation based on the charging key information feature representation to output a first parameter of the predicted possibility of the internal user charging data in the charging preference information and a second parameter of the predicted possibility of the internal user charging data in the charging preference information includes: Based on the charging key information feature representation, the internal user charging data is classified to form an original classification possibility degree parameter of the internal user charging data in each charging preference information; The original classification possibility degree parameter is decomposed to form a first prediction possibility parameter corresponding to each charging preference information of the first internal user charging data and a second prediction possibility parameter corresponding to each charging preference information of the second internal user charging data.
5. The charging data collection method for new energy vehicles as claimed in claim 4, characterized in that: The step of analyzing the second mining error parameter corresponding to the first user charging data based on the first prediction possibility parameter and the second prediction possibility parameter includes: Analyzing first difference data of the predicted possibility of the first user charging data in the target charging preference information based on the first predicted possibility parameter of the target charging preference information and the second predicted possibility parameter of the target charging preference information; Analyzing second difference data of the predicted possibility of the first user charging data between the target charging preference information and the other charging preference information based on the first predicted possibility parameter of the target charging preference information, the second predicted possibility parameter of the target charging preference information, the first predicted possibility parameter of other charging preference information, and the second predicted possibility parameter of other charging preference information; Based on the first difference data of the predicted possibility and the second difference data of the predicted possibility, analyzing a second mining error parameter to be processed for the charging data of the first user, wherein the second mining error parameter to be processed is used to reflect the degree of mining error between the first parameter of the predicted possibility of the charging data of the first user and the second parameter of the predicted possibility of the charging data of the first user in the target charging preference information; A corresponding second mining error parameter is obtained based on the second mining error parameter to be processed.
6. The charging data collection method for new energy vehicles as claimed in claim 4, characterized in that: The step of analyzing the second mining error parameter corresponding to the first user charging data based on the first prediction possibility parameter and the second prediction possibility parameter includes: Based on the first prediction possibility parameter corresponding to each charging preference information of the first internal user charging data and the second prediction possibility parameter corresponding to each charging preference information of the second internal user charging data, analyzing the second mining error parameter to be analyzed corresponding to the first user charging data, the second mining error parameter to be analyzed being used to reflect the mining error degree between the first prediction possibility parameter of the first user charging data and the second prediction possibility parameter of the first user charging data in different charging preference information; A corresponding second mining error parameter is obtained based on the second mining error parameter to be analyzed.
7. A charging data acquisition system for new energy vehicles, characterized in that: include: A network optimization module, used for performing network optimization operations on the pre-built key information mining network to form a target key information mining network corresponding to the key information mining network; A key information mining module, used to perform a key information mining operation on the received user charging data to be processed through the target key information mining network, so as to output a corresponding charging key information feature representation; A mapping output module, used to perform a mapping output operation on the charging key information feature representation through a mapping output network, so as to output a data anomaly characterization parameter corresponding to the user charging data to be processed; a collection confirmation module, configured to perform a data collection confirmation operation on the charging data of the user to be processed based on the data anomaly characterization parameter, and, if it is confirmed that the charging data of the user to be processed is collected, determine the charging data of the user to be processed as the collected charging data of the target user; Wherein, the network optimization module includes: A first network optimization submodule is used to determine an internal user charging data combination corresponding to the first user charging data, wherein the internal user charging data combination includes at least two internal user charging data of the first user charging data, and the at least two internal user charging data belong to charging data of the same user in different time periods; a second network optimization submodule, configured to perform a key information mining operation on the user charging data in the internal user charging data combination through a key information mining network to form a corresponding charging key information cluster, wherein the charging key information cluster includes a charging key information feature representation corresponding to the first user charging data; a third network optimization submodule, configured to analyze, based on the characteristic representation of the key charging information, an internal matching parameter between the internal user charging data of the first user charging data, and an external matching parameter between the internal user charging data and the external user charging data, wherein the external user charging data belongs to a second user charging data other than the first user charging data, the first user charging data and the second user charging data belong to charging data of different users, and the first user charging data and the second user charging data belong to the same time period; a fourth network optimization submodule, configured to analyze a first mining error parameter corresponding to the first user charging data based on the internal matching parameter and the external matching parameter, wherein the first mining error parameter is used to reflect a mining error degree between the internal matching parameter corresponding to the first user charging data and the external matching parameter corresponding to the first user charging data; a fifth network optimization submodule, configured to perform a charging preference information identification operation based on the charging key information feature representation, so as to output a first parameter of the predicted possibility of the internal user charging data in the charging preference information and a second parameter of the predicted possibility of the internal user charging data in the charging preference information; a sixth network optimization submodule, configured to analyze a second mining error parameter corresponding to the first user charging data based on the first prediction possibility parameter and the second prediction possibility parameter, wherein the second mining error parameter is used to reflect a mining error degree between the first prediction possibility parameter of the first user charging data and the second prediction possibility parameter of the first user charging data; A seventh network optimization submodule is used to perform a network optimization operation on the key information mining network based on the first mining error parameter and the second mining error parameter to form a target key information mining network.
8. The charging data acquisition system for new energy vehicles as claimed in claim 7, characterized in that: The fifth network optimization submodule is specifically used for: Based on the charging key information feature representation, the internal user charging data is classified to form an original classification possibility degree parameter of the internal user charging data in each charging preference information; The original classification possibility degree parameter is decomposed to form a first prediction possibility parameter corresponding to each charging preference information of the first internal user charging data and a second prediction possibility parameter corresponding to each charging preference information of the second internal user charging data.
Citation Information
Patent Citations
Lithium battery anomaly detection method based on long and short time memory auto-encoder
CN114565008A
Charging load ultra-short-term prediction method based on charging pile usage degree
CN114638440A
Lithium battery health state estimation method
CN115792637A
Charging abnormal behavior analysis method based on deep learning algorithm
CN115840922A
Data feature processing method and data feature processing apparatus
US20220091818A1
