Charging equipment classification method and related device

Abstract

The invention discloses a charging equipment classification method and a related device. The charging equipment classification method comprises the following steps: acquiring a current transient waveform of current charging equipment; wherein the current transient waveform represents the distribution of the power utilization parameters of the current charging equipment in the charging starting time period; retrieving at least one type of standard transient waveform similar to the current transient waveform in a preset transient waveform set as a reference transient waveform; wherein the transient waveform set comprises various standard transient waveforms of the target charging equipment; extracting transient characteristics of the current transient waveform and each reference transient waveform; and judging whether the current charging equipment is target charging equipment or not based on the transient characteristics of the current transient waveform and each reference transient waveform. According to the scheme, the classification result accuracy can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of load identification, in particular to a charging device classification method, a charging system, an electronic device and a computer readable storage medium. BACKGROUND

[0002] Charging device classification aims to determine whether the charging device is a target charging device, which can be applied to scenarios such as non-standard charging behavior identification and charging device management. Taking the identification of non-standard charging behavior of electric vehicles as an example, in places where charging of electric vehicles is prohibited, charging device classification can determine whether the charging device is an electric vehicle, thereby identifying whether there is electric vehicle charging behavior.

[0003] However, the classification result obtained by the charging device classification method in the related art is not accurate enough. SUMMARY

[0004] The present application provides a charging device classification method, a charging system, an electronic device and a computer readable storage medium, which can solve the problem of low accuracy of the classification result obtained by the charging device classification method in the related art.

[0005] The present application provides a charging device classification method, comprising: obtaining a current transient waveform of a current charging device; wherein the current transient waveform represents the distribution of the power consumption parameter of the current charging device at the charging start period; retrieving at least one type of standard transient waveform similar to the current transient waveform from a preset transient waveform set as a reference transient waveform; wherein the transient waveform set includes various standard transient waveforms of a target charging device; extracting the transient features of the current transient waveform and each reference transient waveform; and determining whether the current charging device is a target charging device based on the transient features of the current transient waveform and each reference transient waveform.

[0006] The present application provides a charging device classification apparatus, comprising: an acquisition module, a retrieval module, an extraction module and a determination module, the acquisition module is configured to obtain a current transient waveform of a current charging device; wherein the current transient waveform represents the distribution of the power consumption parameter of the current charging device at the charging start period; the retrieval module is configured to retrieve at least one type of standard transient waveform similar to the current transient waveform from a preset transient waveform set as a reference transient waveform; wherein the transient waveform set includes various standard transient waveforms of a target charging device; the extraction module is configured to extract the transient features of the current transient waveform and each reference transient waveform; and the determination module is configured to determine whether the current charging device is a target charging device based on the transient features of the current transient waveform and each reference transient waveform.

[0007] The present application provides a charging system, comprising: a charging device and a power supply device, the power supply device is configured to execute the foregoing method.

[0008] The application provides an electronic device, comprising a memory and a processor, the processor is used for executing program instructions stored in the memory to realize the above method.

[0009] The application provides a computer readable storage medium, which stores program instructions, and the program instructions are executed by a processor to realize the above method.

[0010] The above scheme has at least the following technical effects:

[0011] On the one hand, various standard transient waveforms of the target charging device are organized in a transient waveform set, a standard transient waveform similar to the current transient waveform is searched in the transient waveform set as a reference transient waveform, the transient features of the current transient waveform and the reference transient waveform are combined, and it is judged whether the current charging device is the target charging device. Since the various standard transient waveforms in the transient waveform set can accurately express the transient charging information of various target charging devices compared with a single standard transient waveform, the similar standard transient waveform searched from the transient waveform set is used for judgment again, which can overcome the randomness of the transient charging information of the target charging device, avoid misjudgment caused by randomness, thereby reducing the misjudgment probability and improving the classification result accuracy.

[0012] On the other hand, since it is a transient waveform level search, the transient waveform has no loss compared with the transient feature, and the expressed transient charging information is more complete, so the accuracy of the searched reference transient waveform can be improved, and the accuracy of the judgment result can be improved.

[0013] On the other hand, when the target charging device has a new category, the standard transient waveform of the new category can be directly added to the transient waveform set to maintain the generalization ability on various categories of the target charging device, and the dynamic incremental expansion of the transient waveform set can be realized at extremely low cost and fastest speed, thereby improving the efficiency and response speed.

[0014] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, but not limiting the application. BRIEF DESCRIPTION OF DRAWINGS

[0015] The accompanying drawings incorporated in the specification and constituting a part hereof illustrate embodiments consistent with the application and, together with the specification, serve to explain the technical solutions of the application.

[0016] Figure 1 is a structural schematic diagram of an embodiment of the charging system provided by the application;

[0017] Figure 2 is a flowchart of an embodiment of the charging device classification method provided by the application;

[0018] Figure 3is a flowchart of the charging device classification method provided in Embodiment Two of the application;

[0019] Figure 4 is a flowchart of the charging device classification method provided in Embodiment Three of the application;

[0020] Figure 5 is a flowchart of the charging device classification method provided in Embodiment Four of the application;

[0021] Figure 6 is a flowchart of the charging device classification method provided in Embodiment One of the application;

[0022] Figure 7 is a schematic diagram of the current transient waveform and the current steady-state waveform provided in the application;

[0023] Figure 8 is a structural schematic diagram of the charging device classification apparatus provided in Embodiment One of the application;

[0024] Figure 9 is a structural schematic diagram of the electronic device provided in Embodiment One of the application;

[0025] Figure 10 is a structural schematic diagram of the computer readable storage medium provided in Embodiment One of the application. DETAILED DESCRIPTION

[0026] The schemes of the embodiments of the application will be described in detail below with reference to the accompanying drawings.

[0027] In the following description, specific details are set forth in order to provide a thorough understanding of the application. However, persons having ordinary skill in the art will realize that the application can be practiced without some or all of these details. For the purpose of explanation, specific structural and functional details disclosed herein are presented simply to provide a thorough understanding of the features of the application.

[0028] The term "and / or" herein is merely an associated relationship between associated objects, and means that there can be three relationships, for example, A and / or B can mean that there are three cases of A alone, A and B together, and B alone. In addition, the character " / " herein generally means that the front and rear associated objects are in an "or" relationship. In addition, "multiple" herein means two or more than two. In addition, the term "at least one" herein means any one of multiple or any combination of at least two of multiple, for example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0029] The charging device classification method provided in the application can be implemented according to a charging system.

[0030] Figure 1 is a structural schematic diagram of the charging system provided in Embodiment One of the application. As shown in the figure, Figure 1As shown, the charging system includes a charging device and a power supply device.

[0031] The charging device can be any device with power demand, such as an electric bicycle, an electric car, a computer, a mobile phone, etc.

[0032] The power supply device can be any device with power supply capability, such as a socket, an air switch, etc.

[0033] After the charging device and the power supply device establish a connection, the power supply device can supply power to the charging device.

[0034] Figure 2 is a flowchart of an embodiment of the charging device classification method provided by the present application. As shown in the embodiment, the charging device classification method can include the following steps: Figure 2

[0035] S110: Obtain the current transient waveform of the current charging device.

[0036] The current transient waveform represents the distribution of the power consumption parameter of the current charging device at the charging start period.

[0037] The execution subject of the embodiment is a charging device classification apparatus, which can be a power supply device or an electronic device other than the power supply device.

[0038] The power consumption parameter can be current, power, etc. Power is the product of current and voltage, and generally, the voltage is constant.

[0039] The charging start period refers to a first preset time period with the charging device switching point as the first time starting point. The charging start period includes a plurality of continuous time periods, and each time period includes a plurality of sampling time points. For example, each time period is sampled at a sampling frequency of 6.4 kHz and includes 128 sampling time points. For another example, each time period is sampled at a sampling frequency of 12.8 kHz and includes 256 sampling time points.

[0040] The horizontal axis of the current transient waveform represents time, and the vertical axis represents the power consumption parameter, which shows the power consumption parameter at each sampling time point in the charging start period and can express the transient charging information of the current charging device.

[0041] S120: Retrieve at least one type of standard transient waveform similar to the current transient waveform from a preset transient waveform set as a reference transient waveform.

[0042] The transient waveform set includes various standard transient waveforms of the target charging device.

[0043] ​The target charging device can be any charging device, such as a battery car. The target charging device can be subdivided into multiple categories according to brand, battery type (such as lead-acid battery, lithium battery, and lithium iron phosphate battery), and the like.

[0044] The transient charging information of the target charging device has strong randomness, large intra-class difference, and intra-class non-aggregation, that is, the transient charging information of different types of target charging devices has large difference, and therefore a single standard transient waveform cannot accurately express the transient charging information of each type of target charging device. The transient waveform set includes standard transient waveforms of each type of target charging device, and each type of standard transient waveform is used to express the transient charging information of each type of target charging device. Therefore, compared with a single standard transient waveform, the transient waveform set can accurately express the transient charging information of each type of charging device.

[0045] In some embodiments, the at least one type of standard transient waveform similar to the current transient waveform refers to at least one type of standard transient waveform having a similarity higher than a similarity threshold value with the current transient waveform.

[0046] In some embodiments, the at least one type of standard transient waveform similar to the current transient waveform refers to at least one type of standard transient waveform having the highest similarity with the current transient waveform.

[0047] S130: Extracting transient features of the current transient waveform and each reference transient waveform.

[0048] S140: Judging whether the current charging device is the target charging device based on the transient features of the current transient waveform and each reference transient waveform.

[0049] The above scheme has at least the following technical effects:

[0050] On the one hand, it can be understood that the transient charging information of the target charging device has randomness, and a single standard transient waveform cannot accurately express the transient charging information of each type of target charging device. If the current charging device is directly judged to be the target charging device based on the transient features of the current transient waveform and the single standard transient waveform, the probability of misjudgment is high. The above scheme organizes the standard transient waveforms of each type of target charging device in the transient waveform set, retrieves a standard transient waveform similar to the current transient waveform in the transient waveform set as a reference transient waveform, and judges whether the current charging device is the target charging device by combining the transient features of the current transient waveform and the reference transient waveform. Since each type of standard transient waveform in the transient waveform set can accurately express the transient charging information of each type of target charging device compared with a single standard transient waveform, retrieving a similar standard transient waveform from the transient waveform set and then using it for judgment can overcome the randomness of the transient charging information of the target charging device, avoid misjudgment caused by randomness, and thus reduce the probability of misjudgment and improve the accuracy of the classification result.

[0051] On the other hand, it can be understood that, since the transient feature has information loss relative to the transient waveform, the retrieval at the transient feature level is less accurate than the retrieval at the transient waveform level. The above scheme, since it is the retrieval at the transient waveform level, the transient waveform has no loss relative to the transient feature, and the expressed transient charging information is more complete, so as to improve the accuracy of the retrieved reference transient waveform, and further improve the accuracy of the judgment result.

[0052] In another aspect, it can be understood that, if the trained model is used for judgment, when the target charging device has a new category, the model needs to be retrained to maintain the generalization ability on each category of the target charging device, so the efficiency and response speed are low. The above scheme, when the target charging device has a new category, the standard transient waveform of the new category can be directly added to the transient waveform set to maintain the generalization ability on each category of the target charging device, so as to realize the dynamic incremental expansion of the transient waveform set at extremely low cost and the fastest speed, and improve the efficiency and response speed.

[0053] Further, in some embodiments, S120 comprises: judging whether at least one similar standard transient waveform is retrieved; in response to retrieving at least one similar standard transient waveform, taking the at least one similar standard transient waveform as a reference transient waveform and continuing to perform S130-S140; in response to not retrieving a similar standard transient waveform, determining that the current charging device is not the target charging device.

[0054] It can be understood that, if no similar standard transient waveform is retrieved, it means that there is no similar target charging device category to the current charging device, so the current charging device can be directly determined to be not the target charging device, reducing processing overhead. If a similar standard transient waveform is retrieved, it means that there is a similar target charging device category to the current charging device, and it is necessary to further determine whether the current charging device belongs to the similar target charging device category.

[0055] Further, in some embodiments, the transient feature comprises a transient statistical feature. S130 comprises: respectively performing statistics on the current transient waveform and each reference transient waveform to obtain corresponding transient statistical features. The transient statistical feature represents the distribution of each power parameter in the charging start period in the corresponding current transient waveform or reference transient waveform. The statistical method can be calculation by formula / rule.

[0056] In some embodiments, the transient statistical feature comprises at least one of: at least one of a power parameter peak value, a transient duration, and a power parameter trend value. The power parameter peak value refers to the maximum value of the power parameter. The transient duration refers to the time length from the start of rapid change of the power parameter to the relative stability. The power parameter trend value can be the median, mean, mode, etc. of the power parameter.

[0057] In some embodiments, the transient feature includes a transient inductive feature. S130 includes: extracting a transient inductive feature between the current transient waveform and each reference transient waveform respectively. The transient inductive feature characterizes the shape difference between the current transient waveform and the corresponding reference transient waveform. The transient inductive feature is statistically different, and the current transient waveform and the corresponding reference transient waveform can be input into a large model to obtain the transient inductive feature by induction of the large model.

[0058] Figure 3 FIG. 2 is a flowchart of an embodiment of a charging device classification method provided by the present application. The embodiment is a further extension of S140. As shown in the figure, in the embodiment, the charging device classification method can include the following steps: Figure 3

[0059] S141: Based on the transient features of the current transient waveform and each reference transient waveform, determine whether there is a target charging device category to which the current charging device belongs in the target charging device category corresponding to each reference transient waveform.

[0060] In some embodiments, the transient feature includes a transient statistical feature. S141 includes: calculating a first similarity between the transient statistical features of the current transient waveform and each reference transient waveform respectively; determining whether the maximum first similarity is greater than a first similarity threshold; in response to being greater than the first similarity threshold, determining that the target charging device category corresponding to the maximum first similarity is the target charging device category to which the current charging device belongs; and in response to not being greater than the first similarity threshold, determining that there is no target charging device category to which the current charging device belongs.

[0061] In some embodiments, the transient feature includes a transient inductive feature. S141 includes: calculating a second similarity between each transient inductive feature and a standard transient inductive feature; determining whether the maximum second similarity is greater than a second similarity threshold; in response to being greater than the second similarity threshold, determining that the target charging device category corresponding to the maximum second similarity is the target charging device category to which the current charging device belongs; and in response to not being greater than the second similarity threshold, determining that there is no target charging device category to which the current charging device belongs.

[0062] In some embodiments, the transient feature includes a transient statistical feature and a transient inductive feature. S141 includes: determining whether the maximum first similarity is greater than a first similarity threshold and the maximum second similarity is greater than a second similarity threshold; in response to both being greater than, determining that there is; otherwise, determining that there is not. Alternatively, S141 includes: weighting the maximum first similarity and the maximum second similarity to obtain a transient total similarity; determining whether the transient total similarity is greater than a total similarity threshold; in response to being greater than, determining that there is; otherwise, determining that there is not. The total similarity threshold can be equal to 0 or greater than 0.

[0063] ​In some embodiments, the transient features based on the current transient waveform and the reference transient waveforms can be input into a trained large model (LLM, Large Language Model) to obtain a total similarity of the transient.

[0064] In response to the existence of the target charging device category to which the current charging device belongs, S142 is performed; in response to the non-existence of the target charging device category to which the current charging device belongs, S143 is performed.

[0065] S142: determining that the current charging device is the target charging device.

[0066] S143: determining that the current charging device is not the target charging device.

[0067] Further, it can be understood that the transient charging information has an inter-class non-dispersive characteristic, that is, the transient charging information of part of the non-target charging devices and the target charging device can be similar. Therefore, it is still possible to misjudge whether the current charging device is the target charging device only according to the transient charging information. In order to further reduce the misjudgment probability, the above embodiments can be extended as follows:

[0068] Figure 4 is a flowchart of a charging device classification method embodiment provided by the present application. The present embodiment is a further extension of the foregoing embodiments. Wherein S210-S230 is a further extension of S110-S130, and S240 is a further extension of S140. As shown in the figure, Figure 4 The charging device classification method can include the following steps:

[0069] S210: obtaining a current transient waveform of a current charging device, and obtaining a current steady-state waveform of the current charging device.

[0070] The current steady-state waveform represents the distribution of the power consumption parameter of the current charging device in the charging intermediate period.

[0071] The charging intermediate period is located after the charging start period. The charging intermediate period refers to a time period of a second preset duration from a second time starting point. The second time starting point is the first time ending point of the charging start period or a sampling time point located after the first time ending point. The second preset duration can be a time period, and correspondingly, the current steady-state waveform can be a steady-state waveform of a time period. The second preset duration can also be a plurality of continuous time periods, and correspondingly, the current steady-state waveform can be a steady-state waveform of a plurality of continuous time periods.

[0072] In some embodiments, a waveform of the current charging device in a charging intermediate period can be collected, and the collected waveform can be taken as the current steady-state waveform, or a sub-waveform of one time period of the collected waveform can be taken as the current steady-state waveform.

[0073] In some embodiments, S210 includes S211-S214. Details are as follows.

[0074] S211: After the charging start period, collect candidate steady-state waveforms of the current charging device in several continuous periods.

[0075] S212: Obtain power consumption parameter trend values of each candidate steady-state waveform respectively.

[0076] The power consumption parameter trend value represents the fluctuation amplitude of each power consumption parameter in the candidate steady-state waveform.

[0077] The power consumption parameter trend value can be variance, standard deviation, etc.

[0078] S213: Determine whether each power consumption parameter trend value meets a steady-state fluctuation condition.

[0079] The steady-state fluctuation condition includes that the minimum power consumption parameter trend value is less than a trend threshold. Meeting the steady-state fluctuation condition means that the power consumption parameter fluctuation degree is small enough.

[0080] In response to meeting the steady-state fluctuation condition, perform S214; in response to not meeting the steady-state fluctuation condition, return to S211 to repeat performing S211-S213 until the steady-state fluctuation condition is met.

[0081] S214: Obtain the current steady-state waveform based on one of the candidate steady-state waveforms.

[0082] The one of the candidate steady-state waveforms can be any one of the candidate steady-state waveforms. For example, the candidate steady-state waveform is the candidate steady-state waveform of the last period.

[0083] In some embodiments, the one of the candidate steady-state waveforms can be taken as the current steady-state waveform.

[0084] In some embodiments, a time period sub-waveform of the one of the candidate steady-state waveforms can be taken as the current steady-state waveform to reduce the amount of calculation.

[0085] It can be understood that, due to the fact that the power consumption parameters of the current charging device have not really entered a steady state, or the power consumption parameters of other charging devices connected with the power supply device abnormally fluctuate, etc., the power consumption parameters of the current charging device may fluctuate greatly, which weakens the accuracy of the steady state charging information of the current charging device expressed by the current steady state waveform obtained. Through S211-S214, the current steady state waveform with a small enough fluctuation degree can be obtained, and the accuracy of the steady state charging information of the current charging device expressed by the current steady state waveform is improved.

[0086] S220: searching for at least one type of standard transient waveform similar to the current transient waveform in the preset transient waveform set as a reference transient waveform, and searching for at least one type of standard steady state waveform similar to the current steady state waveform in the preset steady state waveform set as a reference steady state waveform.

[0087] The steady state waveform set includes various types of standard steady state waveforms of the target charging device.

[0088] The various types of standard steady state waveforms correspond to one type of target charging device / one target charging device category. The various types of standard steady state waveforms are respectively used to express the steady state charging information of the various types of target charging devices.

[0089] In some embodiments, the at least one type of standard steady state waveform similar to the current steady state waveform refers to at least one type of standard steady state waveform with a similarity higher than a similarity threshold value with the current steady state waveform.

[0090] In some embodiments, the at least one type of standard steady state waveform similar to the current steady state waveform refers to at least one type of standard steady state waveform with the highest similarity with the current steady state waveform.

[0091] S230: extracting transient features of the current transient waveform and each reference transient waveform, and extracting steady state features of the current steady state waveform and each reference steady state waveform.

[0092] In some embodiments, the steady state features include steady state statistical features. S230 includes: respectively performing statistics on the current steady state waveform and each reference steady state waveform to obtain corresponding steady state statistical features. The steady state statistical features represent the distribution of each power consumption parameter in the charging start period in the corresponding current steady state waveform or reference steady state waveform.

[0093] In some embodiments, the steady-state statistical features include at least one of: a Euclidean distance of a coordinate point where a peak value of the power parameter, a maximum power parameter and a minimum power parameter are located, a power parameter waveform factor / phase difference, a waveform kurtosis, a waveform variance, a waveform area, a pulse width. The peak value of the power parameter refers to the maximum value of the power parameter. The power parameter waveform factor refers to a ratio of a power parameter effective value (root mean square value) to a power parameter average value. The waveform area refers to an area formed by the corresponding waveform and the horizontal axis (time axis). The area formed by the waveform and the horizontal axis can be an area formed by the entire waveform and the time axis, or an area formed by a partial waveform (e.g., a half-time period waveform, a one-time period waveform) and the time axis. The waveform variance refers to a variance of each power parameter in the corresponding waveform.

[0094] In some embodiments, the steady-state features include steady-state induction features. S230 includes: extracting steady-state induction features between the current steady-state waveform and each reference steady-state waveform. The steady-state induction features represent shape differences between the current steady-state waveform and the corresponding reference steady-state waveform. The steady-state induction features are statistically different, and the current steady-state waveform and the corresponding reference steady-state waveform can be input into a large model to obtain the steady-state induction features through induction of the large model.

[0095] S240: based on the transient features of the current transient waveform and each reference transient waveform, and the steady-state features of the current steady-state waveform and each reference steady-state waveform, determining whether the current charging device is the target charging device.

[0096] In some embodiments, the transient features include transient statistical features, the steady-state features include steady-state statistical features, and S240 includes: calculating a first similarity between the transient statistical features of the current transient waveform and each reference transient waveform, and a first similarity between the steady-state statistical features of the current steady-state waveform and each reference steady-state waveform; determining whether the maximum first similarity is greater than a first similarity threshold; in response to being greater than the first similarity threshold, determining that a target charging device category corresponding to the maximum first similarity is a target charging device category to which the current charging device belongs; and in response to not being greater than the first similarity threshold, determining that there is no target charging device category to which the current charging device belongs.

[0097] In some embodiments, the transient features include transient induction features, the steady-state features include steady-state induction features, and S240 includes: calculating a second similarity between each transient induction feature and a standard transient induction feature, and a second similarity between each steady-state induction feature and a standard steady-state induction feature; determining whether the maximum second similarity is greater than a second similarity threshold; in response to being greater than the second similarity threshold, determining that a target charging device category corresponding to the maximum second similarity is a target charging device category to which the current charging device belongs; and in response to not being greater than the second similarity threshold, determining that there is no target charging device category to which the current charging device belongs.

[0098] In some embodiments, the transient features include transient statistical features, transient inductive features, the steady features include steady statistical features, steady inductive features. S240 includes: judging whether the maximum first similarity is greater than a first similarity threshold, whether the maximum second similarity is greater than a second similarity threshold; in response to both being greater than, determining that there is; otherwise, determining that there is not. Alternatively, S240 includes: weighting the maximum first similarity and the maximum second similarity to obtain a total similarity; judging whether the total similarity is greater than a total similarity threshold; in response to being greater than, determining that there is; otherwise, determining that there is not.

[0099] In some embodiments, the transient features based on the current transient waveform and each reference transient waveform, the steady features of the current steady waveform and each reference steady waveform can be input into a trained large model (LLM, Large Language Model) to obtain a total similarity.

[0100] For detailed description of the steady part of the present embodiment, please refer to other embodiments, which will not be described here.

[0101] Further, it can be understood that the transient charging information has an inter-class non-dispersion characteristic, that is, the transient charging information of part of the non-target charging device and the target charging device can be similar. Therefore, it is still possible to misjudge whether the current charging device is the target charging device only according to the transient charging information. In order to further reduce the misjudgment probability, the above embodiment can be extended as follows:

[0102] The above scheme, on the one hand, since the inter-class non-dispersion degree of the steady charging information of the target charging device is relatively lower than that of the transient charging information, it can make up for the inter-class non-dispersion characteristic of the transient charging information to some extent. Therefore, judging whether the current charging device is the target charging device by combining the transient waveform and the steady waveform can further reduce the misjudgment probability and improve the accuracy of the classification result. On the other hand, the steady waveform concentrates on subdividing various standard steady waveforms, and judges whether the current charging device is the target charging device by combining the current steady waveform and the similar standard steady waveform. Therefore, the inter-class difference of the steady charging information of the target charging device is considered, which can further reduce the misjudgment probability and improve the accuracy of the classification result.

[0103] Further, it can be understood that the inter-class difference of the steady charging information is small (commonality is strong) compared with the transient charging information of the target charging device. Therefore, in order to reduce the processing overhead, in some embodiments, different from S220, it can be determined whether the current steady waveform is similar to the representative steady waveform of the target charging device, in response to being similar, the representative steady waveform is taken as the reference steady waveform and the subsequent steps are executed; in response to being dissimilar, it is determined that the current charging device is not the target charging device.

[0104] Wherein, represents a steady-state waveform is used to represent various types of standard steady-state waveforms, and each representative steady-state waveform parameter in represents a steady-state waveform can be the average, mode, median, etc. of the corresponding power consumption parameter in each type of standard steady-state waveform.

[0105] Further, in some embodiments, S220 includes: determining whether at least one similar standard steady-state waveform is retrieved; in response to retrieving at least one similar standard steady-state waveform, taking the at least one similar standard steady-state waveform as a reference steady-state waveform and continuing to perform S230 and subsequent steps; and in response to not retrieving a similar standard steady-state waveform, determining that the current charging device is not the target charging device.

[0106] Figure 5 is a flowchart of an embodiment four of the charging device classification method provided by the present application. The present embodiment is a further extension of the foregoing embodiments. As shown in Figure 5 , in the present embodiment, the charging device classification method can include the following steps:

[0107] S310: determining whether the current charging device meets the steady-state feature condition corresponding to the target charging device based on the steady-state statistical features of the current steady-state waveform of the current charging device.

[0108] In some embodiments, a third similarity between the steady-state statistical features of the current steady-state waveform and the standard statistical features can be obtained; it is determined whether the third similarity is greater than a third similarity threshold; in response to being greater than the third similarity threshold, it is determined that the current steady-state waveform meets the steady-state feature condition; and in response to not being greater than the third similarity threshold, it is determined that the current steady-state waveform does not meet the steady-state feature condition. Wherein, the third similarity threshold can be equal to 0, or can be greater than 0.

[0109] In some embodiments, the steady-state statistical features of the current steady-state waveform can be input into a trained large model to obtain the third similarity.

[0110] In response to meeting the steady-state feature condition, S320 is performed. In response to not meeting the steady-state feature condition, S330 is performed.

[0111] S320: performing S120-S140 or S220-S240.

[0112] S330: determining that the current charging device is not the target charging device.

[0113] It can be understood that the steady-state charging information of the target charging device has strong intra-class commonality, is relatively aggregated within the class, and has small intra-class difference. Therefore, on the one hand, the steady-state charging information of the target charging device expressed by a single standard steady-state statistical feature is accurate enough, and the result of "not meeting the steady-state feature condition" obtained by comparing the steady-state statistical feature of the current steady-state waveform with the standard steady-state statistical feature is accurate enough. On the other hand, the large model does not need to be trained on multiple target charging device categories. Even if there is a new target charging category, the large model does not need to be retrained on the new target charging category to maintain the accuracy of the judgment of the large model, and the result of "not meeting the steady-state feature condition" is accurate enough. Therefore, the above scheme determines that the current charging device is not the target charging device in the case of "not meeting the steady-state feature condition".

[0114] For the result of "meeting the steady-state feature condition", since the steady-state charging information of the target charging device has inter-class non-dispersion, that is, the steady-state charging information of part of the non-charging device and the target charging device may be similar, for example, the steady-state charging information of the fast charging mobile phone, the notebook computer and the electric vehicle is similar, the result of "meeting the steady-state feature condition" cannot represent that the current charging device is the target charging device, and it can also be a non-target charging device, which needs to be further judged by S120-S140 or S220-S240 to reduce the probability of misjudgment.

[0115] Therefore, in the case of determining that the steady-state feature condition is not met based on the steady-state statistical feature, it is directly determined that the current charging device is not the target charging device. In the case of meeting the steady-state feature condition, it is not directly determined that the current charging device is the target charging device, but whether it is the target charging device is determined by S120-S140 or S220-S240. Thus, the calculation overhead can be reduced while reducing the probability of misjudgment.

[0116] Further, in some embodiments, based on the transient features of the current transient waveform and each reference transient waveform, a feature item and a complete reason on which the result of "whether the current charging device is the target charging device" is based are obtained. For example, which feature item has a large difference to cause the current charging result to be a non-target charging device. On the one hand, it is helpful for the developer to determine the optimization direction of the algorithm, and on the other hand, it is helpful for the operation and maintenance personnel to identify the fault direction when a fault occurs.

[0117] Further, in some embodiments, the transient waveform set and the steady-state waveform set can be stored in a retrieval library FAISS (Facebook AI Similarity Search).

[0118] For ease of understanding, the charging device classification method provided by the present application is described in the form of a specific example as follows.

[0119] Figure 6 is a flowchart of a specific example of the charging device classification method provided by the present application. As shown in Figure 6 , the charging device classification method comprises:

[0120] S410: obtaining a current transient waveform of a current charging device.

[0121] At the charging start period [T0, T1], the current transient waveform of the current charging device is collected. Wherein, T0 represents the first time start point (switching point), and T1 represents the first time end point.

[0122] S420: obtaining a current steady-state waveform of the current charging device. Specifically as follows:

[0123] S421: collecting the candidate steady-state waveforms 1-M of the continuous charging intermediate period 1 (T1, T2], the charging intermediate period 2 (T2, T3], …, the charging intermediate period M (T M-1 , T M ]. Wherein, T M-1 , T M represent the second time start point and the second time end point of the charging intermediate period M, respectively.

[0124] S422: obtaining the standard deviation of the candidate steady-state waveforms 1-M, respectively. The standard deviation calculation formula of the candidate steady-state waveform j (j∈1~M) is as follows:

[0125]

[0126] Wherein, V j represents the standard deviation of the candidate steady-state waveform j, N represents the number of sampling time points / power in the candidate steady-state waveform j, P i represents the i-th power, represents the average value of the N powers.

[0127] S423: judging whether the minimum value V min =min(V1, V2, …, V M ) of the standard deviations of the candidate steady-state waveforms 1-M is less than the standard deviation threshold value.

[0128] In response to less than the standard deviation threshold value, S424 is entered; in response to not less than the standard deviation threshold value, T1 is updated to T M , and S421-S423 are cyclically executed until less than the standard deviation threshold value.

[0129] S424: taking the sub-waveform of one of the time periods of the candidate steady-state waveform M as the current steady-state waveform.

[0130] Figure 7 is a schematic diagram of the current transient waveform and the current steady-state waveform of the present application.Figure 7 The current steady-state waveform is a time period sub-waveform of the charging intermediate period 1.

[0131] S430: Extract steady-state statistical features of the current steady-state waveform and perform normalization processing.

[0132] The steady-state statistical features include: waveform area composed of the first half time period sub-waveform and the horizontal axis, Euclidean distance of the coordinate points where the maximum power and the minimum power are located, waveform factor, waveform kurtosis of the first half time period sub-waveform, waveform variance, power peak value, and pulse width.

[0133] S440: Classify the steady-state statistical features of the current steady-state waveform using the trained OC-SVM (One-Class SVM) model to obtain a third similarity D t .

[0134] S450: Determine whether the third similarity D t is greater than a third similarity threshold T d .

[0135] In response to D t being greater than T d , proceed to S460; in response to D t not being greater than T d , proceed to S510.

[0136] S460: Search for K types of standard transient-state waveforms similar to the current transient-state waveform in the preset transient-state waveform set, respectively as K reference transient-state waveforms; search for K types of standard steady-state waveforms with the highest similarity to the current steady-state waveform in the preset steady-state waveform set, respectively as K reference steady-state waveforms.

[0137] S470: Extract transient-state statistical features of the current transient-state waveform and each reference transient-state waveform, respectively, and extract transient-state induction features between the current transient-state waveform and each reference transient-state waveform, respectively; extract steady-state statistical features of each reference steady-state waveform, respectively, and extract steady-state induction features between the current steady-state waveform and each reference steady-state waveform, respectively.

[0138] The transient-state statistical features include peak power, transient-state duration, and average power.

[0139] For example, the transient-state statistical features of the current transient-state waveform and the transient-state induction features between the current transient-state waveform and a reference transient-state waveform are as follows:

[0140]

[0141] Among them, the peak power, transient-state duration, and average power are transient-state statistical features, and the waveform difference description is a transient-state induction feature.

[0142] For example, the steady-state statistical features of the current steady-state waveform, and the steady-state inductive features between the current steady-state waveform and a reference steady-state waveform are as follows:

[0143] Wherein, the phase difference, power peak value, pulse width are the steady-state statistical features, and the waveform difference is described as the steady-state inductive feature.

[0144] S480: using the large model to classify based on the transient features of the current transient waveform and each reference transient waveform, and the steady-state features of the current steady-state waveform and each reference steady-state waveform, to obtain the total similarity P llm and the total similarity T p According to the feature item, the complete reason.

[0145] S490: judging whether the total similarity P llm is greater than the total similarity threshold T p Or, F = D t *P llm ( the product of the total similarity P llm and the third similarity D t ) is greater than the final similarity threshold T f .

[0146] In response to greater, enter S500; in response to not greater, enter S510.

[0147] S500: determining that the current charging device is a battery car.

[0148] S510: determining that the current charging device is not a battery car.

[0149] Figure 8 is the structural schematic diagram of an embodiment of the charging device classification apparatus provided by the present application. As shown in Figure 8 , the charging device classification apparatus 60 comprises an acquisition module 61, a retrieval module 62, an extraction module 63, and a judgment module 64.

[0150] The acquisition module 61 is used for acquiring the current transient waveform of the current charging device; wherein, the current transient waveform represents the distribution of the power consumption parameter of the current charging device at the charging start period.

[0151] The retrieval module 62 is used for retrieving at least one type of standard transient waveform similar to the current transient waveform from a preset transient waveform set as the reference transient waveform; wherein, the transient waveform set comprises various types of standard transient waveforms of the target charging device.

[0152] The extraction module 63 is used for extracting the transient features of the current transient waveform and each reference transient waveform.

[0153] The determining module 64 is configured to determine whether the current charging device is the target charging device based on the transient feature of the current transient waveform and the transient features of the reference transient waveforms.

[0154] For further details of this embodiment, please refer to the previous embodiment, which will not be repeated here.

[0155] Figure 9 is a structural schematic diagram of an embodiment of an electronic device of the present application. As shown in Figure 9 , the electronic device 70 includes a memory 71 and a processor 72, and the processor 72 is configured to execute program instructions stored in the memory 71 to implement the steps in any of the method embodiments described above. In a specific implementation scenario, the electronic device 70 can include but is not limited to a microcomputer, a server, and in addition, the electronic device 70 can also include a notebook computer, a tablet computer and other carrying devices, which are not limited here.

[0156] Specifically, the processor 72 is configured to control itself and the memory 71 to implement the steps in any of the method embodiments described above. The processor 72 can also be referred to as a CPU (Central Processing Unit). The processor 72 can be an integrated circuit chip with processing capability. The processor 72 can also be a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. In addition, the processor 72 can be implemented by an integrated circuit chip together.

[0157] Please refer to Figure 10 , Figure 10 is a structural schematic diagram of an embodiment of a computer-readable storage medium of the present application. The computer-readable storage medium 80 has program instructions 81 stored thereon, and the program instructions 81 are executed by a processor to implement the steps in any of the method embodiments described above.

[0158] In some embodiments, the apparatus provided by the embodiments of the present disclosure has functions or includes modules that can be used to execute the methods described in the above method embodiments, and the specific implementation can refer to the description of the above method embodiments. For the sake of brevity, it will not be repeated here.

[0159] The above description of the various embodiments tends to emphasize differences between the various embodiments, and the same or similar elements can be referred to one another, and will not be repeated here for brevity.

[0160] In several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the above-described device implementation is only schematic, for example, the division of the modules or units is only a logical function division, and actual implementation can have another division manner, for example, the units or components can be combined or integrated into another system, or some features can be ignored or not executed. In another image position, the coupling or direct coupling or communication connection between the displayed or discussed elements can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0161] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The above integrated unit can be realized in the form of hardware or in the form of a software functional unit. When the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application, essentially or say the part of the prior art that makes a contribution, or all or part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program codes that can be stored in the medium.

Claims

1. A method for classifying charging devices, characterized in that, include: Obtain the current transient waveform of the current charging device; wherein, the current transient waveform represents the distribution of the power consumption parameters of the current charging device during the charging start period; At least one type of standard transient waveform similar to the current transient waveform is retrieved from a preset transient waveform set and used as a reference transient waveform; wherein, the transient waveform set includes various types of the standard transient waveforms of the target charging device; Extract the transient features of the current transient waveform and each of the reference transient waveforms; Based on the transient characteristics of the current transient waveform and each of the reference transient waveforms, it is determined whether the current charging device is the target charging device.

2. The method according to claim 1, characterized in that, The extraction of transient features from the current transient waveform and each of the reference transient waveforms includes: Statistical analysis is performed on the current transient waveform and each of the reference transient waveforms to obtain corresponding transient statistical features. These transient statistical features characterize the distribution of each power consumption parameter in the corresponding current transient waveform or reference transient waveform during the charging start period; and / or Transient inductive features are extracted between the current transient waveform and each of the reference transient waveforms, and the transient inductive features characterize the shape difference between the current transient waveform and the corresponding reference transient waveform.

3. The method according to claim 1, characterized in that, The step of determining whether the current charging device is the target charging device based on the transient characteristics of the current transient waveform and each of the reference transient waveforms includes: Based on the transient characteristics of the current transient waveform and each of the reference transient waveforms, determine whether the target charging device category to which the current charging device belongs exists in the target charging device category corresponding to each of the reference transient waveforms; In response to the existence of a target charging device category to which the current charging device belongs, the current charging device is determined to be the target charging device; In response to the absence of a target charging device category to which the current charging device belongs, it is determined that the current charging device is not the target charging device.

4. The method according to claim 1, characterized in that, Before determining whether the current charging device is the target charging device based on the transient characteristics of the current transient waveform and each of the reference transient waveforms, the method further includes: Obtain the current steady-state waveform of the current charging device, wherein the current steady-state waveform represents the distribution of the power consumption parameters of the current charging device during the middle period of charging; At least one type of standard steady-state waveform similar to the current steady-state waveform is retrieved from a preset steady-state waveform set as a reference steady-state waveform. The steady-state waveform set includes various types of the standard steady-state waveforms of the target charging device. Extract the steady-state characteristics of the current steady-state waveform and each of the reference steady-state waveforms; The step of determining whether the current charging device is the target charging device based on the transient characteristics of the current transient waveform and each of the reference transient waveforms includes: Based on the transient characteristics of the current transient waveform and each of the reference transient waveforms, and the steady-state characteristics of the current steady-state waveform and each of the reference steady-state waveforms, it is determined whether the current charging device is the target charging device.

5. The method according to claim 4, characterized in that, The extraction of steady-state features of the current steady-state waveform and each of the reference steady-state waveforms includes: Statistical analysis is performed on the current steady-state waveform and each of the reference steady-state waveforms to obtain corresponding steady-state statistical features. These steady-state statistical features characterize the distribution of each power consumption parameter within the corresponding current steady-state waveform or reference steady-state waveform during the intermediate charging period; and / or Steady-state inductive features are extracted between the current steady-state waveform and each of the reference steady-state waveforms. These steady-state inductive features characterize the shape difference between the current steady-state waveform and the corresponding reference steady-state waveform.

6. The method according to claim 4, characterized in that, The step of obtaining the current steady-state waveform of the current charging device includes: After the charging start period, candidate steady-state waveforms of the current charging device are collected over several consecutive periods; The power consumption parameter trend values ​​of each candidate steady-state waveform are obtained respectively, and the power consumption parameter trend values ​​represent the fluctuation amplitude of each power consumption parameter in the candidate steady-state waveform; Determine whether the trend values ​​of each of the power consumption parameters meet the steady-state fluctuation conditions, wherein the steady-state fluctuation conditions include the smallest trend value of the power consumption parameter being less than a trend threshold; In response to satisfying the steady-state fluctuation condition, the current steady-state waveform is obtained based on one of the candidate steady-state waveforms.

7. The method according to claim 1, characterized in that, Before retrieving at least one type of standard transient waveform similar to the current transient waveform from a preset transient waveform set as a reference transient waveform, the method further includes: Based on the steady-state statistical characteristics of the current steady-state waveform of the current charging device, determine whether the current charging device satisfies the steady-state characteristic conditions corresponding to the target charging device; In response to satisfying the steady-state characteristic condition, the steps of searching for at least one type of standard transient waveform similar to the current transient waveform in a preset transient waveform set as a reference transient waveform and subsequent steps are performed. In response to the failure to meet the steady-state characteristic condition, it is determined that the current charging device is not the target charging device.

8. A charging system, characterized in that, It includes a charging device and a power supply device, the power supply device being used to perform the method according to any one of claims 1 to 7.

9. An electronic device, characterized in that, It includes a memory and a processor, the processor being configured to execute program instructions stored in the memory to implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores program instructions, characterized in that the program instructions, when executed by a processor, implement the method of any one of claims 1 to 7.

Images