Surge protection method, device and system, computer equipment, medium and battery

Abstract

The invention relates to a surge protection method, device and system, computer equipment, a medium and a battery, and can obtain output state data of charging equipment, input the output state data into a preset surge risk model for machine learning analysis, and determine a predicted surge risk level of a surge risk in a current state. After this, in combination with the predicted surge risk level, the number of surge suppressors connected between the battery management unit and the charging device is dynamically adjusted to suppress the output surge of the charging device. Through the scheme, the output surge of the charging equipment can be predicted in combination with the output state data, the surge suppression capability can be dynamically adjusted according to the prediction result, the surge risk can be warned in advance, and the charging equipment can be accessed to operate by using a suitable number of surge suppressors. Therefore, the abnormal surge output by the charging equipment can be predicted and protected, the risk that the battery management unit suffers from electrical overstress is greatly reduced, and the ablation condition of the battery management unit is relieved.

Description

Technical Field

[0001] The present application relates to the field of charging protection technology, and in particular to a surge protection method, device, system, computer equipment, medium and battery. Background Art

[0002] With the rapid development of new energy technologies, the coverage rate of new energy powered vehicles has gradually increased, and the demand for charging piles has also gradually increased.

[0003] In related technologies, when charging new energy vehicles with charging piles, abnormal output surges can easily cause the battery management unit to suffer electrical overstress, thereby causing internal ablation problems. Summary of the Invention

[0004] Based on this, there is a need for a surge protection method, device, system, computer equipment, medium and battery to protect against abnormal surges output by charging equipment, reduce the risk of electrical overstress to the battery management unit, and alleviate the occurrence of battery management unit ablation.

[0005] The present application provides a surge protection method, including: obtaining output status data of a charging device; performing machine learning analysis based on the output status data and a preset surge risk model to determine a predicted surge risk level; and dynamically adjusting the number of surge suppressors connected between a battery management unit and the charging device based on the predicted surge risk level.

[0006] In the above solution, multiple surge suppressors are installed at the front end of the battery management unit of the battery to be charged. During the process of the charging device charging the battery to be charged, the output status data of the charging device can be obtained, and this data is input into the preset surge risk model for machine learning analysis to determine the predicted surge risk level of the surge risk in the current state. After that, based on the predicted surge risk level, the number of surge suppressors connected between the battery management unit and the charging device is dynamically adjusted to suppress the output surge of the charging device. Through this solution, the output surge of the charging device can be predicted based on the output status data, and the surge suppression capacity can be dynamically adjusted based on the prediction results. The surge risk can be warned in advance, and the appropriate number of surge suppressors can be connected for operation. In this way, the prediction and protection of abnormal surges output by the charging device can be achieved, greatly reducing the risk of the battery management unit suffering from electrical overstress and alleviating the occurrence of battery management unit ablation.

[0007] In some embodiments, the performing of machine learning analysis based on the output state data and a preset surge risk model to determine a predicted surge risk level includes: performing feature extraction analysis based on the output state data to determine feature vector parameters; and performing machine learning analysis based on the feature vector parameters and a preset surge risk model to determine a predicted surge risk level.

[0008] In the above scheme, after obtaining the output status data, it is necessary to extract features from the output status data and input the feature vector parameters into the preset surge risk model for machine learning analysis. In this way, the input data type can be enriched and the accuracy of the machine learning results can be improved.

[0009] In some embodiments, performing feature extraction analysis based on the output state data to determine feature vector parameters includes: performing feature extraction analysis based on the output state data to determine initial feature vector parameters; performing dimensionality reduction analysis based on the initial feature vector parameters to determine feature vector parameters.

[0010] The above solution first performs feature extraction analysis to determine the initial eigenvector parameters, and then performs dimensionality reduction analysis on these initial eigenvector parameters to determine the eigenvector parameters. This reduces data storage space, reduces the complexity of subsequent calculations, and improves the efficiency of surge protection.

[0011] In some embodiments, the output state data includes output current parameters, output voltage parameters and noise parameters; the feature extraction analysis based on the output state data to determine the initial feature vector parameters includes: performing at least one of time domain analysis, frequency domain analysis and wavelet transform analysis on the output current parameters and output voltage parameters to determine the extracted feature parameters; and determining the initial feature vector parameters based on the extracted feature parameters and the noise parameters.

[0012] The above scheme combines the output current parameters and the output voltage parameters, performs at least one of time domain analysis, frequency domain analysis and wavelet transform analysis, and then determines the initial eigenvector parameters in combination with the noise parameters, which can effectively improve the richness of the initial eigenvector parameters.

[0013] In some embodiments, the output state data further includes temperature compensation parameters and / or aging compensation parameters, and the method further includes: determining initial feature vector parameters based on the extracted feature parameters, and the temperature compensation parameters and / or aging compensation parameters.

[0014] The above scheme, based on the output current parameters, output voltage parameters and noise parameters, further takes into account the temperature compensation parameters and / or aging compensation parameters, thereby increasing the data dimension of the input preset surge risk model and further improving the accuracy of the prediction results.

[0015] In some embodiments, the extracted characteristic parameters include at least one of voltage peak, current peak, voltage rise time, current rise time, voltage pulse density, current pulse density, voltage waveform kurtosis, current waveform kurtosis, voltage total harmonic distortion rate, current total harmonic distortion rate, voltage spectrum center of gravity, current spectrum center of gravity, voltage high-frequency energy ratio, current high-frequency energy ratio, voltage wavelet energy, current wavelet energy, voltage wavelet entropy and current wavelet entropy.

[0016] In the above scheme, the characteristic parameters extracted can be obtained by selecting at least one of time domain analysis, frequency domain analysis and wavelet transform analysis according to the actual scenario or requirements, thereby effectively improving the adaptability of the surge protection method.

[0017] In some embodiments, the higher the predicted surge risk level, the more surge suppressors are connected between the battery management unit and the charging device.

[0018] In the above scheme, when the predicted surge risk level is high, more surge suppressors are connected to perform surge suppression, so as to adapt the surge suppression capability to the predicted surge risk level and improve the accuracy of surge suppression.

[0019] In some embodiments, dynamically adjusting the number of surge suppressors connected between the battery management unit and the charging device based on the predicted surge risk level includes: when the predicted surge risk level is a first level, controlling a first number of surge suppressors to be connected between the battery management unit and the charging device; when the predicted surge risk level is a second level, controlling a second number of surge suppressors to be connected between the battery management unit and the charging device; wherein the risk level of the second level is higher than that of the first level, and the second number is greater than the first number; when the predicted surge risk level is a third level, controlling a third number of surge suppressors to be connected between the battery management unit and the charging device; wherein the risk level of the third level is higher than that of the second level, and the third number is greater than the second number.

[0020] The above scheme divides the predicted surge risk level into three different levels, and connects different numbers of surge suppressors at different levels. In this way, different surge protection strategies can be distinguished with fewer risk levels, thereby improving the reliability of surge protection operation.

[0021] In some embodiments, the method further includes: when the predicted surge risk level reaches the target risk level, switching the backup branch access to collect output status data; when the predicted surge risk level does not reach the target risk level, maintaining the main branch access to collect output status data; wherein the high-frequency noise suppression capability of the backup branch is stronger than that of the main branch.

[0022] The above solution allows for output status data collection using different branch connections. If the predicted surge risk level reaches the target risk level, the backup branch with stronger high-frequency noise suppression capabilities is switched to operation. This not only improves high-frequency noise suppression capabilities but also provides a certain degree of redundancy. If the backup branch fails at the target risk level, the main branch can remain operational.

[0023] In some embodiments, the method further includes: outputting a power reduction charging instruction when the predicted surge risk level reaches a target risk level; and maintaining charging operation at a current charging power when the predicted surge risk level does not reach the target risk level.

[0024] The above solution can output a power reduction charging instruction when the surge risk level is predicted to reach the target risk level, thereby reducing the output power of the charging equipment and improving charging safety.

[0025] In some embodiments, the method further includes: obtaining health status monitoring information of the surge suppressor; and dynamically adjusting the number of surge suppressors connected between the battery management unit and the charging device based on the predicted surge risk level and the health status monitoring information.

[0026] The above solution can also monitor the health status of the surge suppressor. By combining the health status monitoring information and the predicted surge risk level, the surge suppressor with good health status can be connected for operation to improve the reliability of surge suppression.

[0027] In some embodiments, the method further includes: when the mandatory target risk level execution conditions are met, adjusting the number of surge suppressors corresponding to the target risk level connected between the battery management unit and the charging device; when the mandatory target risk level execution conditions are not met, executing the step of dynamically adjusting the number of surge suppressors connected between the battery management unit and the charging device based on the predicted surge risk level.

[0028] The above scheme, when the conditions for executing the mandatory target risk level are met, no longer makes decisions based on the predicted surge risk level. Instead, it directly connects and operates the surge suppressors with the number corresponding to the target risk level, further improving the reliability of surge protection.

[0029] In some embodiments, the forced target risk level execution condition is determined to be met when at least one of the following items is met: the first item: the current pulse density determined according to the output state data is greater than a preset current density threshold; the second item: the voltage pulse density determined according to the output state data is greater than a preset voltage density threshold; the third item: the current change rate determined according to the output state data is greater than a preset current change threshold; the fourth item: the voltage change rate determined according to the output state data is greater than a preset voltage change threshold.

[0030] The above scheme can combine at least one of current pulse density, voltage pulse density, current change rate and voltage change rate to determine whether the mandatory target risk level execution conditions are met, and has a high accuracy in identifying mandatory target risk level execution conditions.

[0031] In some embodiments, the method further includes: updating the preset surge risk model according to the output status data acquired in real time.

[0032] The above solution can also be combined with the output status data and the preset surge risk model for updating, thereby greatly improving the accuracy of the preset surge risk model.

[0033] The present application also provides a surge protection device, including: a data acquisition module for acquiring output status data of a charging device; a risk prediction module for performing machine learning analysis based on the output status data and a preset surge risk model to determine a predicted surge risk level; and a dynamic protection module for dynamically adjusting the number of surge suppressors connected between a battery management unit and the charging device according to the predicted surge risk level.

[0034] The present application also provides a surge protection system, including a surge suppression array circuit, a monitoring component and a controller, wherein the monitoring component is arranged between the surge suppression array circuit and the battery management unit, the surge suppression array circuit includes a normally closed surge component and at least one controllable surge branch, the controllable surge branch is connected in parallel with the normally closed surge component, and the first end formed in parallel is connected to the charging device, and the second end formed in parallel is connected to the battery management unit, the controllable surge branch includes a controllable switch and a surge suppressor connected in series, the monitoring component and the controllable switch are respectively connected to the controller, and the controller is used to execute the steps of the above-mentioned surge protection method.

[0035] In some embodiments, the normally closed surge assembly includes a surge suppressor and a resettable fuse connected in series.

[0036] In the above solution, the normally closed surge component is connected in series with a resettable fuse on the basis of the surge suppressor. The resettable fuse can limit the current when the current is too large, thereby improving the operational reliability of the surge suppression array circuit.

[0037] In some embodiments, the system further includes a varistor and / or a common-mode choke connected in series between the surge suppression array circuit and the charging device.

[0038] The above solution can connect a varistor and / or a common-mode choke in series between the surge suppression array circuit and the charging device to absorb medium and high energy surges and / or suppress common-mode interference, further improving the operational reliability of the surge protection system.

[0039] In some embodiments, the charging device is connected to the surge suppression array circuit via a positive charging line and a negative charging line, and a gas discharge tube is further provided in parallel between the positive charging line and the negative charging line.

[0040] The above solution can also connect the charging line of the charging device and the surge suppression array circuit in parallel with the gas discharge tube to discharge ultra-high energy surges, thereby further improving the operational reliability of the surge protection system in a multi-level protection manner.

[0041] In some embodiments, the system further includes a switching switch, a main branch, and a backup branch, the input end of the switching switch is connected to the surge suppression array circuit, the first output end of the switching switch is connected to the monitoring component through the main branch, the second output end of the switching switch is connected to the monitoring component through the backup branch, the switching switch is connected to the controller, and the high-frequency noise suppression capability of the backup branch is stronger than that of the main branch.

[0042] The above solution allows for output status data collection using different branch connections. If the predicted surge risk level reaches the target risk level, the backup branch with stronger high-frequency noise suppression capabilities is switched to operation. This not only improves high-frequency noise suppression capabilities but also provides a certain degree of redundancy. If the backup branch fails at the target risk level, the main branch can remain operational.

[0043] The present application also provides a battery, including a battery management unit, a battery cell and the above-mentioned surge protection system.

[0044] The present application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned surge protection method when executing the computer program.

[0045] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned surge protection method are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present application. The same reference numerals are used throughout the drawings to represent the same components. In the drawings:

[0047] Figure 1 This is a flow chart of a surge protection method in some embodiments of the present application;

[0048] Figure 2 This is a schematic diagram of the structure of a surge protection system in some embodiments of the present application;

[0049] Figure 3 This is a flow chart of a surge protection method in some other embodiments of the present application;

[0050] Figure 4 This is a schematic diagram of the surge risk level prediction process in some embodiments of the present application;

[0051] Figure 5 This is a schematic diagram of the data processing flow in some embodiments of the present application;

[0052] Figure 6 Schematic diagram of data processing flow in other embodiments of the present application;

[0053] Figure 7 Schematic diagram of the surge protection method flow in some embodiments of the present application;

[0054] Figure 8 This is a flow chart of a surge protection method in some further embodiments of the present application;

[0055] Figure 9 This is a flow chart of a surge protection method in some other embodiments of the present application;

[0056] Figure 10 Schematic diagram of the surge protection method flow in some embodiments of the present application;

[0057] Figure 11 This is a flow chart of a surge protection method in some further embodiments of the present application;

[0058] Figure 12 This is a schematic structural diagram of a surge protection device in some embodiments of the present application;

[0059] Figure 13 Schematic diagram of the structure of a surge protection device in some other embodiments of the present application;

[0060] Figure 14 Schematic diagram of the structure of a surge protection device in some embodiments of the present application;

[0061] Figure 15 This is a schematic structural diagram of a surge protection device in some further embodiments of the present application;

[0062] Figure 16 Schematic diagram of the structure of a surge protection system in some other embodiments of the present application;

[0063] Figure 17 This is a schematic structural diagram of a surge protection system in some embodiments of the present application;

[0064] Figure 18 This is a schematic structural diagram of a surge protection system in some further embodiments of the present application. DETAILED DESCRIPTION

[0065] The following embodiments of the technical solution of the present application will be described in detail with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application and are therefore only examples and are not intended to limit the scope of protection of the present application.

[0066] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned figure descriptions are intended to cover non-exclusive inclusions.

[0067] In the description of the embodiments of this application, the technical terms "first" and "second" are used only to distinguish different objects and should not be understood to indicate or imply relative importance or implicitly specify the quantity, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, the meaning of "plurality" is more than two, unless otherwise clearly and specifically defined.

[0068] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0069] In the description of the embodiments of this application, the term "and / or" is simply a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent the following three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0070] In the description of the embodiments of the present application, the term "multiple" refers to more than two (including two). Similarly, "multiple groups" refers to more than two groups (including two groups), and "multiple pieces" refers to more than two pieces (including two pieces).

[0071] In the description of the embodiments of the present application, unless otherwise expressly specified or limited, technical terms such as "installed," "connected," "connected," and "fixed" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integration; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; internal connections between two components or interactions between two components. Those skilled in the art can understand the specific meanings of the above terms in the embodiments of the present application based on specific circumstances.

[0072] Currently, market developments indicate that batteries are increasingly being used. They are not only used in energy storage systems such as hydropower, thermal power, wind power, and solar power plants, but are also widely used in electric vehicles like electric bicycles, electric motorcycles, and electric cars, as well as in a variety of fields such as aerospace. As battery applications continue to expand, market demand is also growing.

[0073] These battery-powered devices often require recharging via charging equipment when the battery is depleted or nearly depleted. However, during the actual charging process, the battery management unit (BMU) is susceptible to electrical overstress, which can lead to internal corrosion. Research has found that abnormal surges are the root cause of this phenomenon. This can be caused by abnormal surges caused by grid fluctuations or by insulation degradation caused by long-term outdoor exposure of the charging equipment (due to humidity, salt spray, aging, etc.), leading to insulation failure between the low-voltage circuit and the high-voltage circuit. Once the insulation breaks down, the high-voltage electricity directly enters the low-voltage circuit, causing a surge shock.

[0074] To alleviate this phenomenon, a surge suppressor capable of suppressing surges can be deployed at the front end of the battery management unit (BMU), that is, between the BMU and the charging equipment (such as a charging station). Furthermore, given that different surge conditions require different surge suppression capabilities, the output status data of the BMU, which is output by the charging equipment to the battery to be charged, can be monitored to predict the risk level of a surge. Based on the prediction results, different numbers of surge suppressors can be connected to ensure that the actual surge suppression capability is compatible with the predicted surge risk level, extending the service life of the surge suppressors and optimizing the system life of the surge protection system.

[0075] Based on the above considerations, the present application provides a surge protection method. Multiple surge suppressors are installed at the front end of the battery management unit of the battery to be charged. During the process of charging the battery to be charged by the charging device, the output status data of the charging device can be obtained. This data is input into a preset surge risk model for machine learning analysis to determine the predicted surge risk level of the surge risk in the current state. After this, based on the predicted surge risk level, the number of surge suppressors connected between the battery management unit and the charging device is dynamically adjusted to suppress the output surge of the charging device.

[0076] This solution combines output status data to predict charging device output surges, dynamically adjusting surge suppression capabilities based on the predictions. This provides early warning of surge risks and allows for the appropriate number of surge suppressors to be connected and operated. This enables prediction and protection against abnormal charging device output surges, significantly reducing the risk of electrical overstress on the battery management unit and mitigating the risk of battery management unit burnout.

[0077] The surge protection method provided in the embodiments of the present application is used to perform surge prediction and protection between a charging device and a battery management unit of a battery to be charged. The type of charging device is not limited to a single type and can be a DC charging device or an AC charging device. The battery to be charged can be configured in any electrical device, including but not limited to electric vehicles, electric motorcycles, electric bicycles, etc., which are also not limited. For ease of understanding, the following embodiments all assume that the charging device is a charging pile and the battery to be charged is the power battery of an electric vehicle.

[0078] See also Figure 1 , the present application provides a surge protection method, including step 102, step 104 and step 106.

[0079] Step 102: Obtain output status data of the charging device.

[0080] Specifically, a charging device is a device used to output electrical energy to charge a battery to be charged. It can be a charging station or a charging base station. Output status data refers to parameters related to the electrical energy output by the charging device to the battery management unit of the battery to be charged. The type of output status data is not limited to a single type. Any parameter that can be used for surge assessment or changes due to surges can be used. It can be an electrical parameter obtained by direct measurement or a parameter obtained by further analysis and calculation of measured parameters. There is no limitation here.

[0081] There is no single way to obtain output status data. In one embodiment, the data can be collected by configuring relevant detection devices between the charging device and the battery management unit, such as a status monitoring component. The specific method may vary depending on the type of output status data and is not specifically limited.

[0082] Step 104 : Perform machine learning analysis based on the output state data and a preset surge risk model to determine a predicted surge risk level.

[0083] Specifically, the preset surge risk model is a surge risk level prediction model established by pre-analyzing surge risk tests based on the same type of charging equipment and batteries to be charged. The predicted surge risk level is the risk level that will trigger a surge impact under the current charging state. The number of predicted surge risk levels varies depending on the actual scenario and can be configured based on actual needs.

[0084] It should be noted that there is no single method for establishing the preset surge risk model. The preset surge risk model can be trained based on a Long Short-Term Memory (LSTM) network. In other embodiments, the preset surge risk model can also be trained based on a GRU (Gated Recurrent Unit), a convolutional neural network model, a recurrent neural network model, or the like. In other embodiments, the preset surge risk model can be trained using a combination of multiple different models, without limitation.

[0085] Step 106 : Dynamically adjust the number of surge suppressors connected between the battery management unit and the charging device according to the predicted surge risk level.

[0086] Specifically, a surge suppressor is a device that can suppress surges. A battery management unit (BMU) is a device that monitors and manages battery status, converts and transmits the power output by the charging device to the battery unit for charging, and manages the charging and discharging process.

[0087] It should be noted that there is no single type of surge suppressor. It can be a TVS (Transient Voltage Suppressor). In other embodiments, it can also be a metal oxide varistor, a semiconductor discharge tube, a gas discharge tube, a silicon avalanche diode, etc. In other embodiments, it can also be a combination of the above, without specific limitation. To facilitate understanding of the technical solution of this application, the following embodiments can all be understood as surge suppressors being TVS.

[0088] In one embodiment, the surge suppressor can be implemented as a surge suppressor array circuit, located at the front end of the battery management unit (BMU), that is, between the charging device and the BMU. In practical scenarios, the number of connected surge suppressors can be adjusted by simply switching different numbers of surge suppressors in the surge suppressor array circuit.

[0089] For more details, please refer to Figure 2 The surge suppression array circuit 10 includes a normally closed surge component 11 and at least one controllable surge branch 12. The controllable surge branch 12 is connected in parallel with the normally closed surge component 11, with a first end formed by the parallel connection connected to the charging device and a second end formed by the parallel connection connected to the battery management unit. The controllable surge branch 12 includes a controllable switch and a surge suppressor connected in series, and the controllable switch is connected to the controller 30. In this way, when the controllable surge branch 12 is not in operation, the normally closed surge component 11 remains in operation. Under the control of the controller, the number of controllable surge branches 12 in operation can be changed, thereby changing the number of surge suppressors connected between the battery management unit and the charging device.

[0090] In another embodiment, the surge suppression array circuit may include multiple controllable surge branches, and the number of surge suppressors connected between the battery management unit and the charging device may be changed by controlling the number of controllable surge branches put into operation.

[0091] In the above solution, multiple surge suppressors are installed at the front end of the battery management unit of the battery to be charged. During the process of the charging device charging the battery to be charged, the output status data of the charging device can be obtained, and this data is input into the preset surge risk model for machine learning analysis to determine the predicted surge risk level of the surge risk in the current state. After that, based on the predicted surge risk level, the number of surge suppressors connected between the battery management unit and the charging device is dynamically adjusted to suppress the output surge of the charging device. Through this solution, the output surge of the charging device can be predicted based on the output status data, and the surge suppression capacity can be dynamically adjusted based on the prediction results. The surge risk can be warned in advance, and the appropriate number of surge suppressors can be connected for operation. In this way, the prediction and protection of abnormal surges output by the charging device can be achieved, greatly reducing the risk of the battery management unit suffering from electrical overstress and alleviating the occurrence of battery management unit ablation.

[0092] See also Figure 3 In some embodiments, step 104 includes step 302 and step 304 .

[0093] Step 302: Perform feature extraction and analysis based on the output state data to determine feature vector parameters.

[0094] Step 304 : Perform machine learning analysis based on the characteristic vector parameters and the preset surge risk model to determine a predicted surge risk level.

[0095] Specifically, feature extraction and analysis involves analyzing and calculating the output state data to extract characteristic parameters of other charging states. Feature vector parameters are the state parameters ultimately derived by fusing the different feature parameters obtained through feature extraction and analysis. This embodiment does not directly perform learning and analysis on the collected output state data. Instead, it preprocesses the data to extract and fuse the characteristic vector parameters, which are then used for learning and analysis.

[0096] In the above scheme, after obtaining the output status data, it is necessary to extract features from the output status data and input the feature vector parameters into the preset surge risk model for machine learning analysis. In this way, the input data type can be enriched and the accuracy of the machine learning results can be improved.

[0097] See also Figure 4 In some embodiments, step 302 includes step 402 and step 404 .

[0098] Step 402: Perform feature extraction and analysis based on the output state data to determine initial feature vector parameters.

[0099] Step 404: Perform dimensionality reduction analysis based on the initial eigenvector parameters to determine the eigenvector parameters.

[0100] Specifically, the initial feature vector parameters are the feature vector parameters initially obtained after fusing the feature parameters extracted from the output state data. Dimensionality reduction analysis is to analyze the initial feature vector parameters to reduce the data dimension of the initial feature vector parameters.

[0101] It is understood that there is no single method for dimensionality reduction analysis. In one embodiment, PCA (Principal Components Analysis) dimensionality reduction technology can be used to reduce the dimensionality of the initial eigenvector parameters. Specifically, the initial eigenvector parameters are projected into a low-dimensional space through a linear transformation while retaining the main variation information in the data.

[0102] In other embodiments, other methods may be used to reduce the dimension of the initial feature vector parameters, such as linear discriminant analysis, multidimensional scaling, etc., which are not specifically limited and can be selected based on actual needs.

[0103] The above solution first performs feature extraction analysis to determine the initial eigenvector parameters, and then performs dimensionality reduction analysis on these initial eigenvector parameters to determine the eigenvector parameters. This reduces data storage space, reduces the complexity of subsequent calculations, and improves the efficiency of surge protection.

[0104] See also Figure 5In some embodiments, the output state data includes output current parameters, output voltage parameters, and noise parameters; step 402 includes steps 502 and 504 .

[0105] Step 502 : Perform at least one of time domain analysis, frequency domain analysis, and wavelet transform analysis on the output current parameter and the output voltage parameter to determine and extract characteristic parameters.

[0106] Step 504: Determine initial feature vector parameters based on the extracted feature parameters and noise parameters.

[0107] Specifically, the output voltage parameter is the voltage value of the electric energy output by the charging device to the battery management unit, which can be understood as the charging voltage; the output current parameter is the current value of the electric energy output by the charging device to the battery management unit, which can be understood as the charging current.

[0108] Time domain analysis refers to the analysis of signals in the time domain, focusing on how the signal changes over time, such as the amplitude, waveform, and period of the signal. Frequency domain analysis is a method of converting signals from the time domain to the frequency domain for analysis, focusing on the distribution and intensity of different frequency components in the signal. Wavelet transform analysis is a time-frequency analysis method that combines the characteristics of time domain and frequency domain analysis. It analyzes the characteristics of the signal at different times and frequencies by shifting and scaling the wavelet basis functions. The extracted characteristic parameters are the output current parameters and output voltage parameters, which are the parameters extracted after time domain analysis, frequency domain analysis, or wavelet transform analysis.

[0109] Depending on actual needs, the output current parameters and output voltage parameters may be subjected to only one of time domain analysis, frequency domain analysis, and wavelet transform analysis, or may be subjected to multiple of time domain analysis, frequency domain analysis, and wavelet transform analysis, without specific limitation.

[0110] The noise parameter, or the number or frequency of noise events, can be understood as a noise count. In actual analysis, the noise parameter can be directly input into a preset surge risk model for learning and analysis, or the noise parameter can be processed and then input into the preset surge risk model for learning and analysis. There are no specific limitations on this.

[0111] For example, in one embodiment, the noise parameter (noise count value) and the sampling period may be combined for calculation (such as the ratio of the two) to obtain the noise pulse density, and then the initial feature vector parameters may be determined by combining the noise pulse density and the extracted feature parameters.

[0112] The above scheme combines the output current parameters and the output voltage parameters, performs at least one of time domain analysis, frequency domain analysis and wavelet transform analysis, and then determines the initial eigenvector parameters in combination with the noise parameters, which can effectively improve the richness of the initial eigenvector parameters.

[0113] See also Figure 6 In some embodiments, the output state data further includes a temperature compensation parameter and / or an aging compensation parameter. The method further includes step 602 .

[0114] Step 602: determining initial feature vector parameters based on the extracted feature parameters, and the temperature compensation parameters and / or the aging compensation parameters.

[0115] Specifically, the temperature compensation parameter is the temperature value of the surge protection system located between the charging device and the battery management unit when the charging device is charging the rechargeable battery. The aging compensation parameter is a parameter that characterizes the different degrees of device aging. In a more detailed embodiment, a varistor (MOV) is connected in series between the surge suppression array circuit and the charging device. The temperature compensation parameter can be the surface temperature of the varistor collected in real time. Correspondingly, the aging compensation parameter is the aging degree parameter of the varistor.

[0116] The processing method for the output current parameters, output voltage parameters, and noise parameters is as described in the above embodiment and will not be repeated here. The solution of this embodiment, based on the output current parameters, output voltage parameters, and noise parameters, further considers the impact of temperature compensation parameters and / or aging compensation parameters on surge prediction results, and introduces temperature compensation parameters and / or aging compensation parameters for learning and analysis.

[0117] It can be understood that, in one embodiment, the temperature compensation parameters do not need to be further processed, and the initial feature vector parameters can be obtained by combining the temperature compensation parameters collected in real time and extracting the feature parameters and noise parameters for fusion analysis.

[0118] The above scheme, based on the output current parameters, output voltage parameters and noise parameters, further takes into account the temperature compensation parameters and / or aging compensation parameters, thereby increasing the data dimension of the input preset surge risk model and further improving the accuracy of the prediction results.

[0119] In some embodiments, the extracted characteristic parameters include at least one of voltage peak, current peak, voltage rise time, current rise time, voltage pulse density, current pulse density, voltage waveform kurtosis, current waveform kurtosis, voltage total harmonic distortion rate, current total harmonic distortion rate, voltage spectrum center of gravity, current spectrum center of gravity, voltage high-frequency energy ratio, current high-frequency energy ratio, voltage wavelet energy, current wavelet energy, voltage wavelet entropy and current wavelet entropy.

[0120] Specifically, voltage peak refers to the maximum value reached by voltage during a voltage cycle; current peak refers to the maximum value reached by current during a current cycle. Voltage rise time refers to the time required for voltage to rise from one threshold (such as 10%) to another threshold (such as 90%); current rise time refers to the time required for current to rise from one threshold to another. Voltage pulse density refers to the number of voltage pulses per unit time, reflecting the density of pulses in the signal; current pulse density refers to the number of current pulses per unit time. Voltage waveform kurtosis is a statistic that describes the sharpness of the voltage waveform, reflecting the degree of deviation of the voltage waveform from the normal distribution. Current waveform kurtosis is a statistic that describes the sharpness of the current waveform, reflecting the degree of deviation of the current waveform from the normal distribution.

[0121] The voltage total harmonic distortion (THD) is the ratio of the harmonic content to the fundamental content in the voltage signal, while the current total harmonic distortion is the ratio of the harmonic content to the fundamental content in the current signal. The voltage spectrum center of gravity is the center of the energy distribution in the voltage signal spectrum; the current spectrum center of gravity is the center of the energy distribution in the current signal spectrum. The voltage high-frequency energy ratio is the ratio of the high-frequency component energy to the total energy in the voltage signal; the current high-frequency energy ratio is the ratio of the high-frequency component energy to the total energy in the current signal. The voltage wavelet energy is the energy distribution of the voltage signal at different scales, obtained through wavelet transform; the current wavelet energy is the energy distribution of the current signal at different scales, obtained through wavelet transform. The voltage wavelet entropy is the entropy value of the voltage signal based on the wavelet transform; the current wavelet entropy is the entropy value of the current signal based on the wavelet transform.

[0122] It should be pointed out that in one embodiment, after receiving the output current parameters and output voltage parameters collected and sent by the monitoring component, the controller also needs to preprocess the output current parameters and output voltage parameters to obtain normalized voltage and normalized current. Subsequent time domain analysis, frequency domain analysis and wavelet transform analysis are based on the normalized voltage and normalized current.

[0123] Specifically, the normalized voltage is analyzed. After being sampled by the ADC (Analog-to-Digital Converter), the output voltage parameters need to be converted into the original voltage value. , to reflect the instantaneous voltage sampling value; among them, represents the sampling time, Indicates the original value of voltage, Indicates the output voltage parameter, that is, the real-time voltage signal, is the reference voltage value. The original current value is expressed as , reflects the instantaneous current sampling value, where is the output current parameter, that is, the real-time current signal.

[0124] After that, pre-processing is performed to obtain the voltage after power frequency notching ,in, The delay is half of the power frequency cycle, with a typical value of 0.02, reflecting the 50Hz (Hertz) signal notch delay parameter; voltage baseline ,in The sliding window size, the typical number of points is 1000 (can be set to other sizes according to actual needs), reflects the number of sampling points of baseline filtering; remove the baseline voltage , and finally obtain the normalized voltage (Obtained by analyzing multiple sampling data within a sampling period).

[0125] Using the same method as above, after power frequency notching and baseline removal, the normalized current can be obtained. , then the normalized voltage and normalized current Perform time domain analysis, frequency domain analysis and wavelet transform analysis.

[0126] In a more detailed embodiment, for time domain analysis:

[0127] Voltage peak Expressed as , current peak Expressed as The voltage rise time is the time it takes for the voltage to rise from 0.1 times the peak voltage to 0.9 times the peak voltage within a voltage cycle, i.e. , the current rise time is similar and can be expressed as .

[0128] Voltage pulse density According to Calculate the current pulse density According to Calculate; where, is the voltage differential threshold, which is a preset constant, typically 0.2 (can also be set to other values); is the current differential threshold, which is also a preset constant with a typical value of 0.15 (can also be set to other values).

[0129] Voltage waveform kurtosis According to Calculate the current waveform kurtosis According to Calculate, where is the mean value of the voltage signal, is the standard deviation of the voltage signal; is the mean value of the current signal, is the standard deviation of the current signal, which can be obtained by combining the above-mentioned normalized voltage and normalized current analysis, and will not be described in detail.

[0130] In a more detailed embodiment, for frequency domain analysis:

[0131] First, the normalized voltage and normalized current are Fourier transformed to output the complex spectrum of voltage and current, that is, and , which can be expressed as: , , is the Fourier transform operator, which reflects the linear transformation from time domain to frequency domain.

[0132] After this, the voltage total harmonic distortion rate According to The analysis shows that the total harmonic distortion rate of current is According to Analyze and obtain; among them, is the highest harmonic order, reflecting the harmonic analysis order, For the Subharmonic frequency ( ).

[0133] Voltage spectrum center of gravity According to The center of gravity of the current spectrum is obtained through analysis and calculation. According to The voltage high frequency energy ratio is obtained by analysis and calculation. According to The analysis and calculation show that the current high-frequency energy ratio According to Analytical calculations are performed; among them, is the starting frequency of high-frequency analysis, representing the lower limit of the noise frequency band; It is the high-frequency analysis cutoff frequency, representing the upper limit of the noise frequency band.

[0134] In a more detailed embodiment, the wavelet transform is analyzed:

[0135] First, the wavelet coefficients of the normalized voltage and normalized current are calculated, where the voltage wavelet coefficients are , current wavelet coefficient ;in, is the wavelet change operator, representing the multi-scale time-frequency analysis operator; , which belongs to the wavelet decomposition scale to assist in decomposition level indexing.

[0136] After that, the voltage wavelet energy and current wavelet energy are calculated. Taking values ​​1-6 in sequence, the voltage wavelet energy can be calculated respectively - , that is ; Similarly, the current wavelet energy can be calculated separately - , that is .

[0137] Normalize the voltage and current energy to get and . The voltage wavelet entropy and current wavelet entropy are calculated in sequence. The voltage wavelet entropy According to Perform analysis and calculation, current wavelet entropy According to Obtained by analysis and calculation.

[0138] In the above scheme, the characteristic parameters extracted can be obtained by selecting at least one of time domain analysis, frequency domain analysis and wavelet transform analysis according to the actual scenario or requirements, thereby effectively improving the adaptability of the surge protection method.

[0139] There is more than one way to perform machine learning analysis based on output status data and a preset surge risk model. The specific learning and analysis methods vary depending on the preset surge risk model. To facilitate understanding, this article uses LSTM as an example to explain.

[0140] The input feature vector parameters are analyzed based on the LSTM gating mechanism, including:

[0141] Forget Gate: ,in, is the sigmoid activation function (the range is 0-1), is the forget gate weight matrix, Hidden With the input feature vector parameters The concatenated vector, is the forget gate bias vector.

[0142] Input Gate: , ,in, is the input gate weight matrix , is the input gate bias vector, is the candidate state weight matrix, is the candidate state bias vector.

[0143] Cell status update , output gate: , ,in, is the output gate bias vector.

[0144] After that, the risk probability is calculated based on the above output:

[0145]

[0146] in, is the output layer weight matrix, is the hidden state matrix, is the output layer bias matrix, represents the risk probability, 、 and It represents the probability of occurrence of different surge risk levels, and the sum of the three is 1 (taking three predicted surge risk levels as an example).

[0147] Ultimately, probability dominates decision-making, and instructions are produced according to the risk probability matrix, outputting the final predicted surge risk level: .

[0148] It should be noted that in one embodiment, when the temperature compensation parameter is taken into account, the temperature compensation parameter needs to be introduced when calculating the risk probability, and the same analysis as above is used to determine the probability matrix without including the aging compensation parameter. , and in Based on the compensation probability caused by the introduction of aging compensation parameters , correspondingly, the final risk probability matrix at this time for: ,in, , is the MOV aging factor (0-1, 0 = new product state, 1 = scrapped state), just need to get Substituting the probability-driven decision into the above formula, the predicted surge risk level under this embodiment can be determined: .

[0149] In some embodiments, the higher the predicted surge risk level, the more surge suppressors are connected between the battery management unit and the charging device.

[0150] Specifically, a higher predicted surge risk level indicates a more severe surge. In this case, more surge suppressors need to be put into operation to meet the surge suppression requirements.

[0151] In the above scheme, when the predicted surge risk level is high, more surge suppressors are connected to perform surge suppression, so as to adapt the surge suppression capability to the predicted surge risk level and improve the accuracy of surge suppression.

[0152] See also Figure 7In some embodiments, step 106 includes step 702 , step 704 , and step 706 .

[0153] Step 702 : When the predicted surge risk level is the first level, control a first number of surge suppressors to be connected between the battery management unit and the charging device.

[0154] Step 704 : When the predicted surge risk level is the second level, control a second number of surge suppressors to be connected between the battery management unit and the charging device.

[0155] Step 706 : When the predicted surge risk level is the third level, control a third number of surge suppressors to be connected between the battery management unit and the charging device.

[0156] Specifically, the second level has a higher risk level than the first level, and the second number is greater than the first number; the third level has a higher risk level than the second level, and the third number is greater than the second number. For ease of understanding, in one embodiment, the first risk level can be understood as a low risk level, the second risk level as a medium risk level, and the third risk level as a high risk level.

[0157] It should be noted that the first, second, and third quantities are not unique. In one embodiment, the first quantity can be set to 0, and in other embodiments, the first quantity can be set to 1, without specific limitation, as long as it is lower than the second quantity. The third quantity is not unique. In a more detailed embodiment, all surge suppressors can be connected and operated at the third level. In other embodiments, the third quantity can be set to other values, without specific limitation, as long as it is greater than the second quantity.

[0158] For ease of understanding, in one embodiment, the number of surge suppressors is four as an example for explanation. When the predicted surge risk level is the first level, one surge suppressor is connected for operation; when the predicted surge risk level is the second level, two or three surge suppressors are connected for operation; and when the predicted surge risk level is the third level, four surge suppressors are connected for operation.

[0159] The above scheme divides the predicted surge risk level into three different levels, and connects different numbers of surge suppressors at different levels. In this way, different surge protection strategies can be distinguished with fewer risk levels, thereby improving the reliability of surge protection operation.

[0160] See also Figure 8 In some embodiments, the method further includes steps 802 and 804.

[0161] Step 802 : When the predicted surge risk level reaches the target risk level, the standby branch is switched to access the standby branch to collect output status data.

[0162] Step 804 : When the predicted surge risk level does not reach the target risk level, the main branch is maintained connected to collect output status data.

[0163] Specifically, the backup branch has a stronger high-frequency noise suppression capability than the main branch. The main branch, which connects the monitoring component and the surge suppression array circuit to collect output status data when the surge risk level does not reach the target risk level, and the auxiliary branch, which connects the monitoring component and the surge suppression array circuit to collect output status data when the surge risk level reaches the target risk level, also connects the monitoring component and the surge suppression array circuit to collect output status data.

[0164] The structure of the main branch and the auxiliary branch is not unique. As long as the high-frequency noise suppression capability of the auxiliary branch is stronger than that of the main branch, so that more reliable high-frequency noise suppression can be achieved at the target risk level, it is acceptable. For example, in one embodiment, the main branch includes a connecting line, and the auxiliary branch includes a filter circuit. In another embodiment, the main branch may also include a filter circuit, and the auxiliary branch may include a filter circuit with stronger filtering capability, and the specific embodiment is not limited. For example, in a more detailed embodiment, the main branch includes an LC (inductor-capacitor) filter circuit, and the auxiliary branch includes a π-type filter circuit. More specifically, the π-type filter circuit is constructed by magnetic beads and ceramic capacitors, and the specific selection can be based on actual needs.

[0165] It is understood that there is no single method for switching between the main branch and the auxiliary branch. In one embodiment, both branches can be connected to a switch, and the switching between the main branch and the auxiliary branch can be achieved through the switching control of the switch. For example, the input of the switch is connected to the surge suppression array circuit, the first output of the switch is connected to the monitoring component via the main branch, the second output of the switch is connected to the monitoring component via the backup branch, and the switch is connected to the controller. It should be noted that the type of switch is not limited to a single type and can be a relay, a single-pole double-throw switch, etc., without specific limitation.

[0166] The target risk level selection is not exclusive and may be configured differently based on actual needs, without specific limitations. For example, in one embodiment, the highest risk level among the predicted surge risk levels may be used as the target risk level; in another embodiment, the target risk level may be an intermediate predicted surge risk level. For example, if there are three levels, the second level may be used as the target risk level. The configuration can be tailored to actual needs.

[0167] The above solution allows for output status data collection using different branch connections. If the predicted surge risk level reaches the target risk level, the backup branch with stronger high-frequency noise suppression capabilities is switched to operation. This not only improves high-frequency noise suppression capabilities but also provides a certain degree of redundancy. If the backup branch fails at the target risk level, the main branch can remain operational.

[0168] See also Figure 9 In some embodiments, the method further includes step 902 and step 904.

[0169] Step 902: When the predicted surge risk level reaches the target risk level, output a power reduction charging instruction.

[0170] Step 904 : When the predicted surge risk level does not reach the target risk level, the charging operation is maintained at the current charging power.

[0171] Specifically, the power reduction instruction instructs the charging device to reduce the charging power. If the controller detects that the predicted surge risk level has reached the target risk level, it can return a power reduction instruction to the charging device to reduce the charging power. If the predicted surge risk level does not reach the target risk level, that is, the predicted surge risk level is low, the current charging power is maintained to maintain charging efficiency.

[0172] Similarly, the selection of the target risk level is not unique. In one embodiment, the highest risk level among the predicted surge risk levels can be used as the target risk level; in another embodiment, the middle predicted surge risk level can also be used as the target risk level. For example, in the case of three levels, the second level can be used as the target risk level. The specific configuration can be based on actual needs.

[0173] The above solution can output a power reduction charging instruction when the surge risk level is predicted to reach the target risk level, thereby reducing the output power of the charging equipment and improving charging safety.

[0174] See also Figure 10 In some embodiments, the method further includes step 1002 and step 1004 .

[0175] Step 1002: Obtain health status monitoring information of the surge suppressor.

[0176] Step 1004 : Dynamically adjust the number of surge suppressors connected between the battery management unit and the charging device based on the predicted surge risk level and health status monitoring information.

[0177] Specifically, the health status monitoring information is information obtained through real-time monitoring and is used to characterize the health status of each surge suppressor. In actual scenarios, due to differences in the operating time of each surge suppressor, there are certain differences in the health status of each surge suppressor. In order to balance the life differences between the various surge suppressors, or to still provide surge protection when the life of the surge suppressor is exhausted, this embodiment needs to take into account the health status monitoring information of the surge suppressor and dynamically adjust the access surge suppressor. In this way, even at the same predicted surge risk level, the same number of different surge suppressors can be connected; or at different predicted surge risk levels, the same surge suppressor can be connected and operated according to demand.

[0178] For example, in a more detailed embodiment, the controller stores corresponding protection strategies for different predicted surge risk levels. At the same predicted surge risk level, the connected surge suppressor is fixed. However, based on the health monitoring information of the surge suppressor, the protection strategy can be adjusted during actual control to connect different surge suppressors (with the same number) for the same predicted surge risk level.

[0179] The above solution can also monitor the health status of the surge suppressor. By combining the health status monitoring information and the predicted surge risk level, the surge suppressor with good health status can be connected for operation to improve the reliability of surge suppression.

[0180] See also Figure 11 In some embodiments, the method further includes step 112 .

[0181] Step 112 : When the mandatory target risk level execution condition is met, adjust the number of surge suppressors corresponding to the target risk level to be connected between the battery management unit and the charging device.

[0182] If the mandatory target risk level execution condition is not met, step 106 is executed.

[0183] Specifically, the mandatory target risk level execution condition is a condition that is satisfied when the surge protection system is forced to execute a protection operation according to a protection strategy corresponding to the target risk level without undergoing a probabilistic decision.

[0184] Similarly, the selection of the target risk level is not unique. In one embodiment, the highest risk level among the predicted surge risk levels can be used as the target risk level; in another embodiment, the middle predicted surge risk level can also be used as the target risk level. For example, in the case of three levels, the second level can be used as the target risk level. The specific configuration can be based on actual needs.

[0185] Taking the target risk level as the highest risk level as an example, in one embodiment, when it is detected that the mandatory target risk level execution conditions are met, the controller will directly enter the surge protection stage and control the number of surge suppressors corresponding to the target risk level (such as 4) to be connected between the battery management unit and the charging device.

[0186] The above scheme, when the conditions for executing the mandatory target risk level are met, no longer makes decisions based on the predicted surge risk level. Instead, it directly connects and operates the surge suppressors with the number corresponding to the target risk level, further improving the reliability of surge protection.

[0187] In some embodiments, the forced target risk level execution condition is determined to be met when at least one of the following items is met: the first item: the current pulse density determined according to the output status data is greater than the preset current density threshold; the second item: the voltage pulse density determined according to the output status data is greater than the preset voltage density threshold; the third item: the current change rate determined according to the output status data is greater than the preset current change threshold; the fourth item: the voltage change rate determined according to the output status data is greater than the preset voltage change threshold.

[0188] Specifically, the current pulse density and voltage pulse density are determined as described in the above embodiments and will not be further described here. The current change rate is the rate of change of current over time, and the voltage change rate is the rate of change of voltage over time. In actual scenarios, considering that a surge usually causes a sudden change in voltage and current, at least one of the current pulse density, voltage pulse density, current change rate, and voltage change rate can be combined to analyze whether the mandatory target risk level execution conditions are met at this time.

[0189] The values ​​of the preset current density threshold, preset voltage density threshold, preset voltage change threshold, and preset current change threshold are not unique. As long as it indicates that a surge corresponding to the target risk level occurs when the threshold is reached, the configuration is specific to the actual scenario.

[0190] The above scheme can combine at least one of current pulse density, voltage pulse density, current change rate and voltage change rate to determine whether the mandatory target risk level execution conditions are met, and has a high accuracy in identifying mandatory target risk level execution conditions.

[0191] In some embodiments, the method further includes: updating a preset surge risk model according to the output status data acquired in real time.

[0192] Specifically, when performing prediction analysis based on the preset surge risk model, this embodiment can also perform model retraining in combination with the output status data obtained in real time, thereby optimizing the parameters in the preset surge risk model, that is, updating the preset surge risk model.

[0193] The above solution can also be combined with the output status data and the preset surge risk model for updating, thereby greatly improving the accuracy of the preset surge risk model.

[0194] In order to facilitate understanding of the technical solution of the present application, the present application is explained below in conjunction with more detailed embodiments.

[0195] Signal acquisition layer: When charging starts, the output status data of the charging device (charging pile) output to the battery to be charged is obtained in real time, including output voltage parameters, output current parameters and noise parameters. These parameters can be collected by monitoring components and sent to the controller.

[0196] Feature extraction layer: After receiving the output state data, the controller first performs preprocessing to obtain normalized current and normalized voltage. Then, the normalized voltage and current are combined for time domain analysis, frequency domain analysis, and wavelet transform analysis to obtain voltage peak, current peak, voltage rise time, current rise time, voltage pulse density, current pulse density, voltage waveform kurtosis, current waveform kurtosis, voltage total harmonic distortion (THD), current total harmonic distortion (THD), voltage spectrum center of gravity, current spectrum center of gravity, voltage high-frequency energy ratio, current high-frequency energy ratio, voltage wavelet energy, current wavelet energy, voltage wavelet entropy, and current wavelet entropy.

[0197] For the noise parameters, the corresponding noise density parameters are obtained through preprocessing. After that, feature fusion analysis is performed on the above parameters and the noise density parameters to obtain the initial eigenvector parameters, and PCA dimensionality reduction analysis is performed on the initial eigenvector parameters to obtain the eigenvector parameters.

[0198] The voltage change rate and current change rate can also be obtained in the feature extraction layer. At this time, the controller can combine the current pulse density and voltage pulse density to analyze whether the execution conditions of the mandatory target risk level (high risk) are met. If this condition is met, it will directly enter the dynamic protection layer and control the number of surge suppressors corresponding to the high risk level to be connected between the battery management unit and the charging equipment. That is, all TVS in the surge suppressor (TVS) array circuit are connected and operating. In addition, the backup branch connection can be switched to collect output status data and output a power reduction charging instruction to reduce the charging power. If this condition is not met, it will enter the subsequent machine learning layer for analysis, and use the machine learning analysis results for protection control.

[0199] Machine Learning Layer: The controller performs learning analysis based on the preset surge risk model and the input feature vector parameters to determine the risk probabilities corresponding to different risk levels. It then performs a probabilistic decision-making analysis on the resulting risk probabilities to determine the risk level that dominates the decision-making process. This allows for surge risk level prediction and output of the predicted surge risk level.

[0200] Dynamic protection layer:

[0201] In the case of a low risk level, a TVS that controls the surge suppression array circuit is connected between the battery management unit and the charging device; at this time, the main branch is maintained for output status data collection and the current charging power is maintained.

[0202] In the case of a medium risk level, the two TVSs that control the surge suppression array circuit are connected between the battery management unit and the charging device. At this time, the main branch is maintained for output status data collection and the current charging power is maintained.

[0203] In high-risk situations, the four TVSs that control the surge suppression array circuit are fully connected between the battery management unit and the charging device. At this time, the backup branch is switched to collect output status data and a power reduction charging instruction is output to reduce the charging power.

[0204] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily performed in sequence in the order indicated by the arrows. Unless clearly stated herein, the execution of these steps is not strictly limited in order, and these steps can be performed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above may include multiple steps or multiple stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of the steps or stages in other steps or other steps. It is understandable that the various steps in different embodiments can be freely combined as needed, and the various non-contradictory schemes formed by the combination all fall within the scope of protection of this application.

[0205] Based on the same inventive concept, embodiments of the present application also provide a surge protection device for implementing the aforementioned surge protection method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more surge protection device embodiments provided below can be found in the above-described limitations of the surge protection method and will not be further elaborated here.

[0206] See also Figure 12 The present application also provides a surge protection device, including a data acquisition module 122 , a risk prediction module 124 and a dynamic protection module 126 .

[0207] The data acquisition module 122 is used to obtain the output status data of the charging device; the risk prediction module 124 is used to perform machine learning analysis based on the output status data and a preset surge risk model to determine the predicted surge risk level; the dynamic protection module 126 is used to dynamically adjust the number of surge suppressors connected between the battery management unit and the charging device according to the predicted surge risk level.

[0208] In some embodiments, the risk prediction module 124 is further configured to perform feature extraction analysis based on the output state data to determine feature vector parameters; and perform machine learning analysis based on the feature vector parameters and a preset surge risk model to determine a predicted surge risk level.

[0209] In some embodiments, the risk prediction module 124 is further configured to perform feature extraction analysis based on the output state data to determine initial feature vector parameters; and perform dimensionality reduction analysis based on the initial feature vector parameters to determine feature vector parameters.

[0210] In some embodiments, the risk prediction module 124 is also used to perform at least one of time domain analysis, frequency domain analysis, and wavelet transform analysis on the output current parameters and the output voltage parameters to determine the extracted feature parameters; and determine the initial feature vector parameters based on the extracted feature parameters and the noise parameters.

[0211] In some embodiments, the risk prediction module 124 is further configured to determine initial feature vector parameters based on the extracted feature parameters, and the temperature compensation parameters and / or the aging compensation parameters.

[0212] In some embodiments, the dynamic protection module 126 is also used to control a first number of surge suppressors to be connected between the battery management unit and the charging device when the predicted surge risk level is the first level; to control a second number of surge suppressors to be connected between the battery management unit and the charging device when the predicted surge risk level is the second level; and to control a third number of surge suppressors to be connected between the battery management unit and the charging device when the predicted surge risk level is the third level.

[0213] See also Figure 13 In some embodiments, the device further includes a path switching module 132 .

[0214] The path switching module 132 is used to switch the backup branch access to collect output status data when the predicted surge risk level reaches the target risk level; and to maintain the main branch access to collect output status data when the predicted surge risk level does not reach the target risk level.

[0215] See also Figure 14 In some embodiments, the device further includes a power regulation module 142 .

[0216] The power regulation module 142 is configured to output a power reduction charging instruction when the predicted surge risk level reaches the target risk level; and to maintain the current charging power charging operation when the predicted surge risk level does not reach the target risk level.

[0217] In some embodiments, the dynamic protection module 126 is further used to obtain health status monitoring information of the surge suppressor; and dynamically adjust the number of surge suppressors connected between the battery management unit and the charging device based on the predicted surge risk level and the health status monitoring information.

[0218] In some embodiments, the dynamic protection module 126 is also used to adjust the number of surge suppressors connected between the battery management unit and the charging device according to the target risk level when the mandatory target risk level execution conditions are met; when the mandatory target risk level execution conditions are not met, dynamically adjust the number of surge suppressors connected between the battery management unit and the charging device according to the predicted surge risk level.

[0219] See also Figure 15 In some embodiments, the apparatus further includes a model updating module 152 .

[0220] The model updating module 152 is used to update the preset surge risk model according to the output status data obtained in real time.

[0221] Each module in the aforementioned surge protection device may be implemented in whole or in part through software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0222] See also Figure 2 The present application also provides a surge protection system, including a surge suppression array circuit 10, a monitoring component 20 and a controller 30. The monitoring component 20 is arranged between the surge suppression array circuit 10 and the battery management unit. The surge suppression array circuit 10 includes a normally closed surge component 11 and at least one controllable surge branch 12. The controllable surge branch 12 is connected in parallel with the normally closed surge component 11, and the first end formed in parallel is connected to the charging device, and the second end formed in parallel is connected to the battery management unit. The controllable surge branch 12 includes a controllable switch and a surge suppressor connected in series. The monitoring component 20 and the controllable switch are respectively connected to the controller 30. The controller 30 is used to execute the steps of the above-mentioned surge protection method.

[0223] Specifically, the normally closed surge component 11 is a surge suppression device that maintains connection between the charging device and the battery management unit at different predicted surge risk levels. It is understood that to improve operational safety, the surge detector and the normally closed surge component 11 should be grounded, which is not detailed here.

[0224] The monitoring component 20 is a device used to monitor the output status data of the charging device. Its type is not unique and may vary depending on the output status data. In one embodiment, the monitoring component 20 includes at least one of a current detector, a voltage detector (which may be a voltage divider circuit, etc., but is not limited to this), and a temperature detector. Furthermore, in one embodiment, if aging compensation parameters or surge suppressor health monitoring information are required, the monitoring component 20 may also include a device or apparatus for device aging detection. The selection can be based on actual needs.

[0225] The implementation of the surge protection method is as shown in the above embodiments and will not be repeated here. In actual scenarios, the number of surge suppressors connected between the battery management unit and the charging device can be adjusted by controlling the number of controllable surge branches 12 that are turned on.

[0226] This solution combines output status data to predict charging device output surges and dynamically adjusts surge suppression capabilities based on the predictions. This provides early warning of surge risks and allows for the appropriate number of surge suppressors to be connected and operated. This enables prediction and protection against abnormal charging device output surges, significantly reducing the risk of electrical overstress on the battery management unit and mitigating the risk of battery management unit burnout.

[0227] Please refer to Figure 16 In some embodiments, the normally closed surge assembly 11 includes a surge suppressor and a resettable fuse connected in series.

[0228] Specifically, a resettable fuse, also known as a polymer positive temperature coefficient (PPTC) device, has conductive particles that form a conductive path under normal operating conditions, allowing current to flow smoothly. When overcurrent or overheating occurs in the circuit, the polymer matrix expands due to the heat, causing the contact points between the conductive particles to disconnect and the resistance to increase sharply, thereby limiting the flow of current and providing protection. The device can automatically recover after the fault is eliminated.

[0229] It is understandable that in other embodiments, the normally closed surge assembly 11 may also include only a surge suppressor, specifically one surge suppressor, or multiple surge suppressors connected in series and / or in parallel. There is no specific limitation and the selection can be made based on actual needs.

[0230] In the above solution, the normally closed surge component 11 is connected in series with a resettable fuse on the basis of the surge suppressor. The resettable fuse can limit the current when the current is too large, thereby improving the operating reliability of the surge suppression array circuit 10.

[0231] See also Figure 17 In some embodiments, the system further includes a varistor 40 and / or a common mode choke 50 connected in series between the surge suppression array circuit 10 and the charging device.

[0232] Specifically, varistor 40 is a component whose resistance changes with voltage. When the voltage exceeds a certain threshold, the resistance of varistor 40 drops sharply, thereby limiting further voltage increases and providing overvoltage protection. Common-mode choke 50 is an inductive component used to suppress common-mode noise. It generates a reverse magnetic field to offset common-mode noise, thereby reducing its impact on the circuit.

[0233] In this embodiment, a varistor 40 is connected in series with the front end of the PPTC, that is, between the PPTC and the charging device, to absorb medium- to high-energy surges (typically 1kA (kiloamperes) to 10kA) in the protection system. Furthermore, a common-mode choke 50 can be connected in series with the front end of the PPTC to offset common-mode noise.

[0234] In the above solution, a varistor 40 and / or a common-mode choke 50 can be connected in series between the surge suppression array circuit 10 and the charging device to absorb medium and high energy surges and / or suppress common-mode interference, further improving the operational reliability of the surge protection system.

[0235] In some embodiments, a current detector can be connected in series with the output of the surge suppression array circuit 10, that is, between the surge suppression array circuit 10 and the battery management unit; and a voltage detector can be connected in parallel with the output of the surge suppression array circuit 10. A temperature detector can be disposed on the heat dissipation surface of the varistor 40 to monitor the temperature.

[0236] Please continue reading Figure 17 In some embodiments, the charging device is connected to the surge suppression array circuit 10 via a positive charging line (L+) and a negative charging line (L-), and a gas discharge tube 60 is also provided in parallel between the positive charging line and the negative charging line.

[0237] Specifically, the gas discharge tube (GDT) 60 is a high-voltage component that utilizes gas ionization for surge protection. It features high current capacity, high insulation resistance, and low inter-electrode capacitance. This embodiment, by placing a GDT 60 in parallel between the positive and negative charging lines, can dissipate ultra-high-energy surges exceeding 10kA, providing an additional level of protection for battery charging.

[0238] The above solution can also connect the charging line of the charging device and the surge suppression array circuit 10 in parallel with the gas discharge tube 60 to discharge ultra-high energy surges, thereby further improving the operational reliability of the surge protection system in a multi-level protection manner.

[0239] See also Figure 18 In some embodiments, the system further includes a switching switch 71, a main branch 72, and a backup branch 73. The input end of the switching switch 71 is connected to the surge suppression array circuit 10, the first output end of the switching switch 71 is connected to the monitoring component 20 through the main branch 72, the second output end of the switching switch 71 is connected to the monitoring component 20 through the backup branch 73, the switching switch 71 is connected to the controller 30, and the high-frequency noise suppression capability of the backup branch 73 is stronger than that of the main branch 72.

[0240] Specifically, the main branch 72 connects the monitoring component 20 and the surge suppression array circuit 10 when the surge risk level does not reach the target risk level to realize the line for collecting output status data; the auxiliary branch connects the monitoring component 20 and the surge suppression array circuit 10 when the surge risk level reaches the target risk level to realize the line for collecting output status data.

[0241] The structures of the main branch 72 and the auxiliary branch are not unique. As long as the high-frequency noise suppression capability of the auxiliary branch is stronger than that of the main branch 72, so that more reliable high-frequency noise suppression can be achieved at the target risk level, any structure is acceptable. For example, in one embodiment, the main branch 72 includes a connecting line, and the auxiliary branch includes a filter circuit. In another embodiment, the main branch 72 may also include a filter circuit, while the auxiliary branch includes a filter circuit with stronger filtering capability, and the specific embodiment is not limited. For example, in a more detailed embodiment, the main branch 72 includes an LC (inductor-capacitor) filter circuit, and the auxiliary branch includes a π-type filter circuit. More specifically, the π-type filter circuit is constructed using magnetic beads and ceramic capacitors, and the specific selection can be based on actual needs.

[0242] The above solution allows different branches to be connected for output status data collection. If the predicted surge risk level reaches the target risk level, the backup branch 73, which has stronger high-frequency noise suppression capabilities, is switched to operation. This not only improves high-frequency noise suppression capabilities but also provides a certain degree of redundancy. If the backup branch 73 fails at the target risk level, the main branch 72 can remain operational.

[0243] The present application also provides a battery, including a battery management unit, a battery cell and the above-mentioned surge protection system.

[0244] Specifically, a battery unit is a unit used to store electrical energy, which can be a battery cell, a battery pack formed by connecting multiple battery cells in series and / or in parallel, or a battery pack formed by connecting multiple battery packs in series and / or in parallel, without specific limitation.

[0245] The structure and implementation of the surge protection system are as described in the various embodiments and accompanying drawings above and will not be further elaborated here. This solution allows the output surge of the charging device to be predicted in conjunction with output status data during battery charging, and the surge suppression capability to be dynamically adjusted based on the prediction results. This provides early warning of surge risks and allows the appropriate number of surge suppressors to be connected and operated. This enables the prediction and protection of abnormal surges output by the charging device, significantly reducing the risk of electrical overstress on the battery management unit and mitigating the occurrence of battery management unit ablation.

[0246] The present application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps of the surge protection method are implemented:

[0247] Obtain output status data of the charging device; perform machine learning analysis based on the output status data and a preset surge risk model to determine a predicted surge risk level; and dynamically adjust the number of surge suppressors connected between the battery management unit and the charging device based on the predicted surge risk level.

[0248] The present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the following surge protection method are implemented:

[0249] Obtain output status data of the charging device; perform machine learning analysis based on the output status data and a preset surge risk model to determine a predicted surge risk level; and dynamically adjust the number of surge suppressors connected between the battery management unit and the charging device based on the predicted surge risk level.

[0250] In one embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements the following surge protection method steps:

[0251] Obtain output status data of the charging device; perform machine learning analysis based on the output status data and a preset surge risk model to determine a predicted surge risk level; and dynamically adjust the number of surge suppressors connected between the battery management unit and the charging device based on the predicted surge risk level.

[0252] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and they should all be included in the scope of the claims and specification of the present application. In particular, as long as there is no structural conflict, the various technical features mentioned in the various embodiments can be combined in any way. The present application is not limited to the specific embodiments disclosed herein, but includes all technical solutions that fall within the scope of the claims.

Claims

1. A surge protection method, characterized in that: include: Obtain output status data of the charging device; performing machine learning analysis based on the output state data and a preset surge risk model to determine a predicted surge risk level; According to the predicted surge risk level, the number of surge suppressors connected between the battery management unit and the charging device is dynamically adjusted; wherein, the battery management unit and the charging device are connected through a surge suppression array circuit, and the surge suppression array circuit includes a normally closed surge component and at least one controllable surge branch, and the controllable surge branch is connected in parallel with the normally closed surge component; by changing the number of the controllable surge branches put into operation, the number of the connected surge suppressors is changed.

2. The surge protection method according to claim 1, wherein: The performing machine learning analysis based on the output state data and a preset surge risk model to determine a predicted surge risk level includes: Perform feature extraction and analysis based on the output state data to determine feature vector parameters; A machine learning analysis is performed based on the characteristic vector parameters and a preset surge risk model to determine a predicted surge risk level.

3. The surge protection method according to claim 2, wherein: The performing feature extraction and analysis based on the output state data to determine feature vector parameters includes: Perform feature extraction and analysis based on the output state data to determine initial feature vector parameters; A dimensionality reduction analysis is performed based on the initial eigenvector parameters to determine the eigenvector parameters.

4. The surge protection method according to claim 3, wherein: The output state data includes output current parameters, output voltage parameters and noise parameters; and the feature extraction and analysis based on the output state data to determine the initial feature vector parameters includes: Performing at least one of time domain analysis, frequency domain analysis, and wavelet transform analysis on the output current parameter and the output voltage parameter to determine and extract characteristic parameters; Initial feature vector parameters are determined according to the extracted feature parameters and the noise parameters.

5. The surge protection method according to claim 4, characterized in that: The output state data further includes a temperature compensation parameter and / or an aging compensation parameter, and the method further includes: Initial feature vector parameters are determined according to the extracted feature parameters, and temperature compensation parameters and / or aging compensation parameters.

6. The surge protection method according to claim 4 or 5, characterized in that: The extracted characteristic parameters include at least one of voltage peak, current peak, voltage rise time, current rise time, voltage pulse density, current pulse density, voltage waveform kurtosis, current waveform kurtosis, voltage total harmonic distortion rate, current total harmonic distortion rate, voltage spectrum center of gravity, current spectrum center of gravity, voltage high-frequency energy ratio, current high-frequency energy ratio, voltage wavelet energy, current wavelet energy, voltage wavelet entropy and current wavelet entropy.

7. The surge protection method according to any one of claims 1 to 5, characterized in that: The higher the predicted surge risk level, the more surge suppressors are connected between the battery management unit and the charging device.

8. The surge protection method according to any one of claims 1 to 5, characterized in that: The dynamically adjusting the number of surge suppressors connected between the battery management unit and the charging device according to the predicted surge risk level includes: When the predicted surge risk level is the first level, controlling a first number of surge suppressors to be connected between the battery management unit and the charging device; When the predicted surge risk level is a second level, controlling a second number of surge suppressors to be connected between the battery management unit and the charging device; wherein the risk level of the second level is higher than the first level, and the second number is greater than the first number; When the predicted surge risk level is the third level, a third number of surge suppressors are controlled to be connected between the battery management unit and the charging device; wherein the risk level of the third level is higher than that of the second level, and the third number is greater than the second number.

9. The surge protection method according to any one of claims 1 to 5, characterized in that: The method further comprises: When the predicted surge risk level reaches the target risk level, switching the standby branch line to access the output state data collection; When the predicted surge risk level does not reach the target risk level, the main branch is maintained connected to collect output status data; wherein the high-frequency noise suppression capability of the backup branch is stronger than that of the main branch.

10. The surge protection method according to any one of claims 1 to 5, characterized in that: The method further comprises: When the predicted surge risk level reaches a target risk level, outputting a power reduction charging instruction; When the predicted surge risk level does not reach the target risk level, the charging operation is maintained at the current charging power.

11. The surge protection method according to any one of claims 1 to 5, characterized in that: The method further comprises: Obtain health status monitoring information of surge suppressors; The number of surge suppressors connected between the battery management unit and the charging device is dynamically adjusted according to the predicted surge risk level and the health status monitoring information.

12. The surge protection method according to any one of claims 1 to 5, characterized in that: The method further comprises: When the mandatory target risk level execution conditions are met, adjusting the number of surge suppressors corresponding to the target risk level to be connected between the battery management unit and the charging device; When the mandatory target risk level execution condition is not met, the step of dynamically adjusting the number of surge suppressors connected between the battery management unit and the charging device according to the predicted surge risk level is performed.

13. The surge protection method according to claim 12, wherein: The mandatory target risk level execution conditions are determined to be met when at least one of the following is met: Item 1: The current pulse density determined according to the output state data is greater than a preset current density threshold; Item 2: The voltage pulse density determined according to the output state data is greater than a preset voltage density threshold; Item 3: The current change rate determined according to the output state data is greater than a preset current change threshold; Item 4: The voltage change rate determined according to the output state data is greater than a preset voltage change threshold.

14. The surge protection method according to any one of claims 1 to 5, characterized in that: The method further comprises: The preset surge risk model is updated according to the output status data acquired in real time.

15. A surge protection device, characterized in that: include: A data acquisition module, used to obtain output status data of the charging device; a risk prediction module, configured to perform machine learning analysis based on the output state data and a preset surge risk model to determine a predicted surge risk level; A dynamic protection module is used to dynamically adjust the number of surge suppressors connected between the battery management unit and the charging device according to the predicted surge risk level; wherein the battery management unit and the charging device are connected through a surge suppression array circuit, and the surge suppression array circuit includes a normally closed surge component and at least one controllable surge branch, and the controllable surge branch is connected in parallel with the normally closed surge component; by changing the number of the controllable surge branches put into operation, the number of the connected surge suppressors can be changed.

16. A surge protection system, characterized in that: The method comprises a surge suppression array circuit, a monitoring component and a controller, wherein the monitoring component is arranged between the surge suppression array circuit and a battery management unit, the surge suppression array circuit comprises a normally closed surge component and at least one controllable surge branch, the controllable surge branch is connected in parallel with the normally closed surge component, and the first end formed in parallel is connected to the charging device, and the second end formed in parallel is connected to the battery management unit, the controllable surge branch comprises a controllable switch and a surge suppressor connected in series, the monitoring component and the controllable switch are respectively connected to the controller, and the controller is used to perform the steps of the surge protection method described in any one of claims 1 to 14.

17. The surge protection system according to claim 16, wherein: The normally closed surge component includes a surge suppressor and a resettable fuse connected in series.

18. The surge protection system according to claim 16, wherein: The system further includes a varistor and / or a common-mode choke arranged in series between the surge suppression array circuit and the charging device.

19. The surge protection system according to claim 16, wherein: The charging device is connected to the surge suppression array circuit via a positive charging line and a negative charging line, and a gas discharge tube is further provided in parallel between the positive charging line and the negative charging line.

20. The surge protection system according to claim 16, wherein: The system also includes a switching switch, a main branch and a backup branch. The input end of the switching switch is connected to the surge suppression array circuit, the first output end of the switching switch is connected to the monitoring component through the main branch, the second output end of the switching switch is connected to the monitoring component through the backup branch, the switching switch is connected to the controller, and the high-frequency noise suppression capability of the backup branch is stronger than that of the main branch.

21. A battery, characterized in that: The invention comprises a battery management unit, a battery unit and the surge protection system according to any one of claims 16 to 20.

22. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the surge protection method according to any one of claims 1 to 14 are implemented.

23. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the surge protection method according to any one of claims 1 to 14 are implemented.

Images