Charging safety management method and device based on large model, equipment and storage medium

Abstract

The invention relates to the technical field of charging management, in particular to a charging safety management method, device and equipment based on a large model and a storage medium, and the method applied to a cloud comprises the steps that target data are acquired, and the target data comprise real-time charging data of a charging pile; on the basis of the target data, the charging state of each charging pile is detected in real time through a preset charging safety model, a detection result is obtained, and the charging safety model is obtained by training a pre-constructed large model through historical charging data of each charging pile; the detection result is used for indicating whether each charging pile is in an abnormal charging state or not; determining charging control information according to the detection result; and controlling the charging process of each charging pile according to the charging control information. According to the method, the charging safety model is obtained through training based on the general large model, real-time monitoring, anomaly detection and automatic adjustment of the charging process of the charging pile are achieved based on the charging safety model, and the effect is better.

Description

Technical Field

[0001] The present invention relates to the technical field of charging management, and in particular to a charging safety management method, device, equipment and storage medium based on a large model. Background Art

[0002] With the increasing popularity of electric vehicles, charging safety has become increasingly important. Traditional charging management systems rely on simple static thresholds (such as current / voltage limits) and rule-based control strategies, such as triggering overload protection or temperature alarms at preset thresholds.

[0003] With the increase in charging power (such as the power of supercharging piles reaching 480kW) and the increase in complex scenarios (such as high temperature, high humidity, and power grid fluctuations), the safety hazards of the charging process are becoming increasingly prominent. Traditional charging management solutions are difficult to adapt to the complex and changing usage environment. Specifically, the static threshold cannot respond to dynamic factors such as battery aging and sudden changes in ambient temperature and humidity in real time, resulting in an increase in the false alarm rate (research shows that the false alarm rate can reach 28% in extreme environments); multivariable coupling failure: simple strategies cannot analyze the nonlinear correlation between multiple parameters such as current, voltage, and temperature, and the detection accuracy of complex anomalies (such as concurrent overload and overheating) is less than 65%; response delay hazards: the decision-making process based on fixed rules has a delay of hundreds of milliseconds, which makes it difficult to timely block instantaneous high-risk events such as short circuits and arcs. Summary of the Invention

[0004] One purpose of the present invention is to provide a charging safety management method, device, equipment and storage medium based on a large model to solve the technical problem in the related art that the traditional charging management system relies on static thresholds and simple control strategies, is difficult to adapt to complex and changeable usage environments, and results in poor charging safety management effects.

[0005] In a first aspect, an embodiment of the present invention provides a charging safety management method based on a large model, which is applied to a cloud, wherein the cloud is communicatively connected to a plurality of charging piles, and the method includes:

[0006] Acquiring target data, wherein the target data includes real-time charging data of the charging pile;

[0007] Based on the target data, the charging status of each charging pile is detected in real time using a preset charging safety model to obtain a detection result. The charging safety model is obtained by training a pre-built large model using historical charging data of each charging pile. The detection result is used to indicate whether each charging pile is in an abnormal charging state;

[0008] determining charging control information according to the detection result;

[0009] The charging process of each charging pile is controlled according to the charging control information.

[0010] In conjunction with the first aspect, in one possible implementation, before detecting the charging status of each charging pile in real time using a preset charging safety model, the method further includes:

[0011] Acquire historical charging data of each charging pile, wherein the historical charging data includes historical normal charging data and historical abnormal charging data;

[0012] Constructing a sample set based on the historical charging data, wherein the training set includes a normal sample set and an abnormal sample set, wherein the normal sample set is composed of the historical normal charging data of each charging pile, and the abnormal sample set is composed of the historical abnormal charging data of each charging pile;

[0013] Based on the normal sample set and the abnormal sample set, a pre-built large model is trained to obtain the charging safety model.

[0014] In conjunction with the first aspect, in one possible implementation, detecting the charging status of each charging pile based on the target data using a preset charging safety model to obtain a detection result includes:

[0015] Preprocessing the target data;

[0016] Performing principal component analysis on the preprocessed target data to extract features of the target data and obtain target features;

[0017] The target feature is input into the charging safety model, and the charging status of each charging pile is detected by the charging safety model to obtain the detection result output by the charging safety model.

[0018] In conjunction with the first aspect, in a possible implementation, the method further includes:

[0019] Creating a digital twin of each charging pile based on the target data, wherein the digital twin is used to simulate the working status of each charging pile;

[0020] Based on the digital twin, the charging process of each charging pile is simulated and the real-time simulation results are determined;

[0021] The charging control information is optimized according to the real-time simulation result.

[0022] In a second aspect, an embodiment of the present invention further proposes a charging safety management method based on a large model, which is applied to a charging pile, wherein the charging pile is communicatively connected to the cloud, and the method includes:

[0023] Collect real-time charging data from each charging pile to determine target data;

[0024] Sending the target data to the cloud, so that the cloud detects the charging status of each charging pile in real time according to the target data and generates charging control information;

[0025] Receiving charging control information returned by the cloud;

[0026] The charging process of each charging pile is managed according to the charging control information.

[0027] In conjunction with the second aspect, in a possible implementation, managing the charging process of the charging pile according to the charging control information includes:

[0028] Transmitting the charging control information to a scheduling module of the charging pile, wherein the scheduling module parses the charging control information to determine a target instruction, wherein the target instruction is used to respond to an abnormal charging state of the charging pile;

[0029] According to the target instruction, the charging process of the charging pile is managed.

[0030] In conjunction with the second aspect, in one possible implementation, managing the charging process of the target charging pile according to the target instruction includes:

[0031] determining a type of the target instruction, wherein the type includes an overload response instruction, an overheat response instruction, and a short circuit response instruction;

[0032] If the target instruction is an overload response instruction, adjusting the input current of the target charging pile according to a preset threshold;

[0033] If the target instruction is an overheat response instruction, the cooling mechanism of the target charging pile is activated according to a preset adjustment strategy;

[0034] If the target instruction is a short-circuit response instruction, the target charging pile is controlled to shut down, and the communication connection between the target charging pile and the scheduling module is disconnected.

[0035] In a third aspect, an embodiment of the present invention further provides a charging safety management device based on a large model, which is applied to the cloud. The device includes:

[0036] A data acquisition module, configured to acquire target data, wherein the target data includes real-time charging data of the charging pile;

[0037] a data detection module, configured to detect the charging status of each charging pile in real time based on the target data using a preset charging safety model, and obtain a detection result. The charging safety model is obtained by training a pre-built large model using the historical charging data of each charging pile. The detection result is used to indicate whether each charging pile is in an abnormal charging state;

[0038] An instruction generation module, configured to determine charging control information based on the detection result;

[0039] The data transmission module is used to control the charging process of each charging pile according to the charging control information.

[0040] In the fourth aspect, an embodiment of the present invention further proposes a charging management device, which is applied to a charging pile. The charging management device includes a memory and a processor. The memory is connected to the processor. The processor is used to execute one or more computer programs stored in the memory. When the processor executes the one or more computer programs, the charging management device implements the large model-based charging safety management method as described in the second aspect.

[0041] In the fifth aspect, an embodiment of the present invention also proposes a computer-readable storage medium, which stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a charging safety management device based on a large model, the device executes the charging safety management method based on the large model as described in the first aspect. When the program instructions are executed by the processor of the charging management device, the charging management device executes the charging safety management method based on the large model as described in the second aspect.

[0042] The embodiments of the present invention can achieve the following technical effects:

[0043] Based on the method proposed in the embodiment of the present invention, when applied to the cloud, the target data is first obtained, and the target data includes the real-time charging data of the charging pile. Then, the charging status of each charging pile is detected in real time through a preset charging safety model to obtain a detection result. The charging safety model is obtained by training a pre-built large model using the historical charging data of each charging pile. The detection result is used to indicate whether each charging pile is in an abnormal charging state. Finally, the charging control information is determined based on the detection result, and the charging process of each charging pile is controlled according to the charging control information.

[0044] Based on this method, a large model commonly used in the field is used for training to obtain a charging safety model, which is used to replace the traditional static detection scheme, thereby avoiding the drawbacks of management based on static thresholds, accurately predicting and monitoring the charging process, and significantly reducing the risk of safety accidents. At the same time, the charging safety model built and trained based on the large model can support more and more complex control strategies, thereby more reasonably optimizing the charging control process and significantly reducing time and energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0046] Figure 1 A schematic diagram of a charging system provided by an embodiment of the present invention;

[0047] Figure 2 A schematic diagram of a flow chart of a charging safety management method based on a large model provided by an embodiment of the present invention;

[0048] Figure 3 Another flowchart of a charging safety management method based on a large model provided by an embodiment of the present invention;

[0049] Figure 4 A schematic structural diagram of a charging safety management device based on a large model provided by an embodiment of the present invention;

[0050] Figure 5 A schematic structural diagram of a charging management device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0052] It should be noted that, if there is no conflict, the various features in the embodiments of the present invention can be combined with each other and are all within the scope of protection of the present invention. In addition, although the functional modules are divided in the system schematic and the logical order is shown in the flow chart, in some cases, the steps shown or described can be executed in a different order than the module division in the system or the order in the flow chart. Furthermore, the words "first", "second", "third", etc. used in the present invention do not limit the data and execution order, but only distinguish between the same or similar items with basically the same functions and effects.

[0053] For a better explanation of the present invention, please see Figure 1 ,The embodiment of the present invention provides a charging system, such as Figure 1 As shown, the charging system includes a cloud 200 and an edge 100, wherein the edge 100 is provided with a charging station, which includes at least one charging group, such as Figure 1 As shown, the charging station includes a first charging group 10 and a second charging group 20. It is understood that the charging station can also include a larger number of charging groups, such as three or more. The charging groups can be single-layer or multi-layer. For example, if the first charging group 10 is a single-layer charging group, there are no subdivided charging groups within the first charging group 10. If the first charging group 10 is a multi-layer charging group, there are subdivided charging groups within the first charging group 10, such as including a third charging group and a fourth charging group.

[0054] It should be noted that, for the convenience of description, unless otherwise specified, all charging groups in the charging stations described below are single-layer charging groups.

[0055] Each charging group includes at least two charging piles, each charging pile can charge a charging vehicle, for example, Figure 1 As shown, the first charging group 10 includes a first charging pile 11, a second charging pile 12 and a third charging pile 13. The first charging pile 11 is charging a first charging vehicle 101, the second charging pile 12 is charging a second charging vehicle 102, and the third charging pile 13 is charging a third charging vehicle 103.

[0056] The charging groups in the charging station can be divided according to actual needs. For example, the charging piles using the same power supply point can be divided into a charging group. Figure 1 As shown, first charging pile 11, second charging pile 12, and third charging pile 13 use a first power supply point, while fourth charging pile 21, fifth charging pile 22, and sixth charging pile 23 use a second power supply point. The first power supply point and the second power supply point are different power supply points. While a charging pile is charging a vehicle, the vehicle can report its required power to the charging pile in real time, allowing the charging pile to subsequently deliver the power required to the vehicle.

[0057] It is easy to understand that in addition to the charging station, the edge 100 is also equipped with a charging management device and a scheduling module for controlling the charging process of each charging pile. The charging management device is connected to the cloud and the scheduling module respectively. The charging management device collects the real-time charging data of each charging pile and uploads it through communication with the cloud. At the same time, it receives the charging control information sent from the cloud and forwards it to the scheduling module. Each charging pile is equipped with a communication module that can communicate with the scheduling module, so that the scheduling module can control the charging process of each charging pile according to the charging control information sent from the cloud.

[0058] Furthermore, based on the above Figure 1The charging system shown in the figure proposes a charging safety management method based on a large model, which is applied to the above-mentioned cloud. Figure 2 , the method comprising:

[0059] Step S10, acquiring target data, wherein the target data includes real-time charging data of the charging pile;

[0060] Among them, the real-time charging data of each charging pile refers to the real-time working parameters related to the charging process of each charging pile, such as the electrical parameters of each charging pile (such as its current (It), voltage (Vt), power (Pt), etc.), the status parameters of each charging pile (such as its equipment temperature (Tt), battery health status (SoH), charging time (Δt), etc.).

[0061] As a feasible implementation method, the above-mentioned real-time charging data can be collected in real time through built-in sensors (such as current sensors, temperature sensors, and voltmeters) integrated into each charging pile. For example, a Hall current sensor can be used to collect the charging current (It) of the charging pile in real time; a voltage sensor can be used to collect the charging voltage of the charging pile in real time; and a thermocouple or digital temperature sensor can be used to collect the internal temperature (Tt) and ambient temperature of the charging pile in real time. In addition, by communicating with the vehicle's BMS (battery management system), information such as the battery SoC (state of charge), SoH (state of health), and temperature gradient (ΔT / Δt) of the charging vehicle can be obtained.

[0062] Step S20: Based on the real-time charging data, the charging status of each charging pile is detected in real time using a preset charging safety model to obtain a detection result. The charging safety model is obtained by training a pre-built large model using the historical charging data of each charging pile. The detection result is used to indicate whether each charging pile is in an abnormal charging state.

[0063] In this embodiment, a pre-trained large model is used to analyze the real-time charging data of each charging pile in real time, and predictions and judgments are made based on this data to determine whether there are any abnormalities in the charging process.

[0064] Among them, the preset charging safety model can use a timing model (such as LSTM, GRU) to capture dynamic changes, or a classification model (such as SVM, random forest) to identify abnormal patterns, thereby achieving the effect of real-time detection of the charging process of the charging pile.

[0065] The charging safety model can be trained using historical charging data from each charging station. This data should include both normal charging records and annotated abnormal events (such as overloads and short circuits). As a viable implementation, the dataset can be augmented by adding Gaussian noise or time offsets to improve model training.

[0066] In this embodiment, the detection result includes the type of abnormal state that may exist in each charging pile.

[0067] Step S30, determining charging control information according to the detection result;

[0068] Specifically, this embodiment determines corresponding charging control information in combination with the output results of the above-mentioned model. The purpose of the charging control information is to enable the charging pile to control the charging pile with abnormal charging process to end the abnormal charging state by reading the content of the charging control information.

[0069] It is easy to understand that in order for the charging pile to read and execute the charging control information, the above-mentioned charging control information should include at least two parts: device information indicating that individual charging piles are in an abnormal charging state, that is, through this information, the charging station can determine the individual charging piles or charging pile groups that are in an abnormal charging state; control instructions that characterize the specific category of abnormal charging state, that is, through this instruction, the charging pile can perform corresponding actions in response to the instruction, thereby enabling the corresponding charging pile to end the current abnormal charging state and reduce its impact on the station.

[0070] As a feasible implementation method, after obtaining the detection results output by the charging safety model, the abnormal charging pile in the abnormal charging state is first determined according to the detection results, and its abnormal type is identified. Then, according to the preset expert strategy, the corresponding control strategy can be matched for each abnormal type, and the control strategy is converted into a control signal that can be used for the charging pile to obtain the above-mentioned charging control information.

[0071] Step S40: Control the charging process of each charging pile according to the charging control information.

[0072] Specifically, in this embodiment, the charging control information is sent to the edge (corresponding to the execution process in the cloud, which should at least include the charging piles) so that the edge manages the charging process of each charging pile according to the charging control information.

[0073] In this embodiment, after the charging control information is determined, it is returned to the edge. The charging management device at the edge controls the charging progress of each charging pile according to the corresponding charging control information. For example, the charging control information includes {A, B, C}, which includes the exception handling signals corresponding to charging piles a, b, and c, respectively. The edge charging management device extracts the exception handling signals of charging piles a, b, and c by parsing the charging control information, and then generates corresponding control instructions a1, b1, and c1 based on the exception handling signals. The control instructions a1, b1, and c1 are transmitted to charging piles a, b, and c, respectively. Ultimately, charging piles a, b, and c adjust the current charging progress according to the control instructions they received, thus achieving charging safety management for the edge charging piles.

[0074] In this embodiment, before detecting the charging status of each charging pile in real time using a preset charging safety model, the method further includes:

[0075] Obtain historical charging data of each charging pile, wherein the historical charging data includes historical normal charging data and historical abnormal charging data; construct a sample set based on the historical charging data, wherein the training set includes a normal sample set and an abnormal sample set, wherein the normal sample set is composed of the historical normal charging data of each charging pile, and the abnormal sample set is composed of the historical abnormal charging data of each charging pile; train a pre-constructed large model using the sample set to obtain the charging safety model.

[0076] The pre-built large model in this embodiment can be a large model architecture such as a deep neural network (DNN) or a long short-term memory network (LSTM), which is trained using historical normal charging data and historical abnormal charging data in the sample set, and a charging safety model is obtained after the training is completed.

[0077] Taking LSTM as an example, its input gate satisfies:

[0078] i t =σ(W i ·[h t-1 , x t ]+b i );

[0079] Its forget gate satisfies:

[0080] f t =σ(W f ·[h t-1 , x t ]+b f );

[0081] Its output gate satisfies:

[0082] o t =σ(Wo ·[h t-1 , x t ]+b o );

[0083] Its cell state update satisfies:

[0084] C t =f t ·C t-1 +i t tanh(W C ·[h t-1 , x t ]+b C );

[0085] Its final output satisfies:

[0086] h t =o t tanh(C t ).

[0087] In the above formula, σ is the Sigmoid activation function; tanh is the hyperbolic tangent function; W i 、W f 、W o 、W C is the weight matrix; b i 、b f 、b o 、b c 、b i is the bias term; h t-1 is the hidden state of the previous moment; x t The current input data.

[0088] In this embodiment, based on the real-time charging data, the charging status of each charging pile is detected by a preset charging safety model to obtain the detection result, including:

[0089] The target data is preprocessed; principal component analysis is performed on the preprocessed target data to extract features of the target data and obtain target features; the target features are input into the charging safety model, the charging status of each charging pile is detected by the charging safety model, and the detection results output by the charging safety model are obtained.

[0090] In this embodiment, data is cleaned and preprocessed by methods such as denoising, smoothing, and normalization. Specifically, sensor noise can be eliminated by sliding average or wavelet transform, and dimensional differences can be eliminated by Z-score standardization.

[0091] This embodiment uses principal component analysis (PCA) technology to perform feature extraction to identify key influencing factors and patterns in the charging process, and then extracts key features (such as "current change rate" and "temperature gradient") from high-dimensional charging data (such as current, voltage, temperature, battery health status, etc.), reducing redundant information and improving model inference speed.

[0092] Furthermore, in this embodiment, the above method further includes:

[0093] Based on the real-time charging data, a digital twin of each charging pile is created, and the digital twin is used to simulate the working status of each corresponding charging pile; based on the digital twin, the charging process of each charging pile is simulated, and the real-time simulation result is determined; based on the real-time simulation result, the charging control information is optimized.

[0094] This embodiment develops a digital twin of each charging pile (i.e., a real-time digital mirror of the real charging pile) to implement the client logic of the real charging pile in the cloud, including establishing a long connection and implementing charging pile communication protocols such as OCPP and ACMP. At the same time, it uses real-time reported data streams to continuously update the virtual model to ensure the consistency between the virtual world and the physical world, enabling the system to test and optimize safety control strategies in a virtual environment, reducing the trial and error costs of real-world adjustments.

[0095] It is easy to understand that the digital twin of the charging pile built in the cloud in this embodiment is used to map the status of each charging pile (such as charging power, voltage, fault alarm, etc.) in real time, and to participate in system testing and strategy optimization by replacing physical devices with digital ones, thereby avoiding the risks of directly operating real devices.

[0096] The embodiment of the present invention also proposes a charging safety management method based on a large model, which is applied to charging piles. Figure 3 , the method comprising:

[0097] Step A10: collecting real-time charging data of each charging pile to determine target data;

[0098] Step A20: sending the target data to the cloud, so that the cloud detects the charging status of each charging pile in real time according to the target data and generates charging control information;

[0099] Step A30, receiving the charging control information returned by the cloud;

[0100] Step A40: managing the charging process of each charging pile according to the charging control information.

[0101] Among them, target data refers to key information collected from charging piles in real time, which is of direct value to charging status monitoring and control. Compared with real-time charging data, it is valid data that has been further screened and preprocessed, rather than all real-time charging data collected by the original sensor. Its core function is to drive status analysis and control command generation in the cloud.

[0102] In this embodiment, the cloud's processing of real-time charging data and the judgment logic for detecting each charging pile are involved. Please refer to the description of the above embodiments, and this embodiment will not be repeated here.

[0103] Specifically, multiple charging piles are commonly connected to a preset scheduling module. The scheduling module manages the charging progress of each charging pile in real time by communicating with the multiple charging piles. In addition, the scheduling module can also be independently set relative to each charging pile, that is, the scheduling module is integrated on each charging pile. The above-mentioned management of the charging progress of each charging pile according to the charging control information includes:

[0104] The charging control information is transmitted to the scheduling module of the charging pile. The scheduling module determines a target instruction by parsing the charging control information. The target instruction is used to respond to the abnormal charging state of the charging pile; and the charging process of the charging pile is managed according to the target instruction.

[0105] More specifically, managing the charging process of the target charging pile according to the target instruction includes:

[0106] Determine the type of the target instruction, which includes an overload response instruction, an overheat response instruction, and a short-circuit response instruction; if the target instruction is an overload response instruction, adjust the input current of the target charging pile according to a preset threshold; if the target instruction is an overheat response instruction, start the cooling mechanism of the target charging pile according to a preset adjustment strategy; if the target instruction is a short-circuit response instruction, control the target charging pile to shut down, and disconnect the communication connection between the target charging pile and the scheduling module.

[0107] As another aspect of the present invention, embodiments of the present invention further provide a large-scale model-based charging safety management device, which can be applied to the aforementioned cloud. The large-scale model-based charging safety management device can be a software module, comprising several instructions stored in a memory. A processor can access the memory and execute the instructions to implement the large-scale model-based charging safety management method described in each of the aforementioned embodiments.

[0108] In some embodiments, the charging safety management device based on the large model can also be constructed by hardware devices. For example, the charging safety management device based on the large model can be constructed by one or more chips, and the chips can work in coordination with each other to complete the charging safety management method based on the large model described in each embodiment. For another example, the charging safety management device based on the large model can also be constructed by various logic devices, such as a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a single-chip microcomputer, an ARM (Acorn RISC Machine) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination of these components.

[0109] For details, please refer to Figure 4 , Figure 4 The figure shows a schematic diagram of the structure of a charging safety management device based on a large model, which includes:

[0110] The data acquisition module 410 is used to acquire target data, wherein the target data includes real-time charging data of the charging pile;

[0111] A data detection module 420 is configured to detect the charging status of each charging pile in real time based on the target data using a preset charging safety model, and obtain a detection result. The charging safety model is obtained by training a pre-built large model using the historical charging data of each charging pile. The detection result is used to indicate whether each charging pile is in an abnormal charging state;

[0112] An instruction generating module 430 is configured to determine charging control information based on the detection result;

[0113] The data transmission module 440 is used to control the charging process of each charging pile according to the charging control information.

[0114] In a possible implementation, the data detection module 420 is further configured to:

[0115] Obtain historical charging data of each charging pile, wherein the historical charging data includes historical normal charging data and historical abnormal charging data; construct a sample set based on the historical charging data, wherein the training set includes a normal sample set and an abnormal sample set, wherein the normal sample set is composed of the historical normal charging data of each charging pile, and the abnormal sample set is composed of the historical abnormal charging data of each charging pile; train a pre-constructed large model using the sample set to obtain the charging safety model.

[0116] In one possible implementation, the data detection module 420, when used to detect the charging status of each charging pile based on the target data using a preset charging safety model and obtain a detection result, is specifically configured to:

[0117] The target data is preprocessed; principal component analysis is performed on the preprocessed target data to extract features of the target data and obtain target features; the target features are input into the charging safety model, the charging status of each charging pile is detected by the charging safety model, and the detection results output by the charging safety model are obtained.

[0118] In a possible implementation, the instruction generation module 430, when used to determine the charging control information according to the detection result, is further used to:

[0119] Based on the target data, a digital twin of each charging pile is created, and the digital twin is used to simulate the working status of each corresponding charging pile; based on the digital twin, the charging process of each charging pile is simulated, and real-time simulation results are determined; based on the real-time simulation results, the charging control information is optimized.

[0120] This embodiment proposes a charging safety management device based on a large model, which can be applied to the cloud. By calling the above-mentioned data acquisition module, data detection module, instruction generation module and data transmission module, and combining the computing resources of the cloud, the charging safety management method applied to the cloud in the above-mentioned embodiment is implemented, and has all the technical effects of the above-mentioned method embodiment. Moreover, since the control logic of the above-mentioned method embodiment is integrated into the device, the coupling degree of the above-mentioned method at the application level is reduced, and the application effect of the above-mentioned method in actual scenarios is further improved.

[0121] Further, see Figure 5 , Figure 5 FIG2 is a schematic diagram of the structure of a charging management device provided in an embodiment of the present invention, which can be applied to the charging pile described above. The charging management device includes one or more processors 51 and a memory 52. ​​The memory 52 is connected to the one or more processors 51, for example, via a bus.

[0122] The processor 51 is configured to support the charging management device in performing the corresponding functions in the above-mentioned method applied to the charging pile. The processor 51 can be a central processing unit (CPU), a network processor (NP), a hardware chip, or any combination thereof. The above-mentioned hardware chip can be an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The above-mentioned PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0123] The memory 52 is used to store program code, etc. The memory 52 may include volatile memory (VM), such as random access memory (RAM); non-volatile memory (NVM), such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid-state drive (SSD); or a combination of the aforementioned types of memory.

[0124] Memory 52 can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to the large-scale model-based charging safety management method in the embodiments of the present invention. Processor 51 executes the non-volatile software programs, instructions, and modules stored in memory 52 to execute the various functional applications and data processing of the large-scale model-based charging safety management method and the large-scale model-based charging safety management system, thereby implementing the functions of the various modules or units of the large-scale model-based charging safety management method and the large-scale model-based charging safety management system provided in the above-mentioned method embodiments.

[0125] The memory 52 may include a program storage area and a data storage area. The program storage area may store the operating system and application programs required for at least one function. The data storage area may store data generated based on the use of the large-scale model-based charging safety management system. In some embodiments, the memory 52 may optionally include a remote storage device located relative to the processor 51. Such remote storage device may be connected to the large-scale model-based charging safety management device via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0126] One or more modules are stored in the memory 52. ​​When executed by one or more processors 51, the charging safety management method based on the large model in any of the above method embodiments is executed, for example, the method steps described in the above method embodiments are executed to realize the functions of the modules described in the above system embodiments.

[0127] An embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a charging safety management device based on a large model, the device executes the charging safety management method based on a large model applied to the cloud in the above-mentioned embodiment. When the program instructions are executed by the processor of the charging management device, the charging management device executes the charging safety management method applied to the charging pile in the above-mentioned embodiment.

[0128] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0129] The above disclosure is merely a preferred embodiment of the present invention and certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A charging safety management method based on a large model, characterized in that: Applied to a cloud, the cloud is communicatively connected to a plurality of charging piles, and the method includes: Acquiring target data, wherein the target data includes real-time charging data of the charging pile; Based on the target data, the charging status of each charging pile is detected in real time using a preset charging safety model to obtain a detection result. The charging safety model is obtained by training a pre-built large model using historical charging data of each charging pile. The detection result is used to indicate whether each charging pile is in an abnormal charging state; determining charging control information according to the detection result; The charging process of each charging pile is controlled according to the charging control information.

2. The charging safety management method based on a large model according to claim 1 is characterized in that: Before detecting the charging status of each charging pile in real time using a preset charging safety model, the method further includes: Acquire historical charging data of each charging pile, wherein the historical charging data includes historical normal charging data and historical abnormal charging data; Constructing a sample set based on the historical charging data, wherein the training set includes a normal sample set and an abnormal sample set, wherein the normal sample set is composed of the historical normal charging data of each charging pile, and the abnormal sample set is composed of the historical abnormal charging data of each charging pile; Based on the normal sample set and the abnormal sample set, a pre-built large model is trained to obtain the charging safety model.

3. The charging safety management method based on a large model according to claim 1 is characterized in that: The method of detecting the charging status of each charging pile based on the target data by using a preset charging safety model to obtain a detection result includes: Preprocessing the target data; Performing principal component analysis on the preprocessed target data to extract features of the target data and obtain target features; The target feature is input into the charging safety model, and the charging status of each charging pile is detected by the charging safety model to obtain the detection result output by the charging safety model.

4. The charging safety management method based on a large model according to claim 1, characterized in that: The method further comprises: Creating a digital twin of each charging pile based on the target data, wherein the digital twin is used to simulate the working status of each charging pile; Based on the digital twin, the charging process of each charging pile is simulated and the real-time simulation results are determined; The charging control information is optimized according to the real-time simulation result.

5. A charging safety management method based on a large model, characterized in that: Applied to a charging pile, the charging pile is connected to the cloud for communication, and the method includes: Collecting real-time charging data of the charging pile to determine target data; Sending the target data to the cloud, so that the cloud detects the charging status of the charging pile in real time according to the target data and generates charging control information; Receiving charging control information returned by the cloud; The charging process of the charging pile is managed according to the charging control information.

6. The charging safety management method based on a large model according to claim 5 is characterized in that: Managing the charging process of the charging pile according to the charging control information includes: Transmitting the charging control information to a scheduling module of the charging pile, wherein the scheduling module parses the charging control information to determine a target instruction, wherein the target instruction is used to respond to an abnormal charging state of the charging pile; According to the target instruction, the charging process of the charging pile is managed.

7. The charging safety management method based on a large model according to claim 6, characterized in that: Managing the charging process of the target charging pile according to the target instruction includes: determining a type of the target instruction, wherein the type includes an overload response instruction, an overheat response instruction, and a short circuit response instruction; If the target instruction is an overload response instruction, adjusting the input current of the target charging pile according to a preset threshold; If the target instruction is an overheat response instruction, the cooling mechanism of the target charging pile is activated according to a preset adjustment strategy; If the target instruction is a short-circuit response instruction, the target charging pile is controlled to shut down, and the communication connection between the target charging pile and the scheduling module is disconnected.

8. A charging safety management device based on a large model, characterized in that: Applied to the cloud, the device includes: A data acquisition module, configured to acquire target data, wherein the target data includes real-time charging data of the charging pile; a data detection module, configured to detect the charging status of each charging pile in real time based on the target data using a preset charging safety model, and obtain a detection result. The charging safety model is obtained by training a pre-built large model using the historical charging data of each charging pile. The detection result is used to indicate whether each charging pile is in an abnormal charging state; An instruction generation module, configured to determine charging control information based on the detection result; The data transmission module is used to control the charging process of each charging pile according to the charging control information.

9. A charging management device, characterized in that: Applied to a charging pile, the charging management device includes a memory and a processor, the memory is connected to the processor, and the processor is used to execute one or more computer programs stored in the memory. When the processor executes the one or more computer programs, the charging management device executes the large model-based charging safety management method as described in any one of claims 5-7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a charging safety management device based on a large model, the device executes the charging safety management method based on a large model as described in any one of claims 1 to 4. When the program instructions are executed by a charging management device, the charging management device executes the charging safety management method based on a large model as described in any one of claims 5 to 7.

Images