Direct current charging pile risk assessment method, device, equipment and medium

Abstract

The invention relates to a direct current charging pile risk assessment method, device and equipment and a medium, and relates to the technical field of computers. The method comprises the following steps: acquiring operation data of a direct current charging pile; determining a first threshold value and a second threshold value of the direct current charging pile based on equipment preset parameters of the direct current charging pile; wherein the first threshold value is smaller than the second threshold value; based on the fault type, the operation data, the first threshold value and the second threshold value of the DC charging pile, the operation state of the DC charging pile is determined; and based on the operation state of the DC charging pile, determining a risk assessment result of the DC charging pile. By adopting the method, an accurate risk assessment result can be obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer technology, and in particular to a direct current charging pile risk assessment method, device, equipment and medium. BACKGROUND

[0002] The current mainstream direct current charging pile has basic operation parameter monitoring and fault protection functions. The specific technical means include: 1) real-time acquisition of core parameters such as input current, output current, power tube temperature through current sensors (such as Hall sensors) and temperature sensors (such as PT100), and some devices can realize full-phase monitoring of three-phase input current; 2) triggering automatic power-off protection based on the protection threshold preset by the device at the factory, to avoid device burning or electric shock accidents; 3) some operation platforms can collect device operation data, generate alarm information through simple threshold comparison, and prompt the on-site troubleshooting of operation and maintenance personnel. In addition, machine learning technology is gradually introduced in the industry to try to train the model through historical data to realize preliminary prediction of faults.

[0003] Although the existing technology has basic monitoring capability, there are still significant shortcomings in actual application, which is difficult to meet the needs of large-scale and refined operation and maintenance, and ultimately leads to inaccurate risk assessment results of direct current charging piles. SUMMARY

[0004] The present application provides a direct current charging pile risk assessment method, device, equipment and medium, which can obtain accurate risk assessment results.

[0005] To achieve the above purpose, the present application adopts the following technical scheme: In a first aspect, the present application provides a direct current charging pile risk assessment method, comprising: obtaining operation data of a direct current charging pile; determining a first threshold value and a second threshold value of the direct current charging pile based on device preset parameters of the direct current charging pile; wherein the first threshold value is less than the second threshold value; determining the operation state of the direct current charging pile based on the fault type, the operation data, the first threshold value and the second threshold value of the direct current charging pile; determining the risk assessment result of the direct current charging pile based on the operation state of the direct current charging pile.

[0006] In one embodiment, determining the first threshold value and the second threshold value of the direct current charging pile based on the device preset parameters of the direct current charging pile comprises: determining the first threshold value of the direct current charging pile based on the rated parameters of the device and the corresponding preset safety coefficient; determining the second threshold value of the direct current charging pile based on the device threshold value, the device fault parameter and the preset fault coefficient.

[0007] In one embodiment, the operating state of the DC charging pile is determined based on the fault type of the DC charging pile, the operating data, the first threshold value and the second threshold value, including: The operating state of the DC charging pile is determined based on the operating data, the first threshold value and the second threshold value of the DC charging pile; The operating state of the DC charging pile is determined based on the fault type and the operating state of the DC charging pile.

[0008] In one embodiment, the operating state of the DC charging pile is determined based on the operating data, the first threshold value and the second threshold value of the DC charging pile, including: The operating change information of the DC charging pile is determined based on the operating data, the first threshold value and the second threshold value of the DC charging pile; The operating state of the DC charging pile is determined based on the operating change information.

[0009] In one embodiment, the risk assessment result of the DC charging pile is determined based on the operating state of the DC charging pile, including: The risk level corresponding to the operating state is determined based on the operating state of the DC charging pile and the preset mapping relationship table; The risk assessment result of the DC charging pile is determined based on the operating state of the DC charging pile and the risk level.

[0010] In one embodiment, including: The operating data is input into the risk prediction model to obtain risk level change data of the DC charging pile; wherein the risk prediction model is a model constructed based on a random forest algorithm; the risk level change data includes at least one of a risk level change curve, a fault probability and a fault type.

[0011] In one embodiment, the operating data includes at least one of input current, output current and power tube temperature.

[0012] In a second aspect, the present application provides a DC charging pile risk assessment device, including: The acquisition module is used for acquiring the operating data of the DC charging pile; The threshold value determination module is used for determining the first threshold value and the second threshold value of the DC charging pile based on the device preset parameters of the DC charging pile; wherein the first threshold value is smaller than the second threshold value; The state determination module is used for determining the operating state of the DC charging pile based on the fault type, the operating data, the first threshold value and the second threshold value of the DC charging pile; The assessment determination module is used for determining the risk assessment result of the DC charging pile based on the operating state of the DC charging pile.

[0013] In a third aspect, the present application provides a computing device including a memory and a processor; The one or more computer programs include instructions; and when the instructions are executed by the processor, the computing device performs the method according to any one of the first aspect.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium for storing a computer program for performing the method according to any one of the first aspect.

[0015] In a fifth aspect, the present application provides a computer program product comprising one or more computer instructions, which, when executed by a computer, perform the method according to any one of the first aspect.

[0016] From the above technical solutions, the present application has at least the following beneficial effects: In the present application, by obtaining the operation data of the direct-current charging pile, the basic reliability of risk assessment from the source is realized, which is the premise of precise risk assessment; further, based on the preset parameters of the direct-current charging pile, the first threshold and the second threshold of the direct-current charging pile are determined, the threshold customization and gradient are realized, a scientific judgment scale is provided for risk assessment, the accuracy of state classification is ensured, and protection hysteresis or false alarm is avoided; further, based on the fault type, operation data, first threshold and second threshold of the direct-current charging pile, the operation state of the direct-current charging pile is determined, the precise classification of the equipment state is realized; finally, based on the operation state of the direct-current charging pile, the risk assessment result of the direct-current charging pile is determined. The present application provides a basis for determining the operation state of the direct-current charging pile by introducing the first threshold and the second threshold; further, by introducing the operation state, the basis for directly determining the risk of the direct-current charging pile is laid, and the accuracy of the risk assessment result is finally ensured.

[0017] It should be understood that the description of technical features, technical solutions, beneficial effects or similar language in the present application does not imply that all features and advantages can be achieved in any single embodiment. On the contrary, it can be understood that the description of a feature or a beneficial effect means that the specific technical feature, technical solution or beneficial effect is included in at least one embodiment. Therefore, the description of technical features, technical solutions or beneficial effects in the specification does not necessarily refer to the same embodiment. Further, the technical features, technical solutions and beneficial effects described in the present embodiment can be combined in any appropriate manner. Those skilled in the art will understand that the embodiments can be implemented without one or more specific technical features, technical solutions or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects can be identified in specific embodiments that do not embody all embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 An application environment diagram of the DC charging pile risk assessment method provided in the embodiment of the application is shown in FIG. 1; Figure 2 A flowchart of the DC charging pile risk assessment method provided in the embodiment of the application is shown in FIG. 2; Figure 3 A structure block diagram of the DC charging pile risk assessment device provided in the embodiment of the application is shown in FIG. 3; Figure 4 An internal structure diagram of the computer device provided in the embodiment of the application is shown in FIG. 4. DETAILED DESCRIPTION

[0019] The terms “first”, “second”, and “third” and the like in the specification of the application and the accompanying drawings are used to distinguish different objects, and are not used to limit a specific order.

[0020] In the embodiments of the application, the words “exemplary” or “for example” are used to mean serving as an example, instance, or illustration, in any non-limiting and / or illustrative sense. The embodiments or design schemes described in the application as “exemplary” or “for example” should not be construed as being more preferred or advantageous than other embodiments or design schemes. Rather, the use of “exemplary” or “for example” is intended to present related concepts in a concrete manner.

[0021] For the sake of clear and concise description of the following embodiments, first, a brief introduction of related technologies is given: The current mainstream DC charging pile has basic operation parameter monitoring and fault protection functions, and the specific technical means include: 1) real-time acquisition of core parameters such as input current, output current, and power tube temperature through current sensors (such as Hall sensors) and temperature sensors (such as PT100), and some devices can realize full-phase monitoring of three-phase input current; 2) based on the protection threshold preset by the device at the factory, automatic power-off protection is triggered to avoid device burning or electric shock accidents; 3) some operation platforms can collect device operation data, generate alarm information through simple threshold comparison, and prompt the on-site troubleshooting of operation and maintenance personnel. In addition, machine learning technology is gradually introduced in the industry to try to train the model through historical data to realize preliminary prediction of faults.

[0022] Although the prior art has basic monitoring capabilities, there are still significant shortcomings in practical applications, which are difficult to meet the needs of large-scale and refined operation and maintenance. First, the protection threshold is wide and the passive response is lagging. The protection threshold of the existing device is mostly based on the device limit parameter setting, and the protection is triggered only when the fault has occurred and threatens the safety of the device, which cannot identify the sub-healthy state (such as slow parameter degradation and slight component failure), resulting in post-disposal of faults, increasing the risk of equipment damage and operation and maintenance costs. Second, the state risk assessment is fuzzy and the fault location is difficult. The existing monitoring can only judge the binary state of normal / fault, lacks subdivision of risk levels (such as safety, illness, and danger), and cannot associate with specific fault types (such as power tube failure and diode failure), so the operation and maintenance personnel need to check the components one by one, which is low in positioning efficiency. Third, the threshold lacks adaptability and the false alarm rate is high. The fixed protection threshold does not consider factors such as device aging and environmental temperature changes, which may cause false alarms during normal operation or hidden dangers without triggering alarms, interfering with operation and maintenance decisions. Fourth, the data is not fully utilized and lacks predictive ability. Most platforms only store data without deep analysis, and even if machine learning models are introduced, the training data is single in dimension and the threshold system is not dynamically optimized, resulting in inaccurate risk assessment results and inability to achieve pre-prevention.

[0023] In order to make the technical solutions of the present application clearer and easier to understand, the application scenarios of the technical solutions of the present application are introduced below in conjunction with the drawings. As shown in Figure 1 The figure is an application environment diagram provided by an embodiment of the present application.

[0024] In this application environment, the sensor 104 acquires the operating data of the direct current charging pile, such as current data and temperature data, etc. For example, the current data can be acquired by an input side current acquisition device (such as a Hall sensor) to acquire three-phase input current, by an output side current acquisition device to acquire output current, and by a temperature acquisition device (such as a PT100) arranged at a power switch tube (such as an IGBT) to acquire the temperature of each power tube. The server 103 can analyze and calculate according to the data acquired by the sensor 104 to obtain a risk assessment result. Further, the server 103 can transmit the risk assessment result to the terminal 102 for viewing by technicians and operation and maintenance personnel.

[0025] In order to make the technical solutions of the present application clearer and easier to understand, a direct current charging pile risk assessment method provided by an embodiment of the present application is introduced below in conjunction with the above application scenario. As shown in Figure 2 The figure is a flowchart of a direct current charging pile risk assessment method provided in an embodiment of the present application.

[0026] S201, acquiring operating data of a direct current charging pile.

[0027] The running data includes at least one of input current, output current and power tube temperature; optionally, it can also include voltage ripple, power factor, DC bus voltage, switching frequency, radiator temperature and other related parameters, thereby constructing a multi-dimensional and multi-level running state monitoring system.

[0028] For example, high-precision Hall current sensors can be installed on the A, B and C three-phase lines of the AC input side of the DC charging pile, and the sampling frequency can be set to 50 Hz to collect the original input current of the DC charging pile and capture the dynamic change characteristics of the AC side current. A DC current sensor is installed on the positive line of the DC output side, and the sampling frequency can be set to 100 Hz to collect the original output current of the DC charging pile and accurately reflect the current output characteristics of the load side. In addition, a PT100 platinum resistance temperature sensor is attached to the heat dissipation substrate of each Insulated Gate Bipolar Transistor (IGBT, Insulated Gate Bipolar Transistor) power switch tube to collect the original power tube temperature of the DC charging pile. Considering that the temperature change rate is relatively slow, in order to reduce data redundancy and take into account system resources, the sampling frequency can be set to 10 Hz to achieve moderate collection of temperature parameters.

[0029] Further, the original data collected by the sensor can be transmitted to the local controller through the RS485 bus. After receiving the data, the controller first performs preliminary filtering and cleaning of the original data, including rejecting abnormal values (such as interference data with a current instantaneous mutation of more than 3 times the rated value) and filling missing values caused by transmission interference or sampling loss (the arithmetic mean of 5 valid data points before and after the missing position can be used for replacement).

[0030] On this basis, the controller aggregates the cleaned data according to the preset 1-minute aggregation rule to generate representative running data. Specifically, for input current and output current, the maximum value within a 1-minute time window is taken to effectively capture the peak fluctuation during charging and discharging; for power tube temperature, the average value within 1 minute is taken to smooth the instantaneous error caused by measurement noise and reflect the overall trend of temperature change. Finally, a structured running data record is formed, which can be designed as: "timestamp + input current A phase + input current B phase + input current C phase + output current + power tube temperature 1 + … + power tube temperature p", thereby obtaining a standardized and consistent DC charging pile running data set; where p is the total number of power tubes.

[0031] It should be noted that the three-phase input current is obtained through the input-side current acquisition device, the power grid side power quality and load characteristics can be monitored; the output current is obtained through the output-side current acquisition device, the load demand and output stability during the charging process can be mastered in real time; the temperature of each power tube is obtained through the power switch tube temperature acquisition device, which directly reflects the thermal operating state and heat dissipation efficiency of the core power device.

[0032] This parameter acquisition scheme covering the input, output, and component temperature full link realizes comprehensive collection of operating parameters, effectively avoids one-sidedness of evaluation results caused by single data dimension or missing of key parameters. For example, if only current is monitored and temperature parameters are ignored, overheating of power tubes caused by poor heat dissipation or overload operation is easily missed, thereby affecting the safety evaluation conclusion of the equipment. Therefore, the present scheme ensures that the subsequent state evaluation and risk analysis can comprehensively and accurately reflect the overall operating condition of the equipment through multi-parameter collaborative acquisition, not only guarantees the integrity and systematicness of the data dimension, but also ensures the basic reliability of the risk evaluation work from the data source, provides solid and reliable data support for subsequent data analysis, fault diagnosis, life prediction, and other links, and is an important prerequisite for realizing intelligent operation and maintenance and fine management of the direct-current charging pile.

[0033] S202, determining a first threshold value and a second threshold value of the direct-current charging pile based on preset parameters of the equipment of the direct-current charging pile.

[0034] The preset parameters of the equipment include equipment rated parameters, equipment threshold values, and equipment fault parameters; the first threshold value is less than the second threshold value.

[0035] Optionally, the equipment rated parameters refer to the benchmark technical parameters of the direct-current charging pile under standard working conditions, such as rated input current, rated output current, power tube rated working temperature, etc.; the equipment threshold values refer to the protection action triggering parameters built in the control system of the direct-current charging pile, such as input current protection threshold value, output current protection threshold value, etc., which are used to start the equipment self-protection mechanism when the parameters are abnormal; the equipment fault parameters refer to the critical parameters of irreversible failure of the core components in the direct-current charging pile, such as power tube fault temperature, failure current threshold value that cannot be protected, etc., which represent the final boundary of the safe operation of the equipment. The first threshold value is the upper limit boundary of the normal operating state of the direct-current charging pile, which can be determined by superimposing a preset safety coefficient on the equipment rated parameters, and is used to distinguish between normal operation and abnormal initial state; the second threshold value is the critical boundary of the direct-current charging pile approaching failure, which is determined by superimposing a preset failure coefficient on the equipment threshold value and the equipment fault parameter, and is used to distinguish between the sub-healthy state and the dangerous state.

[0036] In one implementation manner, the first threshold value of the direct-current charging pile is determined based on the equipment rated parameters and a corresponding preset safety coefficient; the second threshold value of the direct-current charging pile is determined based on the equipment threshold value, the equipment fault parameter, and a preset failure coefficient.

[0037] wherein the preset safety factor is a safety redundancy factor reserved for adapting the operating parameter fluctuation, and the value needs to balance the misjudgment risk and protection accuracy, usually the current type parameter takes 1.1 (i.e. 110%), and the temperature type parameter takes 1.2 (i.e. 120%) or 0.8 (i.e. 80%) based on the fault temperature; The preset fault factor is a conservative factor set to avoid the risk of protection failure, to ensure that the early warning is triggered before the protection action of the DC charging pile, usually taking 0.9 (i.e. 90%), that is, the second threshold is 90% of the limit parameter; the limit parameter can be the minimum value of the input current protection threshold and the unprotectable fault current threshold, and the minimum value of the output current protection threshold and the unprotectable fault current threshold, etc.

[0038] For example, the device preset parameters can be extracted from the DC charging pile product specification and the equipment management system, such as rated input current 100A, rated output current 120A, power tube rated working temperature 80℃, power tube fault temperature 150℃, input current protection threshold 130A, output current protection threshold 180A, and unprotectable fault current threshold 150A; the coefficients can be set in combination with industry standards (such as GB / T 18487.1-2015 electric vehicle conductive charging system standard) and operation and maintenance experience, such as the preset safety factor corresponding to the current type is 1.1, the preset safety factor corresponding to the temperature type is 1.2, and the preset fault factor is 0.9, etc.

[0039] For example, the first threshold, which can also be called the safety threshold, can be calculated as follows: The first threshold corresponding to the input current = rated input current (100A) × preset safety factor (1.1) = 100A × 1.1 = 110A.

[0040] The first threshold corresponding to the output current = rated output current (120A) × preset safety factor (1.1) = 120A × 1.1 = 132A.

[0041] The first threshold corresponding to the power tube temperature = min (power tube rated working temperature × 1.2, power tube fault temperature × 0.8) = min (80℃ × 1.2, 150℃ × 0.8) = min (96℃, 120℃) = 96℃.

[0042] The second threshold, which can also be called the out-of-bound fault threshold, can be calculated as follows: The second threshold corresponding to the input current = min (input current protection threshold, unprotectable fault current threshold) × preset fault factor = min (130A, 150A) × 0.9 = 117A.

[0043] The second threshold value corresponding to the output current is min (output current protection threshold value, unable to protect fault current threshold value) x preset fault coefficient = min (180 A, 150 A) x 0.9 = 135 A.

[0044] The second threshold value corresponding to the power tube temperature is power tube fault temperature x preset fault coefficient = 150 DEG C x 0.9 = 135 DEG C.

[0045] Optionally, the historical threshold application data of the same type of equipment can also be compared. If the first threshold value of a certain parameter frequently triggers the parameter greater than or equal to the threshold value but no fault (false positive rate > 5%) in actual operation, the safety coefficient is lowered, such as from 1.1 to 1.05. If the second threshold value does not trigger a warning before a fault occurs multiple times (missed report rate > 2%), the fault coefficient is lowered, such as from 0.9 to 0.85, to ensure that the threshold value adapts to the actual operation characteristics of the equipment.

[0046] The setting of the threshold value can also be dynamically adapted. For example, it can be adjusted in stages: new equipment (running for 0 to 2 years) calculates the threshold value according to the standard coefficient; aged equipment (running for more than 5 years) can lower the current safety coefficient to 1.05 and the temperature safety coefficient to 1.15, to avoid false positives due to component performance degradation.

[0047] The setting of the threshold value can also be adapted to the scene: in a high-temperature environment (summer outdoor), the power tube temperature first threshold value can be lowered by 5 DEG C; in a heavy load scenario (continuous full power charging), the current first threshold value can be lowered by 3% to 5%, to improve the relevance of the threshold value in different working conditions.

[0048] It should be noted that the setting of the first threshold value establishes the boundary between safety and sub-health state, and can identify the initial state of parameter abnormality (such as slow rise of power tube temperature to near safety threshold value), realize early warning of equipment performance degradation, and avoid passive situation of triggering protection only at the limit value.

[0049] The setting of the second threshold value (such as input current being 90% of the minimum protection threshold value, power tube temperature being 90% of the fault temperature) establishes the boundary between sub-health and dangerous state, which not only reserves a fault disposal time window for maintenance personnel, but also effectively reduces the possibility of equipment damage, solving the problem of wide traditional protection threshold value setting and slow response.

[0050] In addition, the first threshold value and the second threshold value are based on the rated parameters of the equipment itself and the component failure parameter settings (such as the rated current of different power level charging piles is different, the threshold value is synchronized and differentiated), avoiding false positives caused by the one-size-fits-all general threshold value (such as the threshold value of a high-power device is applied to a small-power charging pile, which is easy to trigger frequent alarms). By defining numerical boundaries, abstract equipment states are converted into quantifiable indicators for comparison. The subsequent steps only need to compare the parameters with the threshold values to determine the state, reducing subjective judgment errors, providing reliable quantitative standards for equipment state determination, and establishing a scientific risk determination boundary, which lays a solid foundation for predictive maintenance and fine operation and maintenance management of direct current charging piles.

[0051] S203, determining the operation state of the direct current charging pile based on the fault type of the direct current charging pile, the operation data, the first threshold value and the second threshold value.

[0052] The operation state includes a first operation state, a second operation state and a third operation state. Optionally, the first operation state can be defined as a safe state (such as G level), indicating that the equipment is healthy and has no hidden dangers, and the operation parameters are stable within the safe range. The second operation state can be defined as a sick state (such as Y level), indicating that the parameters are abnormal but not reaching the dangerous limit, and there is an observable deterioration trend or slight failure. The third operation state can be defined as a dangerous state (such as R level), indicating that the parameters are greater than or equal to the critical value or the core component fails, and emergency measures need to be taken immediately.

[0053] One implementation manner is to determine the working state of the direct current charging pile based on the operation data, the first threshold value and the second threshold value of the direct current charging pile; and determine the operation state of the direct current charging pile based on the fault type and the working state of the direct current charging pile.

[0054] Optionally, the operation change information of the direct current charging pile is determined based on the operation data, the first threshold value and the second threshold value of the direct current charging pile; and the working state of the direct current charging pile is determined based on the operation change information.

[0055] The working state is a preliminary state determination result based on the operation data and the threshold value, and only reflects the operation level at the parameter level (such as stable parameters and ≤ the first threshold value, parameter fluctuation and between the first threshold value and the second threshold value, i.e. the first threshold value , parameter Second threshold value); the operation change information is a dynamic characteristic index of the operation data, including parameter fluctuation amplitude (such as within ±5% for stability, ±10% to 20% for fluctuation, and >20% for shock), change trend (up / down / stable), duration (such as more than 5 minutes of the first threshold value), etc., for describing the dynamic characteristics of parameter change.

[0056] Exemplary, the dynamic indicators can be calculated based on the operating data of the direct current charging pile. The following will be introduced respectively: First, the fluctuation amplitude. The difference between the current input current value and the average value of the previous 5 minutes accounts for, such as the current input current 115A, the average value of the previous 5 minutes input current 110A, the fluctuation amplitude = (115-110) / 110≈4.5%.

[0057] Second, the change trend. The slope of the parameter in the last 10 minutes is calculated by linear fitting, such as positive slope for upward trend, negative slope for downward trend, and slope absolute value <0.1 for stable.

[0058] Third, threshold duration. The operating data > first threshold, or the operating data ≥ second threshold for continuous minutes, such as power tube temperature 98℃ > 96℃, threshold duration has lasted for 3 minutes.

[0059] Further, based on the operating data, the first threshold, the second threshold and the operating change information, it can be divided into 3 types of working states, which can be specifically represented as.

[0060] Working state A: all operating data ≤ first threshold, and fluctuation amplitude ≤ ±5%, change trend stable, such as input current 105A ≤ 110A, fluctuation 3%, trend stable.

[0061] Working state B: operating data between first threshold and second threshold, or fluctuation amplitude 5% to 20%, such as output current 134A, between 132A and 135A, that is , or fluctuation amplitude 10%.

[0062] Working state C: any operating data ≥ second threshold, or fluctuation amplitude > 20%, such as power tube temperature 136℃ ≥ 135℃, or input current fluctuation 25%.

[0063] Further, in combination with the device fault diagnosis system and the operating data characteristics, the fault type can be identified.

[0064] In the case of operating change information consistent with working state A and no sensor alarm signal, the fault type is determined to be no fault.

[0065] In the case of freewheeling diode single tube fault (characteristics: input / output current fluctuation 8%, not greater than or equal to second threshold, no protection action), the fault type is determined to be slight fault.

[0066] In the case of power tube single tube fault (characteristics: temperature 136℃ ≥ 135℃), rectifier diode open phase fault (characteristics: input current A phase is 0, B / C phase is normal), the fault type is determined to be serious fault.

[0067] Further, the final running state can be output based on a cross-determination rule of the fault type and the working state.

[0068] If the fault type is no fault and the working state is A (for example, all parameters are stable and ≤ the first threshold value, no fault), it is determined that the final running state is the first running state (G level).

[0069] If the fault type is a slight fault and the working state is B, or no fault but the working state is B and the threshold duration ≥ 5 minutes (for example, a freewheeling diode fault, the output current fluctuation is 8% and the output current is between the output current first threshold value 132A and the output current second threshold value 135A; or, no fault but the power tube temperature 97℃> 96℃ for 6 minutes), it is determined that the final running state is the second running state (Y level).

[0070] If the fault type is a serious fault (regardless of the working state), or the working state is C (regardless of the fault type) (for example, a power tube single tube fault, temperature 136℃; or no fault but input current 170A≥117A), it is determined that the final running state is the third running state (R level).

[0071] It should be noted that the state determination can also consider the following special scene processing logic.

[0072] In the multi-parameter conflict scene, if the input current is in the working state A (105A≤110A) but the power tube temperature is in the working state C (136℃≥135℃), the working state corresponding to the most serious parameter is used as the criterion, the working state C is determined, and the final running state is the third running state.

[0073] In the short-term over-threshold value scene, if the parameter ≥ the second threshold value and the duration < 1 minute (for example, the output current 140A lasts for 30 seconds and then falls to 134A), and no fault type is identified, the working state B is determined to avoid misjudgment caused by transient interference.

[0074] In the protection action linkage scene, if the self-protection mechanism of the device has been started (for example, tripped due to input open-phase), regardless of the running data, it is directly determined as the third running state to ensure that the emergency state is prioritized.

[0075] It should be noted that by breaking through the traditional binary classification mode of normal / fault, a multi-dimensional determination system integrating fault type, parameter trend and threshold comparison is constructed. For example, even if the current is greater than the first threshold value, the embodiment can distinguish between different situations such as slow rise caused by parameter degradation and oscillation caused by diode fault, effectively avoiding state misjudgment caused by single parameter comparison, solving the problem of misjudging the sub-health of parameter degradation as a safe state, making the state classification more accurate, and realizing the multi-dimensional accurate perception of the device state.

[0076] The fault type recognition is closely associated with the operation state determination, which not only determines the state of the equipment, but also preliminarily diagnoses the cause of the state. For example, when the second operation state is determined, it is recognized that it is caused by the single tube fault of the freewheeling diode, which provides clear fault positioning information for subsequent operation and maintenance disposal, realizes the extension of the function from state early warning to preliminary diagnosis of the cause, significantly improves the operation and maintenance efficiency, and establishes the linkage mechanism of fault positioning and state evaluation.

[0077] The multi-determination mechanism combining parameter comparison, trend analysis and fault type verification effectively improves the reliability of state determination. For example, the third operation state is determined only when the parameter is greater than or equal to the second threshold value and the fault type is power tube fault, which avoids the misjudgment of dangerous state caused by temporary interference leading to temporary threshold value, significantly reduces the false positive rate of the system, and makes the state determination result more consistent with the actual operation condition of the equipment, and builds a rigorous determination logic of multiple verification.

[0078] Special processing rules are set for special scenes such as multi-parameter conflict, temporary threshold value and protection action linkage to ensure the robustness of the determination system under various complex working conditions. This scene adaptive capability enables the embodiment to adapt to various boundary conditions in the actual operation of the direct current charging pile, improves the practicality and reliability of the state evaluation system, and enhances the adaptive processing capability of complex scenes.

[0079] By establishing a standardized and quantitative state classification system, the abstract equipment operation condition is converted into a clear state level, which provides a clear and reliable input basis for the subsequent risk assessment link, ensuring that the final risk level determination can truly reflect the actual operation condition of the equipment, laying a solid foundation for predictive maintenance and refined operation and maintenance management, and providing a clear state basis for risk assessment.

[0080] That is, through multi-dimensional fusion, the equipment state is accurately classified to provide a clear state basis for the final risk assessment result, ensuring that the risk level determination does not deviate from the actual operation condition of the equipment.

[0081] S204, determining the risk assessment result of the direct current charging pile based on the operation state of the direct current charging pile.

[0082] Among them, the risk assessment result is a comprehensive determination conclusion for the operation safety of the direct current charging pile, which is a structured output information including risk level, risk characteristics, disposal suggestion and prediction data.

[0083] Wherein, the risk level refers to the equipment safety level classified according to the risk severity, which is the quantitative representation of the risk assessment result. The risk feature refers to the description of the specific manifestation of the risk, including detailed information such as risk source and development trend. The disposal suggestion refers to the specific operation and maintenance measures and response time requirements proposed for different risk levels. The prediction data refers to the risk development trend and failure probability information predicted based on the model.

[0084] One implementation manner is to determine the risk level corresponding to the operating state based on the operating state of the DC charging pile and the preset mapping relationship table; and determine the risk assessment result of the DC charging pile based on the operating state of the DC charging pile and the risk level.

[0085] Wherein, the preset mapping relationship table is a running state-risk level corresponding rule table established based on equipment failure mechanism analysis and long-term operation and maintenance experience, which ensures the uniformity and repeatability of the risk assessment standard. The construction of this table follows the following principles.

[0086] Scientific principle: based on a large amount of historical failure data and equipment failure mode analysis.

[0087] Practical principle: combined with field operation and maintenance experience and actual disposal capacity.

[0088] Extensibility principle: support adjustment and optimization according to different equipment models and operating environments.

[0089] Optionally, the preset mapping relationship table can be represented as: Table 1 preset mapping relationship table

[0090] Further, the risk level is the equipment safety level classified according to the risk severity, which can correspond to three categories, namely G level (safety level), Y level (illness level) and R level (danger level).

[0091] Wherein, G level (safety level) refers to no failure risk, and the equipment is running healthily; Y level (illness level) refers to the existence of potential risk, showing a progressive deterioration trend; R level (danger level) refers to high risk, which may immediately cause failure or safety accident.

[0092] The risk level determination process adopts a rule-based matching algorithm, and the process can be as follows: State feature extraction, extracting key feature parameters from the operating state; Mapping table query, matching the feature parameters with the preset mapping relationship table; Priority determination, when there are multiple state features, the most serious principle is adopted; Level output, output the finally determined risk level; For example, the operating state of a certain DC charging pile is input current 115A (between input current first threshold 110A and input current second threshold 117A), and the single tube of the freewheeling diode fails. After feature extraction and mapping table matching, it is determined to be the second operating state, corresponding to risk level Y level.

[0093] Optionally, the operating data can also be input into the risk prediction model to obtain the risk level change data of the DC charging pile.

[0094] The risk prediction model is a model constructed based on a random forest algorithm, which can use device historical operating data and failure data as training samples, and can output risk level change trend, failure probability and other prediction results. The risk level change data is the device risk evolution information predicted by the model, including at least one of the risk level change curve, the failure probability and the failure type.

[0095] Optionally, the random forest algorithm is an ensemble learning algorithm that improves prediction accuracy by constructing multiple decision trees and combining their output results. It has the characteristics of anti-overfitting and strong adaptability to nonlinear data, and is suitable for device risk prediction scenarios. The risk level change curve can be the dynamic change trajectory of the risk level in the future period of time (such as 24 hours). The failure probability can be the possibility (quantified as 0% to 100%) of failure in a specific period of time (such as 7 days). The failure type is the most likely failure category (such as power tube failure, diode failure).

[0096] For example, structured mapping rules can be developed in combination with operating state definitions, failure impact levels and operation and maintenance priorities. Optionally, the pre-set mapping relationship table can be iteratively optimized in combination with device models, for example, the second operating state of old equipment can lower the risk level determination standard (such as determining Y level if the parameter is greater than the first threshold for 3 minutes), which adapts to the risk characteristics of aging equipment.

[0097] Further, the operating state can be used as a search keyword to locate the corresponding risk level in the pre-set mapping relationship table. For example, if the device operating state is input current 115A (between input current first threshold 110A and input current second threshold 117A) and the single tube of the freewheeling diode fails, it is determined to be the second operating state, and the risk level Y level is determined by mapping table matching. Optionally, the matching process can support multi-state priority determination. If the device (such as a DC charging pile) has multiple operating characteristics (such as input current greater than or equal to the second threshold and power tube temperature normal) at the same time, the risk level corresponding to the most serious state is used as the reference.

[0098] Furthermore, the operating state details and the risk level can be integrated to supplement the risk feature description and the preliminary treatment suggestion to form a structured risk assessment result, which can specifically include: risk level, operating state details, risk features and preliminary suggestions.

[0099] For example, the risk level is Y (illness level); the operating status details are a single freewheeling diode failure, with an output current of 134A (the first threshold for the output current is 132A, and the second threshold for the output current is 135A), and a fluctuation range of 4%; the risk characteristics are that the parameters are in the critical range, the fault has not spread but there is a possibility of deterioration; the preliminary recommendation is to conduct diode performance testing within 24 hours.

[0100] Optionally, risk level change data can be obtained based on risk prediction models to predict advanced risks. This approach can be data-driven prediction, suitable for medium- to long-term risk evolution analysis. The steps are training-input-prediction, and the specific implementation process can be as follows: Step 1: Construct a risk prediction model (random forest algorithm).

[0101] Using the random forest algorithm as the core, the entire process of data preparation, model training, and optimization is completed.

[0102] Training sample preparation. Obtain input features, such as historical operating data of the device (e.g., real-time values ​​and rates of change of input current, output current, and power transistor temperature), operating status codes (first state = 0, second state = 1, third state = 2), device operating time, and number of historical faults.

[0103] Obtain output labels, such as the risk level for the next 24 hours (G=0, Y=1, R=2), the probability of failure for the next 7 days (0% to 100%), and the type of failure (e.g., power transistor failure = 1, diode failure = 2).

[0104] The data volume needs to meet the following requirements: It includes historical data from at least 1,000 devices of the same model, spanning one year, to ensure that the sample covers all operational scenarios.

[0105] Model training process. The dataset is divided into a training set (for model learning), a validation set (for parameter tuning), and a test set (for performance evaluation) in a 7:2:1 ratio. Decision tree construction: 50-100 decision trees are generated based on the training set (the number needs to balance accuracy and efficiency). 3-5 input features are randomly selected from each branch node of each tree (such as power transistor temperature change rate and operating status encoding), and nodes are split according to the principle of maximizing information gain. Parameter optimization: Key parameters (such as the number of decision trees, maximum depth, and number of split features) are adjusted using the validation set, with the goal of achieving a prediction accuracy ≥90% and a fault type identification accuracy ≥85% on the test set.

[0106] Model deployment: The trained model is encapsulated as an API (Application Programming Interface) interface and integrated into the charging pile management system, supporting real-time data input and result output with a response delay ≤ 1 second.

[0107] It should be noted that the random forest algorithm ensures the accuracy and stability of predictions through the following mechanisms.

[0108] Bootstrap aggregation mechanism: Each decision tree uses a different training subset to increase model diversity; Feature random selection mechanism: When splitting nodes, a random subset of features is selected to reduce the risk of overfitting; Majority voting mechanism: The results of all decision trees are combined to improve prediction stability; Out-of-Bag evaluation mechanism: The performance of the model is evaluated using data not involved in training.

[0109] The specific execution process of the algorithm is as follows: Randomly and with replacement, k sample subsets are drawn from the original training set using the Bootstrap method. A decision tree is built for each sample subset, and when splitting nodes, k features are randomly selected from M features (k < M). Each tree grows completely without pruning. For classification problems, the final result is determined by the voting method; for regression problems, the output is determined by the averaging method.

[0110] Step 2: Input the operating data into the risk prediction model.

[0111] The real-time operating data can be preprocessed in the format required by the risk prediction model and uploaded to the model interface.

[0112] First, data preprocessing is performed, specifically including: normalization and feature engineering.

[0113] The normalization process converts parameters with different magnitudes such as current (0 to 200 A) and temperature (0 to 200 °C) into standardized data in the range of 0 to 1.

[0114] The feature engineering process generates derivative features (such as "maximum temperature difference in the last 10 minutes" and "cumulative duration greater than the first threshold") based on the original data and supplements them to the input feature set.

[0115] An example of the input format is as follows.

[0116] Input features: [input current 115 A (normalized 0.575), output current 134 A (normalized 0.67), power transistor temperature 97 °C (normalized 0.485), temperature change rate 0.3 °C / min, operating status code 1, operating duration 3 years, historical faults 1 time].

[0117] Step 3: Output risk level change data.

[0118] The risk prediction model performs parallel discrimination using multiple decision trees on the input data, and outputs risk level change data after summarizing the results, which can specifically include three types of information.

[0119] Risk level change curve: For example, a line graph with time as the horizontal axis (0 to 24 hours) and risk level as the vertical axis (0=G, 1=Y, 2=R); this line graph can intuitively show the risk development trend and provide a time window for preventive maintenance; for example, the risk level remains at level Y from 0 to 8 hours, and rises to level R from 8 to 12 hours, indicating that the risk is deteriorating rapidly.

[0120] Failure probability: For example, quantified values ​​(accurate to 1%) are accompanied by confidence intervals (e.g., a failure probability of 68%, a 95% confidence level, and a confidence interval of 62% to 74%); a probability ≥80% is a high-risk warning, requiring immediate intervention; 50% to 80% is a medium-risk warning, requiring action within 24 hours; and <50% is a low-risk warning, requiring enhanced monitoring.

[0121] Risk classification: For example, a high-risk warning (≥80%) requires immediate intervention; a medium-risk warning (50% to 80%) requires action within 24 hours; and a low-risk warning (<50%) requires increased monitoring frequency.

[0122] Fault type prediction: Output possible fault types and their corresponding probabilities (e.g., freewheeling diode failure, probability 72%; power transistor aging, probability 18%; poor line contact, probability 10%); providing maintenance personnel with accurate troubleshooting directions and shortening fault location time.

[0123] For example, the probability of output freewheeling diode failure is 72%, the probability of power transistor aging is 18%, and the probability of poor line contact is 10%.

[0124] Step 4: Integrate forecast data to improve risk assessment results.

[0125] The basic risk level can be combined with the forecast data to form a comprehensive risk assessment conclusion on the current situation and trends.

[0126] For example, the current risk level is Y (sick level); the operating status details are a single freewheeling diode failure, with an output current of 134A and a fluctuation range of 4%; the future risk prediction includes the change curve (e.g., it will rise to level R after 8 hours), the failure probability (e.g., a failure probability of 68% in the next 7 days), the possible failure type (e.g., freewheeling diode breakdown), and the handling suggestions (e.g., immediately arrange diode testing and replacement to avoid the failure from escalating).

[0127] Finally, the risk assessment results can be verified and output.

[0128] Optionally, it is also possible to check whether the matching of operating status and risk level conforms to the preset mapping table rules, and whether the predicted data is consistent with the current status logic (e.g., if the current level is G but the predicted level is R after 1 hour, the input data needs to be reviewed). It is also possible to compare the historical data of the same type of equipment. If the predicted failure probability deviates from the actual failure rate by more than 15% under similar operating conditions, the model is triggered to recalibrate to ensure the reliability of the results.

[0129] The output of risk assessment results can be tailored to the needs of the operation and maintenance scenario. For example, the content displayed on the operation and maintenance terminal can include: text conclusions + risk level change curves, highlighting key information such as "R-level warning" and "high probability failure". The backend system can store raw data and risk assessment results, generate historical trajectory ledgers for subsequent threshold optimization and model iteration. It can also include an alarm module, which can trigger audible and visual alarms + SMS notifications when the risk level is R or the failure probability is ≥80%, and push them to the operation and maintenance personnel's mobile devices.

[0130] It should be noted that transforming abstract operational states (such as the second operational state) into concrete risk conclusions (Level Y, to be addressed within 24 hours) directly guides operational actions, realizing the transformation from state description to risk decision-making; using the random forest model to achieve a full-dimensional risk assessment of current risks and future evolution, shifting from post-event repairs to pre-event prevention, realizing an upgrade from real-time risk assessment to trend prediction; and by combining fault type prediction with operational status details, providing precise guidance on causes and solutions for operations and maintenance, reducing operational costs, and achieving optimization from unified standards to precise positioning.

[0131] It should be noted that clearly defining the maintenance strategies for G-level (safe, low priority), Y-level (ill-conditioned, priority maintenance), and R-level (dangerous, immediate shutdown) avoids the problem of traditional risk assessments only assigning levels without action guidelines. This ensures that risk assessment results can directly guide maintenance (e.g., Y-level equipment should be tested within 24 hours, R-level equipment should be immediately powered off). In addition to outputting the risk level, it can also include risk characteristics (e.g., "freewheeling diode failure caused current fluctuations, not reaching the second threshold") and fault location, solving the problem of vague risk descriptions in traditional risk assessments (e.g., only stating that the equipment is at risk without specifying the source of the risk). This facilitates accurate troubleshooting by maintenance personnel and makes risk conclusions more concrete. The risk assessment results (such as the false alarm rate of Class Y equipment and the false alarm rate of Class R equipment) can also be used to feed back into the accumulation of historical data, and be used for the dynamic optimization of the dual threshold (such as lowering the first threshold if the false alarm rate of a certain model of equipment is high in Class Y) and the training of machine learning models, so as to further improve the accuracy of subsequent risk assessments, form a closed loop of assessment, optimization and reassessment, and form a closed loop linkage with subsequent optimization.

[0132] In other words, by transforming operational status into actionable risk conclusions, the accuracy and practicality of risk assessment results are ultimately achieved. This not only clarifies the safety level of equipment but also provides clear guidance for operation and maintenance, fully meeting the core objective of obtaining accurate risk assessment results.

[0133] By forming a complete chain of data collection, threshold construction, status classification, and risk output, each link is interconnected and mutually supportive: the accurate data obtained provides reliable input for subsequent threshold calculation, the scientific threshold provides a judgment standard for subsequent status classification, and the accurate status classification provides a basis for risk rating.

[0134] Through coordinated optimization of each step, the risk assessment results are ultimately ensured to accurately reflect the safety status of the equipment, precisely locate potential faults, and clarify the direction of operation and maintenance. This effectively solves the problems of lagging protection, ambiguous status, and misjudgment of risks in traditional risk assessment, and realizes the accuracy and practicality of risk assessment for DC charging piles.

[0135] The Random Forest algorithm is an algorithm that integrates multiple trees using the concept of ensemble learning. Its basic unit is the decision tree, and it belongs to a major branch of machine learning, namely ensemble learning methods. Each decision tree is a classifier, so for an input sample, N trees will have N classification results. Random Forest integrates all the classification voting results and designates the class with the most votes as the final output.

[0136] A decision tree is a classification algorithm that uses attribute criteria on layers of nodes to classify input data and assign the classification results to leaf nodes. Specifically, a decision tree uses the attribute that has the greatest impact on the classification result in the input dataset as the root node, and then continues to find the attribute with the next greatest impact on the classification result in the classified dataset as the next branch node, until the dataset is classified into the specified categories, and the classification results are assigned to leaf nodes.

[0137] Based on the existing threshold system, a random forest model is trained using historical fault data to achieve trend prediction and fault probability risk assessment of target equipment. Specifically, this involves: generating n decision trees using the training set; before selecting an attribute at each branch node, randomly selecting k attributes (where M is the number of sample attributes) from M attributes as the splitting attribute set for the current node, and splitting the node using the optimal splitting method among these k attributes; inputting actual samples into the n decision trees for discrimination, and summarizing the results of all decision trees through an ensemble strategy (such as voting, probability averaging, etc.) to finally obtain the trend prediction and fault probability risk assessment results for the target equipment.

[0138] The aforementioned risk assessment method for DC charging piles ensures the fundamental reliability of risk assessment from the source by acquiring operational data of the DC charging piles, which is a prerequisite for accurate risk assessment. Furthermore, based on the preset parameters of the DC charging pile equipment, a first and second threshold are determined, achieving threshold customization and gradation. This provides a scientific judgment benchmark for risk assessment, ensuring the accuracy of status classification while avoiding protection lag or false alarms. Moreover, based on the fault type, operational data, and the first and second thresholds, the operating status of the DC charging pile can be determined, achieving accurate classification of equipment status. Finally, the risk assessment result of the DC charging pile can be determined based on its operating status. This solution provides a basis for determining the operating status of the DC charging pile by introducing the first and second thresholds; furthermore, by introducing the operating status, it lays the foundation for directly assessing the risk of the DC charging pile, ultimately ensuring the accuracy of the risk assessment results.

[0139] The above text combined Figures 1 to 2 The risk assessment method for DC charging piles provided in the embodiments of the present invention has been described in detail. The apparatus and equipment provided in the embodiments of the present invention will be described below with reference to the accompanying drawings.

[0140] like Figure 3 As shown in the figure, this is a structural block diagram of a DC charging pile risk assessment device provided in an embodiment of the present invention. The DC charging pile risk assessment device 500 includes: an acquisition module 501, a threshold determination module 502, a status determination module 503, and an assessment determination module 504, wherein: The acquisition module 501 is used to acquire the operating data of the DC charging pile; The threshold determination module 502 is used to determine a first threshold and a second threshold of the DC charging pile based on the preset parameters of the DC charging pile; wherein the first threshold is less than the second threshold. The status determination module 503 is used to determine the operating status of the DC charging pile based on the fault type, operating data, first threshold and second threshold of the DC charging pile. The assessment and determination module 504 is used to determine the risk assessment results of the DC charging pile based on its operating status.

[0141] In one embodiment, the threshold determination module 502 is specifically used for: Based on the equipment's rated parameters and the corresponding preset safety factor, the first threshold of the DC charging pile is determined; The second threshold of the DC charging pile is determined based on the equipment threshold, equipment fault parameters, and preset fault coefficient.

[0142] In one embodiment, the state determination module 503 is specifically used for: The working status of the DC charging pile is determined based on its operating data, the first threshold, and the second threshold. The operating status of a DC charging pile is determined based on its fault type and working status.

[0143] In one embodiment, the state determination module 503 is specifically used for: Based on the operation data of the DC charging pile, the first threshold and the second threshold, the operation change information of the DC charging pile is determined; The operating status of the DC charging pile is determined based on the information on changes in operation.

[0144] In one embodiment, the evaluation and determination module 504 is specifically used for: Based on the operating status of DC charging piles, the risk level corresponding to the operating status is determined according to the preset mapping relationship table. Based on the operating status and risk level of the DC charging pile, the risk assessment results of the DC charging pile are determined.

[0145] In one embodiment, the DC charging pile risk assessment device 500 further includes: The prediction module is used to input operational data into the risk prediction model to obtain risk level change data of DC charging piles; wherein, the risk prediction model is a model built based on the random forest algorithm; the risk level change data includes at least one of the following: risk level change curve, failure probability, and failure type.

[0146] In one embodiment, the operating data includes at least one of input current, output current, and power transistor temperature.

[0147] The DC charging pile risk assessment device 500 according to embodiments of the present invention can correspond to the execution of the method described in the embodiments of the present invention, and the other operations and / or functions of each module / unit of the DC charging pile risk assessment device 500 are respectively for implementing Figure 2 For the sake of brevity, the corresponding processes of each method in the illustrated embodiments will not be described in detail here.

[0148] This invention also provides a computing device. This computing device can be a local computing device or an application server.

[0149] like Figure 4 As shown in the figure, this figure is a schematic diagram of a computing device provided in an embodiment of the present invention. The computing device 700 includes a bus 701, a processor 702, a communication interface 703, and a memory 704.

[0150] The processor 702, memory 704 and communication interface 703 communicate via bus 701.

[0151] The 701 bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0152] The processor 702 can be any one or more of the following processors: central processing unit (CPU), graphics processing unit (GPU), microprocessor (MP), or digital signal processor (DSP).

[0153] The communication interface 703 is used for external communication.

[0154] For example, communication interface 703 can be used to communicate with terminal 102. Communication interface 703 is used to display the risk assessment results to terminal 102.

[0155] The memory 704 may include volatile memory, such as random access memory (RAM).

[0156] The memory 704 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0157] The memory 704 stores executable code, and the processor 702 executes the executable code to perform the aforementioned DC charging pile risk assessment method.

[0158] Specifically, in achieving Figure 3 In the case of the illustrated embodiment, and Figure 3 When the modules or units of the DC charging pile risk assessment device described in the embodiment are implemented through software, the following steps are performed: Figure 3The software or program code required for the functions of each module / unit can be partially or entirely stored in the memory 704. The processor 702 executes the program code corresponding to each unit stored in the memory 704 to perform the aforementioned DC charging pile risk assessment method.

[0159] This invention also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that a computing device can store, or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute the aforementioned DC charging pile risk assessment method.

[0160] This invention also provides a computer program product comprising one or more computer instructions. When the computer instructions are loaded and executed on a computing device, all or part of the processes or functions described in this invention are generated.

[0161] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0162] When the computer program product is executed by a computer, the computer performs any of the aforementioned methods of the DC charging pile risk assessment method. The computer program product can be a software installation package; when any of the aforementioned methods of the DC charging pile risk assessment method is required, the computer program product can be downloaded and executed on the computer.

[0163] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0164] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present invention should be covered within the scope of protection of the present invention.

Claims

1. A risk assessment method for DC charging piles, characterized in that, The method includes: Obtain operational data from DC charging piles; Based on the preset parameters of the DC charging pile, a first threshold and a second threshold of the DC charging pile are determined; wherein the first threshold is less than the second threshold. Based on the fault type, operating data, first threshold and second threshold of the DC charging pile, the operating status of the DC charging pile is determined; Based on the operating status of the DC charging pile, the risk assessment result of the DC charging pile is determined.

2. The method according to claim 1, characterized in that, The preset parameters of the equipment include rated parameters, threshold parameters, and fault parameters. Determining the first and second threshold parameters of the DC charging pile based on these preset parameters includes: Based on the rated parameters of the equipment and the corresponding preset safety factor, the first threshold of the DC charging pile is determined; Based on the device threshold, device fault parameters, and preset fault coefficient, the second threshold of the DC charging pile is determined.

3. The method according to claim 1, characterized in that, The process of determining the operating status of the DC charging pile based on its fault type, operating data, first threshold, and second threshold includes: Based on the operating data of the DC charging pile, the first threshold, and the second threshold, the working status of the DC charging pile is determined; The operating status of the DC charging pile is determined based on its fault type and operating status.

4. The method according to claim 3, characterized in that, Determining the operating status of the DC charging pile based on its operating data, a first threshold, and a second threshold includes: Based on the operating data of the DC charging pile, the first threshold and the second threshold, the operating change information of the DC charging pile is determined; Based on the operational change information, the working status of the DC charging pile is determined.

5. The method according to claim 1, characterized in that, The operating status includes a first operating status, a second operating status, and a third operating status. Determining the risk assessment result of the DC charging pile based on its operating status includes: Based on the operating status of the DC charging pile and the preset mapping relationship table, the risk level corresponding to the operating status is determined; Based on the operating status and risk level of the DC charging pile, the risk assessment result of the DC charging pile is determined.

6. The method according to claim 1, characterized in that, The method further includes: The operational data is input into the risk prediction model to obtain the risk level change data of the DC charging pile; wherein, the risk prediction model is a model constructed based on the random forest algorithm; the risk level change data includes at least one of the following: risk level change curve, failure probability, and failure type.

7. The method according to claim 1, characterized in that, The operating data includes at least one of the following: input current, output current, and power transistor temperature.

8. A risk assessment device for DC charging piles, characterized in that, The device includes: The acquisition module is used to acquire the operating data of the DC charging pile; A threshold determination module is used to determine a first threshold and a second threshold of the DC charging pile based on the preset parameters of the DC charging pile; wherein the first threshold is less than the second threshold. The status determination module is used to determine the operating status of the DC charging pile based on the fault type, operating data, first threshold and second threshold of the DC charging pile; The assessment and determination module is used to determine the risk assessment result of the DC charging pile based on its operating status.

9. A computing device, characterized in that, Including memory and processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program for performing the method as described in any one of claims 1 to 7.

Images