Electric vehicle charging pile fault remote detection method and device, electronic equipment and medium

Abstract

The invention discloses an electric vehicle charging pile fault remote detection method and device, electronic equipment and a medium, and relates to the technical field of computers. The method comprises the following steps: determining an associated charging pile set in a target area, wherein the associated charging pile set and a current charging pile belong to the same power system; acquiring first operation data of the current charging pile and second operation data corresponding to each associated charging pile in the associated charging pile set; calculating a fault evaluation value corresponding to each fault evaluation index of the current charging pile according to the first operation data and each second operation data; calculating a real-time score corresponding to the current charging pile according to each fault evaluation value and different weights corresponding to each fault evaluation index; obtaining historical operation data of the target charging pile set, and calculating a fault score threshold value of the fault charging pile according to the historical operation data; and comparing the real-time score with a fault score threshold value, and determining whether the current charging pile has a fault or not. The fault detection efficiency of the charging pile can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and particularly relates to a method, device, electronic device and medium for remotely detecting faults of electric vehicle charging piles. Background Art

[0002] Currently, with the popularization of new energy, more and more people use electric vehicles. As an important facility for energy supply of electric vehicles, the operating state of the charging pile is extremely important. However, when the charging pile is in a high-load working state for a long time, various faults are likely to occur. Therefore, it is necessary to detect faults of the charging pile to improve the charging efficiency.

[0003] In the related art, the detection of faults of the charging pile mainly relies on manual work, that is, the staff arrives at the site to detect and repair the charging pile only after the charging pile fails, resulting in low efficiency of fault detection of the charging pile. Summary of the Invention

[0004] The present application provides a method, device, electronic device and medium for remotely detecting faults of electric vehicle charging piles, which are used to improve the efficiency of fault detection of the charging pile.

[0005] In the first aspect of the present application, a method for remotely detecting faults of an electric vehicle charging pile is provided, which specifically includes: Determine an associated charging pile set within a target area, where the associated charging pile set and the current charging pile belong to the same power system; Obtain first operation data of the current charging pile and second operation data corresponding to each associated charging pile in the associated charging pile set; According to the first operation data and each second operation data, calculate fault evaluation values corresponding to each fault evaluation index of the current charging pile; According to each fault evaluation value and different weights corresponding to each fault evaluation index, calculate a real-time score corresponding to the current charging pile; Obtain historical operation data of a target charging pile set, and calculate a fault scoring threshold of a faulty charging pile according to the historical operation data. The target charging pile set has the same type as the associated charging pile set, the faulty charging pile and the target charging pile set belong to the same power system, and the faulty charging pile is a charging pile that has failed; Compare the real-time score with the fault scoring threshold to determine whether the current charging pile has failed.

[0006] By adopting the above technical solution, first, analyze the topological structure of the power system in the current area to obtain the power system scope of the current charging pile. According to the power system database within the said scope, the set of associated charging piles in the same power system as the current charging pile can be determined. Then, according to the usage status of the current charging pile and each associated charging pile in the set of associated charging piles, two different acquisition modes can be used to acquire the first operation data of the current charging pile and the second operation data corresponding to each associated charging pile in the set of associated charging piles. Furthermore, based on the first operation data and the second operation data, the fault evaluation value corresponding to the current charging pile is calculated, providing a measurement standard for real-time scoring. Dynamically adjust the weight of each index according to the population distribution characteristics of the fault evaluation index, and perform weighting based on the weight and the fault evaluation value to obtain the real-time score. Then, obtain the historical distribution of the fault scores of the faulty charging piles according to the historical operation data, and calculate the warning threshold and the emergency threshold through this historical distribution of the fault scores. Finally, compare the threshold with the real-time score to achieve the fault detection of the charging pile. By remotely detecting the faults of the charging pile in real time, it is possible to avoid the need to arrive at the scene for detection after a fault occurs, saving time and improving the efficiency of charging pile fault detection.

[0007] Optionally, determining the set of associated charging piles within the target area includes: Obtain the topological structure of the power system within the target area, and determine the power system scope where the current charging pile is located according to the topological structure of the power system; According to the power system database within the power system scope, determine the set of associated charging piles in the same power system as the current charging pile.

[0008] By adopting the above technical solution, first, by obtaining the topological structure of the power system within the target area, the distribution and connection relationships of each power system within the area can be clearly understood. Then, determine the power system scope where the current charging pile is located according to the topological structure. This step ensures the accurate positioning of the current charging pile. Next, according to the power system database within the power system scope, other charging piles in the same power system as the current charging pile can be found. Since these charging piles are in the same power system, they have a certain degree of relevance and similar operating environments. These two technical features cooperate with each other to accurately determine the set of associated charging piles within the target area, solving the problem of how to determine the set of associated charging piles within the target area.

[0009] Optionally, obtaining the first operation data of the current charging pile and the second operation data corresponding to each associated charging pile in the set of associated charging piles includes: Determine the first acquisition mode of the first operation data and the second acquisition mode of the second operation data according to the usage status of the charging pile; Obtain the first operation data of the current charging pile according to the first collection mode, and obtain the second operation data corresponding to each associated charging pile in the associated charging pile set according to the second collection mode.

[0010] By adopting the above technical solution, by dynamically adjusting the data collection mode, it is ensured that the most appropriate operation data can be obtained under different usage states, solving the problem of how to efficiently and accurately obtain the operation data of the current charging pile and associated charging piles. In this way, the operation data of the charging pile can be obtained more accurately and efficiently, thereby providing reliable data support for subsequent fault detection.

[0011] Optionally, the obtaining the first operation data of the current charging pile according to the first collection mode and obtaining the second operation data corresponding to each associated charging pile in the associated charging pile set according to the second collection mode includes: When the usage state is the idle state, determine that both the first collection mode and the second collection mode are the low-frequency collection mode, and use the low-frequency collection mode to collect the first operation data, where the first operation data is the voltage, current, and ambient temperature of the current charging pile; use the low-frequency collection mode to collect the second operation data of each associated charging pile in the associated charging pile set, where the second operation data is the power system-level parameter; When the usage state is the charging state, determine that both the first collection mode and the second collection mode are the high-frequency collection mode, and use the high-frequency collection mode to collect the first operation data, where the first operation data is the multi-dimensional collected current waveform, voltage waveform, temperature gradient, and power factor of the current charging pile; use the high-frequency collection mode to collect the second operation data of each associated charging pile in the associated charging pile set, where the second operation data is the load balancing data; When the usage state is the abnormal state, determine that both the first collection mode and the second collection mode are the high-frequency collection mode, and use the high-frequency collection mode to collect the first operation data, where the first operation data is the full amount of data of the current charging pile; use the high-frequency collection mode to collect the second operation data of each associated charging pile in the associated charging pile set, where the second operation data is the grid harmonic distortion rate and the grounding resistance.

[0012] By adopting the above technical solutions, different acquisition modes can be selected according to the specific usage status, and the data acquisition frequency and content can be flexibly adjusted according to the specific usage status of the charging pile, ensuring accurate and comprehensive operation data can be obtained in different states, so as to achieve precise monitoring and fault detection of the operation status of the charging pile. This not only improves the operation monitoring accuracy of the charging pile, but also provides reliable data support for fault detection and prevention. It can more effectively monitor the operation status of the charging pile, timely detect and handle faults, and improve the reliability and user experience of the charging pile.

[0013] Optionally, the fault evaluation indexes include current harmonic distortion rate, three-phase voltage unbalance degree, power factor deviation value, abnormal coefficient of grounding resistance, and charging efficiency attenuation rate. Calculating the fault evaluation values corresponding to the respective fault evaluation indexes of the current charging pile according to the first operation data and the respective second operation data includes: Calculating the fault evaluation values corresponding to the respective fault evaluation indexes according to the first formula; The first formula is: Wherein, S i is the fault evaluation value of each of the fault evaluation indexes; x i is the detection value of the i-th fault evaluation index of the current charging pile; μ i is the mean value of the i-th fault evaluation index under the historical normal state; σ i is the standard deviation of the i-th fault evaluation index under the historical normal state; β is the environmental coupling coefficient; ΔG i is the offset amplitude between the current harmonic distortion rate of the correlation set and the historical baseline.

[0014] By adopting the above technical solutions, the accuracy and reliability of charging pile fault detection can be improved through quantitative evaluation indexes and formula calculations. By introducing multiple fault evaluation indexes such as current harmonic distortion rate, three-phase voltage unbalance degree, power factor deviation value, abnormal coefficient of grounding resistance, and charging efficiency attenuation rate, and combining historical data and current detection data, the fault evaluation values are calculated using the formula. It can more accurately evaluate the fault situation of the charging pile, improve the accuracy and reliability of fault detection, and avoid the lag and uncertainty of traditional manual detection.

[0015] Optionally, calculating the real-time score corresponding to the current charging pile according to the respective fault evaluation values and different weights corresponding to the respective fault evaluation indexes includes: Dynamically adjusting the weight coefficient corresponding to each fault evaluation index according to the second formula and based on the population distribution characteristics of each fault evaluation index; The second formula is: Among them, the population distribution characteristics include x i , μG i , σG i ; ω i is the weight coefficient of the i-th fault evaluation index; ω i0 is the baseline weight of the i-th fault evaluation index; x i is the deviation degree of the standardized fault evaluation index; μG i is the population mean of the i-th fault evaluation index of the associated charging pile set; σG i is the population dispersion of the i-th fault evaluation index of the associated charging pile set; α is the dynamic sensitivity coefficient; Perform a linear weighted sum on the adjusted weight coefficient and the corresponding fault evaluation value to generate the real-time score.

[0016] By adopting the above technical solution, the operation data of the charging pile is monitored in real time, the standardized deviation degree of each fault evaluation index is calculated, and the weight coefficient is adjusted according to these deviation degrees. Thus, the weight coefficient can reflect the current working state and fault risk of the charging pile. This formula takes into account the population distribution characteristics of the fault evaluation index, including the weight coefficient, baseline weight, standardized deviation degree, population mean, population dispersion, and dynamic sensitivity coefficient. These characteristics work together to ensure that the weight coefficient can be adjusted according to the actual situation, thereby improving the accuracy and real-time performance of fault evaluation. Solve the problem of dynamic adjustment of the weight coefficient in the fault evaluation of the charging pile, ensure that the real-time score is more accurate, and improve the reliability of fault detection.

[0017] Optionally, calculating the fault score threshold of the faulty charging pile according to the historical operation data includes: calculating the historical distribution of the fault scores of the faulty charging pile according to the historical operation data; Determine the fault score threshold according to the percentile, mean, and standard deviation of the historical distribution of the fault scores of the faulty charging pile; The fault score threshold includes multiple levels of thresholds, and the multiple levels of thresholds include: The early warning threshold is used to indicate potential fault risks; The emergency threshold is used to indicate serious fault risks.

[0018] By adopting the above technical solution, calculating the historical distribution of the fault scores of the faulty charging pile can obtain the score data of the charging pile in different states; using the percentile, mean, and standard deviation of these score data can more accurately determine the fault score threshold of the charging pile; the setting of multiple levels of thresholds can respectively indicate the risks of potential faults and serious faults, which helps to give early warnings and handle faults in a timely manner. Solve the problem of how to accurately determine the fault score threshold of the charging pile, thereby improving the accuracy and timeliness of the fault detection of the charging pile.

[0019] In a second aspect of the present application, a charging pile fault detection device is provided, specifically including: A charging pile determination module, configured to determine a set of associated charging piles in the target area that belong to the same power system as the current charging pile; an acquisition data module, which acquires first operation data of the current charging pile and second operation data corresponding to each associated charging pile in the set of associated charging piles; An evaluation value calculation module, which calculates a fault evaluation value corresponding to each fault evaluation index of the current charging pile according to the first operation data and each of the second operation data; A real-time score calculation module, which calculates a real-time score corresponding to the current charging pile according to each of the fault evaluation values and different weights corresponding to each of the fault evaluation indexes; A threshold calculation module, which acquires historical operation data of a target charging pile set, and calculates a fault score threshold of a faulty charging pile according to the historical operation data. The target charging pile set has the same type as the set of associated charging piles, the faulty charging pile belongs to the same power system as the target charging pile set, and the faulty charging pile is a charging pile that has failed; By adopting the above technical solution, first, analyze the power system topology structure of the current area to obtain the power system range of the current charging pile. According to the power system database within the range, a set of associated charging piles in the same power system as the current charging pile can be determined; then, according to the usage status of the current charging pile and each associated charging pile in the set of associated charging piles, two different acquisition modes can be used to acquire the first operation data of the current charging pile and the second operation data corresponding to each associated charging pile in the set of associated charging piles. Furthermore, a fault evaluation value corresponding to the fault evaluation value of the current charging pile is calculated according to the first operation data and the second operation data, providing a measurement standard for the real-time score; dynamically adjust the weight of each index according to the population distribution characteristics of the fault evaluation indexes, and perform weighting according to the weight and the fault evaluation value to obtain the real-time score; then obtain the historical distribution of the fault scores of the faulty charging piles according to the historical operation data, and calculate the warning threshold and the emergency threshold through this historical distribution of the fault scores; finally, compare the threshold and the real-time score, so as to realize the fault detection of the charging pile. The fault of the charging pile is detected in real time in a remote manner, avoiding the need to arrive at the scene for detection after the fault occurs, saving time and improving the efficiency of charging pile fault detection.

[0020] In a third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method described in any one of the above.

[0021] In a fourth aspect of the present application, a computer-readable storage medium is provided. The computer-readable storage medium stores instructions that, when executed, perform the method described in any one of the above. Description of the Drawings

[0022] Figure 1 is a schematic flowchart of a method for remotely detecting faults in an electric vehicle charging pile provided by an embodiment of the present application; Figure 2 is Figure 1 a schematic sub-step flowchart of step S101; Figure 3 is Figure 1 a schematic sub-step flowchart of step S102; Figure 4 is Figure 1 a schematic sub-step flowchart of step S105; Figure 5 is a schematic structural diagram of a device for remotely detecting faults in an electric vehicle charging pile provided by an embodiment of the present application; Figure 6 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

[0023] Description of the reference numerals: 11, charging pile determination module; 12, data acquisition module; 13, evaluation value calculation module; 14, real-time score calculation module; 15, threshold calculation module; 501, processor; 502, communication bus; 503, user interface; 504, network interface; 505, memory. Detailed Embodiments

[0024] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

[0025] In the description of the embodiments of the present application, words such as "for example" or "for illustration" are used to represent examples, illustrations, or explanations. Any embodiment or design solution described as "for example" or "for illustration" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "for example" or "for illustration" is intended to present relevant concepts in a specific manner.

[0026] In the description of the embodiments of the present application, the term "plurality" means two or more. For example, a plurality of systems means two or more systems, and a plurality of screen terminals means two or more screen terminals. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. The terms "comprise", "include", "have" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0027] The present application provides a method for remotely detecting faults in electric vehicle charging piles. Refer to Figure 1 , Figure 1 which is a schematic flowchart of a method for remotely detecting faults in electric vehicle charging piles provided by an embodiment of the present application, including steps S101 to S105. The above steps are as follows: S101: Determine the set of associated charging piles in the target area. The set of associated charging piles belongs to the same power system as the current charging pile. In the embodiments of the present application, the target area may be a certain city, a certain industrial park, a certain parking lot, or the coverage area of a certain power system. In the target area, the current charging pile is the core object for fault detection, and all analyses and calculations are carried out around it. The set of associated charging piles shares the same power resources (such as the same transformer, distribution cabinet, or power line) with the current charging pile, so their operating states will affect each other. The same power system means that these charging piles are connected to the same transformer, distribution cabinet, or power line, so their operating states are affected by the same power system parameters.

[0028] In a feasible implementation, the system first accesses the power topology database or real-time monitoring platform of the power grid company, analyzes the distribution network structure of the target area (including the hierarchical relationship of transformers, buses, and feeders), and traces back the power system node path to which it belongs based on the physical access point of the current charging pile (such as the distribution transformer number or line code). All downstream devices directly or indirectly connected to this node are dynamically identified through a graph traversal algorithm (such as breadth-first search), and a charging pile cluster that shares the same power path and has a high degree of electrical parameter coupling is screened out. At the same time, secondary verification is carried out in combination with the real-time load data of the power system (such as phase synchronization and voltage fluctuation correlation) to exclude misassociations caused by lagging topology data. Finally, through a dynamic update mechanism, the topological changes triggered by events such as power system expansion and line switching are mapped to the set of associated charging piles in real time, ensuring that the cluster members always reflect the real power connection relationship, and then determining the set of associated charging piles in the target area.

[0029] Based on the above embodiments, as an alternative embodiment, S101: The step of determining the associated charging pile set within the target area, where the associated charging pile set and the current charging pile belong to the same power system, may further include the following steps. Please refer to Figure 2 : S1011: Obtain the power system topological structure within the target area, and determine the power system range where the current charging pile is located according to the power system topological structure; In the embodiments of the present application, the power system topological structure refers to the network layout formed by the physical or logical connection relationships of each component in the power grid. Its essence is an abstract description of the energy transmission path within the power system. In the charging pile fault detection, the role of topological structure analysis is to identify the group of devices that have a direct electrical association with the current charging pile. The power system range is a dynamic boundary delimited based on the topological structure, used to define the local power grid environment where the current charging pile is located. The power system range is not fixed but dynamically adjusted according to the real-time power grid operation status and detection target.

[0030] Specifically, by integrating the real-time topological data interface of the power grid enterprise energy management system, parse the physical connection relationship of the distribution network in the target area, and construct the power system topological structure based on the node and edge relationships stored in the graph database. Locate the power system node to which the current charging pile belongs in the graph through the metering point number or access distribution equipment ID of the current charging pile, use the reverse path tracing algorithm to determine its power supply dependence path, and verify the real-time effectiveness of the topological connection in combination with the real-time measurement data, and dynamically delimit the device boundary that has a direct electrical connection or strong coupling relationship with the current charging pile.

[0031] Furthermore, to determine the power system range where the current charging pile is located according to the power system topological structure, the connection relationship between the current charging pile and its associated charging pile set can be found through the process lines of the mutual operation of each circuit in the power system topological structure. For example, if the current charging pile is connected to the #3 outgoing line switch of a certain ring main unit, the system will trace upward along this switch to the upper-level transformer and the corresponding 10kV feeder, and traverse downward all the branch lines under the jurisdiction of this outgoing line. At the same time, by real-time monitoring the voltage phase synchronization (such as phase difference < 0.5 degrees) and load fluctuation correlation (such as Pearson coefficient > 0.85) of the charging pile group on this line, eliminate the temporary topological changes caused by the switching of the tie switch, and finally lock the charging pile set that is in the same power supply island or electrical loop as the current charging pile.

[0032] S1012: Determine the associated charging pile set that is in the same power system as the current charging pile according to the power system database within the power system range.

[0033] In the embodiment of the present application, the power system database is a core information library for storing and managing basic information and dynamic operation data of the power network. Its essence is to digitize key elements such as the physical structure, equipment properties, operating status and historical records of the power grid to form a data center that supports real-time analysis and decision-making.

[0034] Specifically, the physical connection path of the charging pile is located in the topology of the power system database through the asset code or access point identifier (such as the port number of the distribution cabinet or the serial number of the meter). All downstream device nodes covered by the path are extracted, and the set of charging pile devices that share the same power node with the current charging pile and are not blocked by isolation devices are traversed using a graph query language. Furthermore, the static topology relationship is dynamically corrected in combination with the real-time telemetry data of the data acquisition and monitoring system. For example, when a new power supply island is formed in a certain line due to the switching operation of the ring network cabinet, the similarity of the voltage waveform distortion characteristics of the charging piles in the island is detected (such as the harmonic spectrum correlation coefficient ≥ 0.95) or the spatiotemporal correlation of the load mutation events (such as the power change trend matching degree within 5 seconds > 90%), and the pseudo-associated nodes caused by network reconstruction are automatically eliminated, and finally a set of associated charging piles with a real-time strong electrical coupling relationship with the current charging pile is generated.

[0035] S102: Acquire first operating data of the current charging pile and second operating data corresponding to each associated charging pile in the associated charging pile set.

[0036] In this embodiment, the first operating data is the real-time operating parameters of the currently monitored charging pile itself. Its content and collection frequency are dynamically adjusted based on the charging pile's status. The second operating data focuses on the group operating parameters of a set of related charging piles within the same power system. Its collection mode is controlled by the current charging pile status, and horizontal comparison is used to distinguish between independent equipment failures and system-level disturbances.

[0037] Specifically, according to the predefined physical connection relationships in the power system topology database, obtain the list of device identifiers of the associated charging pile set, and then send synchronization acquisition instructions to all charging piles in the list through the power line carrier communication broadcast protocol; according to the clock signal output by the built-in GPS / Beidou dual-mode timing module of the charging pile, obtain a globally unified microsecond-level time reference, and trigger the three-axis current sensor, wide-dynamic-range voltage sampling circuit, and distributed temperature probe array in the integrated data acquisition unit of each pile to synchronously capture current waveform, voltage effective value, and temperature gradient data at a fixed sampling rate of 100 Hz; according to the pre-trained power grid feature extraction model of the on-chip AI coprocessor, obtain key feature vectors such as the kurtosis coefficient of the reduced-dimensional current waveform and the duration of voltage sag events. At the same time, the power quality analysis chip calculates the harmonic distortion rate, three-phase unbalance degree, and power factor offset value in real time based on the original voltage and current signals; according to the time determinacy constraint of the hard real-time Ethernet transmission protocol, obtain a multi-pile data set with aligned timing in the edge computing gateway, and then non-linearly compress the data through an improved compressive sensing algorithm to obtain a feature data packet with a compression ratio of 10:1; according to the blockchain-style data fingerprint generation rule, obtain a data verification chain with anti-tampering characteristics and transmit it to the central data lake through the power dedicated fiber optic ring network; according to the redundant sensor cross-check mechanism, obtain the data validity flag bit; according to the transient overvoltage trigger signal output by the lightning strike detection module, obtain the cache activation control instruction, instantly increase the sampling rate to 1 MHz and last for 100 ms to completely capture the surge current waveform and voltage oscillation process, ensuring that transient fault characteristics are not missed.

[0038] Based on the above embodiments, as an optional embodiment, step S102: obtaining the first operation data of the current charging pile and the second operation data corresponding to each associated charging pile in the associated charging pile set may further include the following steps. Please refer to Figure 3 : S1021: Determine the first acquisition mode of the first operation data and the second acquisition mode of the second operation data according to the usage status of the charging pile; Specifically, a state perception module is locally deployed at the charging pile. By real-time monitoring the charging gun connection status, charging power threshold, and abnormal alarm signals, the operating state machine is automatically switched. When the charging gun is inserted, the Hall sensor of the gun head mechanical locking mechanism detects a change in magnetic flux (the threshold is set to ±5mT), triggering a digital input port interrupt signal. The edge controller immediately reads the PWM duty cycle of the charging gun communication interface and simultaneously starts a power monitoring thread. The instantaneous values of the AC side voltage and current are captured in real-time through a 16-bit ADC at a sampling rate of 1kHz to calculate the instantaneous power. When it is detected that the power mean value exceeds 2kW (such as 2.3 ± 0.1kW) and the volatility < 5% within 5 consecutive sampling periods (i.e., 5 seconds), it is determined as an effective charging state. At this time, the state machine switches from the idle state (S0) to the charging state (S1), triggering the high-speed data acquisition unit to power on. During this process, if the insulation monitoring module detects a sudden drop in the PE line-to-ground impedance (such as from 10MΩ to 0.5MΩ for 100ms) by injecting a 1kHz / 5V square wave, it triggers the hardware comparator to output a low-level alarm signal. After being isolated by an optocoupler, this signal is connected to the external interrupt pin of the controller, forcing the state machine to jump to the abnormal state within 10μs, immediately starting fault recording (recording the current waveform of the 8 cycles before the fault at a sampling rate of 1MHz) and locking the charging relay. When the state machine switches to the idle state, the first acquisition mode is that the edge computing unit starts a low-frequency acquisition thread, periodically reads the effective value of the voltage, root mean square of the current, and temperature sensor data at a sampling rate of 1Hz, and uploads them to the cloud at minute-level intervals through narrowband Internet of Things. If it is detected that the charging gun is connected and the power module is activated, the first acquisition mode is the high-frequency acquisition mode. The high-speed ADC is enabled to synchronously capture the instantaneous waveforms of three-phase voltage / current at a sampling rate of 10kHz, and the harmonic spectrum characteristics are analyzed by calling FFT. At the same time, a synchronous acquisition instruction is broadcast to the associated charging pile set through the time-sensitive network, forcing the second acquisition mode of its edge devices to switch to the same high-frequency mode to ensure that data such as load fluctuations and power factor changes of the associated piles are strictly time-aligned with the waveform acquisition of the current charging pile. When the insulation monitoring or over-temperature protection unit triggers an abnormal signal, the system immediately enters the abnormal state. The first acquisition mode is full-scale data capture, and it triggers the associated charging pile set to switch to the second acquisition mode to synchronously upload the grid-side harmonic distortion rate, zero-sequence current, and breaker operation timing log, and preferentially ensures the real-time transmission bandwidth of key data in the abnormal state through the dynamic load balancing algorithm.

[0039] S1022: Obtain the first operation data of the current charging pile according to the first acquisition mode, and obtain the second operation data corresponding to each associated charging pile in the associated charging pile set according to the second acquisition mode.

[0040] Specifically, in the embodiments of the present application, when the charging pile is in an idle state, its edge controller activates the low-frequency acquisition thread to read the effective values of voltage, current, and ambient temperature sensor data at 1-second intervals, and uploads them to the cloud at a minute-level cycle through the MQTT protocol. At the same time, a low-frequency acquisition instruction is sent to the associated charging pile set to trigger each associated pile to upload power system-level parameters at the same cycle. When the charging pile enters the charging state, the edge device immediately starts the high-speed data acquisition card to synchronously capture the instantaneous values of three-phase current / voltage waveforms at a sampling rate of 10 kHz, and broadcasts an accurate clock synchronization signal to the associated pile set through the time-sensitive network, forcing the associated piles to start the same high-frequency acquisition mode within a 50-μs time window and upload load balancing data in real time (such as dynamic power distribution ratio, current harmonic coupling coefficient between adjacent piles). If the charging pile triggers an over-temperature or insulation anomaly alarm, the system immediately switches to the full-scale acquisition mode, enables the transient recording module with a sampling rate of 1 MHz to record the current mutation waveform from 100 ms before the fault to 500 ms after the fault. At the same time, an emergency acquisition instruction is sent to the associated pile set through the fiber optic ring network, requiring it to upload the power grid harmonic distortion rate spectrum matrix, zero-sequence current trajectory, and breaker operation event sequence within 10 ms, and use the streaming computing engine to perform sliding window correlation analysis on the abnormal coefficient of the grounding resistance of the associated piles (such as calculating the standard deviation mutation rate of the resistance values of the associated set members within 3 seconds before and after the fault moment of the current pile), and prioritize the real-time and integrity of the fault feature data by sacrificing the non-critical data bandwidth (such as pausing the temperature data upload), and finally construct a spatio-temporal association data set of the refined waveform of the current pile and the system-level parameters of the associated piles.

[0041] S103: Calculate the fault evaluation values corresponding to the respective fault evaluation indexes of the current charging pile according to the first operation data and the respective second operation data.

[0042] In the embodiments of the present application, the fault evaluation index corresponds to a class of observable physical quantities or derived parameters. Its essence is to capture a specific pattern of abnormal behavior of the device through data modeling, and the fault evaluation value is a numerical expression of the degree of abnormality in this pattern. Its calculation process needs to combine the current device data with the collaborative analysis of the associated device set to eliminate external environmental interference and highlight the device's own fault signal.

[0043] The calculation of the voltage fluctuation amplitude depends on a high-precision voltage transformer installed at the AC input end to capture the voltage data of the current charging pile in real time at a sampling rate of 1000 times per second. The original voltage waveform undergoes a fast Fourier transform through a 200-millisecond sliding window. After extracting the amplitude sequence of the fundamental wave component, the standard deviation within the window is calculated to obtain the voltage fluctuation amplitude. The calculation of the current harmonic distortion rate depends on the wide-band current sensor (response range of 0 - 10 kHz) on the DC side to obtain the current waveform data sampled 5000 times per second. Harmonic analysis is performed on each 1-second time period to obtain the total harmonic distortion rate. After the harmonic data of the associated pile set is synchronously collected by the power quality monitor, the deviation degree between the current total harmonic distortion rate and the mean value of the associated set is calculated through Z-score standardization. The quantification of the temperature rise rate is achieved by using a PT100 platinum resistance temperature sensor to collect the surface temperature of the power module radiator at a frequency of 1 time per second. A linear regression fit is performed on the temperature sequence for 30 consecutive seconds to obtain the slope as the real-time temperature rise rate. At the same time, temperature rise samples under similar working conditions are matched and screened from the historical database of the associated pile according to the ambient temperature (±5°C) and load current (±10%). A probability distribution model is constructed using kernel density estimation, and the 99% quantile is extracted as the normal upper limit. When the current slope exceeds this upper limit, an evaluation value is generated according to the exceeding ratio. The calculation of the power factor deviation is obtained by using a digital electric meter to obtain the active power and reactive power in real time. According to the current load rate (accurate to the 5% interval), the power factor data of the same load section is extracted from the associated pile set. After calculating the mean value and standard deviation, the deviation multiple of the current value is evaluated.

[0044] Based on the above embodiments, as an alternative embodiment, in step S103: calculating the fault evaluation values corresponding to the respective fault evaluation indicators of the current charging pile according to the first operation data and the respective second operation data, this step may further include the following steps: S1031: Calculate the fault evaluation values corresponding to the respective fault evaluation indicators according to the first formula; The first formula is: Where S i is the fault evaluation value of each of the fault evaluation indicators; x i is the detected value of the i-th fault evaluation indicator of the current charging pile; μ i is the mean value of the i-th fault evaluation indicator under the historical normal state; σ i is the standard deviation of the i-th fault evaluation indicator under the historical normal state; β is the environmental coupling coefficient; ΔG i is the offset amplitude between the current harmonic distortion rate of the associated set and the historical baseline.

[0045] Specifically, the system first extracts features from the first operating data of the current charging pile (such as the current harmonic spectrum sampled at 10 kHz during charging), constructs a multi-dimensional feature matrix by combining the second operating data of the associated charging pile set (such as the average value of the grid harmonic distortion rate collected synchronously), and uses the dynamic baseline calibration technology to perform Z-score normalization calculation on the real-time index value of the current charging pile (such as the current harmonic distortion rate of 7.2%) and its rolling window average value under the historical normal state (such as the average value of 4.1% ± 0.8% under the same working conditions in the past 30 days). At the same time, the population distribution parameter of the associated charging pile set (such as the standard deviation of the harmonic distortion rate of other charging piles under the same feeder of 1.5%) is introduced as an environmental noise compensation factor, and the evaluation formula is dynamically corrected through the adaptive weight algorithm. The fault evaluation value calculation formula of the current charging pile is adjusted to: where S i is the fault evaluation value of each of the fault evaluation indicators; x i is the detected value of the i-th fault evaluation indicator of the current charging pile; μ i is the average value of the i-th fault evaluation indicator under the historical normal state; σ i is the standard deviation of the i-th fault evaluation indicator under the historical normal state; β is the environmental coupling coefficient; ΔG iTo associate the current harmonic distortion rate of the set with the offset amplitude from the historical baseline, so as to strip the influence of system-level interference in the evaluation; the calculation of the environmental coupling coefficient is a data-driven process based on historical fault data and grid operation characteristics, balancing the risks of false alarms and missed alarms: First, extract two types of event data from the historical database. One type is the hardware faults of the charging piles themselves, and the other type is the anomalies caused by grid environment disturbances. For each event, record the detected value of the fault evaluation index of the current charging pile at the time of its occurrence and the population offset of the associated charging pile set. Subsequently, construct an objective function, with the weighted sum of the false alarm rate and the missed alarm rate as the optimization objective. For example, in an industrial park where grid disturbances occur frequently, set the false alarm cost weight to 0.7, and traverse the possible values of β (from 0.1 to 0.5) through grid search, and select the β value that minimizes the total cost. For example, in an industrial park, there is a group of charging piles powered by the same transformer. Among them, charging pile A suddenly triggers an abnormal alarm for the current harmonic distortion rate. The real-time detected current harmonic distortion rate of charging pile A is x1 = 7.2%, the historical normal mean is μ1 = 4.1%, and the standard deviation is σ1 = 0.8%; the offset amplitude ΔG1 of the current harmonic distortion rate of the associated charging pile set (10 charging piles under the same transformer) from the historical baseline is 2.083, and the environmental coupling coefficient β = 0.3. The fault evaluation value calculated by the traditional Z-score is 3.875, far exceeding the preset threshold of 3.0. It is initially determined to be a device fault. Substitute it into the correction formula, and the denominator is expanded to 0.8×(1 + 0.3×2.083) = 1.3. The final fault evaluation value drops to 2.385. This result indicates that the anomaly is more likely to be caused by grid-side interference rather than problems with the charging pile itself, thus solving the risk of misjudgment.

[0046] S104: Calculate the real-time score corresponding to the current charging pile according to each of the fault evaluation values and different weights corresponding to each of the fault evaluation indicators.

[0047] In the embodiment of the present application, the weight represents the influence degree of different fault evaluation indicators on the final fault judgment. Its essence is to dynamically adjust their contributions in the comprehensive score by quantifying the historical relevance between each indicator and the occurrence of the fault. The real-time score is a comprehensive quantitative value obtained by weighted fusion of all indicator evaluation values. Its core role is to compress multi-dimensional device status information into a single value, intuitively reflecting the current fault risk level.

[0048] Specifically, the calculation of real-time scoring is achieved through a multi-stage data fusion and dynamic adjustment mechanism. The core lies in combining the normalized values of different fault evaluation indicators with dynamic weights, and enhancing the sensitivity and accuracy of fault recognition through a non-linear enhancement and threshold self-learning strategy. First, based on thousands of cases in the historical fault database, the system uses the improved entropy weight method to determine the initial weights of each indicator. These weights are not fixed and are updated monthly through an incremental learning algorithm. For example, when the correlation between the harmonic distortion rate and faults of a certain type of charging pile increases due to design changes, its weight can be gradually adjusted from 0.2 to 0.35 to ensure that the model continuously adapts to equipment iteration. In the real-time calculation process, the system generates a score every 5 seconds. After the evaluation values of each indicator are normalized and mapped to the [0, 1] interval, if the value of a certain indicator breaks through the preset abnormal threshold (such as the normalized value of the temperature rise rate exceeding 0.7), the weight dynamic enhancement mechanism is triggered: the weight of this indicator is increased according to a piecewise function. For example, when the evaluation value of the temperature rise rate reaches 0.9, its weight is temporarily increased from 0.67 to 1.02, and the weights of other indicators are compressed proportionally to ensure that the sum is 1. This mechanism enables the system to preferentially amplify the signals strongly related to serious faults when multiple indicators are abnormal at the same time. The normalized evaluation values of each indicator (mapped to the 0-1 interval) are linearly weighted and summed with their corresponding weights to generate the real-time score.

[0049] Based on the above embodiments, as an alternative embodiment, in step S104: calculating the real-time score corresponding to the current charging pile according to each of the fault evaluation values and different weights corresponding to each of the fault evaluation indicators, the following steps may further be included: S1041: calculating the real-time score corresponding to the current charging pile according to each of the fault evaluation values and different weights corresponding to each of the fault evaluation indicators, includes: dynamically adjusting the weight coefficient corresponding to each of the fault evaluation indicators according to the second formula and based on the population distribution characteristics of each of the fault evaluation indicators; The second formula is: wherein, the population distribution characteristics include x i , μG i , σG i ; ω i is the weight coefficient of the i-th fault evaluation indicator; ω i0 is the baseline weight of the i-th fault evaluation indicator; x i is the deviation degree of the standardized fault evaluation indicator; μG i is the population mean of the i-th fault evaluation indicator of the associated charging pile set; σG iis the population dispersion degree of the i-th fault evaluation index of the associated charging pile set; v is the dynamic sensitivity coefficient; Perform a linear weighted sum of the adjusted weight coefficients and the corresponding fault evaluation values to generate the real-time score.

[0050] Specifically, in the embodiment of the present application, in the real-time scoring engine, the system initializes the scoring model based on a predefined index baseline weight matrix (such as the weight of current harmonic distortion rate is 0.3 and the weight of abnormal grounding resistance coefficient is 0.2). The real-time scoring engine receives the fault evaluation values of each index of the current charging pile and the population distribution data of the associated charging pile set (such as the mean and standard deviation of the harmonic distortion rate of other charging piles under the same feeder) in real time. Using the dynamic weight adjustment algorithm, when it is detected that a certain index shows a group shift in the associated set (such as the group mean of three-phase voltage unbalance rises by 15% compared to the historical baseline), according to the formula: Reduce the weight of this index, where ΔG i is the population shift amount, γ is the attenuation factor, v i is the weight coefficient corresponding to the fault evaluation index, v i0 is the baseline weight coefficient corresponding to the fault evaluation index, σG i is the population dispersion degree of the i-th fault evaluation index of the associated charging pile set, and for the index of the current charging pile that significantly deviates from the group characteristics (such as the deviation value of power factor exceeds 3 times the standard deviation of the associated set), according to the formula: Amplify the weight. Finally, generate the real-time score through the normalized weighted sum within a sliding time window window (such as a 30-second window) Generate the real-time score, where Score t is the real-time score, S i is the normalized fault evaluation value, v i is the weight coefficient corresponding to the fault evaluation index. At the same time, introduce the associated set score baseline (such as the 90th percentile of the scores of healthy charging piles in the same topological environment) for secondary calibration to ensure that the scoring result can not only reflect the independent abnormality of the current charging pile, but also couple the influence of system-level environmental fluctuations.

[0051] S105: Calculate the fault scoring threshold of the faulty charging pile according to the historical operation data.

[0052] In the embodiment of the present application, the historical operation data is a set of full-life cycle state records accumulated during the long-term operation of the charging pile. Its essence is an empirical knowledge base formed by continuously observing the normal and abnormal working modes of the equipment. The fault scoring threshold is a dynamic discrimination boundary obtained through the reverse analysis of historical fault events by a machine learning model, and its formation process is similar to extracting a map of fault characteristics from a large amount of historical data.

[0053] Specifically, in the embodiments of the present application, according to the timestamps of historical fault events and the multi-dimensional sensing data stream, the time-varying curve of the current harmonic distortion rate, the three-phase voltage phase angle difference fluctuation matrix, and the grounding resistance temperature coupling parameters in a specific period before the fault are intercepted through a sliding window to obtain the spatio-temporal correlation tensor of the fault precursor characteristics; the current waveform is decomposed at multiple scales according to the wavelet packet energy entropy algorithm, the fundamental frequency band energy attenuation rate and the high-frequency band clustering oscillation intensity are extracted, and the grounding resistance sliding coefficient of variation sequence is converted into a two-dimensional texture map in combination with the Gram angle field to obtain a 128-dimensional hybrid feature vector integrating the physical characteristics in the time-frequency domain and the image pattern features; according to the hidden space compression characteristics of the variational autoencoder, the high-dimensional features are projected onto the 8-dimensional latent space principal components, and at the same time, the differences in different charging pile hardware batches are eliminated through the gradient reversal mechanism in domain adversarial training to obtain a fault mode fingerprint code with cross-device generalization; the heat conduction tensor of the anisotropic diffusion equation is modulated according to the service life of the device and the environmental corrosion index, so that the fault probability cloud in the three-dimensional feature space dynamically extends along the device aging trajectory, and when the real-time score crosses the critical manifold surface in the probability cloud density gradient field, a fault score threshold that adapts to the device life cycle and drifts is obtained.

[0054] Based on the above embodiments, as an alternative embodiment, step S105: calculating the fault score threshold of the faulty charging pile according to the historical operation data may further include the following steps. Please refer to Figure 4 : S1051: Calculate the historical distribution of the fault scores of the faulty charging pile according to the historical operation data; In the embodiments of the present application, the historical distribution of the fault scores is a spatio-temporal probability cloud map formed by quantifying the health status during the entire life cycle of the charging pile, and is composed of the score trajectories of different fault modes in a large amount of historical data.

[0055] Specifically, in the embodiments of the present application, the system screens out all the fault event records confirmed by operation and maintenance from the historical database of the target charging pile set, extracts the real-time scoring sequence in a specific time window before the fault occurs (such as 10 minutes before the fault trigger to the moment of the fault), eliminates the time series misalignment caused by data acquisition delay through the time alignment algorithm, and constructs a probability distribution model of the fault score using Gaussian kernel density estimation; at the same time, the scoring data of healthy charging piles in the same time period is collected as the baseline distribution, and the adversarial generative network is used to simulate the feature boundary between faults and healthy scores, and the aggregation interval of the fault scores is identified (such as the density of fault samples in the score range of 70-90 is 8 times higher than that of healthy samples); considering the dynamic evolution characteristics of the power system, a time decay weight function is introduced to perform weighted smoothing on historical fault events according to the occurrence time, so that the influence weight of recent fault data on the distribution form is increased (such as fault events within three months account for 60% of the weight), and at the same time, the group score drift trend of the associated charging pile set is integrated (such as the year-on-year increase in the scores of charging piles under the same transformer is 12%), and the historical distribution curve adapted to the current power grid environment is reconstructed through the distribution translation compensation algorithm, and finally a fault score historical distribution that can reflect long-term fault laws and adapt to system parameter changes is generated, providing an evolution-aware data basis for threshold setting.

[0056] S1052: Determine the fault score threshold according to the percentile, mean, and standard deviation of the historical distribution of the fault scores of the faulty charging piles.

[0057] The fault score threshold includes multiple levels of thresholds, and the multiple levels of thresholds include: A warning threshold for indicating potential fault risks; An emergency threshold for indicating serious fault risks.

[0058] Specifically, in the embodiments of the present application, according to the full - life - cycle scoring sequence of faulty charging piles stored in the historical database (including continuous data from N hours before the fault occurs to the fault moment), the characteristic evolution trajectory of each fault event is intercepted through a sliding time window to obtain a spatio - temporal matrix including the scoring growth rate, the fluctuation amplitude, and the covariant relationship of correlation parameters; according to the kernel density estimation method, non - parametric fitting is performed on the distribution of a large number of fault trajectories in the three - dimensional feature space to obtain the probability density cloud map of the historical distribution of the fault score, where the topological structure of the cloud map is reduced to a two - dimensional interpretable plane through manifold learning; according to the preset operation and maintenance response level requirements, the contour line is cut along the gradient direction on the probability density cloud map to obtain the initial percentile boundary. For example, the 95th percentile corresponding to the density dropping to 30% of the peak is taken as the candidate value of the early warning threshold, and the 99.7th percentile corresponding to the density dropping to 5% is taken as the emergency threshold baseline; the percentile is dynamically corrected according to the real - time operating environment parameters of the device (such as the acceleration factor of environmental temperature on insulation aging), and by establishing a physical - driving model of temperature - threshold offset, the dynamically corrected percentile boundary after temperature compensation is obtained; according to the mean and standard deviation of the scoring distribution of the historical normal charging - pile population, an outlier scale in the Mahalanobis distance space is constructed, and the percentile boundary is mapped to this space to obtain a relativized threshold that eliminates the baseline fluctuation of the population; the threshold sensitivity is modulated according to the real - time load level of the power grid. When the feeder load rate exceeds 80%, the early warning threshold is lowered by 15% - 20% through a load - threshold coupling equation (including an exponential decay term) to obtain an elastic hierarchical threshold adapted to the system pressure.

[0059] Referring to Figure 5 , the present application also provides a remote detection device 10 for electric - vehicle charging - pile faults, specifically including: A charging - pile determination module 11, configured to determine an associated charging - pile set in the target area that belongs to the same power system as the current charging pile; a data acquisition module 12, configured to acquire the first operation data of the current charging pile and the second operation data corresponding to each associated charging pile in the associated charging - pile set; An evaluation - value calculation module 13, configured to calculate the fault evaluation values corresponding to the various fault evaluation indexes of the current charging pile according to the first operation data and the second operation data; A real - time scoring calculation module 14, configured to calculate the real - time score corresponding to the current charging pile according to the various fault evaluation values and the different weights corresponding to the various fault evaluation indexes; A threshold calculation module 15, configured to acquire the historical operation data of the target charging - pile set, and calculate the fault - scoring threshold of the faulty charging piles according to the historical operation data, where the target charging - pile set has the same type as the associated charging - pile set, the faulty charging piles and the target charging - pile set belong to the same power system, and the faulty charging piles are the charging piles that have failed; Optionally, the charging pile determination module 11 is specifically configured to: Obtain the topology structure of the power system within the target area, and determine the power system range where the current charging pile is located according to the topology structure of the power system; Determine the set of associated charging piles in the same power system as the current charging pile according to the power system database within the power system range.

[0060] Optionally, the data acquisition module 12 is specifically configured to: Determine the first acquisition mode of the first operation data and the second acquisition mode of the second operation data according to the usage status of the charging pile; Obtain the first operation data of the current charging pile according to the first acquisition mode, and obtain the second operation data corresponding to each associated charging pile in the set of associated charging piles according to the second acquisition mode.

[0061] Optionally, the data acquisition module 12 is further specifically configured to: When the usage status is the idle state, determine that both the first acquisition mode and the second acquisition mode are the low-frequency acquisition mode, and use the low-frequency acquisition mode to acquire the first operation data, where the first operation data is the voltage, current, and ambient temperature of the current charging pile; use the low-frequency acquisition mode to acquire the second operation data of each associated charging pile in the set of associated charging piles, where the second operation data is the power system-level parameter; When the usage status is the charging state, determine that both the first acquisition mode and the second acquisition mode are the high-frequency acquisition mode, and use the high-frequency acquisition mode to acquire the first operation data, where the first operation data is the multi-dimensional acquisition current waveform, voltage waveform, temperature gradient, and power factor of the current charging pile; use the high-frequency acquisition mode to acquire the second operation data of each associated charging pile in the set of associated charging piles, where the second operation data is the load balancing data; When the usage status is the abnormal state, determine that both the first acquisition mode and the second acquisition mode are the high-frequency acquisition mode, and use the high-frequency acquisition mode to acquire the first operation data, where the first operation data is the full amount of data of the current charging pile; use the high-frequency acquisition mode to acquire the second operation data of each associated charging pile in the set of associated charging piles, where the second operation data is the grid harmonic distortion rate and the grounding resistance.

[0062] Optionally, the fault evaluation value calculation module 13 is specifically configured to: Calculate the fault evaluation value corresponding to each fault evaluation index according to the first formula; The first formula is: Among them, the fault evaluation indicators include i items, namely the current harmonic distortion rate, three-phase voltage unbalance degree, power factor deviation value, abnormal coefficient of grounding resistance, and charging efficiency decay rate, S i is the fault evaluation value of each of the fault evaluation indicators; x i is the detected value of the i-th fault evaluation indicator of the current charging pile; μ i is the mean value of the i-th fault evaluation indicator under the historical normal state; σ i is the standard deviation of the i-th fault evaluation indicator under the historical normal state; β is the environmental coupling coefficient; ΔG i is the offset amplitude between the current harmonic distortion rate of the association set and the historical baseline.

[0063] Optionally, the real-time scoring calculation module 14 is specifically configured to: Dynamically adjust the weight coefficient corresponding to each of the fault evaluation indicators according to the second formula and based on the population distribution characteristics of each of the fault evaluation indicators; The second formula is: Among them, the population distribution characteristics include x i , μG i , σG i ; ω i is the weight coefficient of the i-th fault evaluation indicator; ω i0 is the baseline weight of the i-th fault evaluation indicator; x i is the deviation degree of the standardized fault evaluation indicator; μG i is the population mean of the i-th fault evaluation indicator of the associated charging pile set; σG i is the population dispersion of the i-th fault evaluation indicator of the associated charging pile set; α is the dynamic sensitivity coefficient; Perform a linear weighted sum of the adjusted weight coefficient and the corresponding fault evaluation value to generate the real-time score.

[0064] Optionally, the fault scoring threshold calculation module 15 is specifically configured to: Calculate the historical distribution of the fault scores of the faulty charging piles according to the historical operation data; Determine the fault scoring threshold according to the percentile, mean value, and standard deviation of the historical distribution of the fault scores of the faulty charging piles. The fault scoring threshold includes a warning threshold and an emergency threshold. The warning threshold is used to indicate potential fault risks, and the emergency threshold is used to indicate serious fault risks.

[0065] It should be noted that: when the device provided in the above embodiment realizes its functions, only the division of the above function modules is used for illustration. In actual applications, the above functions can be allocated to different function modules according to needs, that is, the internal structure of the device is divided into different function modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiments belong to the same concept. For the specific implementation process, please refer to the method embodiments and will not be elaborated here.

[0066] This embodiment also discloses an electronic device 500. Refer to Figure 6 , Figure 6 which is a schematic structural diagram of an electronic device disclosed in an embodiment of the present application. The electronic device 500 may include: at least one processor 501, at least one communication bus 502, a user interface 503, a network interface 504, and at least one memory 505.

[0067] Among them, the communication bus 502 is used to realize the connection and communication between these components.

[0068] Among them, the user interface 503 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 503 may further include a standard wired interface and a wireless interface.

[0069] Among them, the network interface 504 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0070] Among them, the processor 501 may include one or more processing cores. The processor 501 connects various parts within the entire server through various interfaces and lines. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 505, and by calling the data stored in the memory 505, it performs various functions of the server and processes data. Optionally, the processor 501 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor 501 may integrate a combination of one or several of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, user interface, application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 501 and may be implemented separately by a single chip.

[0071] Among them, the memory 505 may include random access memory (RAM) and may also include read-only memory. Optionally, the memory 505 includes a non-transitory computer-readable storage medium. The memory 505 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 505 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store the data involved in the above-mentioned various method embodiments. Optionally, the memory 505 may also be at least one storage device located far from the aforementioned processor 501. Refer to Figure 6 , the memory 505, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for detecting charging pile faults.

[0072] In Figure 6In the electronic device shown, the user interface 503 is mainly used to provide an input interface for the user and obtain the data input by the user; while the processor 501 can be used to call an application program for detecting charging pile faults stored in the memory 505. When executed by one or more processors 501, the electronic device 500 is caused to execute the method of one or more of the above embodiments.

[0073] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be in other sequences or performed simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0074] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0075] In several embodiments provided by this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the displayed or discussed coupling or direct coupling or communication connection between each other can be through some service interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical or other form.

[0076] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0077] In addition, in each embodiment of this application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0078] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of this application. The aforementioned memory includes: various media such as USB flash drives, mobile hard disks, magnetic disks, or optical discs that can store program codes.

[0079] The foregoing are only exemplary embodiments of the present disclosure and should not be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made in accordance with the teachings of the present disclosure still fall within the scope covered by the present disclosure. Those skilled in the art will readily think of other implementation manners of the present disclosure after considering the disclosure of the specification. This application aims to cover any variations, uses, or adaptive changes of the present disclosure, and these variations, uses, or adaptive changes follow the general principles of the present disclosure and include the common general knowledge or conventional technical means in the technical field not described in the present disclosure. The specification and the embodiments are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for remotely detecting faults in an electric vehicle charging pile, characterized in that, Applied to an electronic device, the method includes: determining an associated charging pile set within a target area, where the associated charging pile set and the current charging pile belong to the same power system; Obtaining first operation data of the current charging pile and second operation data corresponding to each associated charging pile in the associated charging pile set; Calculating fault evaluation values corresponding to respective fault evaluation indicators of the current charging pile according to the first operation data and the second operation data; Calculating a real-time score corresponding to the current charging pile according to the respective fault evaluation values and different weights corresponding to the respective fault evaluation indicators; Obtaining historical operation data of a target charging pile set, and calculating a fault score threshold of a faulty charging pile according to the historical operation data, where the target charging pile set is of the same type as the associated charging pile set, the faulty charging pile and the target charging pile set belong to the same power system, and the faulty charging pile is a charging pile that has failed; Comparing the real-time score with the fault score threshold to determine whether the current charging pile has failed.

2. The method according to claim 1, wherein The determining the associated charging pile set within the target area includes: obtaining a power system topology structure within the target area, and determining a power system range where the current charging pile is located according to the power system topology structure; Determining the associated charging pile set that is in the same power system as the current charging pile according to a power system database within the power system range.

3. The method according to claim 1, wherein The obtaining the first operation data of the current charging pile and the second operation data corresponding to each associated charging pile in the associated charging pile set includes: Determining a first acquisition mode of the first operation data and a second acquisition mode of the second operation data according to the usage state of the charging pile; Obtaining the first operation data of the current charging pile according to the first acquisition mode, and obtaining the second operation data corresponding to each associated charging pile in the associated charging pile set according to the second acquisition mode.

4. The method according to claim 3, wherein The obtaining the first operation data of the current charging pile according to the first acquisition mode, and obtaining the second operation data corresponding to each associated charging pile in the associated charging pile set according to the second acquisition mode includes: When the usage state is an idle state, determining that both the first acquisition mode and the second acquisition mode are low-frequency acquisition modes, and acquiring the first operation data using the low-frequency acquisition mode, where the first operation data is the voltage, current, and ambient temperature of the current charging pile; acquiring the second operation data of each associated charging pile in the associated charging pile set using the low-frequency acquisition mode, where the second operation data is power system-level parameters; When the usage state is a charging state, determining that both the first acquisition mode and the second acquisition mode are high-frequency acquisition modes, and acquiring the first operation data using the high-frequency acquisition mode, where the first operation data is the multi-dimensional acquisition current waveform, voltage waveform, temperature gradient, and power factor of the current charging pile; acquiring the second operation data of each associated charging pile in the associated charging pile set using the high-frequency acquisition mode, where the second operation data is load balancing data; When the usage status is an abnormal status, it is determined that both the first acquisition mode and the second acquisition mode are high-frequency acquisition modes, and the first operation data is acquired using the high-frequency acquisition mode. The first operation data is the full amount of data of the current charging pile; the second operation data of each associated charging pile in the associated charging pile set is acquired using the high-frequency acquisition mode. The second operation data is the grid harmonic distortion rate and the grounding resistance.

5. The method according to claim 1, wherein Calculating the fault evaluation values corresponding to the respective fault evaluation indicators of the current charging pile according to the first operation data and the respective second operation data includes: Calculating the fault evaluation values corresponding to the respective fault evaluation indicators according to the first formula; The first formula is as follows: Among them, the fault evaluation indicators include i items, namely the current harmonic distortion rate, three-phase voltage unbalance degree, power factor deviation value, abnormal coefficient of grounding resistance, and charging efficiency attenuation rate, S i is the fault evaluation value of each of the fault evaluation indicators; x i is the detection value of the i-th fault evaluation indicator of the current charging pile; μ i is the mean value of the i-th fault evaluation indicator under the historical normal state; σ i is the standard deviation of the i-th fault evaluation indicator under the historical normal state; β is the environmental coupling coefficient; ΔG i is the offset amplitude between the current harmonic distortion rate of the association set and the historical baseline.

6. The method according to claim 1, characterized in that, Calculating the real-time score corresponding to the current charging pile according to the respective fault evaluation values and the different weights corresponding to the respective fault evaluation indicators includes: Dynamically adjusting the weight coefficient corresponding to each fault evaluation indicator according to the second formula and based on the population distribution characteristics of the respective fault evaluation indicators; The second formula is: Among them, the group distribution characteristics include x i , μG i , σG i ; ω i is the weight coefficient of the i-th fault evaluation index; ω i0 is the baseline weight of the i-th fault evaluation index; x i is the deviation degree of the standardized fault evaluation index; μG i is the group mean of the i-th fault evaluation index of the associated charging pile set; σG i is the group dispersion of the i-th fault evaluation index of the associated charging pile set; α is the dynamic sensitivity coefficient; Performing a linear weighted sum on the adjusted weight coefficient and the corresponding fault evaluation value to generate the real-time score.

7. The method according to claim 1, characterized in that Calculating the fault score threshold of the faulty charging pile according to the historical operation data includes: Calculating the historical distribution of the fault scores of the faulty charging pile according to the historical operation data; Determining the fault score threshold according to the percentile, mean, and standard deviation of the historical distribution of the fault scores of the faulty charging pile. The fault score threshold includes a warning threshold and an emergency threshold. The warning threshold is used to indicate potential fault risks, and the emergency threshold is used to indicate serious fault risks.

8. A charging pile fault detection device, characterized in that, Applied to an electronic device, the apparatus includes: A charging pile determination module, configured to determine an associated charging pile set within a target area. The associated charging pile set and the current charging pile belong to the same power system; A data acquisition module, configured to acquire the first operation data of the current charging pile and the second operation data corresponding to each associated charging pile in the associated charging pile set; An evaluation value calculation module, configured to calculate the fault evaluation values corresponding to the respective fault evaluation indicators of the current charging pile according to the first operation data and the respective second operation data; A real-time score calculation module, configured to calculate the real-time score corresponding to the current charging pile according to the respective fault evaluation values and the different weights corresponding to the respective fault evaluation indicators; A threshold calculation module, configured to acquire the historical operation data of a target charging pile set, and calculate the fault score threshold of the faulty charging pile according to the historical operation data. The target charging pile set has the same type as the associated charging pile set. The faulty charging pile and the target charging pile set belong to the same power system. The faulty charging pile is a charging pile that has failed.

9. An electronic device, characterized in that, Including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions. Both the user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory so that the electronic device executes the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method according to any one of claims 1-7.

Images