A safety evaluation method and a portable detection system for an electric vehicle alternating current charging pile

Abstract

The application discloses a kind of electric vehicle AC charging pile safety evaluation method and portable detection system.Four voltage sampling points are set under the mouth of remote distribution box switch, the mouth of pile side distribution box switch and gun head, voltage and current are collected synchronously, and the impedance of external cable, the voltage drop of pile side switch and the impedance of charging pile are calculated in sections.In the state of power-off, L-PE / N-PE / L-N insulation resistance is detected preferentially, the risk of zero fire line breakdown discharge is evaluated, and if unqualified, the power-on test is terminated.After passing, the pile is controlled to run continuously with rated load, the temperature rise of remote switch, pile side switch and gun head under full load is directly monitored to evaluate thermal risk, and the protection margin is tested through power boundary scanning.The system uses wireless long-distance communication to realize clock synchronization between distribution end and pile end, all evaluation dimensions are integrated into weighted comprehensive score, combined with time dimension degradation trend and horizontal group comparison, and A / B / C / D four-level safety evaluation grade is output.It is suitable for periodic safety evaluation of in-service AC charging pile.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle charging facility safety testing technology, and in particular to a safety evaluation method and portable testing system for electric vehicle AC charging piles. Background Technology

[0002] With the rapid growth of electric vehicle ownership, AC charging piles, as dedicated power supply equipment for Mode 3 connection, are directly related to user safety and power distribution network security. Current national standards GB / T18487.1-2023, GB 39752-2024, and metrological verification procedure JJG 1148-2022 mainly specify the electrical performance, metrological performance, and basic protection function requirements of AC charging piles, but they have shortcomings in the following aspects: First, current testing technologies do not incorporate the front-end power distribution network into the charging pile safety evaluation system. AC charging piles are connected to the power grid through multiple levels of power distribution lines and switching devices. From the remote distribution box to the pile-side distribution box, then to the charging pile itself, and finally to the charging gun output, the impedance and temperature rise of each section of the line directly affect the heat generation and fire risk during the charging process. However, current standards only focus on the charging pile itself and lack online measurement methods for the segmented impedance of the power distribution network. This makes it impossible to distinguish between "external cable heating," "pile-side switch heating," and "charging pile heating," making it difficult to accurately locate the potentially hazardous sections.

[0003] Second, existing technologies do not provide direct safety assessments of critical power devices (such as circuit breakers and contactors) and charging gun heads in power distribution circuits based on measured full-load temperature rises. Under long-term current flow, surface oxidation, and electrochemical reactions between different metal materials, the contact resistance of these devices gradually increases, leading to excessively high localized temperature rises. Current detection methods largely rely on theoretical estimations or indirect resistance measurements, failing to directly monitor the temperature rise curves of the switch body and charging gun heads under actual rated load continuous operation, thus failing to reflect the thermal risks under real heat dissipation conditions.

[0004] Third, existing technologies lack a systematic evaluation of the insulation status of power distribution networks and charging piles. Deterioration of insulation resistance in power distribution lines and connecting devices is a direct cause of leakage current, arcing, and fires caused by sparking between live and neutral wires. In particular, insulation degradation between live and neutral (LN) wires can lead to breakdown discharge under long-term power frequency voltage, forming a continuous arc and igniting surrounding combustibles. Current testing methods do not consider insulation resistance as an independent evaluation dimension, nor have they established a quantitative correlation between "insulation resistance value and breakdown discharge risk," and they do not make insulation testing a mandatory prerequisite for power-on testing.

[0005] Fourth, existing protection function detection only determines "presence" or "operation," without evaluating the "margin" of protection operation. For example, the ratio of overload protection operating current to rated current directly reflects the protection's sensitivity and reliability reserve. Too small a margin easily leads to false tripping, while too large a margin renders the protection meaningless; existing technologies do not provide a quantitative evaluation of this.

[0006] Fifth, current testing methods rely on single-test pass / fail assessments, lacking time-based analysis. The safety status of charging piles, power distribution lines, switching devices, and charging gun heads deteriorates dynamically; indicators such as decreased insulation resistance and increased line impedance are cumulative. The lack of degradation rate calculations based on historical data sequences and remaining safe life predictions makes it impossible to provide early warnings of potential risks.

[0007] Sixth, current testing methods lack cross-sectional group comparisons. Whether a single device's indicators are abnormal requires statistical comparison with similar devices of the same model, batch, and region to accurately identify the problem. Using a single threshold for judgment is prone to misjudgment or missed detection due to individual or environmental differences.

[0008] Seventh, existing testing devices are mostly fixed online monitoring equipment or large-scale laboratory equipment, which are not suitable for portable, rapid on-site testing. For periodic safety inspections of in-service charging piles, a portable system that can be detached after testing is needed. This system should be able to temporarily connect to the power distribution circuit and the charging pile input port, quickly complete the testing, and output evaluation results. In particular, there is often a distance of tens or even hundreds of meters between the power distribution box and the AC charging pile, and there may be walls or metal cabinets obstructing the view. Conventional short-range wireless communication cannot guarantee reliable coverage; therefore, a wireless communication solution suitable for long-distance, high-penetration scenarios is required.

[0009] Therefore, there is an urgent need for a safety evaluation method and portable testing system for AC charging piles that integrates the segmented impedance of the power distribution network, the measured temperature rise of the switch and charging head under full load, the insulation status (including the risk of arcing between live and neutral wires) and the functional protection capabilities into a unified framework, has the ability to analyze degradation trends over time and make horizontal group comparisons, uses long-distance wireless communication to achieve synchronization between the power distribution end and the charging pile end, and is easy to implement on site. Summary of the Invention

[0010] The technical problem to be solved by this invention is that existing AC charging pile safety testing does not cover the segmented impedance evaluation of the front-end power distribution network, does not directly assess the thermal risk of switches and charging gun heads through full-load measured temperature rise, does not systematically evaluate insulation status and arcing risk, does not use insulation testing as a prerequisite for power-on testing, does not evaluate protection function margin, lacks time-dimensional degradation trend prediction and horizontal group anomaly identification, and lacks a portable testing system suitable for on-site surveys and supporting long-distance communication.

[0011] To address the aforementioned technical problems, this invention provides a safety evaluation method for AC charging piles for electric vehicles, comprising the following steps: S1: A first detection unit is set at the lower port of the remote power distribution circuit starting switch of the charging pile to be tested; voltage sampling points of the second detection unit are set at the upper port of the pile-side power distribution box switch, the lower port of the pile-side power distribution box switch (i.e., the charging pile input port), and the output port of the charging gun, respectively; temperature sensors are attached to the body of the remote power distribution circuit starting switch, the body of the pile-side power distribution box switch, and the head of the charging gun, respectively; both the first and second detection units are portable devices and can be removed after the test is completed; S2: Establish clock synchronization between the first detection unit and the second detection unit through long-distance wireless communication, and synchronously collect the voltage at the lower end of the remote switch, the voltage at the upper end of the switch of the pile-side distribution box, the voltage at the lower end of the switch of the pile-side distribution box, the voltage at the output end of the charging gun, and the charging current during the normal charging process of the charging pile. S3: Based on the synchronously acquired data, calculate the external cable impedance, the voltage drop of the pile-side switch, and the charging pile impedance Z_charger respectively; S4: Perform insulation resistance testing on the power distribution network and charging piles (prioritize this step when the power is off), including at least the insulation resistance of the L-PE, N-PE, and LN lines of the power distribution line and the internal insulation resistance of the charging pile; if any insulation resistance is lower than the preset threshold, terminate the subsequent power-on test and directly trigger a safety warning; if the insulation condition is qualified, proceed to the next step. S5: After confirming that the insulation condition is qualified, control the charging pile to run continuously at the rated load or close to the rated load for a preset time. During the operation, continuously collect the surface temperature of the remote switch, the surface temperature of the switch of the pile-side distribution box, and the temperature of the charging gun head, record the temperature rise curve of each monitoring point, calculate the actual steady-state temperature rise and temperature rise rate, and directly assess the thermal risk of each switch and gun head. S6: Perform boundary scan tests based on power capacity on the functional protection capabilities of the charging pile. By progressively increasing the load power, scan the protection action boundary, record the overload protection action current, and calculate the protection margin; perform critical abnormal protection tests on the control guidance signals, including at least two items: CP disconnection and CP grounding; test the emergency stop protection action time. S7: Based on historical detection data, calculate the degradation rate of key safety indicators, predict the remaining safe life, and generate time-dimensional early warning indicators; S8: Perform a horizontal statistical comparison of the current charging pile's various safety indicators with data from a group of similar devices, identify individual abnormal deviations, and generate horizontal comparison indicators; S9: Combining the segmented impedance evaluation, insulation status evaluation, full-load temperature rise evaluation, protection margin evaluation, time-dimensional warning signs, and horizontal comparison signs, all evaluation dimensions are incorporated into a weighted comprehensive score, outputting a unified safety evaluation level and a problem list.

[0012] Further, the clock synchronization in step S2 includes: the second detection unit broadcasts a synchronization trigger frame as a master node of long-distance wireless communication, and the first detection unit receives the frame as a slave node and calibrates the local clock; wireless transmission delay is compensated by bidirectional handshake ranging to ensure that the voltage sampling time alignment error at both ends is ≤5ms; only stable segment data with charging current fluctuation of less than ±2% are used for impedance calculation.

[0013] Further, the formula for calculating the external cable impedance in step S3 is Z_grid = (U1 - U2) / I, where U1 is the voltage at the lower port of the remote distribution box switch, U2 is the voltage at the upper port of the pile-side distribution box switch, and I is the charging current; the voltage drop of the pile-side switch is ΔU_switch = U2 - U3, where U3 is the voltage at the lower port of the pile-side distribution box switch (i.e., the voltage at the charging pile input port); the formula for calculating the charging pile impedance is Z_charger = (U3 - U4) / I, where U4 is the voltage at the output terminal of the charging gun.

[0014] Furthermore, the insulation resistance test described in step S4 is performed under a completely de-energized state. A DC test voltage (500V DC) is injected into the power distribution line and the charging pile through the second detection unit to measure the insulation resistance between L-PE, N-PE, LN and the live circuit in the charging pile to ground. Based on the LN insulation resistance value and the grid voltage stress, the risk of breakdown discharge between the live and neutral wires is assessed. If the LN insulation resistance is less than 1MΩ or the L-PE / N-PE insulation resistance is less than 1MΩ, the insulation is deemed unqualified, all subsequent energized tests are terminated, and the hazard level assessment is directly output.

[0015] Furthermore, the preset duration in step S5 is 15 to 30 minutes to cover the thermal equilibrium establishment process; the actual steady-state temperature rise is the difference between the temperature of each monitoring point and the ambient temperature at the end of the preset duration; based on the comparison between the measured steady-state temperature rise of each monitoring point and the rated temperature rise limit of the corresponding device, the thermal risk level of the device is directly assessed. If the temperature of a certain monitoring point exceeds the rated temperature rise limit of the corresponding device, it is directly determined that the device has a potential thermal risk.

[0016] Furthermore, the boundary scan test based on power capacity in step S6 includes: using the programmable load built into the second detection unit or an external load, gradually increasing the load power from low power, scanning the boundary power point of the charging pile overload protection action, and recording the ratio of the protection action current to the rated current as the protection margin; the abnormal protection test of the control guidance key item includes: simulating CP disconnection to verify whether the charging pile correctly stops charging, and shorting CP to PE through a resistor to verify whether the charging pile correctly identifies the ground fault and stops charging.

[0017] Furthermore, the degradation rate calculation in step S7 is based on linear regression of at least three recent historical detection data; if the absolute value of the degradation rate exceeds the first threshold, it is determined to be slow degradation and the corresponding index score is reduced; if it exceeds the second threshold, it is determined to be rapid degradation and a trend warning flag is triggered; the remaining safe life is calculated by extrapolating the current index value to the danger threshold according to the degradation rate.

[0018] Furthermore, the horizontal statistical comparison in step S8 includes: establishing a benchmark library of similar equipment according to charging pile model, commissioning batch, regional power grid and line length range, and calculating the group mean, standard deviation and percentile of each safety indicator; if a certain indicator of the current charging pile is in the bottom 10% range of the group, it is judged as an individual abnormal deviation and the score of the corresponding indicator is reduced.

[0019] Furthermore, the safety evaluation level mentioned in step S9 is divided into four levels: A, B, C, and D. The problem list only lists the problems found during the detection and the corresponding measured parameters, and does not include rectification suggestions.

[0020] This invention also provides a portable testing system for safety evaluation of AC charging piles for electric vehicles, characterized in that it includes: The first detection unit is used to be temporarily installed at the starting point of the remote power distribution circuit of the charging pile. It has a voltage sampling channel clamped to the lower port of the switch at the starting point of the remote power distribution circuit to collect the voltage at the lower port of the switch and the ambient temperature and humidity data. It also has a temperature sensor attached to the switch body or short wire to collect the surface temperature of the switch and transmits the data through a long-distance wireless communication module. The second detection unit is a charging pile tester with integrated temperature rise monitoring function. It is temporarily installed on the side of the charging pile and has three voltage sampling channels, which are respectively connected to the upper port of the charging pile side distribution box switch, the lower port of the charging pile side distribution box switch (i.e., the charging pile input port), and the output end of the charging gun. It collects the voltage at the upper port of the charging pile side distribution box switch, the voltage at the lower port of the charging pile side distribution box switch, the voltage at the output end of the charging gun, the charging current, the contact resistance of the charging gun head, the insulation resistance (L-PE, N-PE, LN and the insulation to ground inside the pile), the continuity of PE, and the status of the control guidance signal. The second detection unit also has a temperature sensor attached to the body of the charging pile side distribution box switch or the short wire and attached to the charging gun head. The second detection unit has a built-in long-distance wireless communication module to synchronize with the first detection unit. The second detection unit is also used to control the charging pile to run continuously at the rated load for a preset time and continuously record the temperature data of each monitoring point. The second detection unit, acting as the master control node, synchronously triggers the first detection unit to collect data during the charging process. It receives and time-aligns the data from each node, calculates the external cable impedance, the voltage drop of the charging pile switch, and the charging pile impedance Z_charger, and performs insulation resistance testing (power off priority). After the insulation is qualified, it performs a full-load temperature rise test and directly assesses the thermal risk of each switch and charging head. It comprehensively evaluates the segmented impedance, insulation status, full-load temperature rise, protection margin, historical degradation trend, and group horizontal comparison data. All evaluation dimensions are incorporated into a weighted comprehensive score, and the safety evaluation level is output.

[0021] Furthermore, the second detection unit establishes a wireless communication link through LoRa or similar long-range low-power wireless communication methods, broadcasts a synchronization trigger frame, and uses bidirectional handshake ranging to compensate for transmission delay, thereby achieving millisecond-level clock synchronization with the first detection unit; both the first and second detection units are independent portable devices that can be removed after detection is completed.

[0022] Furthermore, the second detection unit incorporates an edge computing module for: calculating the external cable impedance based on the synchronously acquired voltage at the lower end of the remote switch, the upper end voltage of the switch in the pile-side distribution box, and the charging current; calculating the voltage drop of the pile-side switch based on the upper and lower end voltages of the switch in the pile-side distribution box; and calculating the charging pile impedance Z_charger based on the voltage at the lower end of the switch in the pile-side distribution box, the output voltage of the charging gun, and the charging current; injecting DC test voltage into the power distribution line and charging pile and measuring the insulation resistance; assessing the risk of breakdown discharge between the live and neutral wires based on the LN insulation resistance value; controlling the charging pile to operate at rated load for a preset time and continuously collecting temperature rise data of each switch and charging gun, calculating the actual steady-state temperature rise, and directly assessing the thermal risk of each device; calculating the degradation rate and predicting the remaining safe life based on historical detection data sequences; and comparing the current detection data with a benchmark library of similar equipment in the cloud to identify individual abnormal deviations. Beneficial effects

[0023] This invention incorporates the front-end power distribution network into the charging pile safety evaluation system. By simultaneously measuring four voltage points, it achieves segmented calculation of external cable impedance, pile-side switch voltage drop, and charging pile impedance Z_charger. This can accurately distinguish between "external cable overheating hazard," "pile-side switch overheating hazard," and "charging pile overheating hazard," filling the gap in existing standards for segmented assessment of the entire charging link impedance.

[0024] This invention directly monitors the temperature response of key components under actual rated load continuous operation by measuring the full-load temperature rise at the switch and gun head. It eliminates the need to indirectly calculate the contact resistance through pressure difference, avoiding the problem of insufficient measurement accuracy at low resistance values, and realizing a leap from "indirect estimation" to "direct verification".

[0025] This invention makes insulation resistance testing a mandatory prerequisite for all power-on tests. It is performed first in a completely de-energized state. If the insulation fails, the subsequent power-on test is terminated directly. This avoids the safety risks caused by high-power load testing when insulation problems are not eliminated, and fills the gap in the existing field testing process where the "insulation-power-on" sequence is missing.

[0026] This invention uses long-range wireless communication (such as LoRa) to achieve clock synchronization and data transmission between the power distribution end and the charging pile, solving the problem of short-range wireless communication failure caused by long distance and obstruction between the power distribution box and the charging pile. It is suitable for complex site environments such as underground parking lots, roadside charging stations, and scattered layouts in parks.

[0027] This invention proposes a protection margin evaluation method. Based on the power capacity boundary scan test that can be performed on site, the method scans the protection action boundary by gradually increasing the load, quantifies the deviation of the action threshold from the rated value, and evaluates the sensitivity and reliability reserve of the protection. This overcomes the limitation of existing technologies that only make a presence or absence judgment.

[0028] This invention introduces insulation resistance detection and risk assessment of breakdown discharge between live and neutral wires. By measuring the insulation resistance of L-PE, N-PE, and LN, and combining it with the grid voltage stress, the system quantitatively assesses the risk of leakage and sparking caused by insulation degradation.

[0029] This invention introduces a two-dimensional evaluation approach: time-based degradation trend analysis and horizontal group comparison. It calculates degradation rate and remaining safe life through historical data sequences and identifies individual anomalies through statistical distribution of similar equipment, thus achieving a leap from "single-time qualification judgment" to "full life cycle risk prediction".

[0030] This invention employs a portable wireless detection system. All equipment can be disassembled after detection, making it suitable for periodic on-site safety inspections of in-service charging piles. It solves the problems of high cost of fixed online monitoring and inconvenience of carrying laboratory equipment. Attached Figure Description

[0031] Figure 1 is a schematic diagram of the portable detection system architecture provided in an embodiment of the present invention; Figure 2 is a schematic diagram of the four-voltage-point segmented impedance measurement principle provided in an embodiment of the present invention; Figure 3 is a flowchart of the overall safety evaluation method provided in an embodiment of the present invention; Figure 4 is a logic block diagram of the evaluation algorithm provided in an embodiment of the present invention; Figure 5 shows the full-load temperature rise test curve and multi-point cross-verification diagram provided in the embodiment of the present invention; Figure 6 is a schematic diagram of the evaluation report output interface provided in an embodiment of the present invention. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Example 1: System Composition and On-site Installation

[0033] As shown in Figure 1, the portable detection system provided in this embodiment includes two independent portable devices: The first detection unit (Unit A): Temporarily installed in the remote distribution box or cabinet of the charging pile, corresponding to the circuit breaker or switch of the charging pile's power supply circuit. Unit A has a single voltage sampling channel, with a voltage clamp attached to the lower end (outgoing line) of the switch to collect the voltage U1 at the lower end of the switch. Unit A also has a temperature sensor (magnetically attached or clamp-on type), attached to the switch body or its short incoming and outgoing lines to collect the switch surface temperature T1 and the ambient temperature and humidity. Unit A has a built-in long-range wireless communication module (such as a LoRa module) and battery, which can be removed after testing.

[0034] The second testing unit (Unit B, a charging pile tester with integrated temperature rise monitoring function): temporarily installed on the side of the charging pile, it has three voltage sampling channels and multiple temperature monitoring channels. First voltage clamp: clamped to the upper port (incoming end) of the corresponding switch inside the pile-side distribution box, to collect the voltage U2 at the upper port of the switch in the pile-side distribution box; The second voltage clamp is attached to the lower port (outgoing terminal) of the corresponding switch in the pile-side distribution box. This lower port is the AC input port of the charging pile. It collects the voltage U3 at the lower port of the switch in the pile-side distribution box (U3 is the same point as the charging pile input port). Third voltage clamp: clamped to the output end of the charging gun (the base of the gun head or the connection between the gun head and the vehicle) to collect the voltage U4 at the output end of the gun head; Temperature sensor 1: Attached to the switch body or its short wire in the distribution box on the side of the pile, to collect the surface temperature T2 of the switch; Temperature sensor 2: Attached to the charging gun head, it collects the surface temperature T3 of the gun head.

[0035] Unit B also collects the charging current I, the contact resistance of each pin on the charging head (L, N, PE, measured using the four-wire method in micro-ohms), the insulation resistance (L-PE, N-PE, LN and the live circuit within the pile to ground, measured using the 500V DC injection method), the PE continuity resistance, the CP signal status, and the CC signal status. Unit B has a built-in long-range wireless communication module, edge computing module, storage module, programmable load or external load interface, and battery, which can be removed after testing.

[0036] During on-site installation, the testing personnel open the remote distribution box and clamp the voltage clamp of Unit A onto the lower terminal of the target switch, attaching the temperature sensor to the switch body or short wire. At the pile side, the three voltage clamps of Unit B are clamped onto the upper, lower (pile input), and output terminals of the switch in the pile-side distribution box, respectively, and the temperature sensors are attached to the switch and the sampling nozzle. After starting the test, Unit B establishes a long-distance wireless communication link, and Unit A connects and begins synchronous sampling. Example 2: Segmented Impedance Measurement at Four Voltage Points

[0037] As shown in Figure 2, during the normal charging process of the charging pile, Unit B acts as the master node of long-distance wireless communication and broadcasts a synchronization trigger frame, while Unit A acts as the slave node.

[0038] The clock synchronization process and bidirectional handshake ranging compensation are the same as in the previous embodiment, and the target alignment error is ≤5ms.

[0039] After charging begins, Unit B collects U2, U3, U4, and I at a fixed sampling frequency, while simultaneously triggering Unit A to synchronously collect U1 via a long-distance wireless command. Only synchronous data with current fluctuations less than ±2% and a duration of ≥3 seconds are used for impedance calculation.

[0040] External cable impedance (from the bottom of the remote distribution box switch to the top of the pile-side distribution box switch): Z_grid = (U1 - U2) / I Voltage drop of the pile-side switch (from the top to the bottom of the switch in the pile-side distribution box, refer to the record): ΔU_switch = U2 - U3 Charging pile impedance (from the bottom of the switch in the pile-side distribution box to the output of the charging gun): Z_charger = (U3 - U4) / I By performing the above segmented calculations, the impedance anomaly segment can be accurately located: If Z_grid is abnormally large → there is a problem with the external cable (from the remote distribution box to the pile-side distribution box); If ΔU_switch is abnormally large, there is poor contact at the pile-side switch contacts; If Z_charger is abnormally high, there is a problem inside the charging station or in the charging cable. Example 3: Insulation resistance testing and arcing risk assessment (prioritized, under power-off conditions)

[0041] This embodiment is executed first before all power-on tests and is a mandatory requirement of the on-site safety operating procedures.

[0042] Testing process: 1. Disconnect the power supply to the charging station to ensure that the circuit under test is completely de-energized; 2. Unit B injects DC test voltage (500V DC) into the power distribution line and charging pile through the built-in insulation test module or by using a separate insulation test device. 3. Measure the insulation resistance to ground of L-PE, N-PE, LN, and the live circuit inside the pile, respectively; 4. Assess the risk of breakdown discharge between live and neutral wires based on the insulation resistance value of LN and ambient humidity.

[0043] Insulation qualification criteria: L-PE / N-PE ≥ 1MΩ, LN ≥ 1MΩ: Insulation is qualified, and subsequent power-on testing is allowed; If any insulation resistance is < 1MΩ: the insulation is unqualified. The testing process will be terminated immediately, and a Class D hazard assessment will be output directly. The problem list will be marked "Insulation unqualified, power supply prohibited".

[0044] Why insulation is preferred: If the insulation of the external cable or the pile is severely deteriorated, direct power supply may cause leakage, arc discharge or sparking between the live and neutral wires and start a fire. Insulation testing is performed under power-off conditions, which does not affect the accuracy of subsequent full-load temperature rise testing; Avoid conducting high-power load tests before insulation hazards are eliminated, as this could lead to personal injury or equipment safety accidents. Example 4: Full-load temperature rise measurement and direct thermal risk assessment (to be performed after insulation passes inspection)

[0045] This embodiment can only be implemented after confirming that the insulation condition is qualified.

[0046] Test process: 1. Upon restoration of power, Unit B enables the charging station to operate continuously at or near its rated load by controlling the guidance signal (CP) or sending a charging request to the vehicle. 2. Run the test for the preset duration T_test, preferably 15~30 minutes; 3. During operation, Unit A continuously collects the remote switch temperature T1(t), Unit B continuously collects the pile-side switch temperature T2(t) and the nozzle temperature T3(t), and records the ambient temperature T_amb(t). 4. Temperature data is collected in real time to Unit B via long-distance wireless communication, and temperature rise curves for each monitoring point are plotted.

[0047] Calculation of measured parameters: - Actual steady-state temperature rise at each monitoring point: ΔT_i = avg(T_i(t_end-3min : t_end)) - T_amb - Temperature rise rate at each monitoring point: v_heat_i = dT_i / dt Direct thermal risk assessment: Based on the measured steady-state temperature rise at each monitoring point, a comparison is made with the rated temperature rise limit of the corresponding device: Added value: - "Impedance is normal but temperature rise is abnormal": The impedance measurement value of some lines is normal, but the actual measured temperature is higher due to poor local contact or poor heat dissipation conditions, which directly exposes the hidden danger. - Historical comparison: Compare the temperature rise curves of each monitoring point with the historical test curves. If the temperature rise increases year by year under the same current, it indicates that the device is undergoing hidden degradation. Example 5: Functional Protection Capability Test (Field-Implementation Item)

[0048] Boundary scan test based on power capacity (overload protection margin): Since on-site testing cannot adjust the grid input voltage (does not meet the conditions for overvoltage / undervoltage testing), and short-circuit testing is not allowed at the output, this embodiment uses a boundary scan test based on power capacity to evaluate overload protection capability.

[0049] Specific method: Using the programmable load built into Unit B or an external load, start with low power (e.g., 20% of rated power) and gradually increase the load power (or charging current) in fixed steps (e.g., 5% of rated power or 2A current step). Run stably for 30 seconds at each level and record the charging pile status until the charging pile overload protection is activated.

[0050] Record the protection trip current I_trip and rated current I_rated, and calculate the protection margin: M = I_trip / I_rated The ideal margin range is set at 1.1 to 1.5. Margin score calculation formula: S_margin = 100 - |M_actual - M_ideal| / M_ideal × 100 The greater the deviation from the ideal range, the lower the score.

[0051] If the charging pile does not operate within 150% of the rated current, the protection is deemed to have failed or has excessive margin.

[0052] Control guidance critical item anomaly protection test: Considering the limited time for on-site testing, this embodiment selects two of the most critical control guidance anomaly protections for testing: 1. CP Disconnection Test: Simulates CP disconnection at the vehicle plug (by disconnecting the CP circuit via an internal relay in Unit B) to verify whether the charging station stops charging and disconnects the power supply interface within 100ms. CP disconnection is the most common connection fault and directly affects charging safety.

[0053] 2. CP Grounding Test: Short-circuit CP to PE with a 120Ω resistor (simulating a CP grounding fault) to verify whether the charging pile correctly identifies and stops charging. CP grounding is a fault protection item explicitly required by GB / T 34657.1 and belongs to a high-risk fault mode.

[0054] Emergency stop protection test: Press the emergency stop button on the charging pile on site and test the time it takes to cut off the output (it should be ≤0.1s) to evaluate the fault lockout function (it should not automatically recover). Example 6: Time-based and cross-sectional comparative evaluation

[0055] After each test is completed, Unit B uploads all test data to the cloud database to establish a time series of safety profiles for the charging pile.

[0056] Sliding window linear regression (using at least three recent historical data points) was performed on key metrics (Z_grid, ΔU_switch, Z_charger, ΔT1, ΔT2, ΔT3, R_ins, M_protection) to calculate the degradation rate. v = ΔX / Δt Set a first threshold v_threshold and a second threshold 2×v_threshold: - If |v| < v_threshold, the system is considered stable, and K_time = 1.0; - If v_threshold ≤ |v| < 2×v_threshold, it is considered slow degradation, and K_time = 0.95; - If |v| ≥ 2×v_threshold, it is judged as rapid degradation, K_time = 0.90, and the trend warning flag is triggered.

[0057] Remaining safe life prediction: T_remain = (X_danger - X_current) / v Here, X_danger is the danger threshold for this indicator. If T_remain < 3 months, a trend warning is triggered, and the safety evaluation level is directly affected.

[0058] A benchmark library of similar equipment was established in the cloud, with grouping dimensions including: charging pile model, commissioning batch, regional power grid, and line length range (±20%). The group mean μ, standard deviation σ, and percentiles P_10, P_50, and P_90 for each safety indicator in each group were calculated.

[0059] The current percentile P of each indicator of the charging pile in the group is determined by the following rules: - If P ≤ 75%, K_cross = 1.0; - If 75% < P ≤ 90%, K_cross = 0.95; - If P > 90%, it is judged as an individual abnormal deviation, K_cross = 0.90. Example 7: Comprehensive Evaluation and Output

[0060] As shown in Figure 4, all six evaluation dimensions are incorporated into the weighted comprehensive score: Dimension ① Piecewise impedance evaluation: - Z_grid, ΔU_switch, and Z_charger are scored separately to reflect the electrical safety status of each section of the external cable, switch, and charging pile. Dimension ② Full-load temperature rise evaluation: - The measured temperature rises of T1, T2, and T3 are scored separately to reflect the actual thermal risks of each key component. Dimension ③ Insulation status evaluation: - The insulation resistance of L-PE, N-PE, and LN is scored separately to reflect the risk of leakage and arcing. Dimension 4: Protection Margin Assessment - Overload protection margin, CP protection, and emergency stop protection are rated separately. Dimension 5: Time Dimension Evaluation - As a correction factor K_time, it is applied to the scores of all dimensions. Dimension 6: Horizontal Comparison and Evaluation - As a correction factor K_cross, it applies to the scores across all dimensions. Comprehensive scoring formula: S_total = Σ(w_i × S_base,i × K_time,i × K_cross,i) Weighting example: - A. Distribution network safety (including segmented impedance, switch / probe temperature rise, insulation): 35% - B. Electrical safety of the charging station itself (including Z-charger, gun contact resistance, and internal insulation): 25% - C. Functional protection capabilities (including overload margin, CP protection, and emergency stop): 40% Security level classification: - Grade A (Excellent): S_total ≥ 90, and no trend warning; - Grade B (Good): 70 ≤ S_total < 90, or a slight trend warning exists; - Grade C (Qualified): 60 ≤ S_total < 70, or there is a clear trend warning; - Level D (Dangerous): S_total < 60, or any key indicator triggers a trend warning and is expected to reach the danger threshold within 3 months, or there is a fatal defect (such as LN insulation << 1MΩ, PE failure, pile insulation << 1MΩ, switch temperature exceeding the limit, or measured temperature rise seriously deviating from the theoretical value).

[0061] The evaluation output is a structured report, including: overall rating, sub-item scores, measured impedance values ​​for each segment, temperature rise curves for each monitoring point, measured insulation resistance values, time-based warning indicators, horizontal comparison indicators, and a problem list. The problem list only lists the problems found during the inspection and their corresponding measured parameters, and does not include rectification suggestions.

Claims

1. A safety evaluation method for AC charging piles for electric vehicles, characterized in that, Includes the following steps: S1: A first detection unit is set at the lower port of the remote power distribution circuit starting switch of the charging pile to be tested; voltage sampling points of the second detection unit are set at the upper port of the pile-side power distribution box switch, the lower port of the pile-side power distribution box switch (i.e., the charging pile input port), and the output port of the charging gun, respectively; temperature sensors are attached to the body of the remote power distribution circuit starting switch, the body of the pile-side power distribution box switch, and the head of the charging gun, respectively; both the first and second detection units are portable devices and can be removed after the test is completed; S2: Establish clock synchronization between the first detection unit and the second detection unit through long-distance wireless communication, and synchronously collect the voltage at the lower end of the remote switch, the voltage at the upper end of the switch of the pile-side distribution box, the voltage at the lower end of the switch of the pile-side distribution box, the voltage at the output end of the charging gun, and the charging current during the normal charging process of the charging pile. S3: Based on the synchronously acquired data, calculate the external cable impedance, the voltage drop of the pile-side switch, and the charging pile impedance Z_charger respectively; S4: Perform insulation resistance testing on the power distribution network and charging piles (prioritize execution under power-off conditions), including at least the insulation resistance of the L-PE, N-PE, and LN power distribution lines and the internal insulation resistance of the charging piles; if the insulation resistance is lower than the preset threshold, terminate the subsequent power-on test and directly trigger a safety warning; S5: After confirming that the insulation condition is qualified, control the charging pile to run continuously at the rated load or close to the rated load for a preset time. During the operation, continuously collect the surface temperature of the remote switch, the surface temperature of the switch of the pile-side distribution box, and the temperature of the charging gun head, record the temperature rise curve of each monitoring point, calculate the actual steady-state temperature rise and temperature rise rate, and directly assess the thermal risk of each switch and gun head. S6: Perform boundary scan tests based on power capacity on the functional protection capabilities of the charging pile. By progressively increasing the load power, scan the protection action boundary, record the overload protection action current, and calculate the protection margin; perform critical abnormal protection tests on the control guidance signals, including at least two items: CP disconnection and CP grounding; test the emergency stop protection action time. S7: Based on historical detection data, calculate the degradation rate of key safety indicators, predict the remaining safe life, and generate time-dimensional early warning indicators; S8: Perform a horizontal statistical comparison of the current charging pile's various safety indicators with data from a group of similar devices, identify individual abnormal deviations, and generate horizontal comparison indicators; S9: Combining the segmented impedance evaluation, insulation status evaluation, full-load temperature rise evaluation, protection margin evaluation, time-dimensional warning signs, and horizontal comparison signs, all evaluation dimensions are incorporated into a weighted comprehensive score, outputting a unified safety evaluation level and a problem list.

2. The method according to claim 1, characterized in that, The clock synchronization in step S2 includes: the second detection unit broadcasts a synchronization trigger frame as the master node of long-distance wireless communication, and the first detection unit receives the frame as the slave node and calibrates the local clock; wireless transmission delay is compensated by bidirectional handshake ranging to ensure that the voltage sampling time alignment error at both ends is ≤5ms; only stable segment data with charging current fluctuation of less than ±2% are used for impedance calculation.

3. The method according to claim 1, characterized in that, The formula for calculating the impedance of the external cable in step S3 is Z_grid = (U1 - U2) / I, where U1 is the voltage at the lower port of the remote distribution box switch, U2 is the voltage at the upper port of the pile-side distribution box switch, and I is the charging current; the voltage drop of the pile-side switch is ΔU_switch = U2 - U3, where U3 is the voltage at the lower port of the pile-side distribution box switch (i.e., the voltage at the charging pile input port); the formula for calculating the impedance of the charging pile is Z_charger = (U3 - U4) / I, where U4 is the voltage at the output terminal of the charging gun.

4. The method according to claim 1, characterized in that, The insulation resistance test described in step S4 is performed under a completely de-energized state. DC test voltage is injected into the power distribution line and charging pile through the second detection unit to measure the insulation resistance between L-PE, N-PE, LN and the live circuit in the charging pile to ground. Based on the LN insulation resistance value and the grid voltage stress, the risk of breakdown discharge between the live and neutral wires is assessed. If the LN insulation resistance is less than 1MΩ, the insulation is deemed unqualified and all subsequent power-on tests are terminated.

5. The method according to claim 1, characterized in that, The preset duration in step S5 is 15 to 30 minutes to cover the thermal equilibrium establishment process; the actual steady-state temperature rise is the difference between the temperature of each monitoring point and the ambient temperature at the end of the preset duration; based on the comparison between the measured steady-state temperature rise of each monitoring point and the rated temperature rise limit of the corresponding device, the thermal risk level of the device is directly assessed. If the temperature of a monitoring point exceeds the rated temperature rise limit of the corresponding device, it is directly determined that the device has a potential thermal risk.

6. The method according to claim 1, characterized in that, The boundary scan test based on power capacity in step S6 includes: using the programmable load built into the second detection unit or an external load, gradually increasing the load power from low power, scanning the boundary power point of the charging pile overload protection action, and recording the ratio of the protection action current to the rated current as the protection margin; the abnormal protection test of the control guidance key item includes: simulating CP disconnection to verify whether the charging pile stops charging correctly, and shorting CP to PE through a resistor to verify whether the charging pile correctly identifies the ground fault and stops charging.

7. The method according to claim 1, characterized in that, The degradation rate in step S7 is calculated based on linear regression of at least three recent historical detection data; if the absolute value of the degradation rate exceeds the first threshold, it is determined to be slow degradation and the corresponding index score is reduced. If the second threshold is exceeded, it is judged as rapid deterioration and a trend warning sign is triggered; The remaining safe life is calculated by extrapolating the current indicator value to the danger threshold according to the degradation rate.

8. A portable testing system for safety evaluation of AC charging piles for electric vehicles, characterized in that, include: The first detection unit is used to be temporarily installed at the starting point of the remote power distribution circuit of the charging pile. It has a voltage sampling channel clamped to the lower port of the switch at the starting point of the remote power distribution circuit to collect the voltage at the lower port of the switch and the ambient temperature and humidity data. It also has a temperature sensor attached to the switch body or short wire to collect the surface temperature of the switch and transmits the data through a long-distance wireless communication module. The second detection unit is a charging pile tester with integrated temperature rise monitoring function. It is temporarily installed on the side of the charging pile and has three voltage sampling channels, which are respectively connected to the upper port of the charging pile side distribution box switch, the lower port of the charging pile side distribution box switch (i.e., the charging pile input port), and the output end of the charging gun. It collects the voltage at the upper port of the charging pile side distribution box switch, the voltage at the lower port of the charging pile side distribution box switch, the voltage at the output end of the charging gun, the charging current, the contact resistance of the charging gun head, the insulation resistance, the PE continuity, and the status of the control guidance signal. The second detection unit also has a temperature sensor attached to the body of the charging pile side distribution box switch or the short wire and attached to the charging gun head. The second detection unit has a built-in long-distance wireless communication module to synchronize with the first detection unit. The second detection unit is also used to control the charging pile to run continuously at the rated load for a preset time and continuously record the temperature data of each monitoring point. The second detection unit, acting as the master control node, synchronously triggers the first detection unit to collect data during the charging process. It receives and time-aligns the data from each node, calculates the external cable impedance, the voltage drop of the charging pile switch, and the charging pile impedance Z_charger, and performs insulation resistance testing (power off priority). After the insulation is qualified, it performs a full-load temperature rise test and directly assesses the thermal risk of each switch and charging head. It comprehensively evaluates the segmented impedance, insulation status, full-load temperature rise, protection margin, historical degradation trend, and group horizontal comparison data. All evaluation dimensions are incorporated into a weighted comprehensive score, and the safety evaluation level is output.

9. The system according to claim 8, characterized in that, The second detection unit establishes a wireless communication link through LoRa or similar long-range low-power wireless communication methods, broadcasts a synchronization trigger frame, and uses bidirectional handshake ranging to compensate for transmission delay, thereby achieving millisecond-level clock synchronization with the first detection unit. Both the first and second detection units are independent portable devices that can be removed after detection.

10. The system according to claim 8, characterized in that, The second detection unit has a built-in edge computing module, used for: calculating the external cable impedance based on the synchronously acquired voltage at the lower end of the remote switch, the upper end voltage of the switch in the pile-side distribution box, and the charging current; calculating the voltage drop of the pile-side switch based on the voltage at the upper and lower ends of the switch in the pile-side distribution box; and calculating the charging pile impedance Z_charger based on the voltage at the lower end of the switch in the pile-side distribution box, the output voltage of the charging gun, and the charging current; injecting DC test voltage into the power distribution line and the charging pile and measuring the insulation resistance; assessing the risk of breakdown discharge between the live and neutral wires based on the LN insulation resistance value; controlling the charging pile to operate at rated load for a preset time and continuously collecting temperature rise data of each switch and charging gun, calculating the actual steady-state temperature rise, and directly assessing the thermal risk of each device; calculating the degradation rate and predicting the remaining safe life based on historical detection data sequences; and comparing the current detection data with a benchmark library of similar equipment in the cloud to identify individual abnormal deviations.

Images