An intelligent sensor-based phase-to-phase short-circuit fault quick-break measurement method and system

Abstract

This invention relates to the field of power system relay protection and intelligent sensing detection technology, specifically a method and system for instantaneous overcurrent measurement of phase-to-phase short-circuit faults based on intelligent sensors. The method includes: acquiring high-frequency transient electrical waveform data and initial thermal reference data of the surface of the electrical equipment under test during a phase-to-phase short circuit; extracting instantaneous voltage and instantaneous current, calculating instantaneous energy integral data and electrodynamic impact envelope data; inputting a multiphysics twin mapping model, and outputting expected temperature rise curve data and deformation stress curve data; combining the equipment's withstand limit temperature, structural material parameters, and initial thermal reference data to calculate remaining temperature rise margin data, yield strength margin data, and real-time damage-induced disaster risk index; and then comparing these with a preset risk threshold to generate adaptive instantaneous overcurrent decision commands or blocking instantaneous overcurrent commands and continuous monitoring commands. This invention achieves instantaneous overcurrent measurement oriented towards the consequences of equipment damage, while also preventing false tripping and failure to trip.

Description

Technical Field

[0001] This invention relates to the field of power system relay protection and intelligent sensing and detection technology, specifically to a method and system for measuring the instantaneous overcurrent of phase-to-phase short-circuit faults based on intelligent sensors. Background Technology

[0002] Currently, the instantaneous overcurrent protection mechanism for phase-to-phase short-circuit faults in electrical equipment is usually based on current amplitude exceeding limits or setting criteria. When the electrical equipment under test is operating at high temperatures, experiencing frequent transient impacts, or exhibiting complex short-circuit current characteristics, the existing methods cannot measure in real time whether the fault will cause thermal and mechanical damage to the equipment itself. Each action judgment may be based solely on a single electrical quantity to directly trip or block, reducing the accuracy of the instantaneous overcurrent decision. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for instantaneous tripping measurement of phase-to-phase short-circuit faults based on intelligent sensors. This avoids the problem that relying solely on a single electrical quantity cannot provide real-time measurement of thermal and mechanical damage to equipment, and it also makes it easier to achieve accurate and instantaneous tripping decisions that simultaneously prevent false tripping and failure to trip. Specifically, the technical solution of this invention is as follows:

[0004] A method for instantaneous overcurrent measurement of phase-to-phase short-circuit faults based on intelligent sensors includes the following steps:

[0005] S1, uses intelligent sensors to collect high-frequency transient electrical waveform data of the electrical equipment under test when a phase-to-phase short circuit occurs, as well as the initial thermal reference data of the surface of the electrical equipment under test;

[0006] S2, obtain instantaneous energy integral data and electrodynamic impact envelope data based on high-frequency transient electrical waveform data;

[0007] S3, input the instantaneous energy integral data and electrodynamic impact envelope data into the multiphysics twin mapping model that includes the resistive-capacitive thermal network model and the structural dynamics equivalent model, and respectively map to obtain the expected temperature rise curve data and deformation stress curve data;

[0008] S4. Based on the initial thermal baseline data, the preset equipment tolerance limit temperature and the preset structural material parameters, obtain the remaining temperature rise margin data and yield limit margin data. Extract the maximum temperature rise peak in the expected temperature rise curve data and the maximum stress peak in the deformation stress curve data respectively. Obtain the thermal disaster risk degree of the maximum temperature rise peak relative to the remaining temperature rise margin data and the force disaster risk degree of the maximum stress peak in the deformation stress curve data relative to the yield limit margin data. And comprehensively evaluate to obtain the real-time damage disaster risk index.

[0009] S5 compares the real-time damage-causing risk index with the preset risk threshold. When the real-time damage-causing risk index is greater than or equal to the preset risk threshold, an adaptive quick-break decision instruction is generated. When the real-time damage-causing risk index is less than the preset risk threshold, a monitoring status signal indicating that the damage risk has not been reached is generated.

[0010] Preferably, in step S3, the multiphysics twin mapping model includes a resistive-capacitive thermal network model and a structural dynamics equivalent model;

[0011] Among them, the resistance-capacitance thermal network model is used to map instantaneous energy integral data to expected temperature rise curve data; the structural dynamics equivalent model is used to map electrodynamic impact envelope data to deformation stress curve data.

[0012] Preferably, step S4 includes:

[0013] The remaining temperature rise margin data is determined based on the current actual temperature extracted from the initial thermal reference data and the preset equipment tolerance limit temperature.

[0014] Determine the yield strength margin data by combining the current actual temperature and the preset structural material parameters;

[0015] The thermal disaster risk level of the maximum temperature rise peak relative to the remaining temperature rise margin data and the force disaster risk level of the maximum stress peak relative to the yield limit margin data are obtained respectively, and then the real-time damage disaster risk index is determined by combining them.

[0016] Preferably, in step S5, generating a monitoring status signal indicating that the risk of disaster has not been reached includes generating a blocking instantaneous disconnection command and a continuous monitoring command.

[0017] Preferably, after generating the adaptive instantaneous decision instruction, the method further includes:

[0018] Based on the real-time damage risk index and high-frequency transient electrical waveform data, the incremental data of life loss caused by this short-circuit impact on the electrical equipment under test is evaluated, and a condition-based maintenance assessment report is generated accordingly.

[0019] Preferably, after acquiring high-frequency transient electrical waveform data and initial thermal reference data, the method further includes:

[0020] The high-frequency transient electrical waveform data and initial thermal reference data are time-stamped and pre-processed by an edge computing chip deployed on the smart sensor.

[0021] A phase-to-phase short-circuit fault instantaneous overcurrent measurement system based on intelligent sensors includes:

[0022] The data sensing module is used to collect high-frequency transient electrical waveform data of the electrical equipment under test when a phase-to-phase short circuit occurs, as well as the initial thermal reference data of its surface.

[0023] The transient integration module is used to acquire instantaneous energy integral data and electrodynamic impact envelope data based on high-frequency transient electrical waveform data;

[0024] The twin mapping module is used to input instantaneous energy integral data and electrodynamic impact envelope data into the multiphysics twin mapping model to obtain the expected temperature rise curve data and deformation stress curve data of the electrical equipment under test;

[0025] The risk assessment module is used to obtain the remaining temperature rise margin data and yield limit margin data based on the initial thermal baseline data, the preset equipment tolerance limit temperature and structural material parameters, and compare them with the maximum temperature rise peak in the expected temperature rise curve data and the maximum stress peak in the deformation stress curve data to obtain the real-time damage disaster risk index.

[0026] The decision execution module is used to compare the real-time damage risk index with the preset risk threshold, and generate an adaptive quick-break decision command or a monitoring status signal indicating that the risk of disaster has not been reached based on the comparison result.

[0027] Preferably, the decision execution module includes a logic judgment unit;

[0028] The logic judgment unit is used to generate a lockout instantaneous disconnection command and a continuous monitoring command when the real-time damage-causing risk index is less than a preset risk threshold, as a monitoring status signal that the disaster risk has not been reached.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] 1. This invention effectively solves the problem that existing instantaneous overcurrent protection relies solely on a single current amplitude limit or setting criterion and cannot reflect the current thermal state differences of the equipment by using a wideband current and voltage sensor to collect high-frequency transient electrical waveform data when a phase-to-phase short circuit occurs, and combining it with a non-invasive miniature temperature sensor to collect initial thermal reference data of the surface of the electrical equipment under test.

[0031] 2. This invention obtains instantaneous energy integral data by integrating the product of instantaneous voltage and instantaneous current over time, and extracts the envelope of the square of the instantaneous current as the electrodynamic impact envelope data. This can characterize the thermal damage input and mechanical impact input during the short circuit process, avoiding the shortcomings of only looking at the single-point peak current and ignoring the continuous characteristics and impact consequences.

[0032] 3. This invention inputs instantaneous energy integral data and electrodynamic impact envelope data into a preset multiphysics twin mapping model, and outputs expected temperature rise curve data and deformation stress curve data by the resistive-capacitive thermal network model and the structural dynamics equivalent model, respectively. This ensures that the system can quickly convert electrical measurement results into the temperature rise and stress response of the equipment body within the protection time limit, thereby improving the real-time measurement capability of thermal and mechanical damage.

[0033] 4. This invention combines the equipment's tolerance limit temperature, structural material parameters, current actual temperature, temperature decay coefficient, and thermal disaster weighting coefficient to calculate the remaining temperature rise margin, yield limit margin, thermal disaster risk degree, and force disaster risk degree, and forms a real-time damage disaster risk index. This changes the action judgment from whether the current is too large to whether the current impact has approached irreversible damage, thereby improving the accuracy of rapid trip decision.

[0034] 5. This invention generates a blocking instantaneous trip command and a continuous monitoring command when the risk index does not reach the threshold, and generates an adaptive instantaneous trip decision command and sends it to the circuit breaker when the risk index reaches the threshold. It works in conjunction with an edge computing chip to perform timestamp synchronization processing and noise reduction preprocessing, and switches to a conservative mode or backup criteria in cases of abnormal data, synchronization failure, or missing parameters. Through the synergistic effect of risk quantification assessment, low-risk blocking monitoring, and high-risk rapid tripping, it effectively balances the prevention of false tripping and the prevention of failure to trip, and ultimately significantly improves the reliability of instantaneous tripping measurement and protection control of phase-to-phase short-circuit faults under complex operating conditions. Attached Figure Description

[0035] Figure 1 A flowchart illustrating a method for measuring the instantaneous overcurrent of a phase-to-phase short-circuit fault based on a smart sensor, provided in an embodiment of this application;

[0036] Figure 2 This is a schematic diagram of a phase-to-phase short-circuit fault instantaneous overcurrent measurement system based on intelligent sensors provided in an embodiment of this application. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0038] A method for instantaneous overcurrent measurement of phase-to-phase short-circuit faults based on intelligent sensors includes the following steps:

[0039] S1, uses intelligent sensors to collect high-frequency transient electrical waveform data of the electrical equipment under test when a phase-to-phase short circuit occurs, as well as the initial thermal reference data of the surface of the electrical equipment under test;

[0040] S2, obtain instantaneous energy integral data and electrodynamic impact envelope data based on high-frequency transient electrical waveform data;

[0041] S3, input the instantaneous energy integral data and electrodynamic impact envelope data into the multiphysics twin mapping model that includes the resistive-capacitive thermal network model and the structural dynamics equivalent model, and respectively map to obtain the expected temperature rise curve data and deformation stress curve data;

[0042] S4. Based on the initial thermal baseline data, the preset equipment tolerance limit temperature and the preset structural material parameters, obtain the remaining temperature rise margin data and yield limit margin data. Extract the maximum temperature rise peak in the expected temperature rise curve data and the maximum stress peak in the deformation stress curve data respectively. Obtain the thermal disaster risk degree of the maximum temperature rise peak relative to the remaining temperature rise margin data and the force disaster risk degree of the maximum stress peak in the deformation stress curve data relative to the yield limit margin data. And comprehensively evaluate to obtain the real-time damage disaster risk index.

[0043] S5 compares the real-time damage-causing risk index with the preset risk threshold. When the real-time damage-causing risk index is greater than or equal to the preset risk threshold, an adaptive quick-break decision instruction is generated. When the real-time damage-causing risk index is less than the preset risk threshold, a monitoring status signal indicating that the damage risk has not been reached is generated.

[0044] This embodiment provides a fast-break measurement mechanism for phase-to-phase short-circuit faults based on intelligent sensors, such as... Figure 1 As shown; specifically, the main scenario is set as a switch cabinet in a power supply system, in which the copper busbars, insulating supports and outgoing circuit breakers together constitute the electrical equipment under test;

[0045] When the electrical equipment under test is under high load or its initial hot state is close to the upper limit of the rated operating temperature, the traditional instantaneous tripping method that only operates when the current exceeds the limit is prone to false tripping when subjected to external transient interference or load change. Therefore, this solution is adopted to measure in real time whether it will cause substantial damage to the equipment body and make a decision accordingly.

[0046] Specifically, in step S1, a wideband current and voltage sensor is installed near the copper busbars of phase A and phase B to collect a high-frequency transient waveform before and after the phase-to-phase short circuit occurs; a non-invasive miniature temperature sensor is attached to the corresponding position on the outer surface insulation layer of the busbar or the inner side of the cabinet observation window to collect the current thermal reference.

[0047] For ease of explanation, a simplified test data window can be set with a sampling period of four discrete moments t1, t2, t3, and t4. The instantaneous voltages measured at these moments are 10, 8, 5, and 2, and the instantaneous currents are 2, 5, 7, and 3, respectively, thus forming two sets of synchronous sequences. The temperature sensor measures the surface temperature of the equipment at 78°C just before the fault occurs. This value is used as the current thermal starting point, instead of using a fixed ambient temperature.

[0048] In step S2, the instantaneous voltage and instantaneous current are multiplied to obtain the instantaneous power sequence 20, 40, 35, 6 at each moment. This power sequence is then integrated in chronological order. If the distance step size is uniformly set to 1, the cumulative energy integration sequence 20, 60, 95, 101 can be obtained. This integration result reflects the actual electrical energy injected into the device during the short circuit, rather than just considering the single-point peak current. The calculation logic for the instantaneous energy integration data is expressed by the following formula:

[0049]

[0050] in, This is instantaneous energy integral data, in J; The instantaneous voltage extracted from high-frequency transient electrical waveform data, in V; The instantaneous current extracted from high-frequency transient electrical waveform data, in amperes (A). The time variable is the integral time variable, and the unit is seconds (s). The variable is the integral variable, and the unit is seconds (s).

[0051] Meanwhile, the instantaneous current is squared to obtain 4, 25, 49, and 9, and its envelope is extracted. If these four points themselves constitute a single-peak change, the envelope can be simplified to 4, 25, 49, and 49, which are used to characterize the rising and maintaining trend of electrodynamic impact. Thus, one set of data is used to describe thermal damage input, and another set of data is used to describe mechanical impact input.

[0052] In step S3, the energy integral sequence and the electrodynamic impact envelope are input into a preset multiphysics twin mapping model. In this embodiment, the model does not pursue a full solution of the finite element, but adopts an equivalent mapping that can run within the protection time limit. Taking the above data as an example, the model receives the final value of the accumulated energy 101 and the intermediate climbing process, and can output the expected temperature rise curve. For example, the predicted temperature rise values ​​in the following milliseconds are 12℃, 23℃, and 31℃, respectively.

[0053] At the same time, based on the peak value of the electrodynamic impact envelope 49, the output deformation stress curve is given, for example, the corresponding predicted stress is 65MPa, 88MPa, and 91MPa; here, the expectation refers to the most likely temperature rise and stress change trajectory inside the equipment if the current fault continues and evolves according to the measured excitation.

[0054] In step S4, the remaining margin is first calculated based on the current thermal state and equipment parameters; assuming that the allowable temperature limit of the copper busbar insulation system of the switch cabinet is 130℃ and the current measured temperature is 78℃, then the remaining temperature rise margin is 52℃.

[0055] The yield strength margin is obtained by combining structural material parameters. For example, if the upper limit of the withstandable stress after correction is 100 MPa, then by referring to the maximum stress peak of 91 MPa in the deformation stress curve, it can be seen that the mechanical risk reaches 91% of the preset material yield strength. The calculation logic of the remaining temperature rise margin data is expressed by the following formula:

[0056]

[0057] in, This represents the remaining temperature rise margin, expressed in °C. The preset equipment tolerance limit temperature, in °C; The current actual temperature is extracted based on the initial thermal baseline data, in °C.

[0058] The calculation logic for yield strength margin data is expressed by the following formula:

[0059]

[0060] in, This represents the yield strength margin, expressed in MPa. This represents the standard yield strength in structural material parameters, with units of MPa. This is the temperature decay coefficient, measured in MPa / ℃, which is included in the preset structural material parameters; The actual current temperature, in °C.

[0061] Furthermore, the maximum temperature rise of 31°C can be compared with the remaining temperature rise margin of 52°C to obtain the risk level of thermal disaster. The maximum stress peak of 91 MPa was compared with the yield strength margin of 100 MPa to obtain the force-induced disaster risk level. Then, a real-time damage-causing risk index is generated according to preset rules. The calculation logic of the real-time damage-causing risk index is expressed by the following formula:

[0062]

[0063]

[0064]

[0065] in, The degree of risk of thermal disasters is dimensionless. The maximum temperature rise peak value in the expected temperature rise curve data is expressed in °C. This represents the remaining temperature rise margin, expressed in °C. The degree of disaster risk is dimensionless. This represents the maximum stress peak value in the deformation stress curve data, in MPa. This represents the yield strength margin, expressed in MPa. This is a real-time damage-causing risk index, dimensionless. and The preset thermal disaster weighting coefficient is dimensionless and must satisfy the following conditions: ;

[0066] For example, based on the preset thermal disaster weighting coefficients, the risk levels of thermal disasters and mechanical disasters are assigned weighting coefficients of 0.4 and 0.6 respectively. The risk index can then be approximated as follows: This results in a single, actionable real-time risk level.

[0067] In step S5, the real-time damage-causing risk index is compared with a preset risk threshold. Assuming the risk threshold is set to 0.85, if the current calculation result is 0.79, the system will not immediately trip, but will output a monitoring status signal indicating that the damage risk has not been reached. If the waveform amplitude or distortion continues to increase, causing the risk to jump to 0.92, an adaptive trip decision command will be generated and sent to the circuit breaker tripping circuit. The core of this decision-making method is that the judgment basis is no longer whether the current is too large at a certain moment, but whether the current impact has approached irreversible damage after being buffered by the equipment body.

[0068] In the abnormal data processing mechanism, when there are missing data points, sensor clock asynchrony, or short-term noise spikes in the sampled data, the abnormal points are not directly input into the subsequent model; if the current value of a certain sampling point is missing, it is supplemented by interpolation between two adjacent points.

[0069] If the number of consecutive missing points exceeds the preset limit, such as more than 3 discrete points, the current round of instantaneous trip judgment is frozen and the system switches to conservative mode. In conservative mode, the traditional overcurrent instantaneous trip threshold can be used as a backup criterion. If the temperature sensor reading exceeds the physically reasonable range, such as a significant distortion value like -40℃ or 260℃, the temperature frame is discarded and the most recent stable thermal value is used. At the same time, the sensor abnormality marker is recorded.

[0070] During the operation of the electrical equipment under test, a sudden change in high-power load causes a continuous short-circuit impact between phases. Initially, the peak current exceeds the normal operating current threshold, but because the duration is less than the preset time threshold, the initial temperature of the equipment is lower than the equipment's tolerance limit temperature, and there is a preset temperature rise space, the system first outputs a continuous monitoring signal without tripping erroneously. When the second impact is superimposed, the temperature rise prediction and stress prediction rise simultaneously, and the real-time damage and disaster risk index crosses the threshold, at which point the circuit breaker is triggered for instantaneous tripping.

[0071] The purpose of this step is to convert electrical measurement results into risk quantities that are geared towards the consequences of equipment, thereby achieving rapid disconnection measurement that takes into account both preventing malfunction and preventing failure to operate.

[0072] Furthermore, in step S3, the multiphysics twin mapping model includes a resistive-capacitive thermal network model and a structural dynamics equivalent model;

[0073] Among them, the resistance-capacitance thermal network model is used to map instantaneous energy integral data to expected temperature rise curve data; the structural dynamics equivalent model is used to map electrodynamic impact envelope data to deformation stress curve data.

[0074] This embodiment provides a thermodynamic and mechanical dynamic equivalent modeling mechanism for twin mapping. Specifically, in the aforementioned main scenario, if a full-size three-dimensional finite element model is directly used to solve the switch cabinet copper busbar, bolt connection point, and insulating support point by point, although it theoretically has theoretical calculation accuracy, its calculation convergence time exceeds the preset action time limit of the relay protection and cannot meet the real-time requirement of millisecond-level output. Therefore, it is necessary to compress the complex physical process into an equivalent model that can be calculated in real time.

[0075] Specifically, the thermodynamic part adopts a resistance-capacitance thermal network model; the device under test can be abstracted into three thermal nodes: node N1 represents the copper busbar conductor body, node N2 represents the insulating coating layer, and node N3 represents the air inside the cabinet and the area near the wall; a first thermal resistance is set between N1 and N2, and a second thermal resistance is set between N2 and N3, and each node has an equivalent heat capacity; in this way, the instantaneous energy integral is not directly equivalent to the surface temperature rise, but is first injected into N1, and then transferred layer by layer through the thermal resistance;

[0076] In a specific embodiment, if the accumulated input energy in a fault window is 100 units, the model can first raise N1 by 20°C according to a pre-calibrated ratio, and then raise N2 by 12°C and N3 by 5°C through thermal resistance coupling; when the next sampling window injects another 50 units, it does not start from zero, but continues to update on the basis of the residual heat of the previous window, thereby forming a continuous expected temperature rise curve.

[0077] The mechanical dynamics section adopts an equivalent model of structural dynamics; the copper busbar and supporting structure can be simplified as a combination of concentrated mass blocks, springs and dampers; the electrodynamic impact envelope is used as the external force input, and the higher the peak value and the longer it is maintained, the greater the displacement and stress response.

[0078] For example, if the input amplitudes of the electrodynamic envelope in three consecutive time windows are 30, 50, and 20, respectively, the first time window mainly produces the initial amount of elastic deformation, the second time window reaches the maximum structural response, and the third time window enters the decay phase; the model can output a deformation stress curve, such as 40MPa, 90MPa, and 62MPa; due to the introduction of damping, the third time window will not simply be proportionally equal to half of the second time window, but will reflect the inertia and hysteresis of the mechanical system;

[0079] In terms of implementation, the two sub-models can share the same fault time axis, but receive different feature quantities respectively; the former takes energy integral data as the main input, and the latter takes the current square envelope as the main input, and aligns the output at the unified risk assessment time; this can avoid confusing the thermal process and the mechanical process as the same quantity and causing misjudgment.

[0080] As an anomaly handling mechanism, if a certain type of equipment under test lacks complete structural parameters, such as a switchgear whose service life exceeds the preset limit and whose support elastic modulus cannot be accurately obtained, the structural dynamics section can call the template parameter library of the same type of equipment and make corrections according to a conservative amplification factor, such as uniformly increasing the predicted stress by 10%.

[0081] If there is a calibration deviation in the thermal node parameters, the thermal resistance and thermal capacity can be corrected in reverse by recording the historical load temperature rise, so that the deviation between the predicted surface temperature and the measured value under normal operating conditions is kept within a preset range, such as not exceeding 5℃; if the deviation continues to exceed the limit, the mapping result is suspended and the empirical limit curve is switched as a temporary replacement.

[0082] Before the system runs, the structural material parameters such as the cross-sectional dimensions, insulation layer thickness, support spacing, and material yield of the electrical equipment under test are preloaded. Assuming that a phase-to-phase short circuit occurs, the thermal network model quickly determines that the conductor body will experience an instantaneous temperature rise and rapidly absorb heat, while the insulation layer temperature rise will lag behind. At the same time, the mechanical equivalent model determines that a stress peak will appear at the middle support position.

[0083] The purpose of this mechanism is to replace computationally intensive physical simulations with equivalent models that can run in real time, thereby enabling rapid and interpretable output of temperature rise and stress curves.

[0084] Further, step S4 includes:

[0085] The remaining temperature rise margin data is determined based on the current actual temperature extracted from the initial thermal reference data and the preset equipment tolerance limit temperature.

[0086] Determine the yield strength margin data by combining the current actual temperature and the preset structural material parameters;

[0087] The thermal disaster risk level of the maximum temperature rise peak relative to the remaining temperature rise margin data and the force disaster risk level of the maximum stress peak relative to the yield limit margin data are obtained from the expected temperature rise curve data respectively, and the real-time damage disaster risk index is determined by combining them.

[0088] This embodiment provides a risk quantification calculation mechanism; specifically, in the aforementioned scenario, temperature rise curves and stress curves alone are insufficient to directly drive tripping, because the tolerable boundaries are not the same for different equipment, different seasons, and different aging states.

[0089] If the remaining thermal and mechanical tolerance margins of the equipment are not included in the calculation, the same predicted temperature rise of 31°C will be incorrectly considered as equally dangerous under light load in winter and full load in summer. Therefore, this embodiment compares the dynamic margin with the predicted peak value and then synthesizes them into a unified risk index.

[0090] Specifically, in step S41, the current actual temperature is extracted from the initial thermal data of the temperature sensor. In a specific embodiment, if the current temperature is 78°C and the device's tolerance limit temperature is 130°C, then the remaining temperature rise margin is 52°C. The difference here can be calculated by subtracting the current value from the upper limit, which gives the space that can still be increased, rather than the cumulative value of the temperature rise of the device. If the same device is only 45°C in a low-temperature environment, then the corresponding margin will become 85°C. This means that under the condition of inputting the same short-circuit energy integral, the real-time damage risk of the device will be reduced to within the preset safety assessment range.

[0091] In step S42, structural material parameters are introduced to perform temperature correction on the mechanical margin. It should be noted that the standard yield limit here does not refer to the physical limit at normal room temperature, such as 20°C, but is used to simplify the underlying calculation architecture and linearly extrapolate the temperature derating performance curve of the material to the theoretical benchmark value calibrated at 0°C.

[0092] Based on this premise, assuming the standard yield strength of the copper busbar connection component under the above settings is 150 MPa, the temperature attenuation coefficient is 0.5 MPa / ℃, and the current temperature is 78℃, then the yield strength margin after temperature reduction can be directly calculated as follows: After this treatment, the higher the temperature, the lower the upper limit of mechanical tolerance, which is in line with the engineering law that the strength of conductive materials and fastening structures decreases at high temperatures.

[0093] This calculation logic, which directly uses the product of the actual temperature and the temperature decay coefficient and then subtracts them, eliminates the parameter dependence on the temperature difference of the additional reference environment and the underlying calculation steps, reduces the computational load of the embedded processor, and ensures the millisecond-level real-time performance of the fast-break measurement.

[0094] In step S43, the prediction results are dimensionless; if the maximum temperature rise peak obtained by twin mapping is 31℃, the risk degree of thermal disaster is 31 / 52; if the maximum stress peak is 91MPa, the risk degree of force disaster is 91 / 111; through this ratioization process, the data of two different dimensions, temperature rise and stress, are brought into a unified evaluation framework.

[0095] In step S44, thermal disaster weighting coefficients are set and linearly weighted. Assuming that for switchgear that is sensitive to both insulation aging and support deformation, the weighting coefficient for thermal disasters is 0.45, and the weighting coefficient for force-induced disasters is 0.55, then the real-time damage-induced disaster risk index can be calculated as follows: We get 0.7192;

[0096] If another type of equipment is more prone to insulation breakdown, the weighting coefficient corresponding to thermal disasters can be increased by a preset proportional step; if the structural resistance of the equipment is less than the preset impact load and mechanical damage dominates during short circuits, the weighting coefficient corresponding to force disasters can be increased; through weighting configuration, the evaluation logic of different equipment can maintain a unified framework while reflecting differences.

[0097] In the abnormal data processing mechanism, if the remaining temperature rise margin calculation result is less than or equal to 0, it indicates that the equipment has approached or reached the thermal limit before the failure occurs. At this time, the thermal risk ratio calculation is no longer performed to avoid the denominator being zero or negative, which would cause the calculation to fail. Instead, the thermal disaster risk degree is directly set to 1. Similarly, if the yield limit margin is less than or equal to 0, the force disaster risk degree is directly set to 1.

[0098] If the predicted peak value is negative or an obviously unreasonable zero value, such as a model malfunction causing the maximum stress peak value to be -5MPa, then this item is considered an invalid output, and the stable result of the previous time window or a conservative high-value strategy is called; if the sum of the thermal disaster weight coefficients is not equal to 1, then normalization is performed before calculation. For example, the weight coefficients corresponding to thermal disasters (0.6) and force disasters (0.6) will be automatically corrected to 0.5 and 0.5 respectively.

[0099] When the electrical equipment under test is operating at high temperatures continuously, its base temperature has exceeded the historical average. When a phase-to-phase short circuit occurs, although the predicted peak temperature rise is similar to that of the previous few days, the current temperature is higher, the remaining temperature rise margin is compressed, and the mechanical margin after temperature correction also decreases, ultimately leading to a significant increase in the comprehensive risk index.

[0100] The purpose of this step is to convert the predicted absolute temperature rise and absolute stress values ​​into a dynamic risk ratio relative to the current remaining capacity of the equipment, thereby achieving risk assessment coupled with the real-time status of the equipment.

[0101] Furthermore, in step S5, generating a monitoring status signal indicating that the risk of disaster has not been reached includes generating a blocking instantaneous disconnection command and a continuous monitoring command.

[0102] This embodiment provides a blocking and continuous monitoring mechanism for low-risk scenarios. Specifically, in the aforementioned scenario, if the system only has the logic of tripping when the risk exceeds the threshold and maintaining the status quo when the risk does not exceed the threshold, then in cases such as lightning coupling spikes, capacitor bank switching oscillations, or large motor closing inrush current, although it does not operate in this round, the traditional protection channel may still be falsely triggered due to instantaneous overcurrent in the next moment. Therefore, when the risk of disaster has not been reached, it is necessary to explicitly output blocking instantaneous trip commands and continuous monitoring commands to form a complete low-risk handling path.

[0103] Specifically, the instantaneous blocking instruction is used to shield direct tripping triggers based on a single instantaneous current threshold within a preset short-time window, while the continuous monitoring instruction requires the sensing and computing modules to maintain high-frequency sampling and rolling evaluation.

[0104] In a specific embodiment, if a transient shock causes the risk index to reach 0.42 in the first time window, which is below the 0.85 threshold, the system immediately issues a 10ms lockout and continues to recalculate the risk at high frequency. If the risk index drops to 0.20 in the second time window, monitoring continues until the transient ends. If the risk index rises to 0.88 in the third time window, the lockout is automatically released and the system switches to the instantaneous disconnection channel. Thus, the lockout is not a permanent prohibition of action, but rather a means to buy time for the model to complete the consequence determination.

[0105] As an anomaly handling mechanism, if the communication link fails during the lockout period, causing the continuous monitoring results to fail to be refreshed on schedule, the lockout must not be maintained indefinitely. In this case, a maximum lockout duration can be set, such as 10ms or 20ms. Once no new risk value is input after this duration, the system exits the lockout and enters conservative protection mode. If a fault repeatedly stays near the threshold in multiple consecutive time windows, such as fluctuating between 0.83, 0.84, 0.85, and 0.84, a hysteresis range can be introduced, for example, setting the action threshold to 0.85 and the release threshold to 0.80, to prevent frequent jittering of the lockout and action.

[0106] When the electrical equipment under test is subjected to external high-frequency transient induction, the broadband sensor captures a high-amplitude short pulse, but the heat integral is very small and the stress duration is also extremely short, so the overall risk is always below the threshold. Therefore, the system issues a blocking instantaneous trip command to prevent the circuit breaker from tripping erroneously for transient impacts that have not reached the disaster risk threshold, while maintaining continuous monitoring until the transient ends.

[0107] The purpose of this mechanism is to clearly implement a strategy of temporarily not tripping the circuit breaker while maintaining monitoring under low-risk, short-term impacts, thereby improving power supply continuity.

[0108] Furthermore, after generating the adaptive instantaneous decision instruction, it also includes:

[0109] Based on the real-time damage risk index and high-frequency transient electrical waveform data, the incremental data of life loss caused by this short-circuit impact on the electrical equipment under test is evaluated, and a condition-based maintenance assessment report is generated accordingly.

[0110] This embodiment provides a life loss assessment and condition-based maintenance output mechanism after instantaneous tripping. Specifically, in the aforementioned scenario, if the system ends the fault handling immediately after tripping, although the protection action is completed, the equipment operation and maintenance system cannot obtain quantitative data on the equipment life loss caused by this impact, the power outage maintenance requirements, or the basis for determining the replacement of support components or fasteners. Therefore, by further calculating the life loss increment after instantaneous tripping and generating a maintenance report, the protection action and the equipment operation and maintenance closed loop can be connected.

[0111] Specifically, the increase in life loss is strictly based on the peak value of the real-time damage risk index in this fault, as well as the comprehensive calculation of time-domain characteristic indicators such as the duration of short-circuit current and the number of high-frequency distortion oscillations extracted directly from high-frequency transient electrical waveform data.

[0112] In a specific embodiment, it is assumed that the system performs time-domain analysis on the high-frequency transient electrical waveform data and obtains that the actual continuous discharge duration of the short-circuit current waveform is 15ms, and extracts the number of high-frequency distortion oscillations in the transient waveform as 4 times; at the same time, the system obtains that the peak value of the real-time damage and disaster risk index of this short circuit is 0.93.

[0113] The system can call the preset two-dimensional life loss rules: based on the risk index peak of 0.93 in the range of 0.90 to 1.00, it is judged as a high impact level and assigned a basic life loss amount of 0.8%;

[0114] Based on the time-domain characteristics directly extracted from high-frequency transient electrical waveform data, these characteristics are compared and evaluated with preset reference waveform characteristics, such as a reference duration of 10ms and two reference oscillations. For every 1ms exceeding the reference duration, an additional 0.05% lifetime loss is added; for each additional distorted oscillation, an additional 0.02% lifetime loss is added. Therefore, the additional loss calculated based on the underlying waveform data is... ;

[0115] The total lifespan loss increment caused by this short-circuit impact is calculated as follows: The incremental life loss here does not have to be the absolute life value in materials science, but can also be a relative depreciation index used for operation and maintenance ranking. By directly introducing the underlying high-frequency transient electrical waveform data, the life calculation can accurately capture the damage consequences of local high-frequency discharge and long-term heat accumulation that are not fully reflected in the macro equivalent model.

[0116] In terms of report generation, the output can include at least the fault occurrence time, fault phase, current base temperature, peak risk index, waveform duration, waveform distortion count, and recommended maintenance level. Continuing with the above example, if the life loss increment reaches 1.09%, the system can mark the maintenance recommendation as conducting a busbar connection point tightness check and insulation surface thermal damage review after a planned power outage; if the loss is only 0.1%, it can only be recommended to include it in routine inspections.

[0117] In the abnormal data processing mechanism, if the circuit breaker has tripped during the current fault, but some high-frequency transient waveforms are not fully preserved due to insufficient storage window, the lifetime loss calculation can be conservatively calculated based on the obtained risk index peak value and the lower limit of the preserved waveform estimation characteristics, and an incomplete data mark will be added to the report.

[0118] If multiple short-circuit impacts occur within the same day, the damage can be accumulated to form the daily cumulative lifespan loss; when the cumulative value exceeds a preset threshold, such as 3%, the maintenance level will be automatically upgraded; if the equipment already has a historical aging record, the current damage can be added to the historical cumulative value instead of being viewed separately each time;

[0119] Assuming that in a phase-to-phase short circuit accident, after the circuit breaker completes the instantaneous trip, the system determines that although the impact has been cleared in time, the high temperature inside the cabinet at the time of the accident, the high risk index, and the long duration of the high-frequency transient waveform have caused the calculated life loss increment to reach the preset high-level deterioration range. The system then automatically pushes maintenance suggestions containing detailed equipment damage levels to the monitoring system or terminal equipment.

[0120] The purpose of this mechanism is to transform a protection event into quantifiable equipment health information, thereby enabling the extension from fault isolation to condition-based maintenance.

[0121] Furthermore, after acquiring high-frequency transient electrical waveform data and initial thermal reference data, the process also includes:

[0122] The high-frequency transient electrical waveform data and initial thermal reference data are time-stamped and pre-processed by an edge computing chip deployed on the smart sensor.

[0123] This embodiment provides a sensor-end edge preprocessing mechanism. Specifically, in the aforementioned scenario, if the high-frequency electrical waveform and temperature reference are uploaded to the station control layer by different sensors and then synchronized, communication delay will be introduced, and time stamp drift is likely to occur within the short-circuit millisecond window, resulting in a mismatch between the temperature state and the electrical impact, affecting subsequent risk calculations. Therefore, this embodiment deploys an edge computing chip at the smart sensor end to complete timestamp synchronization and noise reduction preprocessing, and sends the results to subsequent steps.

[0124] Specifically, timestamp synchronization can employ a unified clock source or a local alignment method. In a specific embodiment, if the voltage and current channels are sampled at 0ms, 1ms, 2ms, and 3ms, while the temperature channel actually records at -0.2ms, 0.8ms, 1.8ms, and 2.8ms due to sampling delay, the edge chip can shift the entire temperature channel by 0.2ms, making each time point correspond to 0ms, 1ms, 2ms, and 3ms. If the temperature channel sampling frequency is even lower, for example, one point every 2ms, nearest neighbor hold or linear interpolation can be used to ensure that the 1ms and 3ms positions also have usable thermal values.

[0125] Noise reduction preprocessing may include removing glitch points, power frequency background separation, and envelope smoothing; for example, if the original values ​​of a current sequence are 2, 5, 30, and 7, where the amplitude of 30 exceeds the previous and subsequent sampling points and lasts for only one sampling interval, the edge computing chip deployed at the smart sensor end identifies it as high-frequency pulse interference through median filtering and corrects it to a smooth value based on a preset filtering algorithm.

[0126] If the value is not an isolated spike and remains high in subsequent events, it is retained as a true fault rising edge. For temperature signals, a moving average can be used to remove thermoelectric noise. For example, the original readings of 78.1℃, 79.5℃, and 78.2℃ can be smoothed to obtain a stable thermal reference around 78.6℃.

[0127] The reason for preprocessing is that subsequent energy integration is sensitive to waveform spikes. If noise reduction preprocessing is not performed, abnormal sampling points will cause the calculation results of instantaneous power and cumulative energy to exceed the real physical boundary, causing a positive numerical deviation in the risk index. If the time scale is not aligned, the low temperature before the fault may be mistaken for the real thermal state when the fault occurs, resulting in an overestimation of the margin.

[0128] As an anomaly handling mechanism, if the edge chip detects a clock deviation exceeding the preset tolerance, such as exceeding one complete sampling cycle, fine-grained alignment will no longer be performed. Instead, synchronization failure will be directly marked and the upper-level protection logic will be notified to switch to the backup strategy. If the number of effective samples after noise reduction is lower than the preset ratio, such as less than 6 effective points in a 10-point window, fine-grained risk values ​​will not be output in this round, and the output data quality will be marked as insufficient. If the difference before and after filtering is too large, such as the cumulative energy decrease exceeding 50%, both the original waveform and the processed waveform can be retained for post-event traceability and parameter verification.

[0129] Suppose that during an operational failure, the current sensor captures a high-frequency transient of a phase-to-phase short circuit, while the temperature sensor experiences a preset time phase lag due to its sampling frequency being lower than that of the current sensor; the edge chip first aligns the two types of data and removes an occasional noise spike before performing energy integration and risk calculation.

[0130] The purpose of this mechanism is to ensure that the data entering the core algorithm is timely and of usable quality, thereby improving the stability of subsequent risk assessment and rapid decision-making.

[0131] A phase-to-phase short-circuit fault instantaneous overcurrent measurement system based on intelligent sensors includes:

[0132] The data sensing module is used to collect high-frequency transient electrical waveform data of the electrical equipment under test when a phase-to-phase short circuit occurs, as well as the initial thermal reference data of its surface.

[0133] The transient integration module is used to acquire instantaneous energy integral data and electrodynamic impact envelope data based on high-frequency transient electrical waveform data;

[0134] The twin mapping module is used to input instantaneous energy integral data and electrodynamic impact envelope data into the multiphysics twin mapping model to obtain the expected temperature rise curve data and deformation stress curve data of the electrical equipment under test;

[0135] The risk assessment module is used to obtain the remaining temperature rise margin data and yield limit margin data based on the initial thermal baseline data, the preset equipment tolerance limit temperature and structural material parameters, and compare them with the maximum temperature rise peak in the expected temperature rise curve data and the maximum stress peak in the deformation stress curve data to obtain the real-time damage disaster risk index.

[0136] The decision execution module is used to compare the real-time damage risk index with the preset risk threshold, and generate an adaptive quick-break decision command or a monitoring status signal indicating that the risk of disaster has not been reached based on the comparison result.

[0137] This embodiment provides a phase-to-phase short-circuit fault instantaneous overcurrent measurement system based on a modular architecture, such as... Figure 2 As shown; specifically, in the aforementioned substation, the system can be deployed in a single-sided switchgear or managed as a station-level device for multiple-sided switchgear, but each module works continuously around the same data link, avoiding the isolated operation of traditional protection and status monitoring.

[0138] Specifically, the data sensing module is responsible for accessing the broadband current and voltage sensor and the non-invasive miniature temperature sensor; the output of this module is not a simple single value, but a set of data frames organized according to a unified time axis; in a specific embodiment, a data frame can be represented as a voltage sequence, a current sequence and a temperature reference corresponding to the same time window; the data frame is sent to the transient integration module.

[0139] After receiving the above data frame, the transient integration module generates two types of intermediate results: energy integral sequence and current square envelope; these two results represent thermal excitation and mechanical excitation, respectively; the data interface between modules can be implemented using shared memory, on-chip bus or internal register area of ​​protection device to compress processing delay;

[0140] The twin mapping module further maps the intermediate results to the physical response of the equipment; for the same set of inputs, its output can be the expected temperature rise curve and the deformation stress curve; the risk assessment module then calculates the remaining temperature rise margin and yield limit margin based on the equipment parameter library and the current thermal state, and synthesizes the peak values ​​of the two curves into a risk index after normalization.

[0141] The decision execution module is located at the end of the link. After receiving a single risk index, it will take clear action. If the index reaches the threshold, it will generate a circuit breaker trip command. If the threshold is not reached, it will output a monitoring status signal. This module can also record the reason for this round of action for easy traceability. For example, it can record whether the thermal disaster risk level, the force disaster risk level, and the comprehensive real-time damage disaster risk index have reached the action threshold.

[0142] As an exception handling mechanism, if any preceding module fails to output a result within a preset time limit, the following module should not wait indefinitely. For example, if the twin mapping module times out due to a failure to read the parameter library, the risk assessment module can receive the timeout flag and call the conservative experience curve. If the data sensing module detects that a sensor is offline, the system can automatically degrade to the traditional instantaneous overcurrent protection mode. If the output range between modules is abnormal, such as a risk index greater than 1.5, the decision execution module will first perform truncation correction or directly judge it as abnormal data to avoid sending erroneous values ​​directly to the circuit breaker.

[0143] During the operation of the electrical equipment under test, once a phase-to-phase short circuit occurs, each module will process the data continuously in the order of acquisition, integration, mapping, evaluation, and decision-making. Due to the clear system structure, it is easy to implement in an embedded manner and also easy to replicate and deploy in other switch cabinets in the same station.

[0144] The purpose of this system is to translate the aforementioned methods and steps into a chain of functional modules that can be implemented in engineering, thereby achieving closed-loop control from signal sensing to circuit breaker execution.

[0145] Furthermore, the decision execution module includes a logical judgment unit;

[0146] The logic judgment unit is used to generate a lockout instantaneous disconnection command and a continuous monitoring command when the real-time damage-causing risk index is less than a preset risk threshold, as a monitoring status signal that the disaster risk has not been reached.

[0147] This embodiment provides a logic judgment unit set inside the decision execution module. Specifically, in the aforementioned system architecture, if the decision execution module is only used as a simple comparator, it can only output a non-trip control state when the risk is below the threshold, and cannot further constrain the malfunction of other protection channels, nor can it maintain continuous observation of the subsequent evolution of the transient state. Therefore, this embodiment refines the logic judgment unit in this module to output a combination of two instructions with control meaning in low-risk situations.

[0148] Specifically, the logic judgment unit receives at least two inputs: a real-time damage risk index and a current judgment status flag. Its output can be categorized into three states: action state, lockout monitoring state, and abnormal conservative state. In a specific embodiment, if the threshold is 0.85, when the input risk index is 0.62 and the status is normal, the unit outputs a lockout instantaneous trip command and a continuous monitoring command; when the input risk index is 0.91, the unit outputs a trip action command; when the input risk value is missing or the status is abnormal, the unit outputs a conservative protection command. This avoids the unit being merely in a passive waiting state when the risk is low.

[0149] Furthermore, the logic judgment unit can be set with time hold and hysteresis control; taking the above threshold as an example, when the risk index first falls below 0.85, the lockout instantaneous trip command can be held for 5ms; if the risk value remains below 0.80 during the holding period, the lockout can be released naturally and exit high-frequency monitoring; if it rises again to 0.85 or above during the holding period, the lockout is immediately revoked and a trip command is issued; this design can prevent the control signal from frequently flipping when it is near the boundary.

[0150] As an abnormal handling mechanism, if the front-end module does not return a new risk value within the specified time after the continuous monitoring command is issued, the logic judgment unit will not extend the interlocking, but will exit the interlocking protection according to the timeout strategy. If the same device enters the interlocking monitoring state multiple times in a short period of time, such as more than 3 times within 1 minute, an abnormal disturbance frequent flag can be generated to prompt the operation and maintenance side to check for external impact sources of the line or potential insulation deterioration inside the cabinet.

[0151] When subjected to continuous external transient induced impacts, the logic judgment unit repeatedly judges that the risk value is lower than the action threshold, and therefore repeatedly outputs short-term blocking to the circuit breaker side, while commanding the upstream sampling link to maintain high-frequency monitoring; when one of the impacts evolves into a real continuous fault, the risk value crosses the threshold, the blocking is immediately revoked and switched to tripping.

[0152] The purpose of this mechanism is to refine non-actions under low-risk conditions into a combination of interlocking and continuous monitoring with clear control implications, thereby improving the stability and controllability of the system in complex disturbance environments.

[0153] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for instantaneous overcurrent measurement of phase-to-phase short-circuit faults based on intelligent sensors, characterized in that, Includes the following steps: S1, collect high-frequency transient electrical waveform data of the electrical equipment under test when a phase-to-phase short circuit occurs and the initial thermal reference data of the surface of the electrical equipment under test through intelligent sensors; S2, obtain instantaneous energy integral data and electrodynamic impact envelope data based on the high-frequency transient electrical waveform data; S3, input the instantaneous energy integral data and the electrodynamic impact envelope data into a multiphysics twin mapping model that includes a resistive-capacitive thermal network model and a structural dynamics equivalent model, and respectively map to obtain the expected temperature rise curve data and deformation stress curve data; S4. Based on the initial thermal reference data, the preset equipment tolerance limit temperature and the preset structural material parameters, obtain the remaining temperature rise margin data and yield limit margin data. Extract the maximum temperature rise peak value in the expected temperature rise curve data and the maximum stress peak value in the deformation stress curve data respectively. Obtain the thermal disaster risk degree of the maximum temperature rise peak value relative to the remaining temperature rise margin data and the force disaster risk degree of the maximum stress peak value in the deformation stress curve data relative to the yield limit margin data. And comprehensively evaluate to obtain the real-time damage disaster risk index. S5, compare the real-time damage-causing risk index with the preset risk threshold, and when the real-time damage-causing risk index is greater than or equal to the preset risk threshold, generate an adaptive quick-break decision instruction. When the real-time damage risk index is less than the preset risk threshold, a monitoring status signal indicating that the risk of disaster has not been reached is generated.

2. The method for instantaneous overcurrent measurement of phase-to-phase short-circuit faults based on intelligent sensors according to claim 1, characterized in that, In step S3, the multiphysics twin mapping model includes a resistive-capacitive thermal network model and a structural dynamics equivalent model; The resistive-capacitive thermal network model is used to map the instantaneous energy integral data to the expected temperature rise curve data; the structural dynamics equivalent model is used to map the electrodynamic impact envelope data to the deformation stress curve data.

3. The method for instantaneous overcurrent measurement of phase-to-phase short-circuit faults based on intelligent sensors according to claim 1, characterized in that, Step S4 includes: The remaining temperature rise margin data is determined based on the current actual temperature extracted from the initial thermal reference data and the preset equipment tolerance limit temperature. The yield strength margin data is determined by combining the current actual temperature and the preset structural material parameters; The thermal disaster risk level of the maximum temperature rise peak relative to the remaining temperature rise margin data and the force disaster risk level of the maximum stress peak relative to the yield limit margin data are obtained respectively, and then the real-time damage disaster risk index is determined by combining them.

4. The method for instantaneous overcurrent measurement of phase-to-phase short-circuit faults based on intelligent sensors according to claim 1, characterized in that, In step S5, generating a monitoring status signal indicating that the risk of disaster has not been reached includes generating a blocking instantaneous disconnection command and a continuous monitoring command.

5. The method for instantaneous overcurrent measurement of phase-to-phase short-circuit faults based on intelligent sensors according to claim 1, characterized in that, After generating the adaptive quick-break decision instruction, the method further includes: Based on the real-time damage risk index and the high-frequency transient electrical waveform data, the incremental data of life loss caused by this short-circuit impact on the electrical equipment under test is evaluated, and a condition-based maintenance assessment report is generated accordingly.

6. The method for instantaneous overcurrent measurement of phase-to-phase short-circuit faults based on intelligent sensors according to claim 1, characterized in that, After acquiring the high-frequency transient electrical waveform data and the initial thermal reference data, the method further includes: The high-frequency transient electrical waveform data and the initial thermal reference data are time-stamped and pre-processed by an edge computing chip deployed on the smart sensor.

7. A phase-to-phase short-circuit fault instantaneous overcurrent measurement system based on intelligent sensors, characterized in that, To implement the method according to any one of claims 1 to 6, comprising: The data sensing module is used to collect high-frequency transient electrical waveform data of the electrical equipment under test when a phase-to-phase short circuit occurs, as well as the initial thermal reference data of its surface. The transient integration module is used to acquire instantaneous energy integration data and electrodynamic impact envelope data based on the high-frequency transient electrical waveform data; The twin mapping module is used to input the instantaneous energy integral data and the electrodynamic impact envelope data into the multiphysics twin mapping model to obtain the expected temperature rise curve data and deformation stress curve data of the electrical equipment under test; The risk assessment module is used to obtain the remaining temperature rise margin data and yield strength margin data based on the initial thermal baseline data, the preset equipment tolerance limit temperature and structural material parameters, and compare them with the maximum temperature rise peak in the expected temperature rise curve data and the maximum stress peak in the deformation stress curve data to obtain the real-time damage disaster risk index. The decision execution module is used to compare the real-time damage risk index with a preset risk threshold, and generate an adaptive quick-break decision command or a monitoring status signal indicating that the risk of disaster has not been reached based on the comparison result.

8. A phase-to-phase short-circuit fault instantaneous overcurrent measurement system based on intelligent sensors according to claim 7, characterized in that, The decision execution module includes a logic judgment unit; The logic judgment unit is used to generate a lockout instant disconnection command and a continuous monitoring command when the real-time damage-causing risk index is less than the preset risk threshold, as a monitoring status signal that the disaster risk has not been reached.

Images