A method and system for evaluating the operating status of intelligent power metering boxes

Abstract

This invention relates to the field of electrical measurement technology, and more particularly to a method and system for assessing the operational status of intelligent energy metering boxes. The method includes: acquiring the connection section under assessment and simultaneously collecting differential voltage, current, and surface temperature of the section; identifying natural load switching events, calculating the fundamental active power loss and current square index within the stability window before and after the event, and obtaining evidence of the event's resistive slope; matching with the baseline of the operating condition grid frozen during the acceptance period, determining a conservative range for degradation changes by combining measurement link errors and residual thermal model errors, and performing range compatibility verification using two independent events; and outputting status information and maintenance recommendations based on the measurement points, sampling links, calibration, baseline matching, and verification results. This invention improves the accuracy of operational status assessment for intelligent energy metering boxes.

Description

Technical Field

[0001] This invention relates to the field of electrical measurement technology, and in particular to a method and system for evaluating the operating status of an intelligent power metering box. Background Technology

[0002] It should be noted that smart energy metering boxes are installed in low-voltage power distribution circuits. The main incoming line, branch terminals, circuit breaker terminals, and energy meter terminals within the box are responsible for energy transmission, forming multiple series-connected conductive sections. After long-term operation, the contact surfaces of these conductive sections may develop a thickened oxide film, the connecting bolts may creep and loosen, and fretting wear may occur at the connection points. This gradually increases the resistive losses of the conductive sections, potentially leading to localized overheating, wiring faults, or abnormal energy metering in severe cases. Therefore, accurately assessing the operating status of the conductive sections in smart energy metering boxes is crucial for timely detection of connection performance degradation and ensuring the reliability of low-voltage power distribution circuits.

[0003] Therefore, existing technologies typically determine whether there are any abnormalities in the conductive connection section by collecting voltage drop, load current, or temperature rise at the connection point across the conductive connection section and comparing the measurement results with preset fixed thresholds. Since the load current in an actual power supply circuit is constantly fluctuating, the voltage drop across the conductive connection section includes both the normal voltage drop component determined by the load current and the additional loss component caused by connection performance degradation. Directly comparing the voltage drop or temperature rise at different times with thresholds makes it difficult to distinguish between load condition changes and connection performance degradation. Some technical solutions use long-term, continuously updated statistical benchmarks to adapt to load changes, but the slowly accumulating oxidation resistance increase in the connection may be gradually written into the new statistical benchmarks, resulting in a lack of a fixed reference for comparing long-term degradation with the initial acceptance state.

[0004] Specifically, during the actual operation of smart energy metering boxes, a single load switch may be affected by random noise, channel phase error, power frequency variation, spectral leakage, and temperature differences. If a conclusion of conductive connection degradation is drawn solely based on a single measurement result, occasional measurement interference may be misjudged as connection performance degradation. Furthermore, the actual voltage tapping path may cross multiple contact surfaces, short conductors, and the extension of circuit breaker terminals. Directly interpreting the series measurement results of the complete conductive path between two tapping points as the material parameters of a single contact will lead to inconsistencies between the status conclusion and the actual measurement range. For conductive connection sections without measurement points, with mismatched connection currents, or with failed sampling link calibration, there is also a lack of physical basis for outputting a definitive status conclusion. Therefore, the technical problem to be solved is: how to use conductive connection sections with clear physical boundaries as the basic evaluation unit, utilize natural load changes to form resistive status evidence, employ a frozen baseline during the acceptance period to prevent slow degradation from being absorbed by the rolling baseline, and combine conservative error envelopes and independent event verification to reduce misjudgments caused by single measurement interference. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method and system for evaluating the operational status of intelligent power metering boxes.

[0006] In a first aspect, the present invention provides a method for evaluating the operating status of an intelligent energy metering box, employing the following technical solution: A method for evaluating the operational status of a smart energy metering box, comprising the following steps: The evaluation connection segment in the smart energy metering box is determined based on the physical starting point and physical ending point registered before commissioning. A differential voltage sampling channel is established at the fixed conductor positions at both ends of the evaluation connection segment. A current sampling channel and a surface temperature sampling channel corresponding to the evaluation connection segment are also established. The differential voltage, current, and surface temperature of the connection segment are sampled synchronously. Identify natural load switching events from synchronous sampling data. Within the pre-event stability window and post-event stability window of the natural load switching event, calculate the fundamental active power loss index and the current square index based on the complete power frequency cycle. Determine the resistive slope evidence of the event based on the difference between the fundamental active power loss index and the current square index before and after the event. Based on the connection segment surface temperature, current endpoints, and load thermal history of the natural load switching event, the resistive slope evidence of the event is matched with the operating condition baseline formed and frozen during the acceptance period. The conservative possible range of the event resistive slope and the conservative possible range of the degradation change relative to the frozen baseline are determined based on the measurement link error and residual thermal model error. When the lower bound of the conservative possible range of the degradation change exceeds the allowable new resistive change, the evaluated connection segment is identified as a suspected connection segment electrical performance degradation state, and is verified using another independent natural load switching event. When the lower bound of the conservative possible range of the degradation change of both independent natural load switching events exceeds the allowable new resistive change and the two ranges are compatible, the evaluated connection segment is identified as a verified connection segment electrical performance degradation state. Based on the measurement point layout status, sampling link status, calibration status, operating condition baseline matching status, event evidence validity status, and independent event verification results of the evaluated connection segment, the status information of the evaluated connection segment is output.

[0007] Preferably, the evaluated connection segment is a complete conductive path extending from one fixed conductor position to another, the complete conductive path including at least one of a screw crimp surface, a terminal piece, a transition short conductor, and an extension portion of a circuit breaker post; the status information indicates the change in the electrical performance of the complete conductive path relative to the acceptance baseline, and does not indicate the pure contact resistance of a single contact.

[0008] Preferably, the differential voltage sampling channel is connected to the fixed conductor positions at both ends of the evaluated connection segment via a high-impedance differential tap line, and the input impedance of the high-impedance differential tap line is at least two orders of magnitude higher than the equivalent impedance of the main circuit of the evaluated connection segment; the current sensing element of the segment current sampling channel is disposed on the branch conductor connected in series with the evaluated connection segment, and the temperature sensor of the connection segment surface temperature sampling channel is disposed on the surface of the evaluated connection segment; the differential voltage sampling channel and the segment current sampling channel are synchronously sampled according to the same phase reference under the control of a hardware synchronous trigger source.

[0009] Preferably, the natural load switching event is determined based on the effective value of the fundamental current per cycle or the sliding root mean square envelope within the effective bandwidth; when the current in this segment shifts from the first stable level to the second stable level, and there are stable intervals that satisfy the stability criterion before and after the load transition process, the two stable intervals are respectively determined as the stable window before the event and the stable window after the event; only one event resistive slope evidence is generated for a continuous load transition process.

[0010] Preferably, both the pre-event stabilization window and the post-event stabilization window include an integer number of complete power frequency cycles that do not overlap; for each complete power frequency cycle, a current channel unsaturation flag, a voltage channel unsaturation flag, a sample integrity flag, a frequency lock flag, and a spectrum leakage controlled flag are set; only complete power frequency cycles with all flags valid are used to extract the fundamental differential voltage phasor and the fundamental current phasor of this segment.

[0011] Preferably, the operating condition baseline uses the evaluated connection segment identifier, calibration version, connection segment surface temperature range, recent load thermal history range, fundamental current low end point, fundamental current high end point, and verified waveform range as search conditions, and saves the lower bound and upper bound of the conservative possible value of the frozen baseline slope obtained during the acceptance period; when a natural load switching event cannot be matched with the corresponding operating condition baseline, the natural load switching event is determined as an operating condition mismatch event, and the operating condition range extrapolation result or global average slope is not used as a substitute.

[0012] Preferably, the measurement link error includes at least one of voltage gain error, current gain error, phase error, common-mode residual, synchronization timescale error, spectral leakage residual, channel noise, and drift upper bound within the calibration validity period; when the correlation between different error components is not clear, the conservative possible value range of the event resistivity slope is not narrowed in a statistically independent manner.

[0013] Preferably, the allowable additional resistive change is determined based on the ratio of the allowable additional fundamental active power loss limit to the square of the reference current setting; the two independent natural load switching events used for verification satisfy the following conditions: different event numbers, different event occurrence times, and the corresponding pre-event stability windows and post-event stability windows do not overlap.

[0014] Preferably, when the evaluated connection segment has no measurement points, the sampling link calibration fails, the current in this segment does not correspond to the evaluated connection segment, the current polarity verification record is missing, the complete power frequency frequency quality does not meet the requirements, the calibration expires, the measurement drift cannot obtain a reliable upper bound, or the conservative possible value ranges of the degradation change of two independent natural load switching events are incompatible, the status of the electrical performance degradation of the connection segment has been verified will not be output, and the status of insufficient evidence or retest recommendation will be output.

[0015] Secondly, this invention provides an intelligent power metering box operation status evaluation system, which adopts the following technical solution: A smart energy meter box operation status assessment system includes a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned smart energy meter box operation status assessment method is implemented.

[0016] By adopting the above technical solution, a computer program is generated from the above-mentioned method for evaluating the operating status of an intelligent electricity metering box, and stored in a memory for loading and execution by a processor. This allows for the creation of a terminal device based on the memory and processor, making it convenient to use.

[0017] This invention has the following technical effects: By using the evaluated connection segment with a clear physical boundary as the basic evaluation unit and simultaneously collecting differential voltage, segment current, and connection segment surface temperature, this invention utilizes the differential relationship between the fundamental active power loss index and the square index of the segment current within the stability window before and after a natural load switching event to form event resistive slope evidence. This evidence is then compared with the operating condition baseline formed and frozen during the acceptance period under a conservative error envelope, and verified by the conservative possible value range of degradation changes from two independent natural load switching events. Thus, under conditions of load fluctuation, slow degradation, and single interference, this invention reliably distinguishes between normal load changes and connection segment electrical performance degradation, avoiding the absorption of slow degradation by the rolling benchmark and misjudgment by single measurements. It outputs a status conclusion that can be traced back to the location of the evaluated connection segment, the natural load switching event number, and the frozen baseline version number, ultimately improving the accuracy, robustness, and maintainability of the intelligent power metering box's operating status assessment. Attached Figure Description

[0018] Figure 1 This is a flowchart of a method for evaluating the operating status of an intelligent power metering box according to an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0020] This invention discloses a method for evaluating the operating status of an intelligent energy metering box, referring to... Figure 1 This includes steps S1-S4: S1: Register the physical start point, physical end point, voltage tapping line layout path, connection segment identifier, circuit attributes, and current polarity of the connection segment being evaluated, and simultaneously collect the differential voltage across the connection segment being evaluated, the current flowing through the connection segment, and the surface temperature of the connection segment being evaluated.

[0021] It should be noted that the smart energy metering box is installed in the low-voltage power distribution circuit. The main incoming line, branch terminals, external circuit breaker terminals, and energy meter terminals within the box are responsible for energy transmission. These components electrically constitute several series-connected conductive segments. The physical extent of each evaluated connection segment is a complete conductive path extending from one fixed conductor location to another. This path may include screw crimp surfaces, terminal pieces, transition short conductors, and the extended portion of the circuit breaker terminal.

[0022] Since differential voltage measurement actually traverses a complete conductive path consisting of multiple contact surfaces and short conductors connected in series between two voltage taps, the evaluation conclusion output in this embodiment is uniformly expressed as "the change in the electrical performance of the connection segment relative to the acceptance baseline," not as "the pure contact resistance of a single contact." Only when the evaluation object is limited to a connection segment with a clear physical start and end point can subsequent synchronous sampling data, event resistivity slope evidence, and frozen baseline have the same physical orientation. If the evaluation object does not have a defined physical boundary, a clear correspondence cannot be established between the differential voltage and the current in this segment, and any digital processing will lose its physical reference.

[0023] For example, the connection area from the terminal block of the main incoming circuit breaker to the front terminal block of the energy meter is registered as an evaluated connection segment; the evaluation result of the evaluated connection segment corresponds to the complete conductive path between the two fixed conductor positions mentioned above, and its output result cannot be split into two connection segment states without independent voltage tapping ranges.

[0024] The starting and ending points of the evaluated connection sections, as well as the contact surfaces and short conductors included in the path, were all derived from the on-site survey records, installation drawings, circuit ledgers, and manual confirmation records before commissioning.

[0025] It should be noted that before the smart energy metering box is put into operation, the measurement range of each evaluated connection segment must be registered and stored. The registration content should include at least: a unique identifier for the evaluated connection segment; a description of the physical start point location; a description of the physical end point location; the differential voltage tapping line layout path; the connection segment naming information; the circuit attribute of the metering box to which it belongs; the location of the current sampling point for this segment; the current polarity convention; the differential voltage channel number; the current channel number for this segment; and the surface temperature channel number for the connection segment.

[0026] The registration information of the assessed connection segment is invoked during each natural load event evidence generation and baseline matching to ensure that the physical boundaries of the event evidence are consistent with the physical boundaries of the frozen baseline. When the smart energy metering box undergoes physical modifications that change the voltage tapping range of the assessed connection segment, the original frozen baseline cannot be used for the modified connection segment. The modified connection segment requires re-outage acceptance, sampling link calibration, and baseline freezing.

[0027] The registration information mentioned above is derived from on-site survey records, construction and installation records, circuit design drawings, data entry results from monitoring terminal configuration tools, and on-site verification records. Physical modification information is derived from modification records, acceptance records, and re-frozen baseline records.

[0028] It should be noted that, based on the established physical boundaries, high-impedance differential tap lines are led out from the fixed conductor positions at both ends of each evaluated connection segment. These high-impedance differential tap lines guide the millivolt-level voltage drop signal of the connection segment to the differential voltage sampling channel without altering the current path of the main circuit. For ease of understanding, the high-impedance differential tap lines are equivalent to the probes of a high-input-impedance voltmeter, drawing almost no current, and therefore do not significantly change the current distribution of the main circuit.

[0029] In this embodiment, when the equivalent impedance of the main circuit of the evaluated connection section is in the range of tens of microohms to tens of milliohms, the input impedance of the high-impedance differential tap line is set to be at least two orders of magnitude higher than the equivalent impedance of the main circuit, so that the parallel shunt effect caused by the high-impedance differential tap line is within a negligible range.

[0030] Furthermore, a current-limiting resistor and a clamping protection diode are installed at the front end of the differential voltage tapping circuit, and protection parameters are designed based on the potential open-circuit voltage of the evaluated connection section. Simultaneously, electromagnetic isolation measures are implemented for the differential voltage sampling channel to prevent damage to the acquisition circuit from lightning strikes, switching operation overvoltages, or line fault overvoltages.

[0031] When the circuit breaker corresponding to the evaluated connection segment is in the open position, the evaluated connection segment no longer carries current, and its differential voltage signal no longer represents the connection segment voltage drop under power transmission conditions. At this time, the system prohibits the generation of evidence for natural load events and only retains the open / closed status indication information of the evaluated connection segment. After the circuit breaker is reclosed and a stable load current is formed, the operational status assessment is resumed.

[0032] Among them, the differential voltage signal comes from the high-resistance differential tapping line at the fixed conductor position at both ends of the evaluated connection section; the tapping line layout path comes from the pre-commissioning registration information; the current limiting resistor, clamping protection diode and electromagnetic isolation parameters come from the overvoltage risk assessment, channel protection design record and on-site installation record; the circuit breaker opening and closing position information comes from the circuit breaker auxiliary contacts, monitoring terminal switch input or system operation status record.

[0033] It should be noted that the current sampling points in this section strictly correspond to the evaluated connection section. A current sensing element connected in series with the evaluated connection section is installed on the branch conductor to which the evaluated connection section belongs. The total incoming current cannot replace the branch connection section current, nor can the branch connection section current replace the total incoming connection section current. The active power loss of the evaluated connection section depends on the actual current flowing through that connection section. If current samples from adjacent branches or other loops are used in the calculation, a mismatch will occur between the fundamental active power loss and the square of the current in this section.

[0034] In this embodiment, the current flowing into the positive tap of the differential voltage is defined as a forward current. This polarity convention is written into the configuration information of the connection segment being evaluated. When the actual installation direction of the field current sensing element is opposite to the polarity convention, a reverse mapping relationship is fixed in the configuration of the connection segment being evaluated.

[0035] Before initial energization, a low-amplitude test current is used to verify the polarity mapping relationship, and the verification record is maintained. If the polarity mapping is incorrect, the sign of the difference in fundamental active power loss before and after the event will change, causing the resistive slope of the event to lose its correct physical meaning. Therefore, in the absence of polarity verification records, the corresponding evaluated connection section is prohibited from entering the subsequent state determination process.

[0036] The current signal in this section comes from the current sensing element connected in series with the section being evaluated; the installation location, transformation ratio, and channel number of the current sensing element come from the registration information before commissioning; the polarity convention comes from the configuration rules; the reverse mapping relationship comes from the on-site wiring verification results; and the polarity verification record comes from the verification operation performed using a low-amplitude test current before the first energized commissioning.

[0037] It should be noted that the surface temperature measurement points of the connection section are set at the actual surface location of the connection section being evaluated; ambient air temperature cannot directly replace the surface temperature of the connection section. The surface temperature sensor for the connection section can be a contact temperature sensor or a non-contact infrared temperature sensor. When using an infrared temperature sensor, the emissivity of the surface being measured should be calibrated on-site, and the measurement error should be included in the subsequent error budget.

[0038] Among them, the surface temperature signal comes from a contact temperature sensor arranged close to the surface of the connection section being evaluated, or from a non-contact infrared temperature sensor calibrated by emissivity; the temperature channel number comes from the registration information before commissioning; and the infrared emissivity calibration value comes from the on-site calibration record.

[0039] It should be noted that a synchronous sampling link is established, including a differential voltage sampling channel, a current sampling channel for this segment, and a surface temperature sampling channel for the connection segment. The input of the differential voltage sampling channel comes from the high-resistance differential tap lines at both ends of the connection segment being evaluated; the input of the current sampling channel for this segment comes from a current sensing element connected in series with the connection segment being evaluated; and the input of the surface temperature sampling channel for the connection segment comes from a temperature sensor installed close to the surface of the connection segment being evaluated.

[0040] The three sampling channels share the same time scale. The analog-to-digital converters of the differential voltage sampling channel and the current sampling channel are synchronously sampled according to the same phase reference under the control of a hardware synchronous trigger source. In one prototype implementation, the hardware synchronous trigger source consists of a programmable logic device and a phase-locked loop (PLL). Based on the same source power frequency cycle estimation result, the PLL generates a sequence of trigger pulses with equal phase intervals within each complete power frequency cycle, ensuring that the differential voltage channel and the current channel sample at the same phase position.

[0041] As an example, the sampling phase deviation between the two channels is controlled within 0.05 degrees. For a 50Hz power frequency, a 0.05-degree phase deviation corresponds to a time deviation of no more than approximately 2.8 microseconds. Synchronous sampling accuracy is the physical basis for subsequent calculations of the fundamental active power loss index. If there is an uncertain phase deviation between the differential voltage and the current in this segment, a systematic error will be introduced into the calculated active power component, and this error cannot be eliminated simply by increasing the number of samples.

[0042] Among them, the synchronization time stamp comes from the hardware synchronization trigger source; the hardware synchronization trigger source can be composed of a programmable logic device and a phase-locked loop; the same source power frequency period estimation result comes from the frequency tracking module in the phase-locked loop or digital signal processor; the sampling phase deviation and corresponding time deviation come from the design record and calibration record of the hardware synchronization trigger source.

[0043] It should be noted that when the high-impedance differential tap line is far from the monitoring terminal, a remote acquisition unit with synchronous sampling capability can be set up near the evaluated connection section. The remote acquisition unit transmits a synchronization clock and sampling data to the monitoring terminal via optical fiber or shielded twisted-pair cable. The communication delay between the remote acquisition unit and the monitoring terminal is included in the synchronization timescale error budget, and its upper bound is determined by link delay measurement. When a reliable upper bound for the link delay cannot be obtained, the corresponding sampling channel is prohibited from entering the status determination process.

[0044] The communication delay between the remote acquisition unit and the monitoring terminal is derived from the link delay measurement; the synchronization timescale error budget is derived from the link delay measurement results and the hardware synchronization trigger error record.

[0045] It should be noted that the sampling link status of the evaluated connection segment must be checked. A deterministic normal or degraded conclusion is not allowed for the evaluated connection segment if any of the following conditions exist: no differential voltage measurement points are installed on the evaluated connection segment; the current sampling points for this segment do not correspond to those of the evaluated connection segment; surface temperature measurement points for the connection segment are missing; the sampling link calibration fails; current polarity verification records are missing; a reliable upper bound cannot be obtained for the synchronous sampling error; or the physical range of the evaluated connection segment changes but a frozen baseline has not been re-established.

[0046] When the above situation occurs, the evaluated connection segment will be placed in an insufficient evidence state. This state means that the system will not guess, infer, or output a normal or degenerate conclusion, but will only suggest that additional measurement points, recalibration, re-verification of polarity, or refreezing of the baseline are required.

[0047] The above inspection results were obtained from the measurement point configuration table, current sampling point registration information, temperature channel configuration, calibration record, polarity verification record, synchronization error budget record, and physical modification and baseline freezing record, respectively.

[0048] Through the above processing, this step obtains a differential voltage sampling sequence with a unified time scale, a current sampling sequence for this segment, and a surface temperature sampling sequence for the connection segment, as well as the unique identifier of the evaluated connection segment, physical boundary information, current polarity information, and sampling link status.

[0049] The data packets obtained specifically include at least: a unique identifier for the evaluated connection segment; a description of the physical start and end points; a differential voltage sampling sequence; a current sampling sequence for the current segment; a surface temperature sampling sequence for the connection segment; a unified time scale; a current polarity mapping relationship; differential voltage channel number, current channel number for the current segment, and temperature channel number; the validity status of the sampling link; a calibration version number or calibration record number; and, if a remote acquisition unit exists, an upper limit for the link delay.

[0050] S2: Identify natural load switching events from synchronous sampling data, calculate the fundamental active power loss index and the current square index based on the complete power frequency cycle within the stable window before and after the event, and determine the resistive slope evidence of the event based on the index difference before and after the event.

[0051] It should be noted that the purpose of this step is to identify naturally occurring load switching events from continuously synchronized sampling data, and to calculate the fundamental active power loss index and the current square index within two stable windows before and after the event. This generates an event resistive slope evidence that reflects the resistive state of the evaluated connection segment. This evidence is not an absolute resistance value, but rather the ratio of the change in active power loss caused by load changes to the change in the current square, used for subsequent comparison with the frozen baseline to determine whether the electrical performance of the connection segment has degraded relative to the acceptance baseline.

[0052] Specifically, natural increases and decreases in power supply load cause the voltage drop across the evaluated connection segment to change synchronously with the current in that segment, while the resistive increment signal generated by connection segment degradation is superimposed on the load fluctuation. Based on this, the load transition process of the current in that segment from the first stable level to the second stable level is identified according to the effective value of the fundamental current per cycle or the sliding root mean square envelope within the effective bandwidth.

[0053] When there is a first stable interval that satisfies the stability criterion before the load transition process and a second stable interval that satisfies the stability criterion after the load transition process, the first stable interval is determined as the pre-event stable window, the second stable interval is determined as the post-event stable window, and the load transition process is determined as a natural load switching event.

[0054] Natural load switching events do not limit the direction of current change in the current segment. Both load increases and decreases can form natural load switching events, and their direction is stored separately as an event attribute. A single continuous load transition process generates only one event resistive slope evidence; it is not allowed to split multiple local changes in the same load transition process into multiple independent pieces of evidence.

[0055] In this embodiment of the invention, the stability criterion is determined based on the relative rate of change of the effective value of the fundamental current per cycle within the stability window. When the current in this segment within the window continues to rise or fall without forming a stable plateau, the candidate load transition process cannot constitute a natural load switching event.

[0056] Among them, the effective value or moving root mean square envelope of the fundamental current per cycle is derived from the current sampling sequence and unified time scale of this segment; the stability criterion is derived from the monitoring terminal configuration parameters or manually set values; the event occurrence time and event number are automatically identified and generated by this step based on the change of current envelope; the unique identifier of the evaluated connection segment is derived from the registration information.

[0057] Furthermore, the pre-event stability window and the post-event stability window each consist of an integer number of complete power frequency cycles. Cycles containing load transition phenomena are discarded and not included in the calculation of pre- and post-event window indices. The sampling periods covered by the pre-event stability window and the post-event stability window do not overlap.

[0058] Among them, the start and end times of the pre-event stability window and the post-event stability window are derived from the identified event occurrence time and the stability platform judgment result; the integer complete power frequency cycle definition is derived from the same source power frequency cycle estimation result; the fluctuation amplitude of the current index within the window is derived from the calculation of the effective value of the fundamental wave of the current sampling sequence of this segment; the stability criterion threshold is derived from the monitoring terminal configuration parameters or manually set values; the window non-overlapping check result is generated by this step based on the window boundary comparison.

[0059] For the complete power frequency cycle involved in the calculation, equal-phase sampling is used. Equal-phase sampling means that the position of each sampling moment relative to the initial phase of the power frequency cycle remains consistent across different power frequency cycles. When the grid frequency shifts, the boundary of the complete power frequency cycle is determined based on the same-source period estimation results, and then equal-phase sampling or resampling is performed to ensure that each complete power frequency cycle covers the complete 360-degree phase range.

[0060] In one prototype, 128 equal-phase sampling points are set for each complete power frequency cycle. For a 50Hz power frequency, the corresponding sampling rate is 6400 points per second. This number of sampling points is only for illustrating the implementation method and can be adjusted according to the sampling channel bandwidth, anti-aliasing filtering characteristics, and the computing power of the digital signal processor.

[0061] Among them, the equal-phase sampling sequence comes from the hardware synchronous trigger source and phase-locked loop; the same-source power frequency period estimation result comes from the frequency tracking module; the complete cycle definition result is calculated by this step based on the same-source period estimation result; the number of equal-phase sampling points per complete cycle comes from the monitoring terminal configuration parameters; the resampling operation when the frequency shifts is executed by the digital signal processor, whose input is the original sampling sequence and the period estimation result.

[0062] Furthermore, a cycle-by-cycle quality flag is set for each complete power frequency cycle. This flag includes at least: a flag indicating that the current channel is not saturated; a flag indicating that the differential voltage channel is not saturated; a flag indicating sample integrity; a flag indicating that the power frequency is locked; a flag indicating that spectral leakage is controlled; and a flag indicating that anti-aliasing conditions are met. Only when all of the above quality flags are valid is the corresponding complete power frequency cycle allowed to participate in subsequent fundamental phasor calculations. When any quality flag is invalid, the corresponding cycle is completely discarded and not counted in the number of valid cycles in the stability window.

[0063] Among them, the current channel unsaturation indicator and the differential voltage channel unsaturation indicator are derived from the range comparison results of the sampling channel; the sample integrity indicator is derived from the sample count and missing sample detection results of the sampling sequence; the power frequency lock indicator is derived from the lock status of the phase-locked loop or frequency tracking module; the spectrum leakage controlled indicator is derived from the comparison results of the residual leakage residual calculated after frequency domain extraction and the set ratio; the anti-aliasing condition met indicator is derived from the anti-aliasing filter design record and channel bandwidth configuration.

[0064] In one implementation, a Discrete Fourier Transform (DFT) is performed on the isophase sampling sequence for each complete power frequency cycle to extract the fundamental differential voltage phasor and the fundamental current phasor for that segment. In another implementation, a Hanning window is applied to the isophase sampling sequence before performing a DFT, and adjacent spectral line interpolation is used to correct the amplitude and phase errors caused by the power frequency offset. In yet another implementation, a full-phase spectrum analysis method is employed, utilizing the phase consistency of overlapping samples to suppress spectral leakage caused by non-integer period truncation.

[0065] Regardless of which fundamental frequency extraction method is used, the residual spectral leakage after extraction should be within the range that the subsequent error envelope can cover. When the leakage-related indicators exceed the preset proportion of the rated range of a given sampling channel, the corresponding cycle will be marked as a leakage-exceeding cycle and discarded.

[0066] It should be noted that the purpose of extracting the fundamental component in this embodiment is not to calculate the long-term cumulative electrical energy, but to calculate the change in fundamental active power loss in the short-time stability window before and after a natural load switching event. Therefore, the fundamental differential voltage phasor and the fundamental current phasor must come from the same evaluated connection segment, the same complete power frequency cycle, and the same set of equal-phase sampling times.

[0067] Let the first Within the first stable window A complete power frequency cycle, the first The instantaneous value of the differential voltage at each equal-phase sampling point is The instantaneous value of the current in this segment is The number of equal-phase sampling points for each complete power frequency cycle is ; The fundamental frequency is extracted from the differential voltage sampling sequence, and the calculation expression for the effective value phasor of the fundamental differential voltage is as follows: In the formula, Indicates the first Within the first stable window The effective value phasor of the fundamental differential voltage of a complete power frequency cycle; It represents the imaginary unit.

[0068] The fundamental wave is extracted from the current sampling sequence of the same set of equiphase sampling points, and the calculation expression of the effective value phasor of the fundamental wave current segment is as follows: In the formula, Indicates the first Within the first stable window The effective value phasor of the fundamental current in a complete power frequency cycle; This represents an exponential function with the natural constant as its base.

[0069] Furthermore, the calculation expression for the fundamental active power loss index of the calculation window for all effective and complete power frequency cycles within the stable window is as follows: In the formula, Indicates the first The fundamental active power loss index for a stable window; This indicates the number of effective complete power frequency cycles participating in the average. Indicates the conjugate of the phasor of the fundamental current segment; This indicates taking the real part of a complex number.

[0070] As a prototype setting, the number of effective complete power frequency cycles participating in averaging within the stable window can be set to 20. This value can be adjusted according to the stability of the window and the duration of the event.

[0071] Using only the fundamental active component can reduce the impact of reactive induced voltage components and harmonic losses on resistive state evidence. If in-phase crosstalk exists due to changes in ferromagnetic structure or circuit layout, it is included in the error envelope as reactance crosstalk error.

[0072] Furthermore, calculate the square index of the current in the same stable window segment: In the formula, Indicates the first The current square index for a stable window segment is used; the current square index is used instead of the current average index because the resistive loss of the conductive connection segment is approximately proportional to the current square index. Pairing the window fundamental active power loss index with the current square index can reduce the impact of different load levels.

[0073] For the For each natural load switching event, the stability window before the event and the stability window after the event are recorded in chronological order of the actual time. Let the first The fundamental active power loss indices for the pre-event stabilization window and the post-event stabilization window of a natural load switching event are respectively... and The formula for calculating the differential index of the fundamental active power loss of the event is: In the formula, Indicates the first Differential index of fundamental active power loss for a natural load switching event.

[0074] Let the first The current square index of the current segment before and after a natural load switching event is as follows: and The expression for calculating the squared difference index of the current in this event segment is: In the formula, Indicates the first The event segment current squared difference index for a natural load switching event.

[0075] When the resistive performance of the evaluated connection segment is approximately stable before and after the event, and other operating conditions in the two stable windows are comparable, the fundamental active power loss differential index and the current square differential index of the event segment approximately satisfy the following: Therefore, calculate the first The expression for calculating the resistive slope of a natural load switching event is: In the formula, Indicates the first The event resistive slope corresponding to each natural load switching event.

[0076] The event resistivity slope reflects the resistive loss level of the evaluated connection section under the current connection section surface temperature and load thermal history conditions. However, the event resistivity slope itself is not the degradation increment relative to the acceptance state. The degradation increment relative to the acceptance state is obtained by comparing the event resistivity slope with the matched frozen condition baseline in step S3.

[0077] When the event segment current square difference index When the absolute value is insufficient to overcome the numerical resolution capability of the sampling channel, the event resistive slope is not calculated, and the event is not passed to the subsequent freeze baseline matching process.

[0078] Furthermore, each piece of event resistive slope evidence is associated with at least the following information: the unique identifier of the evaluated connection segment; the natural load switching event number; the time of the event; the direction of the event; the stability window boundary before the event; the stability window boundary after the event; the complete power frequency cycle number involved in the calculation; the cycle-by-cycle quality identifier; the surface temperature of the connection segment; the current endpoints of the segment before and after the event; the fundamental active power loss differential index of the event; the squared differential index of the current in the segment of the event; and the resistive slope of the event.

[0079] By saving the above information, it is possible to confirm in the subsequent review process whether the two pieces of event evidence originated from two independent natural load switching events with non-overlapping sampling windows.

[0080] A single continuous load transition process generates only one event resistive slope evidence; multiple local changes within the same load transition process cannot be split into multiple independent pieces of evidence. This rule ensures the independence of each piece of evidence in subsequent statistical determinations.

[0081] In one optional implementation, when the current stability within the stable window is relatively high, but the window duration is insufficient to obtain a preset number of complete power frequency cycles, the number of complete power frequency cycles participating in the averaging can be appropriately reduced, but it cannot be lower than the lower limit of the number that guarantees the stability of the window-level indicators. When the number of effective complete power frequency cycles participating in the averaging is lower than the lower limit, the corresponding stable window is determined as an unusable window, and no event resistive slope evidence is generated.

[0082] Through the above processing, this step obtains a natural load switching event evidence data packet, which includes at least: event number; unique identifier of the evaluated connection segment; event occurrence time; event direction; pre-event stability window boundary; post-event stability window boundary; fundamental active power loss index of the pre-event window; fundamental active power loss index of the post-event window; current square index of the pre-event window segment; current square index of the post-event window segment; fundamental active power loss differential index; current square differential index of the event segment; event resistive slope; event evidence availability status; summary of cycle-by-cycle quality identifiers; number of valid complete power frequency cycles participating in the averaging; sampling sequence within the window and window boundary storage records.

[0083] S3: Based on the surface temperature of the connection section, current endpoints, and load thermal history matching acceptance period of the natural load switching event, the operating condition baseline is formed and frozen. Based on the measurement link error and residual thermal model error, the conservative possible range of degradation change relative to the frozen baseline is determined, and the range compatibility is verified using two independent natural load switching events.

[0084] It should be noted that evidence from a single event is affected by noise, errors, and differences in load endpoints, and cannot be directly used as a conclusion confirming degradation. Meanwhile, if the system continuously incorporates new measurement data as a baseline during operation, the slowly accumulating contact oxidation resistance will be gradually written into the new baseline, and long-term degradation will never be detected. Therefore, this step employs two constraints: first, using the operating condition baseline formed and frozen during the acceptance period; and second, using a conservative interval verification based on bounded error budgeting. These two constraints work together to prevent false alarms caused by single disturbances and to prevent slow degradation from being swallowed up by the rolling baseline.

[0085] In this implementation, the baseline freeze attribute is not a manually set management rule, but a constraint rigidly enforced by both data storage and firmware execution logic. After the baseline data is written to a separate protected storage area, the online evaluation program only has read access to that storage area, and the firmware performing the online evaluation does not contain write functions to that storage area; during normal operation, any suspected degradation data cannot trigger an in-situ update of the baseline data. Only after a new shutdown acceptance test, troubleshooting maintenance, or connector replacement is a new baseline established manually using a dedicated maintenance tool, and the calibration version number is updated. A baseline identification code is generated synchronously during baseline establishment, which corresponds one-to-one with the baseline data and is referenced in every subsequent status output message. Therefore, the freeze attribute cannot be bypassed by online data flow at the implementation level, nor can it be overwritten by simple parameter update commands.

[0086] Specifically, after installation acceptance or repair and inspection are completed, independent connection resistance checks and link calibration are performed first to confirm that the evaluated connection segment is in good baseline condition. Subsequently, a multi-condition operating condition grid is established for this segment. This grid uses connection segment identification, calibration version, terminal surface temperature range, recent load thermal history range, fundamental current low-end endpoint, fundamental current high-end endpoint, and verified waveform range as common search criteria. Each covered operating condition grid stores a conservative possible range of baseline slope values ​​obtained during the acceptance period. ,in To define a conservative lower bound for the possible values ​​of the frozen baseline slope, To freeze the upper bound of the conservative possible value of the baseline slope, both are statistically obtained and saved from multiple benchmark events during the acceptance period. During normal operation, rolling updates are absolutely prohibited, and any suspected degradation data must not be incorporated into the baseline; only after a new shutdown acceptance, defect elimination maintenance, or replacement of connectors is a new baseline established and the calibration version number updated manually.

[0087] Furthermore, the terminal temperature measuring points should be arranged on the actual connection section surface; air temperature cannot directly replace the contact point temperature. As a design example for the R&D prototype, the temperature matching bandwidth is taken as... This value must be recalibrated during prototype verification, taking into account temperature measurement point errors, thermal response time constants, and the temperature sensitivity of the resistive parameters in that section. Only after forming a bounded residual thermal error term can it be used as the official setting value. If the surface temperature reflected by the temperature measurement point does not represent the true temperature of the internal contact surface of that section, a temperature difference correction value and its boundary must be obtained through laboratory temperature rise calibration tests, and the residual part of this boundary must be included in subsequent calculations. The terminal surface temperature measurement point can be a contact temperature sensor or a non-contact infrared temperature sensor; when using infrared temperature measurement, the emissivity of the surface being measured should be calibrated on-site to ensure that the temperature measurement error is within the budget.

[0088] Specifically, when searching the operating condition grid, matching is performed based on the numerical range of the high and low current endpoints, regardless of the actual direction of load switching. Load increases or decreases are only stored separately as event direction information. The resistive slope of the event is still calculated according to the actual chronological order of the events, and reverse load events are not mistakenly rejected due to reverse direction. If no verified record covering its terminal temperature status, current endpoints, and thermal history conditions can be found in the operating condition grid for the current event, it cannot be extrapolated or replaced by the global average slope. This event can only be retained as a mismatch event, waiting for a new natural load event that meets the conditions to be matched again. The range of current endpoints covered in the operating condition grid is determined by the load scan records during the acceptance period. When the current exceeds the verified range during actual operation, the event is marked as an operating condition mismatch and is not included in the degradation judgment.

[0089] It should be noted that the principle for establishing the error envelope is conservative boundedness, rather than treating the statistical standard deviation of repeated measurements as a hard boundary. Voltage gain error, current gain error, phase error, common-mode residual, synchronization timescale error, spectral leakage residual, channel noise, and the upper bound of drift within the calibration validity period in the measurement link are all formed into bounded envelope components based on calibration records. Because these components may be correlated, they cannot be synthesized using statistical independence when the correlation is unclear. Therefore, the resulting interval is not called the expanded uncertainty or confidence interval, but only the range of conservatively possible values. The sources of values ​​for each error component include: the maximum permissible error given in the factory calibration certificate, the upper bound of deviation given in the field calibration record, the product of the sensitivity coefficient and the temperature change range obtained from the temperature drift test, and the upper bound of the interference residual observed in the electromagnetic compatibility test. These error components form specific interval boundaries in each window calculation and propagate item by item with each sample.

[0090] Specifically, the conservative possible range of fundamental active power loss for each stable window before and after the event is given by the aforementioned error components. The range of conservatively possible squared values ​​of the current in this segment subscript Indicates the lower bound, subscript This indicates the upper bound. These intervals are formed by propagating the error budget item by item during window calculation, and their physical meaning is: under the measurement conditions of this window, the true value will not exceed this range according to the existing error budget.

[0091] The specific method for obtaining the conservatively conservative possible ranges of the fundamental active power loss difference and the squared difference of the current in the current segment of the event by subtracting from the most unfavorable direction is as follows: The specific expression for the conservative possible range of the fundamental active power loss difference obtained by subtracting from the most unfavorable direction is as follows: In the formula, This represents the lower bound of the conservative possible value of the change in the fundamental active power loss of the event. This serves as a conservative upper bound for its possible values. By employing cross-subtraction, the possible range of the difference is never artificially narrowed due to the sharing of hardware between the two measurements.

[0092] The conservative possible range of the squared difference of the current in the corresponding segment event is handled in the same way: In the formula, This represents the lower bound of the conservative possible value of the square change in the current during this event segment. This indicates the upper bound of its conservative possible values. If and If the range contains zero values, or if the square of the current change cannot overcome the numerical resolution, it indicates that the load change of the natural load event is insufficient and the resistive slope cannot be reliably calculated. This event is marked as an unusable event and exits the decision path.

[0093] Calculate the conservative possible range of the resistive slope of the event when the current squared difference interval does not contain zero and the numerical resolution condition is met.

[0094] Furthermore, the conservative possible range of the event resistive slope is calculated only when the current squared difference interval does not contain zero and the numerical resolution condition is met. This range is obtained by exhaustively taking the minimum and maximum values ​​of the quotients at the four endpoints.

[0095] The conservative lower bound for the event resistive slope is: In the formula, This represents the lower bound of the conservative possible value of the resistive slope of the event; Take the minimum value.

[0096] The upper bound of the conservative possible value of the event resistive slope is: In the formula, This represents the upper bound of the conservative possible values ​​of the resistive slope of the event; This indicates taking the maximum value.

[0097] If the upper bound exceeds the measurement chain failure boundary, or the corresponding error budget is no longer within the effective calibration range, the event is directly deemed invalid, and it is not allowed to claim to narrow the interval by increasing the number of cycles involved in the averaging.

[0098] Furthermore, after obtaining the conservative possible range of the resistive slope for the current event, this range is subtracted from the frozen baseline range within the matched operating condition cell, and the upper bound of the residual thermal model error, which is not yet included in the power and current squared error budget, is incorporated. .

[0099] The physical meaning of the upper bound of the residual thermal model error includes: the residual error caused by the temperature difference between the temperature measurement point and the actual contact point, the nonlinear model error of the resistive parameter drift with temperature in this segment, and the model residual when the linear weak temperature change assumption is insufficient.

[0100] In one feasible implementation, The values ​​should be obtained through specialized thermal model calibration tests, the steps of which include: The typical load thermal process of the evaluated connection section was simulated under laboratory conditions, and the curve of the difference between the surface temperature measurement points and the reference temperature at the contact point was recorded over time. The resistive slope reference value of the section was measured at multiple temperature levels, and the nonlinear term residuals were obtained by comparing the difference between the linear extrapolated value and the measured value. The maximum value of each of the above residuals within the verification temperature range was taken, and a safety margin was added as the threshold value. The provisional upper limit.

[0101] Based on this, multiple sets of historical event data from when this segment was in a healthy state should be used in actual operation to verify whether the measured degradation range is within the boundary defined by the provisional upper limit; if there is a breach, the limit should be further increased. Alternatively, reduce the operating condition temperature matching bandwidth until all health status history events do not exceed the limit.

[0102] Only if the residual term has a verifiable and credible upper bound should it be allowed to participate in the calculation; otherwise, a quantitative degradation judgment should not be made. When verification is used to determine... In such cases, the events described herein shall not be used as evidence for the review of subsequent suspected degradation events, in order to ensure the independence of the review events.

[0103] Therefore, the conservative lower bound for the degradation change of the current event relative to the frozen baseline is: In the formula, This represents the lower bound of the conservative possible value of the resistive slope for the current event; This represents the upper bound of the conservative possible values ​​for the slope of the frozen baseline in the work case cell matched by the current event; This represents the lower bound of the conservative possible value for the amount of degradation change relative to the frozen baseline.

[0104] The upper bound of the conservative possible value of the corresponding degradation change is: In the formula, This indicates the lower bound of the conservative possible values ​​for the slope of the matched frozen baseline; This represents the upper bound of the conservative possible values ​​of the resistive slope for the current event; This represents the upper bound of the conservative possible value for the amount of degradation change relative to the frozen baseline.

[0105] Since different events share the same calibration and baseline error source, these common errors should not be reduced according to the principle of statistical independence between events, and a conservative envelope should be maintained when the correlation is unclear.

[0106] Specifically, the maintenance threshold is not set as an absolute upper limit of the initial resistance, but rather as the allowable amount of additional resistive change. The allowable amount of additional resistive change is derived from the given reference current and the allowable additional fundamental active power loss using the following formula: In the formula, This indicates the allowable amount of additional resistive change; This indicates the reference current set for the rated operating range of the evaluated connection section; The engineering design limits for allowing additional fundamental active power loss in this connection section.

[0107] The two values ​​need to be determined in advance, taking into account the rated capacity of the metering box, thermal constraints, and acceptance verification range. As an example of calculation for a research and development prototype, we can take... , ,but Ohms; This example value is for illustrative purposes only and does not represent the acceptable limit for any supplied product.

[0108] It should be noted that, The selection of the appropriate rating should be determined based on the heat dissipation conditions, allowable temperature rise, and heat resistance rating of the insulation material of the connection section being evaluated, rather than simply applying the rated power of the entire metering box. This method only requires that the rated power of the metering box be used. Pre-calibrated and stored with the baseline, it does not require adaptive adjustment during operation.

[0109] Specifically, when the degradation change of the first effective natural load event is conservatively bounded by a lower bound... When the change in resistance is significantly greater than the allowable amount, the evaluated connection segment enters a state of suspected electrical performance degradation. The system records evidence of this event, but must not directly output a confirmation conclusion.

[0110] To output a confirmation conclusion, another independent natural load event with a different number must be obtained as verification evidence. Furthermore, the stability sampling windows before and after the verification event must not overlap with the initial suspected event. The conservative range for the degradation change of the verification event must also be calculated independently using the exact same procedure. Reverse load events can be used as a priority verification opportunity, but the positive resistive slope direction of a reverse load event alone cannot prove degradation.

[0111] Furthermore, the criteria for confirmation upon review are: the lower bound of the conservative possible value of the degradation change of both the initial suspected event and the reviewed event is strictly greater than the allowable amount of new resistive change, and the intervals of the two independent pieces of evidence are compatible.

[0112] The compatibility criterion is: In the formula, and These represent the lower and upper bounds of the conservative possible values ​​for the degradation change of the initial suspected event, respectively. and These represent the lower and upper bounds of the conservative possible values ​​for the degradation change of the review event, respectively.

[0113] Satisfying this formula indicates that two independent pieces of evidence point to the same degradation range, and the conclusion that the electrical performance degradation of the evaluated connection segment has been verified can be output. Compatibility between the two ranges only satisfies the verification condition; it does not authorize reducing the error budget, nor does it allow modification of calibration parameters to meet the judgment.

[0114] If the window is unavailable, calibration expires, drift cannot obtain a credible upper bound, observation times out, or two pieces of evidence contradict each other, the connection segment will remain in a suspected state, outputting an insufficient evidence status and providing a retest suggestion. The suspected state will not be automatically cleared by a single invalid event; when evidence contradicts each other, the cause should not be directly attributed to channel drift, and calibration should not be modified accordingly. The suspected state can only be lifted after manual retesting shows that the segment's status has returned to acceptable and the baseline is refrozen.

[0115] It should be noted that during the review and confirmation process, the event... and events The corresponding natural load events must indeed be two distinct load transition processes. The criteria for differentiation are different event numbers, different event times, and no overlap between the stability windows before and after the events. If the two events are actually artificially separated from the same continuous load switching process, they are not independent verification relationships and cannot be used as verification evidence. This independence requirement, combined with the rule that only one piece of evidence is generated per load transition process, eliminates the possibility of constructing false verification pairs by artificially increasing the number of events.

[0116] Through the above processing, this step obtains a status determination data packet, which includes at least: a unique identifier of the evaluated connection segment; the baseline version and baseline identifier code used; the matched condition grid record; the event number; the conservative possible range of the event resistivity slope; the conservative possible range of the degradation change relative to the frozen baseline; the allowed new resistivity change; the current status; maintenance recommendations; if entering a suspected status, record the evidence of the initial suspected event; if the review is completed, record the interval compatibility criterion results of the initial suspected event and the reviewed event; if the evidence is insufficient, record the cause and retest recommendations.

[0117] The current status should include at least the following clearly defined categories: insufficient evidence, operating condition mismatch, suspected electrical performance degradation of the connection section, and verified electrical performance degradation of the connection section. Connection sections without measurement points should not be given a conclusion of normal or degradation; connection sections with expired calibration, unavailable windows, or missing matching baselines should be treated as insufficient evidence. Even if a certain section shows confirmed degradation, it should not be used to infer that other connection sections of the entire metering box are qualified or failed. The baseline, operating condition, and error envelope of each evaluated connection section are independent of each other.

[0118] S4: Based on the measurement point layout status, sampling link status, calibration status, operating condition baseline matching status, event evidence validity status, and interval compatibility verification results of the evaluated connection section, output status information and maintenance suggestions carrying the location identifier of the evaluated connection section, natural load switching event number, and frozen baseline version number.

[0119] It should be noted that all the above judgments must be grounded in the physical meaning of the change in the electrical performance of the connection section relative to the acceptance baseline. A general health score for the entire metering box output is not provided, nor are tasks related to fault root cause location, metering error evaluation of the entire box, or power distribution. Any abstract score detached from the measured physical quantities of the evaluated connection section will prevent maintenance personnel from determining the signal source. Therefore, only synchronous sampling samples, evidence of natural load events, baseline version information, and status judgment criteria can be transmitted between modules.

[0120] In the output information, the status conclusion must carry a clear identifier of the location of the evaluated connection segment, the event number, and the baseline version number, so that maintenance personnel can trace back to the specific physical path and original event data that led to the conclusion.

[0121] Specifically, the system independently outputs the following information for each evaluated connection segment with deployed measurement points: the location identifier of the evaluated connection segment, the baseline version used, the event number, the conservative possible range of the event resistive slope, the conservative range of the degradation change relative to the frozen baseline, the current status, and maintenance recommendations.

[0122] Furthermore, to ensure that the frozen baseline and baseline version information cannot be bypassed in the actual system, this embodiment also provides a preferred system-side implementation: The differential voltage and corresponding segment current synchronous acquisition module corresponds to the steps of defining the evaluated connection segment and establishing the synchronous sampling link; the event evidence generation module corresponds to the steps of generating differential resistive evidence for natural load events; the frozen baseline and calibration record module is responsible for saving, retrieving and maintaining the operating condition grid; the interval discrimination and status output module is responsible for error envelope calculation, threshold comparison, independent verification and status transition.

[0123] The event evidence generation module outputs the event number, the stability window index before and after the event, and the event resistivity slope evidence, which serve as the retrieval input for the frozen baseline and calibration record module. Only after the interval discrimination and status output module receives the matched frozen baseline interval can it truly complete the calculation of the degradation change interval.

[0124] Each baseline record stored in the frozen baseline and calibration record module also includes the baseline freezing time, the original acceptance record number at the time of freezing, the unique identifier of the evaluated connection segment using the baseline, and digital summary verification information of the baseline data.

[0125] When the interval discrimination and status output module reads the baseline record, it first verifies the digital digest verification information; if the verification fails, it prohibits the use of the baseline record as the basis for judgment and outputs the baseline data abnormality information.

[0126] Therefore, any software anomalies or storage media failures during operation will not result in conclusions about the status based on a tampered baseline. Only when a baseline passes all three checks—integrity verification, version validity, and operational condition matching—can it be used to calculate degradation changes.

[0127] This implementation method employs a combined logic of differential resistive evidence of natural load events and frozen baselines with conservative error envelopes, achieving technical effects that cannot be achieved by a single method. If only a single resistive slope is compared with a threshold, channel noise, residual phase error, and occasional disturbances during load switching can all trigger false alarms. If only a frozen baseline is used without two-event interval compatibility verification, a single accidental disturbance can cause an erroneous alarm. This method eliminates load fluctuation interference through stable load event segmentation, prevents slow degradation from being absorbed by rolling data by freezing the baseline, and suppresses the influence of random errors through interval compatibility verification. This enables the smart energy metering box to reliably and verifiably determine the changes in the electrical performance of the connection section relative to the acceptance baseline under unattended operation conditions.

[0128] If any one of the above three methods is missing, the remaining methods cannot independently achieve the same level of discriminative reliability: without event segmentation, the resistive slope evidence itself cannot be separated from load fluctuations; without a frozen baseline, even with high-precision event evidence, there is no initial reference for comparison; without interval compatibility verification, the error envelope of a single event is insufficient to distinguish between real degradation and random disturbances.

[0129] This implementation method uses the following device embodiment to achieve the above method: Each of the above modules can be implemented through a monitoring terminal with a sampling front-end, a digital signal processor, and a communication interface. The voltage channel of the sampling front-end is designed with a range based on the millivolt-level connection section voltage drop; the current channel selects the transformer ratio based on the rated current of that section; and the temperature channel uses a contact-type temperature sensor attached to the surface of the connection section. The digital signal processor performs stable window division, fundamental phasor calculation, event resistive slope calculation, operating condition grid retrieval, and interval operation. The communication interface sends the output information to the maintenance master station.

[0130] When all evaluated connection segments share the same monitoring terminal, the combination, polarity, and temperature sensor number of each channel are independently registered in the configuration information to prevent data crosstalk between different connection segments. The status message reported by the communication interface should include at least one or more of the following: the location identifier of the evaluated connection segment, the baseline version number, the event number, the current status, and maintenance recommendations, so that the maintenance master station can make maintenance decisions directly based on the message without having to retrieve the original sampling data again.

[0131] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for evaluating the operating status of an intelligent electricity metering box, characterized in that, include: The evaluation connection segment in the smart energy metering box is determined based on the physical starting point and physical ending point registered before commissioning. A differential voltage sampling channel is established at the fixed conductor positions at both ends of the evaluation connection segment. A current sampling channel and a surface temperature sampling channel corresponding to the evaluation connection segment are also established. The differential voltage, current, and surface temperature of the connection segment are sampled synchronously. Identify natural load switching events from synchronous sampling data. Within the pre-event stability window and post-event stability window of the natural load switching event, calculate the fundamental active power loss index and the current square index based on the complete power frequency cycle. Determine the resistive slope evidence of the event based on the difference between the fundamental active power loss index and the current square index before and after the event. Based on the connection segment surface temperature, current endpoints, and load thermal history of the natural load switching event, the resistive slope evidence of the event is matched with the operating condition baseline formed and frozen during the acceptance period. The conservative possible range of the event resistive slope and the conservative possible range of the degradation change relative to the frozen baseline are determined based on the measurement link error and residual thermal model error. When the lower bound of the conservative possible range of the degradation change exceeds the allowable new resistive change, the evaluated connection segment is identified as a suspected connection segment electrical performance degradation state, and is verified using another independent natural load switching event. When the lower bound of the conservative possible range of the degradation change of both independent natural load switching events exceeds the allowable new resistive change and the two ranges are compatible, the evaluated connection segment is identified as a verified connection segment electrical performance degradation state. Based on the measurement point layout status, sampling link status, calibration status, operating condition baseline matching status, event evidence validity status, and independent event verification results of the evaluated connection segment, the status information of the evaluated connection segment is output.

2. The method for evaluating the operating status of an intelligent energy metering box according to claim 1, characterized in that, The evaluated connection segment is a complete conductive path extending from one fixed conductor position to another, and the complete conductive path includes at least one of screw crimping face, terminal piece, transition short conductor and circuit breaker post extension; the status information indicates the change in electrical performance of the complete conductive path relative to the acceptance baseline, and does not indicate the pure contact resistance of a single contact.

3. The method for evaluating the operating status of an intelligent energy metering box according to claim 1, characterized in that, The differential voltage sampling channel is connected to the fixed conductor positions at both ends of the evaluated connection segment via a high-impedance differential tap line. The input impedance of the high-impedance differential tap line is at least two orders of magnitude higher than the equivalent impedance of the main circuit of the evaluated connection segment. The current sensing element of the segment current sampling channel is disposed on the branch conductor connected in series with the evaluated connection segment. The temperature sensor of the connection segment surface temperature sampling channel is disposed on the surface of the evaluated connection segment. The differential voltage sampling channel and the segment current sampling channel are synchronously sampled according to the same phase reference under the control of a hardware synchronous trigger source.

4. The method for evaluating the operating status of an intelligent energy metering box according to claim 1, characterized in that, The natural load switching event is determined based on the effective value of the fundamental current per cycle or the sliding root mean square envelope within the effective bandwidth. When the current in this segment shifts from the first stable level to the second stable level, and there are stable intervals that satisfy the stability criterion before and after the load transition process, the two stable intervals are respectively determined as the stable window before the event and the stable window after the event. Only one event resistive slope evidence is generated for a continuous load transition process.

5. The method for evaluating the operating status of an intelligent energy metering box according to claim 1, characterized in that, Both the pre-event stabilization window and the post-event stabilization window include an integer number of complete power frequency cycles that do not overlap with each other; for each complete power frequency cycle, a current channel unsaturation flag, a voltage channel unsaturation flag, a sample integrity flag, a frequency lock flag, and a spectrum leakage controlled flag are set; only complete power frequency cycles with all flags valid are used to extract the fundamental differential voltage phasor and the fundamental current phasor of this segment.

6. The method for evaluating the operating status of an intelligent energy metering box according to claim 1, characterized in that, The operating condition baseline uses the evaluated connection segment identifier, calibration version, connection segment surface temperature range, recent load thermal history range, fundamental current low end point, fundamental current high end point, and verified waveform range as search criteria, and saves the lower bound and upper bound of the conservative possible value of the frozen baseline slope obtained during the acceptance period. When a natural load switching event cannot be matched with the corresponding operating condition baseline, the natural load switching event is identified as an operating condition mismatch event, and the operating condition range extrapolation result or the global average slope is not used as a substitute.

7. The method for evaluating the operating status of an intelligent energy metering box according to claim 1, characterized in that, The measurement link error includes at least one of voltage gain error, current gain error, phase error, common-mode residual, synchronization timescale error, spectral leakage residual, channel noise, and drift upper bound within the calibration validity period; when the correlation between different error components is not clear, the conservative possible value range of the event resistive slope is not narrowed in a statistically independent manner.

8. The method for evaluating the operating status of an intelligent energy metering box according to claim 1, characterized in that, The allowable additional resistive change is determined based on the ratio of the allowable additional fundamental active power loss limit to the square of the reference current setting; the two independent natural load switching events used for verification meet the requirements of different event numbers, different event occurrence times, and non-overlapping pre-event stability windows and post-event stability windows.

9. The method for evaluating the operating status of an intelligent energy metering box according to claim 1, characterized in that, When the evaluated connection segment has no measurement points, the sampling link calibration fails, the current in this segment does not correspond to the evaluated connection segment, the current polarity verification record is missing, the complete power frequency frequency quality does not meet the requirements, the calibration expires, the measurement drift cannot obtain a reliable upper bound, or the conservative possible value ranges of the degradation change of two independent natural load switching events are incompatible, the status of the electrical performance degradation of the connection segment has not been verified, and the status of insufficient evidence or retest recommendation is output.

10. A smart energy metering box operation status evaluation system, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions that, when executed by the processor, implement a method for evaluating the operating status of an intelligent energy metering box according to any one of claims 1-9.

Images