A method and system for testing the performance of a zinc oxide varistor

Abstract

This invention relates to the field of condition monitoring technology for electrical equipment in power systems, specifically to a method and system for testing the performance of zinc oxide resistive elements. The method includes continuously applying an adjustable DC bias voltage across the zinc oxide resistive element under test; periodically injecting an impulse current into the zinc oxide resistive element according to a preset injection cycle during the continuous application of the DC bias voltage; acquiring DC leakage current parameters and impulse residual voltage parameters of the zinc oxide resistive element before and after the impulse current application, and acquiring condition recovery parameters after the DC bias voltage is removed; based on the acquired DC leakage current parameters, impulse residual voltage parameters, and condition recovery parameters, constructing a multi-dimensional accelerated aging model, and quantifying the degree of degradation of the zinc oxide resistive element and predicting its remaining lifespan through the multi-dimensional accelerated aging model. According to the solution of this invention, it is possible to realistically and accurately simulate the complex operating conditions of a DC power system and achieve multi-dimensional, high-precision condition monitoring.

Description

Technical Field

[0001] This invention generally relates to the field of condition monitoring technology for electrical equipment in power systems. More specifically, this invention relates to a method and system for testing the performance of zinc oxide resistor sheets. Background Technology

[0002] With the advancement of global energy transition and the construction of new power systems, photovoltaic power generation, centralized / distributed energy storage systems, flexible direct current transmission networks (VSC-HVDC), and ultra-high voltage direct current transmission networks (UHVDC) have been applied more widely than ever before. In these high-voltage and ultra-high-voltage direct current systems, zinc oxide (ZnO) varistors with nonlinear and non-ohmic characteristics serve as core electrical components of overvoltage protection devices (such as surge arresters, damping resistors, and overvoltage limiters). Their performance directly determines the robustness of the entire power transmission and distribution system's safety defenses.

[0003] Zinc oxide resistors possess extremely high nonlinear coefficients, large current capacity, and microampere-level steady-state leakage current characteristics. However, the operating environment of DC power systems is extremely unique and harsh. During long-term operation, the resistors must withstand continuous, unidirectional DC bias voltage stress. Under these conditions, defects such as zinc ions or oxygen vacancies in the internal non-uniform grain boundary layer will migrate and accumulate locally at one grain boundary, leading to distortion and unidirectional attenuation of the potential barrier height. Simultaneously, while bearing this long-term steady-state DC bias stress, the power grid frequently experiences occasional atmospheric lightning overvoltage surges or operational overvoltage surges caused by system circuit breaker closing. This combined electrical stress condition of "steady-state DC bias + transient high-current pulse" can lead to severe localized hot spots and crystallographic degradation of the grain boundary layer lattice within the zinc oxide resistor, subsequently causing a surge in DC leakage current, residual voltage runaway, and even thermal runaway explosion.

[0004] Currently, the traditional methods for testing and evaluating the performance of zinc oxide resistors largely rely on AC testing systems. These AC testing systems primarily simulate operating conditions under power frequency AC environments (50Hz / 60Hz) to periodically verify the reference voltage and resistive current components of the resistors. However, in the context of the widespread adoption of DC systems, traditional AC testing systems have revealed significant limitations and physical defects.

[0005] Because the grain boundary barrier is subjected to an alternating electric field under AC voltage, the polarization charge and defect ions oscillate symmetrically in the positive and negative half-cycles, and cannot form a unidirectional grain boundary defect ion drift effect under a DC electric field.

[0006] Therefore, traditional AC aging tests or single-stress pulse impact tests cannot accurately reproduce and simulate the real combined operating conditions of "DC bias + periodic / occasional overvoltage impact" faced by zinc oxide resistors in DC systems. This leads to a disconnect between the degradation assessment curves obtained from the tests and actual field operation, making it impossible to accurately and sensitively capture the microscopic performance degradation of the resistors in the early stages of degradation, often resulting in misjudgments of lifespan and failing to meet the high reliability requirements of new DC power systems. Summary of the Invention

[0007] In order to overcome the above-mentioned defects of the prior art, the present invention provides a zinc oxide resistive sheet performance testing method and testing system that can realistically and accurately simulate the complex operating conditions of DC power systems, realize multi-dimensional condition monitoring, and provide high-precision life prediction.

[0008] This invention overcomes the uncertainty of single test indicators by introducing an accelerated aging test method involving multiple stresses (DC bias + transient impact) and simultaneously extracting multi-dimensional electrical and thermal desorption characteristic parameters over a time span from milliseconds to hours. This allows for extremely sensitive characterization of the early fatigue properties of polycrystalline structures within nonlinear materials. To this end, this invention provides the following multiple solutions.

[0009] In a first aspect, the present invention provides a method for testing the performance of zinc oxide resistors, comprising the following steps: Step S1: Continuously apply an adjustable DC bias voltage across the zinc oxide resistive element being tested to simulate the steady-state voltage conditions during the operation of a DC power system. Step S2: During the continuous operation of the DC bias voltage, an impulse current is periodically injected into the zinc oxide resistor according to a preset injection cycle to simulate the occasional overvoltage intrusion condition in the DC power system. Step S3: Obtain the DC leakage current parameters and impact residual voltage parameters of the zinc oxide resistor before and after the impact current is applied, and collect the state recovery parameters after the DC bias voltage is removed; Step S4: Based on the collected DC leakage current parameters, impact residual voltage parameters, and state recovery parameters, a multi-dimensional accelerated aging model is constructed. The degradation degree of the zinc oxide resistor is quantified and its remaining life is predicted through the multi-dimensional accelerated aging model.

[0010] Furthermore, in step S2, the injection cycle of the impact current is not less than once per hour, in order to accelerate the localized accumulation of micro-defects in the grain boundary layer inside the zinc oxide resistor sheet and the accumulation of thermoelectric stress within a limited experimental time.

[0011] Further, in step S3, the DC leakage current parameter is obtained in the following way: Real-time acquisition of current changes flowing through the zinc oxide resistor; The DC leakage current parameters include the difference between the instantaneous values ​​of the DC leakage current before and after the impact. The calculation formula is: ; Among them, I L,before I represents the instantaneous leakage current before a single impact injection. L,after This represents the instantaneous leakage current after a single impact injection. The DC leakage current parameters also include the long-term drift after multiple consecutive impacts. The calculation formula is: ; Among them, I L,0 I is the initial leakage current before the start of the test cycle. L,N This represents the leakage current value after the Nth impact and when the system reaches a steady state.

[0012] Furthermore, in step S3, the method for obtaining the impact residual pressure parameter is as follows: Obtain the waveform of the terminal voltage across the zinc oxide resistor at the instant of the impact current injection; The impact residual pressure parameter includes the residual pressure value U at the peak of the impact current. res Characterized as: U res =U(t peak ); Among them, t peak This represents the peak moment of the inrush current waveform; The impact residual pressure parameter also includes the residual pressure attenuation slope k during continuous impact. d and residual pressure offset δ res The calculation formulas are as follows: k d =(U res,0 -U res,N ) / N; δ res =(U res,N -U res,0 ) / U res,0 ×100%; Among them, U res,0 U is the initial residual pressure value at the time of the first impact. res,N This is the measured residual pressure value at the Nth impact.

[0013] Furthermore, in step S3, the method for obtaining the state recovery parameters includes: After completing the preset number of impact injections or reaching a stage test node, the DC bias voltage is cut off, and the zinc oxide resistor is left to stand in a field-free state for at least 5 minutes to allow the polarized charges and non-equilibrium charge carriers inside it to undergo thermal relaxation and self-recovery due to the micro-thermoelectric effect. Reapply a DC bias voltage of the same amplitude as in step S1, and collect the recovered leakage current value I after stabilization. L,f ; Calculate the leakage current recovery rate η rec : η rec =I L,f / I L,0 ×100%; When η rec When ≤120%, the zinc oxide resistor is determined to be in the reversible defect recovery stage; when η rec When the value is greater than 120%, it is determined that the internal grain boundary layer structure of the zinc oxide resistor has undergone irreversible permanent lattice distortion and thermal degradation failure.

[0014] Further, in step S4, the multi-dimensional accelerated aging model introduces a degradation index DI to quantify the overall performance degradation state of the zinc oxide resistor. The degradation index DI is constructed by weighting and summing the DC bias time parameter, the number of impacts parameter, the permanent increment parameter of leakage current parameter, and the residual voltage offset parameter through their respective parameter weight functions; the calculation formula is: ; In the formula, To accumulate DC bias time, To accumulate the number of impacts suffered, For the permanent increment of leakage current, is the residual pressure offset rate; w1, w2, w3, and w4 are the weighting coefficients of each parameter, and satisfy the constraint condition: w1+w2+w3+w4=1.

[0015] Furthermore, in step S2, the impulse current includes a pulse current that meets the standard lightning impulse waveform or the switching overvoltage impulse waveform, and the amplitude of the impulse current is dynamically adjusted and set within the range of 1kA to 20kA according to the nominal voltage of the sample under test and the actual operating conditions.

[0016] Furthermore, the multi-dimensional accelerated aging model establishes an empirical correlation equation between the degradation index DI and the main independent variables, cumulative DC bias time and number of impacts, through least squares fitting: DI = a0 + a1·t dc +a2·N imp +ε; In the formula, a0 is the correction coefficient for the constant term, a1 is the DC voltage stress aging rate constant, a2 is the pulse impact stress aging rate constant, and ε is the random measurement noise error term. An overdetermined system of equations is constructed by collecting multiple sets of historical test sample data, and the least squares estimated solution is obtained using matrix solving. ; In the formula, Let be the vector of coefficients to be determined, A be the structure matrix constructed from the sample data of the independent variables, and b be the observation vector constructed from the sample data of the measured degradation index.

[0017] Furthermore, after obtaining the empirical correlation equation, a DC operating condition tolerance coefficient is introduced to predict the remaining life of the zinc oxide resistor, wherein the remaining life T... life The prediction formula is: T life =(DI threshold -DI0) / (a1+a2·f imp ); In the formula, DI threshold The preset decommissioning and deactivation threshold is DI0, which is the real-time degradation index calculated at the end of the current test cycle. imp The frequency of the impact pulse injected into the test system or actual operating environment.

[0018] In a second aspect, the present invention provides a zinc oxide resistor performance testing system, comprising: a DC high-voltage power supply module for providing a high-precision DC bias voltage with an adjustable output voltage within a preset range and maintaining its steady-state output during a long-cycle accelerated aging test; an impulse current generation module for periodically generating impulse current pulses with preset waveforms and preset amplitude characteristics and efficiently injecting them into the zinc oxide resistor; a multi-channel data acquisition module connected to the zinc oxide resistor and related sensors for synchronously recording leakage current parameters, residual voltage parameters, and state recovery parameters at high frequency during the combined action of DC bias and impulse pulses; and a processing module connected to the multi-channel data acquisition module, the DC high-voltage power supply module, and the impulse current generation module for executing the zinc oxide resistor performance testing method described above.

[0019] The beneficial effects of this invention are as follows: According to the scheme of this invention, a comprehensive accelerated aging mathematical model is constructed by extracting multi-dimensional features covering microscopic ion drift and macroscopic nonlinear performance, such as leakage current difference, permanent residual increment, residual voltage decay slope, and field-free self-recovery rate. Simultaneously, this invention simulates the real physical boundary conditions of photovoltaic and converter station surge arresters under long-term unidirectional DC electric fields, occasional lightning strikes, or closing overvoltages during actual service, ensuring that the aging mechanism and grain boundary barrier decay path of the material are completely consistent with actual field conditions. This significantly improves the sensitivity and accuracy of testing, enabling the capture of metastable degradation characteristics of the test sample.

[0020] Furthermore, by introducing a DC operating condition tolerance coefficient and degradation index (DI) based on least squares curve fitting, this invention provides a very solid basis for the selection of large quantities of nonlinear varistor elements, the screening of factory quality, and the quantitative prediction of the remaining service life of on-grid equipment. Attached Figure Description

[0021] The objectives, features, and advantages of this invention will become readily apparent upon reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of the invention are shown by way of example and not limitation, and the same or corresponding reference numerals denote the same or corresponding parts, wherein: Figure 1 A flowchart illustrating a method for testing the performance of zinc oxide resistors according to an embodiment of the present invention is shown schematically. Figure 2 This diagram schematically illustrates the composition of a zinc oxide resistor performance testing system according to an embodiment of the present invention. Figure 3 A schematic diagram illustrating a detailed flowchart of a single test cycle according to an embodiment of the present invention is provided. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0024] Figure 1 A flowchart illustrating a method for testing the performance of zinc oxide resistors according to an embodiment of the present invention is shown.

[0025] like Figure 1As shown, step S1: An adjustable DC bias voltage is continuously applied across the two ends of the zinc oxide resistor being tested to simulate the steady-state voltage condition during the operation of a DC power system.

[0026] Step S2: During the continuous application of the DC bias voltage, an impulse current is periodically injected into the zinc oxide resistor according to a preset injection cycle to simulate the occasional overvoltage intrusion condition in the DC power system. The injection cycle of the impulse current is no less than once per hour, in order to accelerate the localized accumulation of micro-defects in the grain boundary layer inside the zinc oxide resistor and the accumulation of thermoelectric stress within a limited experimental time.

[0027] In some embodiments, the impulse current includes a pulse current that meets the standard lightning impulse waveform or the switching overvoltage impulse waveform, and the amplitude of the impulse current is dynamically adjusted in the range of 1kA to 20kA according to the nominal voltage of the sample under test and the actual operating conditions.

[0028] Step S3: Obtain the DC leakage current parameters and impact residual voltage parameters of the zinc oxide resistor before and after the impact current is applied, and collect the state recovery parameters after the DC bias voltage is removed.

[0029] In some embodiments, the DC leakage current parameter is obtained by: acquiring the current change flowing through the zinc oxide resistor in real time; the DC leakage current parameter includes the difference between the instantaneous values ​​of the DC leakage current before and after the impact. The calculation formula is: ; Among them, I L,before I represents the instantaneous leakage current before a single impact injection. L,after This represents the instantaneous leakage current after a single impact injection. The DC leakage current parameters also include the long-term drift after multiple consecutive impacts. The calculation formula is as follows: ; Among them, I L,0 I is the initial leakage current before the start of the test cycle. L,N This represents the leakage current value after the Nth impact and when the system reaches a steady state.

[0030] Furthermore, the impact residual voltage parameter is obtained by acquiring the terminal voltage waveform across the zinc oxide resistor at the instant of impact current injection; the impact residual voltage parameter includes the residual voltage value U at the peak of the impact current. res Characterized as: U res =U(t peak ); Among them, t peakThis represents the peak moment of the inrush current waveform; The impact residual pressure parameter also includes the residual pressure attenuation slope k during continuous impact. d and residual pressure offset δ res The calculation formulas are as follows: k d =(U res,0 -U res,N ) / N; δ res =(U res,N -U res,0 ) / U res,0 ×100%; Among them, U res,0 U is the initial residual pressure value at the time of the first impact. res,N This is the measured residual pressure value at the Nth impact.

[0031] The method for obtaining the state recovery parameters includes: after completing a preset number of impact injections or reaching a stage test node, cutting off the DC bias voltage and letting the zinc oxide resistor sheet stand in a field-free state for at least 5 minutes to allow the polarization charge and non-equilibrium charge carriers inside it to undergo thermal relaxation and self-recovery due to the micro-thermoelectric effect; reapplying the DC bias voltage of the same amplitude as in step S1, and collecting the leakage current recovery value I after stabilization. L,f ; Calculate the leakage current recovery rate η rec : η rec =I L,f / I L,0 ×100%; When η rec When ≤120%, the zinc oxide resistor is determined to be in the reversible defect recovery stage; when η rec When the value is greater than 120%, it is determined that the internal grain boundary layer structure of the zinc oxide resistor has undergone irreversible permanent lattice distortion and thermal degradation failure.

[0032] Step S4: Based on the collected DC leakage current parameters, impact residual voltage parameters, and state recovery parameters, a multi-dimensional accelerated aging model is constructed. The degradation degree of the zinc oxide resistor is quantified and its remaining life is predicted through the multi-dimensional accelerated aging model.

[0033] In some embodiments, this multi-dimensional accelerated aging model introduces a degradation index (DI) to quantify the overall performance degradation state of the zinc oxide resistive element. The degradation index (DI) is constructed by weighting and summing the DC bias time parameter, the number of impacts parameter, the permanent increment of leakage current parameter, and the residual voltage offset parameter through their respective parameter weighting functions; the calculation formula is as follows: ; In the formula, To accumulate DC bias time, To accumulate the number of impacts suffered, For the permanent increment of leakage current, is the residual offset rate; w1, w2, w3, and w4 are the weighting coefficients of each parameter, and satisfy the constraint condition: w1 + w2 + w3 + w4 = 1. In this formula, the parameters marked with the symbol "^" represent the reverse or forward normalized mapping values ​​of the corresponding measured parameters within the preset limit reference range.

[0034] The above-mentioned multi-dimensional accelerated aging model establishes an empirical correlation equation between the degradation index DI and the main independent variables, namely, the cumulative DC bias time and the number of impacts, through least squares fitting: DI = a0 + a1·t dc +a2·N imp +ε; In the formula, a0 is the constant term correction coefficient, a1 is the DC voltage stress aging rate constant, a2 is the pulse impact stress aging rate constant, and ε is the random measurement noise error term. An overdetermined system of equations was constructed by collecting multiple sets of historical experimental sample data, and the least squares estimated solution was obtained using matrix solving. ; In the formula, Let be the vector of coefficients to be determined, A be the structure matrix constructed from the sample data of the independent variables, and b be the observation vector constructed from the sample data of the measured degradation index.

[0035] After obtaining the empirical correlation equation, a DC operating condition tolerance coefficient is introduced to predict the remaining life of the zinc oxide resistor. The remaining life T is... life The prediction formula is: T life =(DI threshold -DI0) / (a1+a2·f imp ); In the formula, DI threshold The preset decommissioning and deactivation threshold is DI0, which is the real-time degradation index calculated at the end of the current test cycle. imp The frequency of the impact pulse injected into the test system or actual operating environment.

[0036] To achieve the above method, the present invention also provides a zinc oxide resistor performance testing system. For example... Figure 2 As shown, the system includes a DC high-voltage power supply module, an inrush current generation module, a multi-channel data acquisition module, and a processing module.

[0037] The DC high-voltage power supply module is used to provide a high-precision DC bias voltage with an adjustable output voltage within a preset range and maintain its steady-state output during long-cycle accelerated aging tests. To meet the requirements of simultaneous series / parallel or differentiated bias stress accelerated aging tests on large batches of zinc oxide resistance sheets of different thicknesses and nominal voltage levels, the DC high-voltage power supply module in this embodiment is designed in hardware as an intelligent power array consisting of multiple independent power supply sub-units switched via a series-parallel matrix switch.

[0038] Each power supply subunit is an independently controllable digital high-voltage source, which integrates an ARM-based microcontroller that can receive voltage command instructions from the processing module via RS485 or CAN bus, enabling independent and smooth adjustment of the output voltage of each unit between 0 and 5kV.

[0039] These power supply sub-units are topologically combined using an intelligent control switch array, enabling both series output of ultra-high DC bias up to tens of thousands of volts and parallel output of large currents. This achieves a wider voltage / current regulation range and automatically meets the composite aging test requirements of multi-section series-connected resistor sheet integral assemblies for ±800kV and above UHVDC transmission systems. The output terminals of each power supply unit are securely connected to the two ends of the zinc oxide resistor sheet test specimen installed in the shielded test chamber via specially designed double-shielded cross-linked polyethylene high-voltage cables, ensuring high insulation and zero-loss transmission of steady-state energy.

[0040] The impulse current generating module is used to periodically generate impulse current pulses with preset waveforms and preset amplitudes, and efficiently inject them into the zinc oxide resistive element.

[0041] In order to simulate the extreme high-energy operating overvoltage or multiple lightning strike conditions that the surge arresters of UHV converter stations may encounter, the impulse current generator in this system adopts a modular topology architecture with multiple independent impulse current generator units connected in parallel.

[0042] Each generator unit contains an independent charging current-limiting impedance, a high-energy-storage, low-self-inductance pulse capacitor, and an all-solid-state transistor (GTO / IGBT) or a fast gas discharge valve switch. Through parallel connection, the units achieve multi-channel in-phase coordinated discharge under unified timing synchronous trigger pulse control, thereby converging to generate a huge surge current with a peak amplitude of up to 20kA or even greater.

[0043] Meanwhile, the device is equipped with an intelligent waveform adjustment inductor chain based on stepper motor regulation. This chain automatically switches the parameters of the connected forming elements according to the test specifications input from the host computer, thus smoothly generating standard lightning impulse waveforms, short-wavefront lightning waveforms, and long-duration operational overvoltage impulse waveforms on a single device. The total output of the impulse current generator is connected to a zinc oxide resistor via a low-inductance, wide-width copper busbar, minimizing lead inductance and ensuring that the impulse current at the steep leading edge is injected into the resistor with 100% distortion-free operation.

[0044] A multi-channel data acquisition module, connected to the zinc oxide resistor and related sensors, is used to synchronously record leakage current parameters, residual voltage parameters, and state recovery parameters at high frequency during the combined action of DC bias and impulse pulse. To completely resolve the crosstalk problem of strong electromagnetic pulse interference (EMP) during high-voltage, high-current impulses on weak current monitoring signals, the multi-channel data acquisition device in this system adopts a distributed high-fidelity data acquisition architecture.

[0045] The device consists of multiple independent distributed acquisition nodes located at different physical locations on the test site (such as the sample clamp, the grounding busbar, and the environmental monitoring point). Each acquisition node is independently encapsulated in a fully sealed pure copper silver-plated shielded box, and contains a high-performance high-speed data acquisition card, a differential preamplifier conditioning circuit, and an electro-optical conversion module.

[0046] The node responsible for acquiring leakage current is connected to the aforementioned high-precision Hall current sensor or coaxial non-inductive shunt via a very short coaxial cable; the node responsible for acquiring residual voltage waveform is connected to the output of a wideband RC voltage divider. The analog electrical signals acquired by each node undergo high-bandwidth A / D digitization locally, and are then modulated into optical signals and transmitted to the main control room via an anti-interference dual-ring fiber optic network. This distributed fiber optic acquisition structure completely severs the electrical connection between the test site and the main control room, eliminates the risk of ground potential backflash, and greatly improves the absolute accuracy of the acquired parameters and the system reliability.

[0047] The processing module, which is connected to the multi-channel data acquisition module, the DC high-voltage power supply module, and the impulse current generation module, is used to perform the zinc oxide resistor performance test method as described above.

[0048] In this embodiment, the processing module is a model computing unit located in the main control room. In terms of hardware, it is preferably a high-performance industrial server or a dedicated cloud computing platform composed of multi-core Xeon processors, which has extremely powerful floating-point computing capabilities and large-capacity high-speed cache.

[0049] A deterministic timing communication network is established with the aforementioned distributed data acquisition device via a high-speed fiber optic Ethernet switch. This network is responsible for real-time retrieval of data packets synchronously uploaded by each node, containing millisecond-level residual voltage waveforms and long-cycle leakage current trends. The server houses a model calculation core engine written in a high-level programming language. Upon receiving a signal indicating the end of a test cycle, it immediately initiates a multi-dimensional accelerated aging model algorithm, a least-squares kinetic curve fitting algorithm, and a lifetime prediction matrix derivation to achieve real-time, multi-dimensional quantitative assessment of the degradation degree of the zinc oxide resistor.

[0050] The overall implementation principle of the system in this embodiment is that the four core hardware modules—DC power supply array, multi-stage impact generator, distributed optical fiber DAQ system, and central computing server—are formed into a tight organic whole and a closed-loop control link through an industrial Ethernet network.

[0051] During system operation, the server uniformly issues control timing sequences. The DC high-voltage power supply module first switches in and applies DC steady-state stress, and the multi-channel acquisition device begins baseline monitoring. Subsequently, the server triggers the impulse current generator to inject a large current pulse at a predetermined time scale, and the DAQ device synchronously records the residual voltage and instantaneous leakage current mutations within microseconds. After the impulse is completed, the power supply module cuts off the voltage according to the program control instructions, and after a preset resting time, it is powered on again to collect recovery state data. Finally, all full-time-series, cross-timescale feature data are seamlessly integrated into the model calculation unit of the central server to execute the accelerated aging model algorithm.

[0052] The implementation of this system has significantly improved the automation level, overall testing efficiency, and scientific accuracy of degradation prediction in the performance testing of zinc oxide resistors.

[0053] Example 1 Step S101: Apply bias voltage and control polarization stability.

[0054] First, the zinc oxide resistance element (hereinafter referred to as the resistance element), after undergoing non-destructive surface cleaning and defect-free side insulation layer inspection, is placed in a special test fixture inside a constant temperature and humidity test chamber. Shielded high-voltage leads are used for the test leads to eliminate interference from stray electromagnetic fields in the space on the acquisition of microampere-level leakage current.

[0055] Start the DC high-voltage power supply module equipped with this system. In this embodiment, the high-voltage power supply is preferably a switching power supply or an ultra-stable linear high-voltage source with an intelligent voltage loop and current loop dual closed-loop negative feedback control system. Its output voltage ripple coefficient is strictly controlled within 0.05%, the output voltage resolution can reach 0.1V, and the voltage control accuracy error is better than ±0.1%.

[0056] Securely connect the positive and negative terminals of the DC high-voltage power supply to both ends of the zinc oxide resistance element test sample using wires. Based on the physical specifications of the test sample and its actual target grid connection conditions, calculate and dynamically adjust the applied reference bias voltage value U according to the following DC bias voltage setting formula. dc : ; Among them, U 1mA The quasi-static terminal voltage corresponding to the two ends of the zinc oxide resistor sample under test when the DC leakage current flowing through it just reaches 1mA in its factory or initial state is commonly referred to as the DC reference voltage. U The bias electric field amplification voltage coefficient is specifically introduced, and its value is strictly limited to the range of [1.0, 1.2].

[0057] In this embodiment, in order to realistically simulate and moderately accelerate the long-term steady-state rated operation of the test sample in the surge arrester of the converter station of the ultra-high voltage / extra-high voltage DC system, the voltage coefficient k is... U The typical setting is 1.05. When applying this k... U After applying the voltage, subsequent tests cannot be performed immediately. The DC bias voltage must be maintained stably across the test sample for at least 30 minutes.

[0058] The purpose of this step is to allow the polycrystalline structure inside the zinc oxide resistor (composed of conductive zinc oxide grains, a high-resistivity bismuth-rich phase grain boundary layer, and a spinel second phase) to reach thermal and dielectric polarization equilibrium under a constant electric field during the initial 30 minutes. This eliminates the component of the transient capacitive displacement current in the polycrystalline structure, and the current recorded by the high-precision sensor is the most accurate initial purely resistive DC leakage current I. L,0 .

[0059] Step S102: Composite impact injection based on long-period timing.

[0060] The aforementioned DC bias voltage U dc The transient inrush current generator in this system is activated under the condition of continuous and uninterrupted application.

[0061] In this embodiment, the impulse current generator preferably employs a capacitor discharge impulse current generator based on a high-energy, low-inductance pulse capacitor bank, a high-pressure adjustable ball gap trigger, and a waveform shaping wavefront / tail resistor array. The impulse current generator establishes timing communication with the main control computer via a fiber optic distributed bus.

[0062] According to the preset timing sequence of the accelerated aging test, high-energy transient impact current pulses are periodically injected into the continuously pressure-bearing resistor. In order to ensure that the microscopic damage suffered by the grain boundary layer inside the material can form an effective fatigue accumulation effect, and to prevent the heat and charge from completely relaxing and dissipating due to the excessively long interval between two pulses, the present invention strictly limits the injection cycle of the impact current (i.e., the time interval between two adjacent impact pulses) to not less than once per hour (for example, set to 1 hour / time in this accelerated test).

[0063] Based on specific surge arrester lightning protection standards and on-site line overvoltage measurements, the injected impulse current pulse waveform is selected from standard lightning impulse waveforms (apparent wavefront time T1 = 8μs ± 10%, apparent wavetail time T2 = 20μs ± 10% of an 8 / 20μs pulse current) or switching overvoltage impulse waveforms (such as 30 / 60μs large current-carrying pulses). The absolute amplitude of the impulse current exhibits adjustable characteristics, with a linear adjustment range covering 1kA to 20kA. In this embodiment, for the high-voltage DC test specimen, the impulse current amplitude is fixed at I. peak =5kA.

[0064] Step S103: High-frequency multi-channel parameter acquisition spanning from milliseconds to hours.

[0065] Throughout the entire lifecycle of each wave of impact current injection (including before impact, the moment of impact, after impact, and the period of rest), the multi-channel data acquisition device (an ultra-high-speed DAQ card with 16-bit vertical resolution, a single-channel sampling rate of not less than 1MHz, and completely independent A / D synchronous sampling of each channel) equipped with the system works in conjunction with various sensors to extract three-dimensional parameters.

[0066] DC leakage current parameter acquisition: A wideband, microsecond-level response Hall effect current sensor or a non-inductive coaxial shunt is connected in series in the grounding circuit of the test specimen. The acquisition is performed during the extremely short window before the impulse occurs and after the impulse pulse has dissipated and the voltage at the test specimen terminals has returned to steady state U. dc At the set time intervals, the instantaneous value I of the DC leakage current before the impact is extracted. L,before Instantaneous value of DC leakage current I after impact L,after The instantaneous difference in DC leakage current before and after a single impact is calculated in real time based on the aforementioned formula. .

[0067] Simultaneously, throughout the entire test, the DAQ system continuously records the overall current drift trend after N consecutive impacts. The long-term drift after these consecutive impacts is then calculated, representing the permanent increase in leakage current. This increment can extremely sensitively reflect the degree of permanent and irreversible decay of the internal grain boundary barrier due to the impact of high-energy thermoelectric combined electrical stress.

[0068] Impact residual voltage parameter acquisition: A high-bandwidth, high-ratio precision RC composite high-voltage divider (or a high-frequency response resistor divider) is connected in parallel across the two ends of the zinc oxide resistor under test. At the instant of the large impulse current pulse injection, the DAQ card captures the entire transient voltage waveform response across the sample at a high sampling rate of over 1MHz. This accurately locates and locks the peak moment of the impulse current (i.e., t). peak The absolute value of the sample terminal voltage at (time) is taken as the residual voltage U under that impact. res =U(t peak As the number of impacts N increases, the processing module calculates the residual voltage decay slope k of the zinc oxide resistor in each stage based on the acquired historical residual voltage dataset. d (Unit: kV / impact) and residual pressure offset rate δ relative to the initial single impact. res .

[0069] Typically, before severe degradation or the formation of microscopic local short-circuit channels, the residual voltage of zinc oxide resistors will show a gradual downward trend (i.e., the offset rate is negative), which reflects the forced breaking of the grain boundary barrier and the drastic change in macroscopic nonlinear resistance characteristics.

[0070] State recovery parameter branch acquisition: When the accelerated test reaches a specific observation node (e.g., the cumulative impact of a preset wave is completed, or a certain stage of continuous testing ends), the main control host computer issues a programmable command to instantly disconnect the DC high-voltage power supply module and cut off the DC bias voltage applied to both ends of the sample, thereby allowing the sample to enter a completely electric field-free thermodynamic self-recovery stage. This invention requires that the resting time of the resistor in the zero-electric-field state be strictly controlled within a range of not less than 5 minutes (e.g., in this embodiment, the resting self-recovery time is precisely set to 8 minutes). After the resting period, the control system re-controls the DC high-voltage power supply module to rapidly apply the same reference bias voltage value U as in step S101 to the sample. dc The final leakage current recovery value I after re-pressure stabilization is read using the data acquisition card. L,f Next, the corresponding current recovery rate η is calculated. rec .

[0071] Safety classification is performed based on pre-established material physical property judgment criteria: if the calculated η rec ≤120% indicates that although the resistor sheet had previously experienced a high-current impact, the non-equilibrium carriers and defect ions within its grain boundaries could return to their metastable in-situ state through thermal diffusion after being left to stand without an electric field, which is a reversible aging stage; if η recIf the value is greater than 120%, it indicates that the internal structure of the material has undergone irreversible, avalanche-like permanent degradation, and a safety alarm must be activated or the resistor must be discarded.

[0072] Step S104: Construct a multi-dimensional accelerated aging model and obtain numerical solutions based on a multivariable matrix.

[0073] After acquiring the aforementioned full set of multidimensional time-series feature data, the processing module in this system (such as an industrial server deployed with MATLAB or high-performance computing components) formally constructs a multidimensional accelerated aging model. This model introduces the Degradation Index (DI) as a dimensionless core state measurement indicator, and its mathematical matrix expression contains weight functions corresponding to four core independent variable parameters.

[0074] To eliminate the distortion caused by the differences in dimensions and numerical magnitudes between the physical quantities in the model, the four core parameters are first normalized and mapped: DC bias time (Hourly value), Number of impacts (Number of times), permanent increment of leakage current (μA value) and residual pressure offset rate (Absolute percentage value).

[0075] Each parameter is projected onto the dimensionless interval [0,1] using the following normalization formula: ; In the formula, x min and x max These represent the minimum and maximum values ​​of the corresponding parameters within the baseline reference range obtained in laboratory historical extreme destructive aging tests, respectively.

[0076] Next, based on the above formula for calculating the degradation index, a multivariate weighted operator is used to sum the results, and the current degradation index DI is calculated.

[0077] In the formula, w1, w2, w3, and w4 are recommended parameter weight coefficients pre-assigned by the system based on different resistive material formulations (such as high-current formulations and low-residual-voltage formulations) using the Analytic Hierarchy Process (AHP), and strictly satisfy the completeness constraint w1+w2+w3+w4=1. In the general preferred scheme of this embodiment, the recommended initial weight matrix ratio is set as follows: w1=0.2 (time weight), w2=0.3 (impact number weight), w3=0.3 (current permanent increment weight), and w4=0.2 (residual voltage offset rate weight).

[0078] To achieve accurate extrapolation and prediction of unknown future service life, this model uses least squares fitting to establish the degradation index DI and the bias time as the core fundamental independent variable. Number of impacts The evolutionary dynamics correlation equation between them.

[0079] Multiple sets of [t] collected at different time points dc N imp Historical multidimensional data vectors (DI) are used to construct an overdetermined linear equation system. Within the matrix space, the generalized inverse matrix is ​​calculated, thereby obtaining the coefficient vector to be determined. The least squares estimate of the optimal solution.

[0080] Where A represents the measured data of the independent variables at each experimental node [1, t]. dc N imp The matrix is ​​a Vandermonde-type structure matrix formed by converging and arranging elements, where b is the degradation index DI calculated from the actual measurements at each corresponding observation point. k The column vector formed.

[0081] After successfully calculating the material-specific aging rate factors a1 and a2, the processing module introduces a DC operating condition tolerance coefficient to quantitatively predict the remaining life of the sample under the current combined stress environment. The remaining life prediction formula (unit: hours, h) has been explained above.

[0082] In the formula, DI threshold The highest critical threshold representing the generally accepted complete loss of performance of a nonlinear element, necessitating its retirement and disconnection (a scientifically recommended setting in this system is 0.8); DI0 is the real-time degradation index at the completion of the current test wave; f imp The pulse injection frequency (times / hour) is specified in the corresponding network area or established system test specifications, for example, f in this accelerated test. imp =1).

[0083] Through this numerical model, the system can directly output a clear remaining service life, thereby providing data support for the precise operation and maintenance of nonlinear electrical components.

[0084] Application examples

[0085] Application conditions: Zinc oxide resistors for surge arresters in converter stations of 500kV UHVDC transmission systems.

[0086] Sample specifications: outer diameter Φ=70mm, height H=32mm, nominal voltage Un=1.6kV / piece, DC reference voltage U1=2.0kV, 20 pieces in series per group.

[0087] Test objective: To simulate a resistor under combined DC bias and lightning strike conditions, complete 50 cumulative aging cycles, and predict its remaining lifespan. Figure 3 As shown, the testing method in this invention is verified through a single test cycle.

[0088] Step 1: Test preparation and initial parameter recording, as shown in Table 1.

[0089] Table 1. Test Preparation and Initial Parameter Records

[0090] Step 2: Apply DC bias voltage.

[0091] Calculate the DC bias voltage using the following formula: Udc=1.05×U1=1.05×2000V=2100V; After applying a 2100V DC bias and stabilizing for 30 minutes, record the initial leakage current: I L,0 =45μA.

[0092] Step 3: Impact injection test (excerpt of 50 complete processes).

[0093] Impulse waveform: Standard lightning impulse 8 / 20μs; Impulse current amplitude: I peak =5kA; Impact interval: 1h / test. The impact test process data is recorded in Table 2.

[0094] Table 2. Impact Test Process Data Records (Selected from 5 Representative Time Nodes)

[0095] After the 50th impact, calculate the permanent increment of the leakage current: ; Next, calculate the residual voltage attenuation slope: k d =(7.82-7.58) / 50=0.0048kV / time; Simultaneously, calculate the residual pressure offset rate for the 50th time: δ res =(7.58-7.82) / 7.82×100%=-3.07%.

[0096] Step 4: Status recovery test.

[0097] After the 50th impact, disconnect the DC bias, let it stand for 8 minutes (meeting the requirement of ≥5min), and then reapply 2100V.

[0098] Collect the leakage current after it has stabilized: I L,rec =50.2μA.

[0099] The recovery rate is calculated using the following formula: η rec=50.2 / 45.0 × 100% = 111.6%; Judgment: The recovery rate of 111.6% is less than the 120% threshold, indicating that the resistor is in the reversible aging stage and has not yet experienced irreversible failure.

[0100] Step 5: Calculate the Deterioration Index (DI).

[0101] Using the limit value of 100 accelerated tests in the laboratory as a reference range, the parameters were normalized, and the results are shown in Table 3.

[0102] Table 3. Normalized Calculation Table of Deterioration Index (DI) Parameters

[0103] Substituting the recommendation weights (w1=0.2, w2=0.3, w3=0.3, w4=0.2), calculate DI: DI=0.2×0.417+0.3×0.500+0.3×0.274+0.2×0.384; DI=0.083+0.150+0.082+0.077=0.392; DI=0.392 <DI threshold =0.8, judgment: the resistor is qualified and can continue to be used.

[0104] Step Six: Remaining Life Prediction.

[0105] The least squares fit obtained from multiple previous experimental samples yielded: a1 = 0.0025h ﹣1 a2 = 0.004 times ﹣1 Impact frequency f imp =1 time / hour.

[0106] According to T life The calculation formula yields the following result: T life =(0.8-0.392) / (0.0025+0.004×1)=0.408 / 0.0065≈62.8h.

[0107] Under the current ±800kV DC operating conditions, this zinc oxide resistor (ZnO-HVDC-001) is estimated to withstand approximately 62.8 hours of combined stress, equivalent to approximately 63 lightning strikes. It is recommended to conduct a re-inspection at 30 hours (approximately 50% of the resistor's remaining lifespan) to ensure the reliable operation of the converter station's surge arresters.

[0108] While various embodiments of the invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of the invention. The appended claims are intended to define the scope of protection of the invention and therefore cover modular compositions, equivalents, or alternatives within the scope of these claims.

Claims

1. A method for testing the performance of zinc oxide resistors, characterized in that, Includes the following steps: Step S1: Continuously apply an adjustable DC bias voltage across the zinc oxide resistive element being tested to simulate the steady-state voltage conditions during the operation of a DC power system. Step S2: During the continuous operation of the DC bias voltage, an impulse current is periodically injected into the zinc oxide resistor according to a preset injection cycle to simulate the occasional overvoltage intrusion condition in the DC power system. Step S3: Obtain the DC leakage current parameters and impact residual voltage parameters of the zinc oxide resistor before and after the impact current is applied, and collect the state recovery parameters after the DC bias voltage is removed; Step S4: Based on the collected DC leakage current parameters, impact residual voltage parameters, and state recovery parameters, a multi-dimensional accelerated aging model is constructed. The degradation degree of the zinc oxide resistor is quantified and its remaining life is predicted through the multi-dimensional accelerated aging model.

2. The method for testing the performance of zinc oxide resistors according to claim 1, characterized in that: In step S2, the injection cycle of the impact current is not less than once per hour, in order to accelerate the localized accumulation of micro-defects in the grain boundary layer inside the zinc oxide resistor sheet and the accumulation of thermoelectric stress within a limited experimental time.

3. The method for testing the performance of zinc oxide resistors according to claim 1, characterized in that: In step S3, the DC leakage current parameter is obtained as follows: Real-time acquisition of current changes flowing through the zinc oxide resistor; The DC leakage current parameters include the difference between the instantaneous values ​​of the DC leakage current before and after the impact. The calculation formula is: ； Among them, I L,before I represents the instantaneous leakage current before a single impact injection. L,after This represents the instantaneous leakage current after a single impact injection. The DC leakage current parameters also include the long-term drift after multiple consecutive impacts. The calculation formula is: ； Among them, I L,0 I is the initial leakage current before the start of the test cycle. L,N This represents the leakage current value after the Nth impact and when the system reaches a steady state.

4. The method for testing the performance of zinc oxide resistors according to claim 1, characterized in that: In step S3, the impact residual pressure parameter is obtained as follows: Obtain the waveform of the terminal voltage across the zinc oxide resistor at the instant of the impact current injection; The impact residual pressure parameter includes the residual pressure value U at the peak of the impact current. res It is characterized as follows: U res =U(t peak ); Among them, t peak This represents the peak moment of the inrush current waveform; The impact residual pressure parameter also includes the residual pressure attenuation slope k during continuous impact. d and residual pressure offset δ res The calculation formulas are as follows: k d =(U) res,0 -U res,N ) / N; δ res =(U res,N -IN res,0 ) / IN res,0 ×100%; Among them, U res,0 U is the initial residual pressure value at the time of the first impact. res,N This is the measured residual pressure value at the Nth impact.

5. The method for testing the performance of zinc oxide resistors according to claim 1, characterized in that: In step S3, the methods for obtaining the state recovery parameters include: After completing the preset number of impact injections or reaching a stage test node, the DC bias voltage is cut off, and the zinc oxide resistor is left to stand in a field-free state for at least 5 minutes to allow the polarized charges and non-equilibrium charge carriers inside it to undergo thermal relaxation and self-recovery due to the micro-thermoelectric effect. Reapply a DC bias voltage of the same amplitude as in step S1, and collect the recovered leakage current value I after stabilization. L,f ; Calculate the leakage current recovery rate η rec : η rec =I L,f / I L,0 ×100%; When η rec When ≤120%, the zinc oxide resistor is determined to be in the reversible defect recovery stage; when η rec When the value is greater than 120%, it is determined that the internal grain boundary layer structure of the zinc oxide resistor has undergone irreversible permanent lattice distortion and thermal degradation failure.

6. The method for testing the performance of zinc oxide resistors according to claim 1, characterized in that: In step S4, the multi-dimensional accelerated aging model introduces a degradation index DI to quantify the overall performance degradation state of the zinc oxide resistor. The degradation index DI is constructed by weighting and summing the DC bias time parameter, the number of impacts parameter, the permanent increment parameter of leakage current parameter, and the residual voltage offset parameter through their respective parameter weight functions; the calculation formula is: ； In the formula, To accumulate DC bias time, To accumulate the number of impacts suffered, For the permanent increment of leakage current, is the residual pressure offset rate; w1, w2, w3, and w4 are the weighting coefficients of each parameter, and satisfy the constraint condition: w1+w2+w3+w4=1.

7. The method for testing the performance of zinc oxide resistors according to claim 1, characterized in that: In step S2, the impulse current includes a pulse current that meets the standard lightning impulse waveform or the switching overvoltage impulse waveform. The amplitude of the impulse current is dynamically adjusted and set within the range of 1kA to 20kA according to the nominal voltage of the sample under test and the actual operating conditions.

8. The method for testing the performance of zinc oxide resistors according to claim 6, characterized in that: The multidimensional accelerated aging model establishes an empirical correlation equation between the degradation index DI and the main independent variables, cumulative DC bias time and number of impacts, through least squares fitting: ； In the formula, a0 is the constant term correction coefficient, a1 is the DC voltage stress aging rate constant, a2 is the pulse impact stress aging rate constant, and ε is the random measurement noise error term. An overdetermined system of equations was constructed by collecting multiple sets of historical experimental sample data, and the least squares estimated solution was obtained using matrix solving. ； In the formula, Let be the vector of coefficients to be determined, A be the structure matrix constructed from the sample data of the independent variables, and b be the observation vector constructed from the sample data of the measured degradation index.

9. The method for testing the performance of zinc oxide resistors according to claim 8, characterized in that: After obtaining the empirical correlation equation, a DC operating condition tolerance coefficient is introduced to predict the remaining life of the zinc oxide resistor. The remaining life T life The prediction formula is: ； In the formula, DI threshold The preset decommissioning and deactivation threshold is DI0, which is the real-time degradation index calculated at the end of the current test cycle. imp Injecting frequency of impact pulses to test the system or actual operating environment.

10. A performance testing system for zinc oxide resistors, characterized in that, include: The DC high voltage power supply module is used to provide a high-precision DC bias voltage with an adjustable output voltage within a preset range and to maintain its steady-state output during long-cycle accelerated aging tests. The impulse current generating module is used to periodically generate impulse current pulses with preset waveform and preset amplitude characteristics, and efficiently inject them into the zinc oxide resistive sheet; A multi-channel data acquisition module, connected to the zinc oxide resistor and related sensors, is used to synchronously record leakage current parameters, residual voltage parameters and state recovery parameters at high frequency during the combined action of DC bias and impulse pulse. The processing module, which is connected to the multi-channel data acquisition module, the DC high-voltage power supply module and the impulse current generation module, is used to perform the zinc oxide resistor performance test method as described in any one of claims 1 to 9.

Images