All-condition equivalent test system and method for capacitor for direct-current power transmission

Abstract

The invention relates to the technical field of capacitor testing, in particular to a full-working-condition equivalent testing system and method for a capacitor for direct-current power transmission. The method comprises the following steps: applying a single-frequency AC component to a capacitor, and obtaining an inherent single-frequency loss baseline of the capacitor under single-frequency excitation; determining harmonic coupling loss according to the single-frequency loss baseline; strong interaction harmonic pairs are screened out according to harmonic coupling loss, and additional loss is measured again under the condition that controlled thermal gradient disturbance and mechanical resonance excitation are applied to calibrate a thermal field gain factor and a mechanical vibration gain factor; and based on the thermal field gain factor and the mechanical vibration gain factor, evaluating the total loss of the full-working-condition test to obtain full-working-condition test data. According to the method, accurate evaluation and contribution degree decomposition of the loss characteristics of the capacitor under the multi-physical-field full working condition are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of capacitor testing, and in particular to a full-operating-condition equivalent testing system and method for capacitors used in direct current transmission. Background Art

[0002] In actual grid operation, DC transmission capacitors withstand not only DC voltage but also various harmonic voltages and currents generated by power electronics and nonlinear loads. These harmonic components interact nonlinearly, causing additional losses within the capacitor, forming local hotspots, accelerating insulation aging, and, in severe cases, even leading to equipment failures and system failures. However, existing capacitor testing methods suffer from the following technical drawbacks: Traditional capacitor testing systems mostly use single-frequency or simple superposition testing methods, which cannot truly simulate the complex harmonic environment found in actual power grids. These testing systems typically only provide the fundamental wave or a few harmonic components, and the amplitude and phase of each harmonic component cannot be precisely controlled, resulting in significant discrepancies between test results and actual capacitor performance. In reality, in complex harmonic environments, significant nonlinear interactions exist between different frequency components, and the additional losses caused by these effects account for a significant proportion of the total losses. Existing analysis methods are unable to decompose and quantify these nonlinear interactions, resulting in an incomplete and inaccurate assessment of capacitor loss characteristics. Summary of the Invention

[0003] Based on this, the present invention provides a full-operating-condition equivalent testing system and method for capacitors for direct current transmission to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above-mentioned purpose, a full-operating-condition equivalent testing method for a capacitor for DC transmission is provided, comprising the following steps: Step S1: applying a single-frequency AC component to the capacitor, obtaining the inherent loss tangent value of the capacitor under the single-frequency excitation, and generating a single-frequency loss baseline; Step S2: Select a harmonic combination based on the single-frequency loss baseline; perform a difference calculation on the theoretical superposition loss in the single-frequency loss baseline and the actual total power loss of the capacitor when the harmonic combination is synchronously injected, and determine the positive value as the harmonic coupling loss; Step S3: Based on the harmonic coupling loss, a strongly interacting harmonic pair is selected and used as the excitation source. Under the conditions of applying controlled thermal gradient perturbation and mechanical resonance excitation, the additional loss is remeasured to calibrate the thermal field gain factor and the mechanical vibration gain factor; Step S4: Based on the thermal field gain factor and the machine vibration gain factor, the total loss of the full-condition test is evaluated, and the loss deviation value is compared with the preset standard loss value. If the loss deviation value is within the preset range, the test result is locked and packaged into full-condition test data. If the loss deviation value is not within the preset range, the test calibration request is executed, and the subsequent test process is automatically suspended, and all excitation outputs are immediately cut off.

[0005] The present invention further provides a full-operating-condition equivalent testing system for a DC power transmission capacitor, which executes the above-mentioned full-operating-condition equivalent testing method for a DC power transmission capacitor. The full-operating-condition equivalent testing system for a DC power transmission capacitor comprises: The inherent loss acquisition module is used to apply a single-frequency AC component to the capacitor to obtain the inherent loss tangent value of the capacitor under single-frequency excitation and generate a single-frequency loss baseline; The harmonic coupling analysis module is used to select a harmonic combination based on a single-frequency loss baseline. The difference between the theoretical superposition loss in the single-frequency loss baseline and the actual total power loss of the capacitor when the harmonic combination is synchronously injected is calculated, and the positive value is determined as the harmonic coupling loss. The multi-condition test module is used to screen out strongly interacting harmonic pairs based on harmonic coupling loss and use them as excitation sources. Under the conditions of applying controlled thermal gradient perturbations and mechanical resonance excitations, the additional losses are remeasured to calibrate the thermal field gain factor and the mechanical vibration gain factor. The full-condition verification control module is used to evaluate the total loss of the full-condition test based on the thermal field gain factor and the machine vibration gain factor, and compare the loss deviation value with the preset standard loss value. If the loss deviation value is within the preset range, the test result is locked and packaged into full-condition test data. If the loss deviation value is not within the preset range, the test calibration request is executed and the subsequent test process is automatically suspended to cut off all excitation outputs.

[0006] The beneficial effects of the present invention are: On the one hand, by generating a single-frequency loss baseline through a variable-step-size frequency sweep method, and on this basis, accurately calculating the harmonic coupling loss by applying multi-frequency synchronous excitation, the present invention can effectively reveal and quantify the nonlinear interaction between harmonics that is ignored in traditional testing methods; further, by performing a difference operation between the measured coupling loss and the theoretical superposition loss, the additional loss caused by the nonlinear characteristics of the medium can be accurately stripped out, avoiding the evaluation error caused by the simple linear superposition of single-frequency test results, and significantly improving the accuracy of capacitor loss assessment in complex power grid harmonic environments.

[0007] On the other hand, the present invention innovatively introduces controlled thermal gradient disturbance and mechanical resonance excitation, and derives the thermal field gain factor and mechanical vibration gain factor respectively, thereby decoupling and quantifying the coupling effect of the electric-thermal-mechanical multi-physical fields; among them, through regional temperature control and multi-axial vibration excitation, it can truly simulate the complex stresses that the capacitor is subjected to due to uneven heat dissipation and external vibration during actual operation, solving the limitation of traditional testing methods that can only be tested in a single, ideal environment, so that the test results can more comprehensively reflect the actual performance degradation law of the capacitor under all working conditions.

[0008] On the other hand, the present invention obtains the total loss of the full-operating condition test by comprehensively evaluating the basic loss, interaction loss and coupling amplification loss, and compares it with the standard value to form a closed-loop verification and evaluation mechanism. By decomposing the total loss into basic loss, interaction loss and coupling amplification loss, the present invention captures the key performance bottlenecks of the capacitor under actual operating conditions, thereby more accurately evaluating its operating margin and life; in addition, after evaluating the total loss of the full-operating condition, it can provide accurate data support for the design optimization, condition inspection and grid harmonic control of the capacitor, effectively ensuring the long-term safe and stable operation of the entire DC transmission system. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 1. A schematic flow chart of the steps of the full-operating-condition equivalent testing method for capacitors for DC transmission according to the present invention; The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0010] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. It is obvious that the embodiments described are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts are within the scope of protection of the present invention.

[0011] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0012] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0013] To achieve this, please refer to Figure 1 The present invention provides a full-operating-condition equivalent testing method for a capacitor for direct current transmission, comprising the following steps: Step S1: applying a single-frequency AC component to the capacitor, obtaining the inherent loss tangent value of the capacitor under the single-frequency excitation, and generating a single-frequency loss baseline; In an embodiment of the present invention, before the test begins, the test environment is first set to a baseline state, that is, the temperature of each part of the capacitor housing is stabilized at 40.0°C through a regional temperature control unit, and the piezoelectric actuator array is ensured to be in a silent and vibration-free state. On this basis, a rated DC bias voltage is applied to the capacitor through a programmable AC / DC power supply, and a small AC signal component of a specific single frequency is superimposed. The preset harmonic frequency range (for example, the 3rd to 50th harmonics) will be automatically scanned point by point. After stable operation at each frequency point, the loss tangent value corresponding to the frequency will be collected by a high-precision loss measurement bridge. All collected "frequency-loss tangent value" data pairs are eventually integrated into a queryable lookup table, namely the single-frequency loss baseline.

[0014] In one implementation of the present invention, assume the capacitor's rated DC voltage is 800 kV. First, an 800 kV DC bias is applied, then a 150 Hz (3rd harmonic) AC component with an amplitude of 1% of the rated voltage is superimposed. After the operating condition stabilizes for 3 minutes, the loss tangent value at this frequency is collected to be 0.00012. Subsequently, the 150 Hz signal is removed, and a 250 Hz (5th harmonic) AC component is applied. After stabilization, the loss tangent value is collected to be 0.00018. This process is repeated for all frequencies of interest, and the resulting single-frequency loss baseline data table is, for example: {(150 Hz, 0.00012), (250 Hz, 0.00018), (350 Hz, 0.00025), ...}.

[0015] Step S2: Select a harmonic combination based on the single-frequency loss baseline; perform a difference calculation on the theoretical superposition loss in the single-frequency loss baseline and the actual total power loss of the capacitor when the harmonic combination is synchronously injected, and determine the positive value as the harmonic coupling loss; In this embodiment of the present invention, a pair of harmonic frequencies of interest, i.e., a harmonic combination, is selected from a single-frequency loss baseline. Based on this pair of frequencies, their corresponding base loss values ​​are retrieved from the single-frequency loss baseline and summed to obtain the theoretical combined loss without considering any interaction. Subsequently, the power supply is controlled to maintain the same DC bias while simultaneously superimposing the AC components of this pair of harmonic frequencies. The actual total power loss of the capacitor is measured. If the difference between the theoretical combined loss and the actual total power loss is positive, the positive value is determined to be harmonic coupling loss caused by nonlinear interactions between the harmonics.

[0016] In one implementation of the present invention, a harmonic combination (150 Hz, 250 Hz) is selected. Assume that based on the single-frequency loss baseline and capacitor parameters, the power loss under 150 Hz single-frequency excitation is calculated to be 1.2 W, and the power loss under 250 Hz single-frequency excitation is calculated to be 1.8 W. The theoretical superposition loss is then: 1.2W+1.8W=3.0W; Subsequently, 150Hz and 250Hz AC components were injected into the capacitor simultaneously, and the actual total power loss was measured to be 3.5W. The harmonic coupling loss is then calculated as follows: Harmonic coupling loss = 3.5W-3.0W = 0.5W; Therefore, the interaction between the 150Hz and 250Hz harmonics produces an additional loss of 0.5W.

[0017] Step S3: Based on the harmonic coupling loss, a strongly interacting harmonic pair is selected and used as the excitation source. Under the conditions of applying controlled thermal gradient perturbation and mechanical resonance excitation, the additional loss is remeasured to calibrate the thermal field gain factor and the mechanical vibration gain factor; In an embodiment of the present invention, the harmonic coupling loss calculated in the previous step is compared with the theoretical superposition loss. If the harmonic coupling loss is greater than a certain percentage of the theoretical superposition loss (for example, 15%), the harmonic pair is determined to be a strongly interactive harmonic pair. A strongly interactive harmonic pair is selected as a fixed electrical excitation source, and its harmonic coupling loss under a reference environment is recorded as the interactive reference loss. First, a controlled thermal gradient disturbance is applied (for example, hot on the top and cold on the bottom), and the new additional loss is measured under this condition, which is recorded as the additional loss under thermal disturbance. The ratio of the additional loss to the interactive reference loss is the thermal field gain factor. Then, the reference temperature is restored, and a controlled mechanical resonance excitation is applied. The additional loss is also measured, and the machine vibration gain factor is calculated.

[0018] In one implementation of this embodiment of the present invention, for the harmonic combination (150 Hz, 250 Hz), the harmonic coupling loss is 0.5 W, and the theoretical superposition loss is 3.0 W. Because 0.5 W > (3.0 W × 15%), that is, 0.5 W > 0.45 W, this is determined to be a strongly interacting harmonic pair. The 0.5 W harmonic coupling loss is the interaction reference loss.

[0019] 1. Thermal Field Calibration: This electrical excitation was applied while creating a 20°C temperature difference between the top and bottom of the capacitor. The measured additional loss (additional loss due to thermal disturbance) at this point was 0.8W. The thermal field gain factor is: 0.8W / 0.5W = 1.6.

[0020] 2. Mechanical vibration calibration: Return to the reference temperature and apply mechanical vibration via the piezoelectric actuator array. The measured additional loss (additional loss under mechanical vibration) is 0.6W. The mechanical vibration gain factor is: 0.6W / 0.5W = 1.2.

[0021] Step S4: Based on the thermal field gain factor and the machine vibration gain factor, the total loss of the full-condition test is evaluated, and the loss deviation value is compared with the preset standard loss value. If the loss deviation value is within the preset range, the test result is locked and packaged into full-condition test data. If the loss deviation value is not within the preset range, the test calibration request is executed, and the subsequent test process is automatically suspended, and all excitation outputs are immediately cut off.

[0022] In this embodiment of the present invention, a specific target full-operation condition is first defined, encompassing all harmonic components, thermal gradients, and mechanical vibration conditions. The predicted total loss under this target condition, i.e., the full-operation condition total loss, is calculated. This total loss is composed of three components: the sum of all single-frequency losses (basic loss), the sum of the baseline interaction losses of all harmonic pairs (interaction loss), and the additional loss resulting from amplifying the interaction losses by the thermal field gain factor and the mechanical vibration gain factor (coupling amplification loss). This full-operation condition total loss is then compared with a preset standard loss value provided by the manufacturer or determined based on historical data, and a loss deviation is calculated. If the deviation is within the allowable range (e.g., less than 5% of the standard value), the test passes, and the analysis graph containing each loss component is packaged as the full-operation condition test data. If the deviation exceeds the range, calibration is determined for this test, triggering an immediate test calibration request, automatically suspending subsequent processes, and shutting off all excitation outputs to ensure safety.

[0023] In one implementation of the embodiment of the present invention, it is assumed that the target full operating condition includes (150 Hz, 250 Hz) harmonics, a 20°C temperature difference, and 1g, 100 Hz vibration.

[0024] The total loss evaluation of the full-load test is as follows: Basic loss = 3.0W; Interaction loss = 0.5W; Coupling amplification loss = interaction loss × (thermal field gain factor - 1) + interaction loss × (machine vibration gain factor - 1) = 0.5W × (1.6-1) + 0.5W × (1.2-1) = 0.3W + 0.1W = 0.4W; Total power loss in full-load test = 3.0W + 0.5W + 0.4W = 3.9W.

[0025] Assuming the preset standard loss value of the capacitor is 4.0W, the loss deviation value is: |3.9W-4.0W|=0.1W.

[0026] The preset allowable range is 5% of the standard value, that is, 4.0W×5%=0.2W.

[0027] Since 0.1W is within the preset range (±0.2W), the test is determined to have passed, the current test result is locked, and it is packaged into full-condition test data.

[0028] It should be noted that if the deviation exceeds the range, it is determined that the test needs to be calibrated, and the test calibration request is immediately triggered, and the subsequent process is automatically suspended. Specifically: When the predicted deviation is not less than 5% of the predicted total loss, a calibration request flag is generated and written synchronously to the trailing edge of the current test cycle; After the calibration request flag takes effect, a parameter backtracking action is immediately triggered, causing the test process to return to step S1 and forcing the single-frequency loss baseline to be regenerated with half the original frequency interval; After the parameter backtracking action is completed, the calibration request flag and the current prediction deviation are encapsulated into a calibration log frame and pushed into the system maintenance queue in sequence; When the system maintenance queue is full, the subsequent test process will be automatically suspended and all stimulus outputs will be cut off.

[0029] Preferably, step S1 includes the following steps: Step S11: superimposing an AC component with an initial frequency of 50 Hz and an amplitude of 1% of the rated voltage on the preset rated DC bias voltage; Step S12: applying an AC component to the capacitor and collecting the first loss tangent value, which is recorded as the first frequency loss point; Step S13: Taking the first frequency loss point as a reference, using 50 Hz as the step size for increasing the frequency of the AC component, and performing frequency point scanning by sequentially replacing the AC component frequency; wherein, after each replacement, instantaneous forced air cooling is immediately applied to the lead terminals of the capacitor; Step S14: collecting the loss tangent value at each frequency point, subtracting it from the loss value at the previous frequency point, and calculating the loss change rate; Step S15: Compare the loss change rate with a preset response threshold; if it is less than the preset response threshold, mark the frequency point as a flat frequency point; if it is greater than the preset response threshold, mark the frequency point as a sensitive frequency point and reduce the frequency scan step to 5 Hz; Step S16: Connect all the flat frequency points and sensitive frequency points and their corresponding loss tangent values ​​in frequency order to draw a single-frequency loss baseline.

[0030] In one implementation of the present invention, assuming the rated DC voltage of the capacitor is 800 kV, the power supply output is set to 800 kV DC, and a sinusoidal AC voltage with a frequency of 50 Hz and a peak-to-peak value of 16 kV (i.e., an amplitude of 8 kV, which is 1% of 800 kV) is superimposed.

[0031] In one implementation of the present invention, after applying a voltage containing a 50 Hz AC component and stabilizing for 3 minutes, the loss tangent value collected and recorded by the loss measurement bridge is 0.00008. This data point (50 Hz, 0.00008) is stored as the first frequency loss point.

[0032] In this embodiment of the present invention, a frequency sweep is automatically performed, starting at the frequency of the first frequency loss point (50 Hz) and with an initial step size of 50 Hz. After each frequency point is measured, the power supply removes the AC component of the current frequency and replaces it with the AC component of the next frequency point (current frequency + 50 Hz). Between each frequency switch, a cooling unit located near the capacitor terminals is controlled to perform a brief (e.g., 10 seconds) forced air cooling to quickly eliminate localized heat buildup from the previous frequency point test, ensuring that the thermal initial conditions for each measurement are essentially consistent.

[0033] In one implementation of the present invention, after completing the 50 Hz measurement, the AC component is switched to 100 Hz. During this switching process, the lead terminals are forced to cool with air for 10 seconds. After completing the 100 Hz measurement, the frequency is switched to 150 Hz and air-cooled again, repeating the cycle.

[0034] In one implementation of the present invention, assuming that the loss value measured at 100 Hz is 0.00010 and the loss value at the previous frequency point (50 Hz) is 0.00008, the loss change rate at 100 Hz is calculated as follows: Loss change rate = (0.00010-0.00008) / 0.00008=0.25, or 25%.

[0035] In one implementation of the present invention, the preset response threshold is 20%. The calculated loss change rate at 100 Hz is 25%, which is greater than 20%. Therefore, 100 Hz is marked as a sensitive frequency point, and the next test frequency point is automatically set to 105 Hz instead of the originally planned 150 Hz.

[0036] It should be noted that the preset response threshold is determined based on the loss peak region estimated based on the dielectric polarization relaxation time constant of the capacitor dielectric material, so as to improve the accuracy of capturing the characteristic frequency.

[0037] In this embodiment of the present invention, after completing an adaptive step scan across the entire preset frequency range (e.g., up to 2500 Hz), all marked flat and sensitive frequency points and their corresponding tangent loss data are sorted in ascending order of frequency. Finally, these discrete data points are connected with a smooth curve to form a visual spectrum, which serves as the final single-frequency loss baseline.

[0038] In one implementation of the present invention, assume that the resulting data point sequence is {(50 Hz, 0.00008), (100 Hz, 0.00010), (105 Hz, 0.00011), (110 Hz, 0.00013), ..., (150 Hz, 0.00020), ...}. These data points are plotted as a curve with frequency on the horizontal axis and loss tangent on the vertical axis. The curve shows a significantly higher data point density near 100 Hz than in other regions, accurately depicting the characteristics of the loss-sensitive region.

[0039] Preferably, step S2 includes the following steps: Step S21: At the beginning of the current test cycle, the first harmonic frequency and the second harmonic frequency are selected in sequence based on the single-frequency loss baseline; wherein the first harmonic frequency is greater than the second harmonic frequency; Step S22: taking the first harmonic frequency and the second harmonic frequency as a harmonic combination; Step S23: Obtain the loss tangent value of each single frequency corresponding to the harmonic combination, calculate its linear superposition equivalent loss, and obtain the theoretical superposition loss; Step S24: In the current test cycle, the power supply is controlled to inject two-frequency AC components into the capacitor by synchronously superimposing the harmonic combination on the DC bias, and the actual total power loss of the capacitor is measured after the injection is stable; Step S25: performing a difference operation on the actual total power loss and the theoretical superposition loss, and determining the obtained positive value as the harmonic coupling loss.

[0040] In this embodiment of the present invention, the test control system initiates an automated harmonic pair traversal process at the beginning of the current test cycle. This process generates the harmonic frequency pairs to be tested based on preset rules, based on all frequency points recorded in the single-frequency loss baseline. To ensure systematic and non-repetitive traversal, the program sequentially selects a frequency as the first harmonic frequency (fi) and, from frequencies less than fi, selects a second harmonic frequency (fj).

[0041] In one implementation of this embodiment of the present invention, assume that the single-frequency loss baseline contains the frequency points {150 Hz, 250 Hz, 350 Hz, ...}. At the beginning of a test cycle, the program first selects 350 Hz as the first harmonic frequency, then selects 250 Hz and 150 Hz as the second harmonic frequencies, thereby generating the frequency pairs (350 Hz, 250 Hz) and (350 Hz, 150 Hz) to be tested. In the next cycle, the program selects 250 Hz as the first harmonic frequency and 150 Hz as the second harmonic frequency, generating the frequency pair (250 Hz, 150 Hz) to be tested.

[0042] In one implementation of the present invention, after selecting 350 Hz and 250 Hz, a harmonic combination is generated, whose data content is {f1: 350, f2: 250}. This combination will be sent to the test sequence queue and wait for execution.

[0043] In the embodiment of the present invention, after receiving the harmonic combination, the two frequency values ​​in the combination are immediately used as indexes to query the single-frequency loss baseline data, and the loss tangent values ​​of the two frequencies under the action of the two frequencies are extracted respectively ( and ). Then, based on known parameters such as the rated voltage (U_rated) and capacitance (C) of the capacitor, the formula Calculate the power loss for each frequency individually (P1 and P2). Finally, add these two power loss values ​​to get the total loss assuming no interaction, which is the theoretical superposition loss.

[0044] In this embodiment of the present invention, the two frequency values ​​of the harmonic combination are transmitted to a programmable AC / DC power supply. While maintaining the rated DC bias voltage, the power supply's internal arbitrary waveform generator precisely synthesizes a composite waveform containing the two AC components and applies it to the capacitor. A preset stabilization time (e.g., 3 minutes) is also allowed to ensure the capacitor reaches a new electro-thermal equilibrium. After stabilization, a high-precision power analyzer or loss measurement bridge performs a comprehensive measurement, directly deriving the total active power consumed by the capacitor under this dual-frequency excitation. This value is recorded as the actual total power loss.

[0045] In one implementation of the embodiment of the present invention, the data in the previous step is calculated: Actual total power loss = 5.0W; theoretical superposition loss = 4.3W; harmonic coupling loss = 5.0W-4.3W = 0.7W.

[0046] The positive value of 0.7 W is determined to be the harmonic coupling loss generated by the harmonic combination (350 Hz, 250 Hz), and is stored in association with the harmonic combination.

[0047] In another implementation of the embodiment of the present invention, two frequency values ​​in the harmonic combination are obtained as inputs, and the measured loss response is used as output to construct a dual-frequency input response data pair; Using dual-frequency input response data pairs, based on the Volterra series expansion method, the second-order Volterra kernel function that characterizes the nonlinear interaction of the two frequency components is extracted to generate the second-order interaction kernel function. The second-order interaction kernel function is evaluated at two frequencies in the harmonic combination, and its result is directly determined as the harmonic coupling loss.

[0048] Preferably, calibrating the thermal field gain factor in step S3 includes: When the harmonic coupling loss is greater than 15% of the theoretical superposition loss, the harmonic pair is determined to be a strong interaction harmonic pair; otherwise, it is determined to be a weak interaction harmonic pair and is eliminated; Using a strong interaction harmonic pair as an excitation source electrical signal, and obtaining the frequency value of the harmonic pair as an interaction reference loss; When applying an excitation source electrical signal, a thermal gradient perturbation is applied to the capacitor housing area. The thermal gradient perturbation sets a target temperature of 70°C at the top of the capacitor and a target temperature of 50°C at the bottom. The heating power is monitored and adjusted in real time until the temperature difference between the top and bottom stabilizes at 20°C ± 1°C. Re-measure the total thermal power loss of the capacitor under the condition of maintaining the thermal gradient disturbance and the excitation source electrical signal; The difference between the total power loss of thermal stimulation and the theoretical superposition loss corresponding to the strong interaction harmonic pair is calculated to obtain the additional loss under thermal disturbance; The ratio of the additional loss to the mutual reference loss under thermal disturbance is calculated to obtain the thermal field gain factor.

[0049] In one implementation of an embodiment of the present invention, a threshold judgment logic is built in, and the threshold is set to 15%. If the calculated harmonic coupling loss is greater than 15% of the theoretical superposition loss, the harmonic combination is marked as a strong interaction harmonic pair and retained in the subsequent test sequence. On the contrary, if it is less than or equal to 15%, it is marked as a weak interaction harmonic pair. For the harmonic combination (350Hz, 250Hz), its harmonic coupling loss is 0.7W, and the theoretical superposition loss is 4.3W. Calculation: 4.3W×15%=0.645W. Since 0.7W>0.645W, the combination is judged as a strong interaction harmonic pair and added to the thermal field gain factor calibration task queue.

[0050] In one implementation of this embodiment, a strongly interacting harmonic pair is extracted from the task queue, and its corresponding frequency values ​​(for example, 350 Hz and 250 Hz) are set as the fixed electrical excitation for subsequent testing, or the excitation source electrical signal. For example, (350 Hz, 250 Hz) is selected as the excitation source electrical signal. Simultaneously, the interaction reference loss is locked to 0.7 W. This 0.7 W value represents the additional loss caused purely by electrical interaction and serves as the "zero point" for measuring the impact of subsequent physical field perturbations.

[0051] In an embodiment of the present invention, while continuously applying an excitation source electrical signal, a regional temperature control unit is activated. This unit issues instructions to the heating / cooling modules located at the top and bottom of the capacitor according to a preset thermal gradient perturbation scheme. Temperature data is fed back in real time by multiple temperature sensors arranged at the top and bottom of the housing, forming a closed-loop PID control loop. The power of the top heating module and the bottom cooling module is continuously adjusted until the average temperature at the top stabilizes at 70°C±0.5°C and the average temperature at the bottom stabilizes at 50°C±0.5°C, thereby ensuring that the temperature difference between the two is precisely maintained within the target range of 20°C±1°C.

[0052] In another implementation of the present invention, when the top temperature is monitored to be 68°C and the bottom temperature is 52°C, the PID controller increases the PWM duty cycle of the top heating wire and reduces the driving current of the bottom Peltier cooler until the sensor readings reach the target values ​​of 70°C and 50°C, respectively, and remain stable within ±0.5°C.

[0053] In one implementation of an embodiment of the present invention, confirming that the thermal gradient has stabilized and the excitation source electrical signal is still being applied will trigger the power analyzer to perform a new power loss measurement. Assuming a stable temperature difference of 70°C at the top and 50°C at the bottom, the power analyzer measures the total power loss of the capacitor to be 5.4W, which is recorded as the total thermal power loss.

[0054] In one implementation of the present invention, the total power loss due to thermal shock is 5.4 W, and the corresponding theoretical superposition loss is 4.3 W. The additional loss under thermal disturbance is calculated as follows: Additional loss under thermal disturbance = 5.4W-4.3W=1.1W.

[0055] In one implementation of the present invention, the additional loss under thermal disturbance is 1.1 W, and the cross-reference loss is 0.7 W. The thermal field gain factor is calculated as follows: Thermal field gain factor = 1.1W / 0.7W≈1.57.

[0056] This result shows that a temperature difference of 20°C amplifies the additional interaction loss of the (350Hz, 250Hz) pair of harmonics by about 57%.

[0057] Preferably, applying thermal gradient disturbance in the capacitor housing region further comprises: A first temperature control sequence is set for the upper area of ​​the capacitor housing, with the target temperature fluctuating sinusoidally between 60°C and 80°C at a frequency of 0.1 Hz. A second temperature control sequence is set for the lower area of ​​the capacitor housing, with the target temperature fluctuating between 60°C and 40°C at a frequency of 0.1 Hz in an anti-phase sinusoidal pattern; The phase difference between the first temperature control timing sequence and the second temperature control timing sequence is fixed at 180 degrees; The temperature difference value and the corresponding total thermal power loss are synchronously recorded according to each control sampling point in the first temperature control sequence and the second temperature control sequence to form a dynamic loss sequence.

[0058] In this embodiment of the present invention, a dynamic target temperature instruction sequence, namely the first temperature control sequence, is sent to the regional temperature control unit. This sequence specifies that the target temperature of the upper region of the capacitor housing is no longer a fixed value, but instead varies sinusoidally between 60°C and 80°C, centered at 70°C. The frequency of this sinusoidal fluctuation is set to 0.1Hz, meaning the period is 10 seconds.

[0059] In one implementation of the embodiment of the present invention, the target temperature of the first temperature control sequence is It can be described by the following formula: ; in is the time (seconds), is the value of pi. =0s, the target temperature is 70°C; = 2.5s, the target temperature reaches a peak value of 80°C; at t = 7.5s, it reaches a valley value of 60°C.

[0060] In this embodiment of the present invention, a second temperature control sequence is set for the lower area of ​​the housing in conjunction with the temperature control sequence for the upper area. This sequence sets the target temperature for the lower area to fluctuate sinusoidally between 60°C and 40°C, centered at 50°C, with the same oscillation frequency of 0.1Hz. Crucially, the phase of this sequence is set to be exactly opposite to that of the first temperature control sequence.

[0061] In one implementation of the embodiment of the present invention, the target temperature of the second temperature control sequence is It can be described by the following formula: ; exist = 2.5s, when the upper temperature reaches a peak value of 80°C, the lower temperature reaches a valley value of 40°C; =7.5s, when the lower temperature reaches a peak value of 60°C, the upper temperature reaches a valley value of 60°C.

[0062] In another implementation of the embodiment of the present invention, the generation of two time series can be expressed as: and , in . Equivalent to , thus achieving perfect anti-phase control. This means that when the upper area is in the fastest heating stage, the lower area is in the fastest cooling stage.

[0063] In the embodiment of the present invention, at each sampling moment during the entire cycle of the dynamic thermal gradient perturbation, two actions are performed synchronously: first, the instantaneous values ​​of the temperature sensors at the upper and lower parts of the housing are read and the instantaneous temperature difference (ΔT(t)) is calculated; second, the power analyzer is triggered to collect the instantaneous total thermal power loss ( ). For each pair (ΔT(t), ) data as a record point and stored in a sequence. After completing one or more complete temperature fluctuation cycles, this sequence containing a large number of instantaneous data points constitutes a dynamic loss sequence.

[0064] It should be noted that this dynamic loss series is different from the single thermal field gain factor obtained from the previous static test. It can reveal the response characteristics of the loss to the dynamic temperature difference, including the existence of the delay effect (hysteresis loop). For example, when the temperature difference is the same 30°C, the loss measured during the heating process is different from the loss measured during the cooling process.

[0065] Preferably, calibrating the machine vibration gain factor in step S3 includes: The capacitor is restored to the reference temperature, and a mechanical resonance excitation of 10 Hz to 2000 Hz is applied through the piezoelectric actuator array to dynamically identify the mechanical resonance frequency. Apply a constant amplitude mechanical vibration to the capacitor shell with the mechanical resonance frequency as the center frequency, inject the excitation source electrical signal synchronously during the vibration period, and read the additional loss under mechanical vibration; The ratio of the additional loss under computer vibration to the mutual reference loss is used to obtain the machine vibration gain factor.

[0066] In one implementation of an embodiment of the present invention, after completing the calibration of the thermal field gain factor, the heating is first stopped and the cooling system is started to restore the overall temperature of the capacitor to a reference temperature, for example, 25°C±1°C. After the temperature stabilizes, the test control system starts the piezoelectric actuator array connected to the capacitor base. The control signal generator outputs a linear sweep signal to drive the actuator array to generate mechanical vibration that increases linearly from 10Hz to 2000Hz, and the sweep rate is set to 10Hz / s. During this process, multiple acceleration sensors installed at key positions of the capacitor housing monitor the vibration response in real time.

[0067] In one implementation of this embodiment, a logarithmic frequency sweep was performed from 10 Hz to 2000 Hz. When the excitation frequency reached 455 Hz, the vibration acceleration amplitude detected by the sensor reached a maximum value of 3.5 g, significantly higher than the response at other frequencies. Therefore, 455 Hz was determined and stored as the mechanical resonant frequency of the capacitor.

[0068] In one implementation of the present invention, a piezoelectric actuator array was instructed to continuously vibrate a capacitor at a frequency of 455 Hz and an acceleration amplitude of 1 g. Simultaneously, an excitation source signal of (350 Hz, 250 Hz) was applied to the capacitor. After stabilization, the measured total power loss was 5.1 W. This corresponds to a theoretical superimposed loss of 4.3 W. Therefore, the additional loss under machine vibration is calculated as follows: Additional loss under machine vibration = 5.1W-4.3W=0.8W.

[0069] In one implementation of the present invention, the additional loss under machine vibration is 0.8 W, and the cross-reference loss is 0.7 W. The machine vibration gain factor is calculated as follows: Machine vibration gain factor = 0.8W / 0.7W≈1.14.

[0070] This result shows that at a resonant frequency of 455 Hz and an amplitude of 1 g, mechanical vibration amplifies the additional loss of the harmonic pair (350 Hz, 250 Hz) by approximately 14%. This mechanical vibration gain factor will be stored and used together with the thermal field gain factor in the final full-operation loss assessment.

[0071] Preferably, dynamically identifying the mechanical resonance frequency point includes: To control the piezoelectric actuator in the X-axis direction, a first sweep frequency excitation sequence is independently generated, and the excitation frequency is set to increase logarithmically from 10 Hz to 2000 Hz; To control the piezoelectric actuator in the Y-axis direction, a second sweep frequency excitation sequence is independently generated. The excitation frequency is set to decrease in an antilogarithmic pattern from 2000 Hz to 10 Hz, and the starting point of the second sweep frequency excitation sequence is aligned in time with the center point of the first sweep frequency excitation sequence. The response amplitude of the capacitor housing is captured using an accelerometer; At each frequency step of the first sweep excitation sequence, if the acceleration response amplitude in the X-axis direction exceeds a preset threshold, the current excitation frequency and response amplitude are pushed into the X-axis resonance candidate stack as a data pair; At each frequency step of the second sweep excitation sequence, if the acceleration response amplitude in the Y-axis direction exceeds a preset threshold, the current excitation frequency and response amplitude are pushed into the Y-axis resonance candidate stack as a data pair; After both frequency sweep sequences are completed, all data pairs in the X-axis resonance candidate stack and the Y-axis resonance candidate stack are compared, and the frequency point with the maximum response amplitude is determined as the mechanical resonance frequency point.

[0072] In one implementation of the embodiment of the present invention, it is assumed that the entire frequency sweep process lasts for 100 seconds. The frequency of the first frequency sweep excitation sequence is The formula can be Generate, where is the time from 0 to 100 seconds. =50s, the frequency is about 141Hz, and =90s, the frequency has reached about 1012Hz, reflecting the characteristics of logarithmic scanning.

[0073] In this embodiment of the present invention, a second swept-frequency excitation sequence is independently generated for the piezoelectric actuator subarray controlling Y-axis vibration. The excitation frequency of this sequence is set to follow an anti-logarithmic pattern, decreasing from 2000 Hz to 10 Hz, with dense sweeps at high frequencies and sparse sweeps at low frequencies. To optimize the testing process and stagger the excitation peaks in the two axes, the start time of the second swept-frequency excitation sequence is precisely timed to align with the center of the first swept-frequency excitation sequence.

[0074] In one implementation of the embodiment of the present invention, the total duration of the first sweep frequency excitation sequence is 100 seconds, and the center point is = 50 seconds. Therefore, the second sweep excitation sequence is =50 seconds, and lasts for 100 seconds. =150 seconds. This timing arrangement avoids high-frequency or low-frequency excitation of the two axes at the same time, which helps to reduce the instantaneous power demand on the power supply.

[0075] In this embodiment of the present invention, two orthogonal accelerometers, mounted on the capacitor housing and used to measure X-axis and Y-axis vibrations, respectively, operate continuously throughout the entire frequency sweep excitation process. These sensors convert the captured vibration signals into voltage signals. After signal conditioning and high-speed ADC sampling, they extract the vibration response amplitude (typically expressed in g, representing the acceleration due to gravity) corresponding to the current excitation frequency in real time using methods such as a fast Fourier transform (FFT) or digital filtering.

[0076] In this embodiment of the present invention, during the execution of the first sweep excitation sequence, the response amplitude of the X-axis accelerometer is compared in real time with a preset response threshold (e.g., 0.5g). This threshold is used to initially filter out significant vibration peaks and ignore minor background vibrations.

[0077] In one implementation of this embodiment of the present invention, when the X-axis excitation frequency sweeps to 125 Hz, the X-axis sensor measures a response amplitude of 0.8 g, which is greater than the 0.5 g threshold. A data pair (125 Hz, 0.8 g) is immediately generated and pushed into the X-axis resonance candidate stack.

[0078] In this embodiment, the Y-axis accelerometer response amplitude is compared to the same threshold value during the second sweep excitation sequence (t = 50s to 150s) using the same processing as the X-axis. If the threshold is exceeded, the current Y-axis excitation frequency and response amplitude are packaged into a data pair and pushed into the Y-axis resonance candidate stack.

[0079] In another implementation of this embodiment, when the Y-axis excitation frequency sweeps to 455 Hz (where t ≈ 118 s), the Y-axis sensor measures a response amplitude of 3.5 g, significantly exceeding the 0.5 g threshold. A data pair (455 Hz, 3.5 g) is generated and pushed into the Y-axis resonance candidate stack.

[0080] It should be noted that to avoid strong coupling interference between the two axial excitations at the same time, the start time of the second sequence is set to the center point of the first sequence, that is, starting at t = 50s. In this way, when the X-axis excitation sweeps to the mid-frequency range, the Y-axis excitation starts sweeping at the high frequency range, achieving temporal staggered excitation.

[0081] In this embodiment of the present invention, after the entire frequency sweep (up to t = 150 seconds) is completed, all data pairs in the X-axis resonance candidate stack and the Y-axis resonance candidate stack are traversed, and the response amplitudes in each data pair are compared to find the global maximum. After the maximum response amplitude is found, the excitation frequency in that data pair is extracted and ultimately determined as the primary mechanical resonant frequency of the entire capacitor structure under the test conditions.

[0082] Preferably, in step S4, evaluating the total loss of the full-condition test based on the thermal field gain factor and the machine vibration gain factor includes: Extract and accumulate the loss values ​​of all corresponding harmonics from the single-frequency loss baseline to generate the basic loss component; Traverse all harmonic pairs from the harmonic coupling loss and accumulate their corresponding harmonic additional losses to generate the interaction loss component; Calculate the coupling amplification loss component based on the interaction loss component according to the thermal field gain factor and the machine vibration gain factor; The basic loss component, interaction loss component, and coupling amplification loss component are combined in numerical order to evaluate the total loss of the full-condition test.

[0083] In one implementation of the present invention, assume that the harmonic list of the target operating condition is {150Hz, 250Hz, 350Hz}. From the single-frequency loss baseline, it is found that the power loss corresponding to 150Hz is 1.2W; the power loss corresponding to 250Hz is 1.8W; and the power loss corresponding to 350Hz is 2.5W. The basic loss component is calculated as follows: Basic power loss component = 1.2W + 1.8W + 2.5W = 5.5W.

[0084] In one implementation of the embodiment of the present invention, based on the harmonic list {150 Hz, 250 Hz, 350 Hz}, the harmonic pairs (150 Hz, 250 Hz), (150 Hz, 350 Hz), and (250 Hz, 350 Hz) are generated. The results show that the harmonic coupling loss of (150 Hz, 250 Hz) is 0.5 W; the harmonic coupling loss of (150 Hz, 350 Hz) is 0.6 W; and the harmonic coupling loss of (250 Hz, 350 Hz) is 0.7 W. The interaction loss component is calculated as follows: Interaction loss component = 0.5W + 0.6W + 0.7W = 1.8W.

[0085] In one implementation of an embodiment of the present invention, it is assumed that the target operating condition sets a temperature difference of 20°C and a vibration at a resonant frequency of 1g. The corresponding thermal field gain factor is found to be 1.6, and the machine vibration gain factor is 1.14. The interaction loss component is 1.8W. The coupled amplification loss component is calculated as follows: coupled amplification loss component = 1.8W × (1.6-1) + 1.8W × (1.14-1) = 1.8W × 0.6 + 1.8W × 0.14 = 1.08W + 0.252W = 1.332W.

[0086] In an embodiment of the present invention, after all loss components are calculated, the three components - basic loss component, interaction loss component, and coupling amplification loss component - are arithmetically summed. The final result obtained is an accurate assessment value of the total loss of the capacitor under the target full operating conditions, that is, the total loss of the full operating condition test.

[0087] Preferably, in step S4, if the loss deviation value is within a preset range, locking the test result and packaging it into full-condition test data includes: Calculate the loss deviation value by using the preset capacitor standard loss value and the total loss of the full-condition test; When the loss deviation value is less than 5% of the total loss of the full-condition test, the test status is marked as verification passed at the end of the current test cycle; After the verification pass flag takes effect, the current test results are locked to construct a full-condition effect matrix, thereby encapsulating it into full-condition test data.

[0088] In this embodiment of the present invention, after evaluating the total loss from the full-condition test, a preset standard loss value for the capacitor model under the same or similar operating conditions is retrieved from an internal database or external file. This standard value can be provided by the manufacturer or set based on extensive historical test data or authoritative simulation results. The calculation unit subtracts the estimated total loss from the full-condition test from this preset standard loss value and takes the absolute value. The resulting difference is determined as the loss deviation value.

[0089] In one implementation of this embodiment of the present invention, the loss deviation value is 0.332 W. The tolerance threshold is calculated as: 8.632 W × 5% = 0.4316 W. Since 0.332 W < 0.4316 W, the condition is met, and the status bit Test_Status is set to 0x01 (representing "verification passed") at the end of the current cycle.

[0090] In an embodiment of the present invention, once the system detects that the verification pass flag is set, it will immediately trigger the final data packaging program. This program will lock all key data obtained in the current evaluation cycle to prevent it from being modified by subsequent operations. Next, the program will organize this data in a structured manner to construct a full-operating-condition effect matrix. This matrix clearly displays the contribution of each harmonic, each physical field perturbation, and their interactions to the total loss in the form of a multidimensional array or nested data structure. Finally, this matrix, along with basic test information (such as capacitor model, test time, target operating condition parameters, etc.), is packaged into an independent, portable digital file, which is the final delivered full-operating-condition test data.

[0091] In another implementation of the embodiment of the present invention, the full operating condition effect matrix may be constructed as follows: The absolute values ​​of harmonic coupling loss, thermal effect loss, and mechanical effect loss are taken as the first column data of the matrix; The respective loss contribution rates are taken as the second column of the matrix; The harmonic frequency, temperature gradient and mechanical vibration frequency used in the test are taken as the third column data of the matrix; The three columns of data are combined into a structured data, which is determined as the full operating condition effect matrix.

[0092] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0093] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A full-operating-condition equivalent test method for capacitors for DC transmission, characterized in that: The following steps are involved: Step S1: applying a single-frequency AC component to the capacitor, obtaining the inherent loss tangent value of the capacitor under the single-frequency excitation, and generating a single-frequency loss baseline; Step S2: Select a harmonic combination based on the single-frequency loss baseline; perform a difference calculation on the theoretical superposition loss in the single-frequency loss baseline and the actual total power loss of the capacitor when the harmonic combination is synchronously injected, and determine the positive value as the harmonic coupling loss; Step S3: Based on the harmonic coupling loss, a strongly interacting harmonic pair is selected and used as the excitation source. Under the conditions of applying controlled thermal gradient perturbation and mechanical resonance excitation, the additional loss is remeasured to calibrate the thermal field gain factor and the mechanical vibration gain factor; Step S4: Based on the thermal field gain factor and the machine vibration gain factor, the total loss of the full-condition test is evaluated, and the loss deviation value is compared with the preset standard loss value. If the loss deviation value is within the preset range, the test result is locked and packaged into full-condition test data. If the loss deviation value is not within the preset range, the test calibration request is executed, and the subsequent test process is automatically suspended, and all excitation outputs are immediately cut off.

2. The full-operating-condition equivalent testing method for capacitors for DC transmission according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: superimposing an AC component with an initial frequency of 50 Hz and an amplitude of 1% of the rated voltage on the preset rated DC bias voltage; Step S12: applying an AC component to the capacitor and collecting the first loss tangent value, which is recorded as the first frequency loss point; Step S13: Taking the first frequency loss point as a reference, using 50 Hz as the step size for increasing the frequency of the AC component, and performing frequency point scanning by sequentially replacing the AC component frequency; wherein, after each replacement, instantaneous forced air cooling is immediately applied to the lead terminals of the capacitor; Step S14: collecting the loss tangent value at each frequency point, subtracting it from the loss value at the previous frequency point, and calculating the loss change rate; Step S15: Compare the loss change rate with a preset response threshold; if it is less than the preset response threshold, mark the frequency point as a flat frequency point; if it is greater than the preset response threshold, mark the frequency point as a sensitive frequency point and reduce the frequency scan step to 5 Hz; Step S16: Connect all the flat frequency points and sensitive frequency points and their corresponding loss tangent values ​​in frequency order to draw a single-frequency loss baseline.

3. The full-operating-condition equivalent testing method for capacitors for DC transmission according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: At the beginning of the current test cycle, the first harmonic frequency and the second harmonic frequency are selected in sequence based on the single-frequency loss baseline; wherein the first harmonic frequency is greater than the second harmonic frequency; Step S22: taking the first harmonic frequency and the second harmonic frequency as a harmonic combination; Step S23: Obtain the loss tangent value of each single frequency corresponding to the harmonic combination, calculate its linear superposition equivalent loss, and obtain the theoretical superposition loss; Step S24: In the current test cycle, the power supply is controlled to inject two-frequency AC components into the capacitor by synchronously superimposing the harmonic combination on the DC bias, and the actual total power loss of the capacitor is measured after the injection is stable; Step S25: performing a difference operation on the actual total power loss and the theoretical superposition loss, and determining the obtained positive value as the harmonic coupling loss.

4. The full-operating-condition equivalent testing method for capacitors for DC transmission according to claim 3, characterized in that: The thermal field gain factor calibration in step S3 includes: When the harmonic coupling loss is greater than 15% of the theoretical superposition loss, the harmonic pair is determined to be a strong interaction harmonic pair; otherwise, it is determined to be a weak interaction harmonic pair and is eliminated; Using a strong interaction harmonic pair as an excitation source electrical signal, and obtaining the frequency value of the harmonic pair as an interaction reference loss; When applying an excitation source electrical signal, a thermal gradient perturbation is applied to the capacitor housing area. The thermal gradient perturbation sets a target temperature of 70°C at the top of the capacitor and a target temperature of 50°C at the bottom. The heating power is monitored and adjusted in real time until the temperature difference between the top and bottom stabilizes at 20°C ± 1°C. Re-measure the total thermal power loss of the capacitor under the condition of maintaining the thermal gradient disturbance and the excitation source electrical signal; The difference between the total power loss of thermal stimulation and the theoretical superposition loss corresponding to the strong interaction harmonic pair is calculated to obtain the additional loss under thermal disturbance; The ratio of the additional loss to the mutual reference loss under thermal disturbance is calculated to obtain the thermal field gain factor.

5. The full-operating-condition equivalent testing method for capacitors for DC transmission according to claim 4, characterized in that: Applying thermal gradient perturbations in the capacitor case region also includes: A first temperature control sequence is set for the upper area of ​​the capacitor housing, with the target temperature fluctuating sinusoidally between 60°C and 80°C at a frequency of 0.1 Hz. A second temperature control sequence is set for the lower area of ​​the capacitor housing, with the target temperature fluctuating between 60°C and 40°C at a frequency of 0.1 Hz in an anti-phase sinusoidal pattern; The phase difference between the first temperature control timing sequence and the second temperature control timing sequence is fixed at 180 degrees; The temperature difference value and the corresponding total thermal power loss are synchronously recorded according to each control sampling point in the first temperature control sequence and the second temperature control sequence to form a dynamic loss sequence.

6. The full-operating-condition equivalent testing method for capacitors for DC transmission according to claim 4, characterized in that: In step S3, the calibration of the machine vibration gain factor includes: The capacitor is restored to the reference temperature, and a mechanical resonance excitation of 10 Hz to 2000 Hz is applied through the piezoelectric actuator array to dynamically identify the mechanical resonance frequency. Apply a constant amplitude mechanical vibration to the capacitor shell with the mechanical resonance frequency as the center frequency, inject the excitation source electrical signal synchronously during the vibration period, and read the additional loss under mechanical vibration; The ratio of the additional loss under computer vibration to the mutual reference loss is used to obtain the machine vibration gain factor.

7. The full-operating-condition equivalent testing method for capacitors for DC transmission according to claim 6, characterized in that: Dynamic identification of mechanical resonance frequencies includes: To control the piezoelectric actuator in the X-axis direction, a first sweep frequency excitation sequence is independently generated, and the excitation frequency is set to increase logarithmically from 10 Hz to 2000 Hz; To control the piezoelectric actuator in the Y-axis direction, a second sweep frequency excitation sequence is independently generated. The excitation frequency is set to decrease in an antilogarithmic pattern from 2000 Hz to 10 Hz, and the starting point of the second sweep frequency excitation sequence is aligned in time with the center point of the first sweep frequency excitation sequence. The response amplitude of the capacitor housing is captured using an accelerometer; At each frequency step of the first sweep excitation sequence, if the acceleration response amplitude in the X-axis direction exceeds a preset threshold, the current excitation frequency and response amplitude are pushed into the X-axis resonance candidate stack as a data pair; At each frequency step of the second sweep excitation sequence, if the acceleration response amplitude in the Y-axis direction exceeds a preset threshold, the current excitation frequency and response amplitude are pushed into the Y-axis resonance candidate stack as a data pair; After both frequency sweep sequences are completed, all data pairs in the X-axis resonance candidate stack and the Y-axis resonance candidate stack are compared, and the frequency point with the maximum response amplitude is determined as the mechanical resonance frequency point.

8. The full-operating-condition equivalent testing method for capacitors for DC transmission according to claim 1, characterized in that: In step S4, the total loss of the full-condition test is evaluated based on the thermal field gain factor and the machine vibration gain factor, including: Extract and accumulate the loss values ​​of all corresponding harmonics from the single-frequency loss baseline to generate the basic loss component; Traverse all harmonic pairs from the harmonic coupling loss and accumulate their corresponding harmonic additional losses to generate the interaction loss component; Calculate the coupling amplification loss component based on the interaction loss component according to the thermal field gain factor and the machine vibration gain factor; The basic loss component, interaction loss component, and coupling amplification loss component are combined in numerical order to evaluate the total loss of the full-condition test.

9. The full-operating-condition equivalent testing method for capacitors for DC transmission according to claim 1, characterized in that: In step S4, if the loss deviation value is within the preset range, the test result is locked and packaged into full-condition test data including: Calculate the loss deviation value by using the preset capacitor standard loss value and the total loss of the full-condition test; When the loss deviation value is less than 5% of the total loss of the full-condition test, the test status is marked as verification passed at the end of the current test cycle; After the verification pass flag takes effect, the current test results are locked to construct a full-condition effect matrix, thereby encapsulating it into full-condition test data.

10. A full-operation equivalent test system for capacitors used in direct current transmission, characterized in that: For executing the full-operating-condition equivalent testing method for a DC power transmission capacitor according to claim 1, the full-operating-condition equivalent testing system for a DC power transmission capacitor comprises: The inherent loss acquisition module is used to apply a single-frequency AC component to the capacitor to obtain the inherent loss tangent value of the capacitor under single-frequency excitation and generate a single-frequency loss baseline; The harmonic coupling analysis module is used to select a harmonic combination based on a single-frequency loss baseline. The difference between the theoretical superposition loss in the single-frequency loss baseline and the actual total power loss of the capacitor when the harmonic combination is synchronously injected is calculated, and the positive value is determined as the harmonic coupling loss. The multi-condition test module is used to screen out strongly interacting harmonic pairs based on harmonic coupling loss and use them as excitation sources. Under the conditions of applying controlled thermal gradient perturbations and mechanical resonance excitations, the additional losses are remeasured to calibrate the thermal field gain factor and the mechanical vibration gain factor. The full-condition verification control module is used to evaluate the total loss of the full-condition test based on the thermal field gain factor and the machine vibration gain factor, and compare the loss deviation value with the preset standard loss value. If the loss deviation value is within the preset range, the test result is locked and packaged into full-condition test data. If the loss deviation value is not within the preset range, the test calibration request is executed and the subsequent test process is automatically suspended to cut off all excitation outputs.

Images