MMC DC cable online monitoring injection frequency adaptive adjustment method

Abstract

The application discloses a kind of self-adaptive adjustment methods of MMC DC cable on-line monitoring injection frequency, including steps S1.Working condition parameter real-time measurement;S2.Based on the initial detection frequency of theoretical optimal detection frequency assignment;Calculate the theoretical optimal detection frequency F opt Required to maintain target dielectric loss angle under current working condition;The theoretical optimal detection frequency F opt Is directly assigned to initial detection frequency f0;S3.Dielectric loss angle is measured and extracted;S4.Based on deviation comparison iterative search calculation determines final detection frequency f, for maintaining the constancy of dielectric loss angle under current working condition;The application eliminates the cross interference of temperature and working condition fluctuation on insulation parameter measurement, ensures that dielectric loss angle is always dynamically locked in target working area, with the significant advantages of strong anti-interference ability, high precision of weak feature extraction, effectively avoiding state misjudgment and missed judgment.

Description

Technical Field

[0001] This invention relates to the field of high-voltage DC cable insulation monitoring, and specifically to an adaptive adjustment method for the injection frequency of online monitoring of MMC DC cables. Background Technology

[0002] High-voltage direct current (HVDC) cables are crucial physical carriers for modular multilevel converter (MMC) flexible DC transmission systems. This application refers to them as MMC DC cables (or simply "cables"), and their insulation reliability directly impacts power grid safety. During long-term operation, the insulating dielectric materials of MMC DC cables (such as cross-linked polyethylene, XLPE) are prone to aging and deterioration under the combined effects of thermal, electrical, and mechanical stresses, leading to partial discharge and even dielectric breakdown, posing a serious threat to the safe and stable operation of the power system. Therefore, achieving accurate online monitoring of the insulation status of MMC DC cables is of great significance. The dielectric loss angle, as a core parameter reflecting the polarization loss and conductivity loss of MMC DC cable insulation, is widely used for the quantitative assessment of insulation aging.

[0003] To achieve simple and sensitive online monitoring of the insulation status of MMC DC cables, existing solutions propose to utilize the flexible control capability of the MMC converter to inject a low-frequency AC detection signal on the DC side, and estimate the dielectric loss angle of the cable insulation by analyzing the leakage current of the cable shield. For example, Chinese Patent No. CN120490704B discloses a method and system for online monitoring of the status of high-voltage DC cables based on low-frequency injection of MMC. It adopts a dual closed-loop control mode consisting of an external voltage loop and an internal current loop, so that the MMC outputs a target voltage superimposed with a low-frequency AC signal to the DC cable.

[0004] However, the above technologies still have certain limitations: studies have shown that the volume resistivity of cable insulation materials is affected by operating temperature and exhibits obvious nonlinear variation characteristics, causing the equivalent insulation resistance of the cable to fluctuate widely with operating conditions (for example, under typical conditions, the cable insulation resistance decreases by an order of magnitude for every 10°C increase in temperature); since the existing low-frequency injection monitoring methods use a fixed single detection frequency for signal injection and feature extraction, they are difficult to adapt to such dynamic changes in operating conditions.

[0005] Specifically, when cables are under high load and high temperature conditions, the natural decay of their insulation resistance will lead to a significantly larger measured dielectric loss angle. If the condition assessment is based solely on the dielectric loss angle at a single frequency and its fixed threshold, normal fluctuations in parameters caused by changes in operating conditions may be misjudged as severe aging or moisture defects in the insulation material itself, thus triggering false alarms or even unnecessary power outages for maintenance. Conversely, under low temperature and light load conditions, fixed detection frequencies often fall into areas with low detection sensitivity, which can easily mask early insulation degradation characteristics, leading to misjudgments. Summary of the Invention

[0006] In view of this, the purpose of this invention is to provide an adaptive adjustment method for the injection frequency of online monitoring of MMC DC cables, which eliminates the cross-interference of temperature and operating condition fluctuations on insulation parameter measurement, ensures that the dielectric loss angle is always dynamically locked in the target working area, and has significant advantages such as strong anti-interference ability, high accuracy of weak feature extraction, and effective avoidance of false and false state judgments.

[0007] The technical solution adopted in this invention is as follows:

[0008] An adaptive adjustment method for online monitoring of injection frequency in MMC DC cables includes the following steps:

[0009] S1. Real-time measurement of operating parameters: Real-time acquisition of the operating temperature T and operating field strength E of the MMC DC cable;

[0010] S2. Assigning the initial detection frequency based on the theoretically optimal detection frequency: Based on real-time operating parameters and combined with pre-set parameters of the MMC DC cable insulation equivalent model, the target dielectric loss angle δ is calculated under the current operating conditions. goal The required theoretical optimal detection frequency F opt ;Assign it to the initial detection frequency f0;

[0011] S3. Measurement and extraction of dielectric loss angle: A low-frequency AC detection signal is injected onto the DC-side output voltage of the MMC with the initial detection frequency f0, and the dielectric loss angle δ is measured.

[0012] S4. Determine the final detection frequency f based on deviation comparison iterative search calculation: compare the dielectric loss angle δ with the target dielectric loss angle δ goal Deviation comparison is performed, and the final detection frequency f is determined through iterative search.

[0013] Preferably, in step S1, the operating temperature T is obtained by fiber optic temperature measurement or thermocouple sensor; the operating field strength E is calculated based on the MMC DC side output voltage and the insulation layer thickness of the MMC DC cable.

[0014] Preferably, in step S2, the insulation equivalent model parameters of the MMC DC cable include: establishing a parallel equivalent circuit model of the MMC DC cable, and the cable insulation resistance R. V It is connected in parallel with the cable insulation capacitance C, where,

[0015] Define the physical quantity τ as the intrinsic time constant of the cable insulation material:

[0016] In the formula, R is the cable insulation resistance R. V ;

[0017] Let the injection voltage of the low-frequency AC detection signal be U, and its angular frequency be ω. Then the resistive component of the leakage current is I. R =U / R, the capacitive component of the leakage current I C =UωC;

[0018] The dielectric loss factor is defined using the following formula:

[0019] In the formula, R is the cable insulation resistance R. V tanδ is the dielectric loss factor, which is the tangent of the dielectric loss angle δ.

[0020] Dielectric loss factor and cable insulation resistance R V The relationship between the cable insulation capacitance C and the angular frequency ω of the injected low-frequency AC detection signal is as follows:

[0021] ;

[0022] The loss angle δ of the target medium is obtained by calculating using the above relationship. goal The required low-frequency AC detection signal angular frequency ω is then used to calculate the theoretical optimal detection frequency F using the following formula. opt :

[0023] .

[0024] Preferably, the formula for calculating the volume resistivity of the cable is as follows:

[0025] ;

[0026] In the formula, The volume resistivity of the cable insulation material is the same as the cable volume resistivity. α is the volume resistivity of the cable insulation material at 0℃; α is the temperature coefficient of volume resistivity; T is the operating temperature; β is the stress coefficient of volume resistivity; E is the electric field strength of the cable insulation layer, i.e., the operating field strength.

[0027] The cable insulation resistance RV The calculation formula is as follows:

[0028] ;

[0029] In the formula, r i r is the outer radius of the cable insulation layer. c The radius of the cable conductor;

[0030] The formula for calculating the cable insulation capacitance C is as follows:

[0031] ;

[0032] In the formula, ε0 is the minimum dielectric constant of the cable under vacuum conditions; ε r is the relative permittivity of the cable.

[0033] Preferably, the target dielectric loss angle δ is set through the following process. goal :

[0034] A sensitivity coefficient S is introduced into the parameters of the equivalent model of MMC DC cable insulation; wherein, the sensitivity coefficient S is defined as the degree of response of the dielectric loss angle change rate dδ to the cable insulation characteristic change rate dτ, and the calculation formula used is as follows:

[0035] ;

[0036] The lower the angular frequency ω of the low-frequency AC detection signal, the higher the sensitivity coefficient S; the target medium loss angle δ goal Substitute the dielectric loss angle δ into the above calculation formula, and set the target dielectric loss angle δ with the goal of obtaining a higher sensitivity coefficient S. goal .

[0037] Preferably, the sensitivity coefficient S adopts the following equivalent formula:

[0038] ;

[0039] When the dielectric loss angle δ reaches 45°, sin(2δ) reaches its peak value, resulting in the highest sensitivity coefficient S. This is combined with measures to avoid exceeding the extreme value and entering the nonlinear saturation region due to changes in the intrinsic time constant of the cable insulation material. The target dielectric loss angle δ... goal The range is set between 10° and 40°.

[0040] Preferably, in step S3, after the low-frequency AC detection signal is superimposed and injected, it is transmitted through a cable and a leakage current is generated in the cable shielding layer; the voltage and leakage current of the injected low-frequency AC detection signal are collected simultaneously, and the amplitude and phase information of the low-frequency fundamental component are extracted by Fast Fourier Transform (FFT) analysis and processing, and the dielectric loss angle δ is calculated based on the phase difference.

[0041] Preferably, in step S4, the dielectric loss angle δ calculated in step S3 is compared with the target dielectric loss angle δ. goal Compare and calculate the deviation Δδ:

[0042] Δδ=δ-δ goal Then, based on the direction and magnitude of the deviation, the bisection frequency optimization mechanism is activated.

[0043] Preferably, the bisection frequency optimization mechanism includes the following optimization calculation process:

[0044] When the cable insulation shows signs of aging, the measured dielectric loss angle δ is greater than the target dielectric loss angle δ. goal If the current detection frequency is too low, it indicates that the detection frequency is too low. Based on the inverse relationship between the dielectric loss angle and the frequency, the detection frequency should be increased according to the following formula:

[0045] ;

[0046] In the formula, f n+1 f is the injection frequency of the detection signal in the (n+1)th cycle; n The frequency injected into the detection signal during the nth cycle is denoted by Δf, which is the frequency adjustment step size. The frequency adjustment step size Δf is taken as the theoretically optimal detection frequency F. opt The value corresponding to the least significant bit;

[0047] When the measured dielectric loss angle δ is less than the target dielectric loss angle δ goal If this indicates that the current detection frequency is too high, the following procedure should be followed to reduce the detection frequency:

[0048] First, determine the frequency search interval: use the current frequency as the upper bound f of the search interval. high Previously, when the measured dielectric loss angle δ was still greater than the target dielectric loss angle δ... goal The frequency corresponding to the time is used as the lower bound f of the search interval. low Construct the initial search interval [f] low ,f high ];

[0049] Then the midpoint frequency f of the search interval is calculated. mid That is: f mid = (f low + fhigh ) / 2, and inject a low-frequency AC detection signal at this midpoint frequency;

[0050] By collecting leakage current and performing Fourier analysis, the dielectric loss angle at this midpoint frequency was measured.

[0051] If the measured dielectric loss angle δ corresponding to the midpoint frequency is still less than the target dielectric loss angle δ goal This indicates that the optimal frequency is too low, and the upper bound of the search interval is updated to the current midpoint frequency, which is: f high = f mid If the measured dielectric loss angle δ corresponding to the midpoint frequency is greater than the target dielectric loss angle δ... goal This indicates that the optimal frequency is too high. Therefore, the lower bound of the search interval is updated to the current midpoint frequency, which is f. low = f mid .

[0052] Preferably, the bisection frequency optimization mechanism further includes the following iterative loop process:

[0053] The process iterates through the following steps: midpoint frequency selection, injection, measurement, comparison, and boundary update. The frequency search interval is gradually narrowed using a bisection method. When the measured dielectric loss angle δ is greater than the target dielectric loss angle δ... goal When the relative deviation converges to within the preset threshold, it is determined that the optimization process has met the accuracy requirements. This means that the output detection frequency is the final detection frequency that maintains a constant dielectric loss angle under the current operating conditions. This detection frequency will be used as the benchmark parameter for online monitoring and evaluation of the insulation status of MMC DC cables, and the iteration loop will automatically terminate.

[0054] The working mechanism of this invention mainly lies in proposing an MMC-driven adaptive adjustment method for cable insulation detection frequency. This method incorporates the real-time operating temperature T and electric field strength (i.e., operating field strength E) of the cable into the detection frequency decision model, and adopts a composite control strategy that couples feedforward prediction and iterative search approximation to dynamically adjust the injection frequency of the low-frequency detection signal (i.e., the final detection frequency f of the low-frequency AC detection signal); wherein, the system operates at the theoretically optimal frequency F. opt Using the initial benchmark, the deviation range between the measured dielectric loss angle and the target value is continuously narrowed through iterative search (preferably using a bisection frequency optimization mechanism), thereby achieving accurate adaptive frequency matching.

[0055] The present invention has the following main advantages:

[0056] I. Strong anti-interference capability for insulation condition assessment: This invention realizes adaptive adjustment of cable insulation detection frequency driven by MMC. In practical applications, a temperature-frequency adaptive modulation function curve library can be constructed based on this. The insulation condition can be diagnosed by comparing the positional relationship between the actual detection frequency and the health and aging reference curves. Compared with the method that relies on a single loss angle absolute value threshold, this mechanism completely eliminates the interference of temperature and electric field fluctuations on the measurement, effectively avoiding false alarms and missed early degradation under extreme operating conditions.

[0057] II. High Precision of Adaptive Frequency Adjustment: This invention adopts a composite control strategy that combines feedforward calculation with iterative (preferably bisection method iteration) approximation. The theoretical optimal frequency calculated from real-time temperature data is used as the feedforward quantity. The frequency search range is dynamically adjusted by iteratively comparing the deviation between the measured dielectric loss angle and the target value. This strategy has the advantages of strong convergence stability and high optimization accuracy, effectively overcoming the adjustment lag caused by sudden changes in operating conditions, ensuring that the detection frequency converges quickly to the optimal value, and keeping the dielectric loss angle dynamically locked in the target working area.

[0058] This invention achieves the target dielectric loss angle δ goal By setting the angle in the high-gain linear rising region (e.g., preferably between 10° and 40°), the dielectric loss angle is always dynamically locked in the high-gain linear rising region. Compared with traditional fixed frequency or extreme point pursuit methods, this setting avoids the amplification of the measurement error of the dielectric loss angle at small angles and avoids the risk of entering nonlinear saturation near the extreme point due to the change of the intrinsic time constant of the cable insulation material. Attached Figure Description

[0059] Figure 1 This is a flowchart illustrating the steps of the adaptive adjustment method for online monitoring of injection frequency of MMC DC cable according to a specific embodiment of this application.

[0060] Figure 2 This is a schematic diagram of the simulation model structure of the ±30kV / 18MW MMC flexible DC transmission system built in a specific embodiment of this application;

[0061] Figure 3 This is a graph showing the relationship between the measured dielectric loss angle and the injected signal frequency in a specific embodiment of this application;

[0062] Figure 4 The temperature-injection signal frequency adaptive modulation function curve is established by the adaptive adjustment method based on injection frequency in the specific implementation of this application.

[0063] Figure 5 This is a graph showing the relationship between temperature and injected signal frequency when the cable experiences different degrees of insulation degradation in a specific embodiment of this application. Detailed Implementation

[0064] Please refer to the above. Figure 1 As shown, this embodiment provides an adaptive adjustment method for online monitoring of the injection frequency of an MMC DC cable, including the following steps.

[0065] S1. Real-time measurement of operating parameters: Real-time acquisition of the operating temperature T and operating field strength E of the MMC DC cable; Preferably, in this step S1, the operating temperature T is obtained by fiber optic temperature measurement or thermocouple sensor; the operating field strength E is calculated based on the output voltage of the MMC DC side and the insulation layer thickness of the MMC DC cable. This calculation method is common knowledge in the field and will not be specifically described in this application embodiment.

[0066] S2. Assigning the initial detection frequency based on the theoretically optimal detection frequency: Based on the real-time operating parameters obtained in step S1, and combined with the pre-set parameters of the MMC DC cable insulation equivalent model, calculate the theoretically optimal detection frequency F required to maintain the target dielectric loss angle under the current operating conditions. opt The theoretically optimal detection frequency F opt The value is directly assigned to the initial detection frequency f0, i.e., f0 = F. opt ;

[0067] Preferably, in step S2, the equivalent model parameters for the insulation of the MMC DC cable include:

[0068] Establish a parallel equivalent circuit model for the MMC DC cable, and determine the cable insulation resistance R. V It is connected in parallel with the cable insulation capacitance C, where the physical quantity τ is defined as the intrinsic time constant of the cable insulation material:

[0069] In the formula, R is the cable insulation resistance R. V ;

[0070] Let the injection voltage of the low-frequency AC detection signal be U, and its angular frequency be ω. Then the resistive component of the leakage current is I. R =U / R, the capacitive component of the leakage current I C =UωC;

[0071] The dielectric loss factor is defined using the following formula:

[0072] In the formula, R is the cable insulation resistance R. V tanδ is the dielectric loss factor, which is the tangent of the dielectric loss angle δ.

[0073] Dielectric loss factor and cable insulation resistance R V The relationship between the cable insulation capacitance C and the angular frequency ω of the injected low-frequency AC detection signal is as follows:

[0074] ;

[0075] The loss angle δ of the target medium is obtained by calculating using the above relationship. goal The required low-frequency AC detection signal angular frequency ω is then used to calculate the theoretical optimal detection frequency F using the following formula. opt :

[0076] .

[0077] Preferably, in this embodiment, the formula for calculating the volume resistivity of the cable is as follows:

[0078] In the formula, The volume resistivity of the cable insulation material is the same as the cable volume resistivity. α is the volume resistivity of the cable insulation material at 0℃; T is the operating temperature; β is the stress coefficient of the volume resistivity; and E is the electric field strength of the cable insulation layer, i.e., the operating field strength.

[0079] Preferably, in this embodiment, the cable insulation resistance R V The calculation formula is as follows:

[0080] In the formula, r i r is the outer radius of the cable insulation layer. c The radius of the cable conductor;

[0081] Preferably, in this embodiment, the formula for calculating the cable insulation capacitance C is as follows:

[0082] In the formula, ε0 is the minimum dielectric constant of the cable under vacuum conditions; ε r is the relative permittivity of the cable.

[0083] In this embodiment, for example, under 50Hz power frequency excitation, the dielectric loss factor (tanδ) of cross-linked polyethylene (XLPE) cable insulation is usually extremely small. By performing vector decomposition on the leakage current, under such a small dielectric loss factor, the resistive leakage current component reflecting insulation degradation is only in the nanoampere to microampere range, while the capacitive component characterizing the background characteristics is as high as the milliampere range. The difference in amplitude between the two is as high as 4 to 5 orders of magnitude. This extreme imbalance causes the weak resistive aging characteristics to be completely "submerged" by the massive capacitive current. Traditional power frequency or higher frequency detection methods are simply unable to accurately extract and quantify this characteristic. To address the difficulty in identifying and quantifying the resistive component, this embodiment analyzes the parameters of the MMC DC cable insulation equivalent model (equivalent to the equivalent circuit of a power cable). By utilizing the inverse relationship between AC voltage frequency and the cable's equivalent capacitive reactance, reducing the injected voltage frequency can effectively increase the cable's equivalent capacitive reactance, thereby reducing the capacitive component in the leakage current. This adjustment allows the resistive and capacitive components of the leakage current to reach a relatively balanced state, which helps improve the measurement accuracy of the resistive component.

[0084] Preferably, in this embodiment, the target dielectric loss angle δ is set through the following process. goal :

[0085] A sensitivity coefficient S is introduced into the parameters of the equivalent model of MMC DC cable insulation. The sensitivity coefficient S is defined as the degree of response of the dielectric loss angle change rate dδ to the cable insulation characteristic change rate dτ, and the calculation formula is as follows:

[0086] ;

[0087] The lower the angular frequency ω of the low-frequency AC detection signal, the higher the sensitivity coefficient S; the target medium loss angle δ goal Substitute the dielectric loss angle δ into the above calculation formula, and set the target dielectric loss angle δ with the goal of obtaining a higher sensitivity coefficient S. goal .

[0088] More preferably, in this embodiment, the sensitivity coefficient S adopts the following equivalent formula:

[0089] Among them, when the dielectric loss angle δ reaches 45°, sin(2δ) reaches its peak value, making the sensitivity coefficient S reach its highest value, representing the highest detection sensitivity of the dielectric loss angle. However, this operating point is easily affected by the intrinsic parameters of the insulation material (including the cable insulation resistance R / R). VThe product of the cable insulation capacitance C) is the change in the intrinsic time constant τ of the cable insulation material. This change can cause the cable insulation material to exceed its extreme value and enter the nonlinear saturation region, leading to a sharp decrease in sensitivity and causing quantization errors. Since the optimal injection frequency selection strategy should ensure that the cable dielectric loss angle is precisely locked at an optimal fixed angle, preferably, in this embodiment, it is combined with avoiding the change in the intrinsic time constant of the cable insulation material from exceeding its extreme value and entering the nonlinear saturation region; the target dielectric loss angle δ goal The range is set between 10° and 40°, which is within the high-gain linear increase range of the sensitivity coefficient S. This range is set mainly based on two considerations: First, the intrinsic parameters of the insulation materials of different cables are different, so a certain adaptation range needs to be reserved; second, the detection efficiency is taken into account. If the target value of the cable dielectric loss angle is set too high, although the sensitivity is the highest, the corresponding detection frequency is too low, which will significantly prolong the single detection cycle and affect the timeliness of the cable insulation condition determination.

[0090] S3. Measurement and extraction of dielectric loss angle: The MMC controller injects a low-frequency AC detection signal onto the DC side output voltage of the MMC with an initial detection frequency f0, and the dielectric loss angle δ is obtained by actual measurement.

[0091] It should be noted that the initial detection frequency f0 in this step will serve as the starting operating frequency of the MMC low-frequency AC injection signal, and also as the central reference point for the subsequent iterative search interval. This can significantly reduce the number of iterations for subsequent closed-loop regulation and improve the system's response speed.

[0092] Preferably, in step S3, after the low-frequency AC detection signal is injected, it is transmitted through a cable and a leakage current is generated in the cable shielding layer; the voltage and leakage current of the injected low-frequency AC detection signal are collected simultaneously, and the amplitude and phase information of the low-frequency fundamental component are extracted by Fast Fourier Transform (FFT) analysis and processing, and the dielectric loss angle δ is calculated based on the phase difference.

[0093] It should be further explained that the specific process of extracting the amplitude and phase information of the low-frequency fundamental component based on the voltage and leakage current of the injected low-frequency AC detection signal, and calculating the dielectric loss angle δ based on the phase difference (including the Fast Fourier Transform (FFT) analysis and processing process) is common knowledge and conventional technical means for those skilled in the art. In order to save space, the embodiments of this application will not be described in detail.

[0094] S4. Determine the final detection frequency f based on deviation comparison iterative search calculation: compare the dielectric loss angle δ with the target dielectric loss angle δ goal Deviation comparison is performed, and the final detection frequency f is determined by using the correction of the dielectric loss angle deviation as the target and through iterative search. This frequency is used to maintain the dielectric loss angle constant under the current operating conditions.

[0095] Preferably, in step S4, the dielectric loss angle δ calculated in step S3 is compared with the target dielectric loss angle δ. goal Compare and calculate the deviation Δδ:

[0096] Δδ=δ-δ goal Then, based on the direction and magnitude of the deviation, the bisection frequency optimization mechanism is activated.

[0097] Preferably, in this embodiment, the bisection frequency optimization mechanism includes the following optimization calculation process:

[0098] When the cable insulation shows signs of aging, the measured dielectric loss angle δ is significantly greater than the target dielectric loss angle δ. goal If the current detection frequency is too low, it indicates that the detection frequency is too low. Based on the inverse relationship between the dielectric loss angle and the frequency, the detection frequency should be increased according to the following formula:

[0099] In the formula, f n+1 f is the injection frequency of the detection signal in the (n+1)th cycle; n The frequency injected into the detection signal during the nth cycle is denoted by Δf, which is the frequency adjustment step size. The frequency adjustment step size Δf is taken as the theoretically optimal detection frequency F. opt The value corresponding to the least significant bit; for example, when the theoretically optimal detection frequency is calculated to be 0.005Hz, the frequency adjustment step size Δf is taken as 0.001Hz (i.e., 10). -3 (on the order of Hz). This step size selection strategy ensures that the accuracy of frequency adjustment matches the theoretical calculation value, while avoiding convergence oscillations caused by excessively large step sizes or low computational efficiency caused by excessively small step sizes. Increasing the frequency will increase the equivalent capacitive reactance and reduce the capacitive current component, thereby causing the dielectric loss angle to converge toward the target value.

[0100] When the measured dielectric loss angle δ is less than the target dielectric loss angle δ goal If this indicates that the current detection frequency is too high, the following procedure should be followed to reduce the detection frequency:

[0101] First, determine the frequency search interval: use the current frequency as the upper bound f of the search interval. high Previously, when the measured dielectric loss angle δ was still greater than the target dielectric loss angle δ... goal The frequency corresponding to the time is used as the lower bound f of the search interval. low Construct the initial search interval [f] low ,f high ];

[0102] Then the midpoint frequency f of the search interval is calculated. mid That is, f mid = (flow + f high ) / 2, and inject a low-frequency AC detection signal at this midpoint frequency;

[0103] By collecting leakage current and performing Fourier transform (i.e., Fast Fourier Transform, FFT) analysis, the dielectric loss angle at the midpoint frequency was measured.

[0104] If the measured dielectric loss angle δ corresponding to the midpoint frequency is still less than the target dielectric loss angle δ goal This indicates that the optimal frequency is too low, and the upper bound of the search interval is updated to the current midpoint frequency, which is f. high = f mid If the measured dielectric loss angle δ corresponding to the midpoint frequency is greater than the target dielectric loss angle δ... goal This indicates that the optimal frequency is too high. Therefore, the lower bound of the search interval is updated to the current midpoint frequency, which is f. low = f mid .

[0105] More preferably, in this embodiment, the bisection frequency optimization mechanism further includes the following iterative loop process:

[0106] The process iterates through the following steps: midpoint frequency selection, injection, measurement, comparison, and boundary update. The frequency search interval is gradually narrowed using a bisection method. When the measured dielectric loss angle δ is greater than the target dielectric loss angle δ... goal When the relative deviation converges to within the set preset threshold (e.g., set at 0.5%, which can be set according to the actual accuracy requirements, and this embodiment does not impose any special restrictions on it), it is determined that the optimization process has met the accuracy requirements. This means that the output detection frequency at this time is the final detection frequency that maintains the dielectric loss angle constant under the current operating conditions (i.e., maintains the target dielectric loss angle). This detection frequency will be used as the benchmark parameter for online monitoring and evaluation of the insulation status of MMC DC cables, and the iteration loop will be automatically terminated.

[0107] To enable those skilled in the art to better understand the technical solutions of this invention, based on the above implementation schemes, the following specific embodiments will be proposed in conjunction with the accompanying drawings of the embodiments of this invention. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0108] In the following specific embodiment, a simulation analysis is carried out using a ±30 kV / 18 MW MMC flexible DC transmission system as an example. The simulation model structure diagram is shown below. Figure 2 As shown, it should be noted that... Figure 2The simulation model used is a well-known simulation model structure in the power field. The MMC sending-end controller, MMC sending-end converter valve, filter, flexible DC transmission cable, MMC receiving-end converter valve, and MMC receiving-end controller are all well-known device structures. To save space, this application's implementation section will not elaborate on these components, as they are all common knowledge to those skilled in the art. During system operation, while maintaining the rated DC voltage output on the MMC DC side, a low-frequency AC detection signal is superimposed. The MMC controller (regardless of whether it is at the receiving or sending end) first calculates the theoretically optimal detection frequency F by combining the cable's real-time operating temperature (i.e., operating temperature T) and electric field strength (i.e., operating field strength E). opt The initial detection frequency f0 is directly assigned to the value; then, based on the measured calculation result of the dielectric loss angle δ, the final detection frequency f of the low-frequency AC detection signal is dynamically adjusted in a closed loop until the cable dielectric loss angle accurately reaches the target value. In this specific embodiment, the target dielectric loss angle δ... goal Set at 20° to complete the adaptive adjustment process of the detection frequency.

[0109] The following specific embodiments use a severely aging condition with an operating temperature T=35℃, an operating field strength E=3.75kV / mm, and a 40% decrease in cable insulation resistance as an example to illustrate in detail the frequency adaptive adjustment calculation process based on the bisection method (i.e., the bisection method frequency optimization mechanism). For the measured relationship between dielectric loss angle and frequency, please refer to [link to relevant documentation]. Figure 3 As shown in Table 1 below, the frequency adaptive adjustment process of the bisection method is illustrated.

[0110] Table 1

[0111]

[0112] Note: The measured dielectric loss angles in Table 1 above are recorded and displayed with three decimal places and rounded to the nearest whole number.

[0113] During the coarse frequency adjustment phase, the system gradually increases the detection frequency in fixed steps of 0.001 Hz, as shown in Table 1. When the injection frequency reaches 0.007 Hz, the measured dielectric loss angle is 20.307°, still slightly larger than the target value (set to 20°). When the frequency continues to increase to 0.008 Hz, the measured dielectric loss angle drops to 19.179°, less than the target value (target deviation is -0.821°). This jump indicates that the optimal frequency for maintaining the target dielectric loss angle must be between 0.007 Hz and 0.008 Hz. At this point, the system automatically locks this frequency range as the initial search boundary (i.e., the lower bound f) for the bisection method. low =0.007Hz, upper bound f high =0.008Hz), and initiate the bisection iterative optimization mechanism, the specific process of which is as follows:

[0114] First, calculate the midpoint frequency f of the search interval. mid = (0.007+0.008) / 2 = 0.0075Hz (corresponding to the 6th cycle in Table 1). After injecting this frequency, the measured dielectric loss angle is 19.345°, which is still less than the target value, indicating that the optimal frequency is too low. The system updates the upper limit of the search interval to 0.0075Hz.

[0115] Then, take the midpoint f of the new interval again. mid =(0.007+0.0075) / 2=0.00725Hz (which corresponds to the 7th cycle in Table 1). The measured dielectric loss angle is 19.773°, which is still less than the target value. The system continues to update the upper limit to 0.00725Hz.

[0116] Finally, take the midpoint f. mid =(0.007+0.00725) / 2=0.007125Hz (corresponding to the 8th cycle in Table 1). At this time, the measured dielectric loss angle is 20.034°, which has converged to the set accuracy threshold range. The system determines that the iteration is over and outputs 0.007125Hz as the final detection frequency, which is the final detection frequency f=0.007125Hz of the low-frequency AC detection signal. This final detection frequency has exceeded the aging threshold curve. This process fully demonstrates the efficiency and reliability of the bisection strategy adopted in this embodiment in eliminating operating condition interference and achieving accurate quantitative assessment of aging degree.

[0117] To further illustrate the technical effects of the embodiments of this application, based on the embodiments described above, the implementation method of this application further establishes a temperature-injection signal frequency adaptive modulation function curve, where the temperature is the operating temperature T, and the injection signal frequency is the corresponding determined final detection frequency f. The function curve can be further referred to... Figure 4 As shown, Figure 4 The curve of the healthy group (i.e.) Figure 4 The marked "healthy insulation resistance test frequency" curve represents the test frequency required to maintain a constant dielectric loss angle for a cable with good insulation at different temperatures. The aging group curve (i.e.,...) Figure 4 The marked "Aging Threshold Insulation Resistance Detection Frequency" curve represents the frequency requirement of a cable that has undergone insulation aging under the same conditions.

[0118] In practical applications, by real-time monitoring of the final detection frequency f required to maintain the target dielectric loss angle and comparing it with a pre-established curve library, the insulation status can be accurately judged: when the actual detection frequency f is between the healthy curve and the aging curve, it indicates that the cable insulation status is normal; when the detection frequency f exceeds the aging curve, it indicates that even if the detection frequency is increased, it is still difficult to maintain the target dielectric loss angle, indicating that the cable insulation has aged significantly and needs to be repaired immediately.

[0119] exist Figure 4 Based on the content shown, this embodiment further simulates the insulation status monitoring process of cables under different temperatures (25℃, 35℃, 45℃) and different aging degrees:

[0120] While keeping the equivalent circuit insulation capacitance unchanged, the cable insulation resistance is reduced by 10%, 20%, and 40% respectively to characterize the gradual deterioration of insulation performance. During the test, the system adaptively adjusts the detection frequency to keep the cable dielectric loss angle stable at 20°.

[0121] For the test results of the detection frequency at which the system finally converges under different aging levels, please refer to [link / reference]. Figure 5 As shown; through this Figure 5 The test results clearly reveal the frequency shift patterns under different aging degrees and the engineering application advantages of this invention:

[0122] Specifically Figure 5 The discrete data points show that as the reduction rate of cable insulation resistance increases (from 10%, 20% to 40%), the detection frequency required to maintain the target dielectric loss angle shows an upward trend. This pattern enables the embodiments of this application to effectively overcome the defects of misjudgment and missed judgment in traditional fixed-frequency monitoring methods: under high-temperature conditions, traditional methods have difficulty distinguishing between the natural decay of insulation resistance with temperature and the actual aging of insulation materials, which easily leads to misjudgment.

[0123] The embodiments of this application employ a frequency adaptive adjustment mechanism to precisely maintain the dielectric loss angle within the high-gain linear operating range of 20°, fundamentally eliminating the interference of environmental fluctuations on insulation parameter measurements and ensuring the accuracy and reliability of cable insulation condition assessment. Simultaneously, this mechanism endows the system with excellent early warning capabilities; when the real-time detection frequency gradually approaches the aging threshold curve from the health baseline curve, the system can issue an early warning of insulation degradation, gaining valuable time for preventative maintenance of the equipment.

[0124] It should be noted that the changes in cable insulation resistance in the specific embodiments provided in this application are simulated by changing the value of the insulation resistance R in the equivalent model of the DC cable (the insulation resistance R is set in the DC cable). These are all conventional technical means for those skilled in the art, and this application will not elaborate on them in detail.

[0125] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0126] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. An adaptive adjustment method for online monitoring of injection frequency in MMC DC cables, characterized in that, Includes the following steps: S1. Real-time measurement of operating parameters: Real-time acquisition of the operating temperature T and operating field strength E of the MMC DC cable; S2. Assigning the initial detection frequency based on the theoretically optimal detection frequency: Based on real-time operating parameters and combined with pre-set parameters of the MMC DC cable insulation equivalent model, the target dielectric loss angle δ is calculated under the current operating conditions. goal The required theoretical optimal detection frequency F opt ;Assign it to the initial detection frequency f0; S3. Measurement and extraction of dielectric loss angle: A low-frequency AC detection signal is injected onto the DC-side output voltage of the MMC with the initial detection frequency f0, and the dielectric loss angle δ is measured. S4. Determine the final detection frequency f based on deviation comparison iterative search calculation: compare the dielectric loss angle δ with the target dielectric loss angle δ goal Deviation comparison is performed, and the final detection frequency f is determined through iterative search.

2. The adaptive adjustment method for online monitoring of injection frequency of MMC DC cable according to claim 1, characterized in that, In step S1, the operating temperature T is obtained by fiber optic temperature measurement or thermocouple sensor; the operating field strength E is calculated based on the MMC DC side output voltage and the insulation layer thickness of the MMC DC cable.

3. The adaptive adjustment method for online monitoring of injection frequency of MMC DC cable according to claim 1, characterized in that, In step S2, the insulation equivalent model parameters of the MMC DC cable include: establishing a parallel equivalent circuit model of the MMC DC cable, and the cable insulation resistance R. V It is connected in parallel with the cable insulation capacitance C, where, Define the physical quantity τ as the intrinsic time constant of the cable insulation material: In the formula, R is the cable insulation resistance R. V ; Let the injection voltage of the low-frequency AC detection signal be U, and its angular frequency be ω. Then the resistive component of the leakage current is I. R =U / R, the capacitive component of the leakage current I C =UωC; The dielectric loss factor is defined using the following formula: In the formula, R is the cable insulation resistance R. V tanδ is the dielectric loss factor, which is the tangent of the dielectric loss angle δ. Dielectric loss factor and cable insulation resistance R V The relationship between the cable insulation capacitance C and the angular frequency ω of the injected low-frequency AC detection signal is as follows: ； The loss angle δ of the target medium is obtained by calculating using the above relationship. goal The required low-frequency AC detection signal angular frequency ω is then used to calculate the theoretical optimal detection frequency F using the following formula. opt : 。 4. The adaptive adjustment method for online monitoring of injection frequency of MMC DC cable according to claim 3, characterized in that, The formula for calculating the volume resistivity of a cable is as follows: ； In the formula, The volume resistivity of the cable insulation material is the same as the cable volume resistivity. α is the volume resistivity of the cable insulation material at 0℃; α is the temperature coefficient of volume resistivity; T is the operating temperature; β is the stress coefficient of volume resistivity; E is the electric field strength of the cable insulation layer, i.e., the operating field strength. The cable insulation resistance R V The calculation formula is as follows: ； In the formula, r i r is the outer radius of the cable insulation layer. c The radius of the cable conductor; The formula for calculating the cable insulation capacitance C is as follows: ； In the formula, ε0 is the minimum dielectric constant of the cable under vacuum conditions; ε r is the relative permittivity of the cable.

5. The adaptive adjustment method for online monitoring of injection frequency of MMC DC cable according to claim 3, characterized in that, The target dielectric loss angle δ is set through the following process. goal : A sensitivity coefficient S is introduced into the parameters of the equivalent model of MMC DC cable insulation; wherein, the sensitivity coefficient S is defined as the degree of response of the dielectric loss angle change rate dδ to the cable insulation characteristic change rate dτ, and the calculation formula used is as follows: ； The lower the angular frequency ω of the low-frequency AC detection signal, the higher the sensitivity coefficient S; the target medium loss angle δ goal Substitute the dielectric loss angle δ into the above calculation formula, and set the target dielectric loss angle δ with the goal of obtaining a higher sensitivity coefficient S. goal .

6. The adaptive adjustment method for online monitoring of injection frequency of MMC DC cable according to claim 5, characterized in that, The sensitivity coefficient S is expressed by the following equivalent formula: ； When the dielectric loss angle δ reaches 45°, sin(2δ) reaches its peak value, resulting in the highest sensitivity coefficient S. This is combined with measures to avoid exceeding the extreme value and entering the nonlinear saturation region due to changes in the intrinsic time constant of the cable insulation material. The target dielectric loss angle δ... goal The range is set between 10° and 40°.

7. The adaptive adjustment method for online monitoring of injection frequency of MMC DC cable according to claim 1, characterized in that, In step S3, after the low-frequency AC detection signal is superimposed and injected, it is transmitted through a cable and a leakage current is generated in the cable shielding layer. The voltage and leakage current of the injected low-frequency AC detection signal are collected simultaneously, and the amplitude and phase information of the low-frequency fundamental component are extracted by Fast Fourier Transform (FFT) analysis. The dielectric loss angle δ is calculated based on the phase difference.

8. The adaptive adjustment method for online monitoring of injection frequency of MMC DC cable according to claim 1, characterized in that, In step S4, the dielectric loss angle δ calculated in step S3 is compared with the target dielectric loss angle δ. goal Compare and calculate the deviation Δδ: Δδ=δ-δ goal Then, based on the direction and magnitude of the deviation, the bisection frequency optimization mechanism is activated.

9. The adaptive adjustment method for online monitoring of injection frequency of MMC DC cable according to claim 8, characterized in that, The bisection frequency optimization mechanism includes the following optimization calculation process: When the cable insulation shows signs of aging, the measured dielectric loss angle δ is greater than the target dielectric loss angle δ. goal If the current detection frequency is too low, it indicates that the detection frequency is too low. Based on the inverse relationship between the dielectric loss angle and the frequency, the detection frequency should be increased according to the following formula: ； In the formula, f n+1 f is the injection frequency of the detection signal in the (n+1)th cycle; n The frequency injected into the detection signal during the nth cycle is denoted as Δf, which is the frequency adjustment step size. The frequency adjustment step size Δf is taken as the theoretically optimal detection frequency F. opt The value corresponding to the least significant bit; When the measured dielectric loss angle δ is less than the target dielectric loss angle δ goal If this indicates that the current detection frequency is too high, the following procedure should be followed to reduce the detection frequency: First, determine the frequency search interval: use the current frequency as the upper bound f of the search interval. high Previously, when the measured dielectric loss angle δ was still greater than the target dielectric loss angle δ... goal The frequency corresponding to the time is used as the lower bound f of the search interval. low Construct the initial search interval [f] low ,f high ]; Then the midpoint frequency f of the search interval is calculated. mid (f) mid = (f low + f high ) / 2), and inject a low-frequency AC detection signal at this midpoint frequency; By collecting leakage current and performing Fourier analysis, the dielectric loss angle at this midpoint frequency was measured. If the measured dielectric loss angle δ corresponding to the midpoint frequency is still less than the target dielectric loss angle δ goal This indicates that the optimal frequency is too low, and the upper bound of the search interval is updated to the current midpoint frequency (f). high = f mid If the measured dielectric loss angle δ corresponding to the midpoint frequency is greater than the target dielectric loss angle δ... goal This indicates that the optimal frequency is too high, so the lower bound of the search interval is updated to the current midpoint frequency (f). low = f mid ).

10. The adaptive adjustment method for online monitoring of injection frequency of MMC DC cable according to claim 9, characterized in that, The bisection frequency optimization mechanism also includes the following iterative loop process: The process iterates through the following steps: midpoint frequency selection, injection, measurement, comparison, and boundary update. The frequency search interval is gradually narrowed using a bisection method. When the measured dielectric loss angle δ is greater than the target dielectric loss angle δ... goal When the relative deviation converges to within the preset threshold, it is determined that the optimization process has met the accuracy requirements. This means that the output detection frequency is the final detection frequency that maintains a constant dielectric loss angle under the current operating conditions. This detection frequency will be used as the benchmark parameter for online monitoring and evaluation of the insulation status of MMC DC cables, and the iteration loop will automatically terminate.

Images