Capacitor health state evaluation method and equipment based on acoustic emission frequency domain characteristics

Abstract

The invention discloses a metallized film capacitor health state assessment method and system based on acoustic emission frequency domain characteristics, a medium and equipment, and the method comprises the steps: building a discharge ultrasonic DC monitoring experiment circuit and an aging test loop; the method comprises the following steps: acquiring metallized film capacitors with different health degrees through an accelerated aging experiment, disassembling the metallized film capacitors in different health states, connecting the metallized film capacitors to a discharge ultrasonic monitoring experiment circuit, and fixing an acoustic emission sensor; collecting discharge ultrasonic signals of the metallized film capacitor within one hour through an oscilloscope; performing frequency domain analysis on the discharge ultrasonic signal, counting two frequency domain characteristics, namely a frequency range corresponding to the highest peak value and the number of peak values, in a spectrogram, and classifying in each frequency domain characteristic; and calculating the proportion distribution of the ultrasonic signal frequency domain characteristics of the metallized film capacitor in different health states to obtain the relative change rate of the frequency domain characteristic proportion, and predicting the failure probability of the capacitor by using a logistic regression model to realize online monitoring of the health state of the metallized film capacitor.

Description

Technical Field

[0001] The present invention relates to the technical field of metallized film capacitors, and in particular to a method, system, medium and equipment for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics. Background Art

[0002] As a key component in the flexible DC transmission converter valve, metallized film capacitors provide the system with functions such as reactive power compensation, voltage stability regulation and power flow control, helping to optimize system stability and power quality, while reducing transmission losses and improving system response speed and efficiency. Compared with traditional power capacitors, metallized film capacitors have more stringent operating conditions. Their operating field strength is several times that of parallel AC capacitors, more than twice that of traditional DC filter capacitors, and has a greater harmonic content. At the same time, they have higher reliability, greater current carrying capacity and other technical indicators.

[0003] For metallized film capacitors, the IEC standard 60384-17 is used in the project, and a 5% drop in the capacitance value of the capacitor is used as the failure standard, so the error in the capacitance value monitoring of metallized film capacitors cannot be greater than 1%. Due to the limitations of algorithms and equipment, many studies have difficulty meeting the requirements for the monitoring accuracy of metallized film capacitors, so there are few results that can be applied in engineering. At the same time, some online monitoring methods will have an impact on the system circuit. For example, using the injection method to measure the capacitance value or ESR value will introduce harmonics into the system, which is not conducive to the stable operation of the system. Therefore, it is of great significance to study more accurate online monitoring methods that have as little impact on the system as possible to improve the reliability and safety of the system.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the invention and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the invention

[0005] The present invention provides a method, system, medium and equipment for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics, thereby improving judgment precision and accuracy.

[0006] A method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics includes:

[0007] Step S1, constructing a discharge ultrasonic DC monitoring experimental circuit and an aging test circuit;

[0008] Step S2, obtaining metallized film capacitors with different health levels through accelerated aging experiments, disassembling the metallized film capacitors in different health states, connecting them to the discharge ultrasonic monitoring experimental circuit, and fixing the acoustic emission sensor;

[0009] Step S3, in the discharge ultrasonic DC monitoring experimental circuit, a DC voltage is applied across the metallized film capacitor, and a discharge ultrasonic signal of the metallized film capacitor within one hour is collected by an oscilloscope;

[0010] Step S4, performing frequency domain analysis on the discharge ultrasonic signal, counting two frequency domain features, namely, the frequency range corresponding to the highest peak value and the number of peak values ​​in the spectrum graph, and performing classification within each frequency domain feature;

[0011] Step S5, calculate the distribution of the frequency domain characteristics of the ultrasonic signal of the metallized film capacitor in different health states, obtain the relative change rate of the frequency domain characteristics, and use the logistic regression model to predict the probability of capacitor failure, so as to realize online monitoring and evaluation of the health status of the metallized film capacitor.

[0012] In the method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics, step S2 includes:

[0013] Step S2.1: In the accelerated aging test loop, the test temperature is set to 65°C, the humidity is 45%RH, and the test voltage is 1.4U N , after each 12-hour aging cycle, the capacitance value of the metallized film capacitor is measured using a bridge to obtain metallized film capacitors in different health states;

[0014] Step S2.2: After the metallized film capacitor has aged to different predetermined percentages of health, the capacitor is taken out and connected to a discharge ultrasonic DC monitoring experimental circuit;

[0015] Step S2.3: Fix the acoustic emitter using ultrasonic coupling agent and fasteners, and arrange the acoustic emitter at the grounding end of the metallized film capacitor.

[0016] In the method for evaluating the health status of a metallized film capacitor based on the frequency domain characteristics of acoustic emission, the selected acoustic emitter has a frequency response characteristic in the frequency band of 120kHz to 300kHz, and the amplifier uses a PXPA3 amplifier to amplify the detected discharge ultrasonic signal by 100 times while filtering out noise to improve the signal-to-noise ratio.

[0017] In the method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics, step S3 includes:

[0018] Step S3.1: Apply 1.4U across the metallized film capacitor N The DC voltage of the discharge ultrasonic DC monitoring experimental circuit is kept constant at 20℃ and 45%RH.

[0019] Step S3.2: After the pressurization is stabilized, the discharge ultrasonic signal generated by the metallized film capacitor within one hour is collected by an oscilloscope, and the corresponding data is saved.

[0020] In the method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics, step S4 includes:

[0021] Step S4.1: Perform Fourier transform analysis on the discharge ultrasonic signal to obtain a spectrum diagram.

[0022] Step S4.2: Classify the frequency range corresponding to the highest peak in the spectrum diagram, and divide the frequency domain into three frequency bands: 0-100kHz, 100-200kHz, and greater than 200kHz. The highest peak corresponds to 0-100kHz, 100-200kHz, and greater than 200kHz. Statistically calculate the proportion of discharge signals in these three frequency bands.

[0023] Step S4.3: Classify the number of peaks in the frequency domain into single peaks and multiple peaks, and count the respective proportions of single peaks and multiple peaks in the discharge signal within one hour.

[0024] In the method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics, step S5 includes:

[0025] Step S5.1: Collect frequency domain feature proportion data under different health states of metallized film capacitors, the frequency domain feature proportion data includes the frequency range proportion corresponding to the highest peak and the peak quantity proportion, and calculate the relative change rate of the frequency domain feature proportion. The relative change rate formula is: 00%,

[0026] Step S5.2: Use the logistic regression model to establish a prediction model for capacitor failure, with the relative change rate as the input feature and the health status as the target variable. The prediction model outputs the probability P (Y=1) that each sample is in a failure state. The form of the prediction model is:

[0027]

[0028] Among them, Y is a binary response variable, failure status, 1 means failure, 0 means health, X1 and X2 are the relative change rates of the two frequency characteristics, β0 is the intercept term, β1 and β2 are the coefficients of the characteristics,

[0029] Step S5.3: Set a discrimination threshold. During the monitoring process, count the frequency domain feature proportion of the ultrasonic discharge signal and compare it with the historical data, calculate the corresponding change rate, and input the change rate as input data into the prediction model to obtain the probability P(Y=1). By comparing the probability with the set threshold, the health status of the capacitor is judged. If P(Y=1)>threshold, it is judged as a failure; otherwise, it is judged as healthy.

[0030] A health status assessment system for metallized film capacitors based on acoustic emission frequency domain characteristics includes:

[0031] Aging test circuit, which is used for: accelerated aging experiment to obtain metallized film capacitors with different health levels;

[0032] The discharge ultrasonic DC monitoring experimental circuit is used to: connect to metallized film capacitors in different health states;

[0033] The acoustic emission sensor is used to: be fixed to the metallized film capacitor, and the discharge ultrasonic monitoring experimental circuit collects the discharge ultrasonic signal of the metallized film capacitor within one hour;

[0034] A frequency domain analysis unit, which is used to: perform frequency domain analysis on the discharge ultrasonic signal, count two frequency domain features, the frequency range corresponding to the highest peak value and the number of peak values ​​in the spectrum graph, and perform classification within each frequency domain feature;

[0035] The calculation unit is used to calculate the distribution of the frequency domain characteristics of the ultrasonic signal of the metallized film capacitor in different health states, obtain the relative change rate of the frequency domain characteristics, and use the logistic regression model to predict the probability of capacitor failure, so as to realize online monitoring and evaluation of the health status of the metallized film capacitor.

[0036] In the system, the computing unit is a central processing unit.

[0037] A computer storage medium includes computer instructions, which, when executed on a computer, cause the computer to execute the method described.

[0038] An electronic device, comprising:

[0039] A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein:

[0040] When the processor executes the program, the method described is implemented.

[0041] Compared with the prior art, the present invention has the following advantages: The present invention adopts acoustic emission technology as a non-destructive detection method. The method can monitor the capacitor under normal operating conditions, and analyze the changes in the number of peaks in the frequency domain characteristics and the frequency range where the frequency domain peaks are located by real-time acquisition of acoustic emission signals. Compared with traditional detection methods, this method avoids the introduction of harmonics and other electrical interference, ensures that the performance of the capacitor during operation is not affected, and does not interfere with the normal operation of the system. This non-destructive detection method makes continuous monitoring possible and improves the safety and effectiveness of capacitor health status assessment. The method has high sensitivity and can capture small changes in the capacitor during operation in real time. By accurately analyzing the number of peaks in the frequency domain characteristics and the frequency range where the frequency domain peaks are located, the method can identify early signs of potential faults. This highly sensitive detection capability enables corresponding preventive measures to be taken before the capacitor deteriorates significantly, thereby effectively reducing the risk of failure and extending the service life of the capacitor. The method has good integrability and can be efficiently connected with existing monitoring systems and equipment. This method only requires the addition of an acoustic emission signal acquisition device and corresponding data processing software to achieve real-time assessment of the health status of metallized film capacitors without making major changes to the existing structure of the power system. This integration process is not only simple and fast, but also improves the system's intelligent monitoring capabilities and the management efficiency of power equipment, which helps to achieve comprehensive equipment health management. As the capacitor gradually ages, the discharge mechanism changes significantly, which is manifested as a jump in frequency domain characteristics. For example, when the health state approaches 95%C0, the number of peaks and frequency range in the frequency domain may change significantly, indicating a transition from local discharge to global breakdown. This change is closely related to the weakening of electrical stress, material degradation and self-healing ability, and this method has high detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] By reading the detailed description of the preferred specific embodiments below, various other advantages and benefits of the present invention will become clear to those of ordinary skill in the art. The drawings in the specification are only for the purpose of illustrating the preferred embodiments and are not considered to be limitations of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative work. Moreover, the same reference numerals are used to represent the same components throughout the drawings.

[0043] In the attached picture:

[0044] Figure 1 It is a schematic diagram of the discharge ultrasonic DC monitoring experimental circuit;

[0045] Figure 2 It is a schematic diagram of the accelerated aging experimental circuit;

[0046] Figure 3 It is a schematic diagram of the frequency range of 0-100kHz corresponding to the highest peak;

[0047] Figure 4 It is a schematic diagram showing that the frequency range corresponding to the highest peak is 100-200kHz;

[0048] Figure 5 It is a schematic diagram that the frequency range corresponding to the highest peak is greater than 200kHz;

[0049] FIG6 (a) and FIG6 (b) are schematic diagrams of classification of the number of peaks in the frequency domain, wherein FIG6 (a) is a schematic diagram of classification when the number of peaks in the frequency domain is single-peaked, and FIG6 (b) is a schematic diagram of classification when the number of peaks in the frequency domain is multi-peaked;

[0050] Figure 7 This is a schematic diagram of the distribution of the frequency range corresponding to the highest peak in the frequency domain;

[0051] Figure 8 It is a schematic diagram of the distribution of the number of peaks in the frequency domain.

[0052] The present invention is further explained below in conjunction with the accompanying drawings and embodiments. DETAILED DESCRIPTION

[0053] The specific embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the specific embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0054] It should be noted that certain words are used in the specification and claims to refer to specific components. Those skilled in the art should understand that technicians may use different nouns to refer to the same component. This specification and claims do not use the difference in nouns as a way to distinguish components, but use the difference in the functions of the components as the criterion for distinction. As mentioned throughout the specification and claims, "including" or "comprising" is an open term, so it should be interpreted as "including but not limited to". The subsequent description of the specification is a preferred embodiment of the present invention, but the description is based on the general principles of the specification and is not intended to limit the scope of the present invention. The scope of protection of the present invention shall be determined by the attached claims.

[0055] To facilitate understanding of the embodiments of the present invention, further explanation will be given below by taking specific embodiments as examples in conjunction with the accompanying drawings, and each of the accompanying drawings does not constitute a limitation on the embodiments of the present invention.

[0056] like Figures 1 to 8As shown, the method for evaluating the health status of metallized film capacitors based on acoustic emission frequency domain characteristics includes the following steps:

[0057] Step S1, constructing a discharge ultrasonic DC monitoring experimental circuit and an aging test circuit;

[0058] Step S2, obtaining metallized film capacitors with different health levels through accelerated aging experiments, disassembling the metallized film capacitors in different health states, connecting them to the discharge ultrasonic monitoring experimental circuit, and fixing the acoustic emission sensor;

[0059] Step S3, in the discharge ultrasonic DC monitoring experimental circuit, a DC voltage is applied across the metallized film capacitor, and a discharge ultrasonic signal of the metallized film capacitor within one hour is collected by an oscilloscope;

[0060] Step S4, performing frequency domain analysis on the discharge ultrasonic signal, counting two frequency domain features, namely, the frequency range corresponding to the highest peak value and the number of peak values ​​in the spectrum graph, and performing classification within each frequency domain feature;

[0061] Step S5, calculate the distribution of the frequency domain characteristics of the ultrasonic signal of the metallized film capacitor in different health states, obtain the relative change rate of the frequency domain characteristics, and use the logistic regression model to predict the probability of capacitor failure, so as to realize online monitoring and evaluation of the health status of the metallized film capacitor.

[0062] In a preferred embodiment of the method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics, step S2 includes:

[0063] Step S2.1: In the accelerated aging test loop, the test temperature is set to 65°C, the humidity is 45%RH, and the test voltage is 1.4U N , after each 12-hour aging cycle, the capacitance value of the metallized film capacitor is measured using a bridge to obtain metallized film capacitors in different health states;

[0064] Step S2.2: After the metallized film capacitor has aged to different predetermined percentages of health, the capacitor is taken out and connected to a discharge ultrasonic DC monitoring experimental circuit;

[0065] Step S2.3: Fix the acoustic emitter using ultrasonic coupling agent and fasteners, and arrange the acoustic emitter at the grounding end of the metallized film capacitor.

[0066] In a preferred embodiment of the method for evaluating the health status of a metallized film capacitor based on the frequency domain characteristics of acoustic emission, the selected acoustic emitter has a frequency response characteristic in the frequency band of 120kHz to 300kHz, and the amplifier uses a PXPA3 amplifier to amplify the detected discharge ultrasonic signal by 100 times while filtering out noise, thereby improving the signal-to-noise ratio.

[0067] In a preferred embodiment of the method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics, step S3 includes:

[0068] Step S3.1: Apply 1.4U across the metallized film capacitor N The DC voltage of the discharge ultrasonic DC monitoring experimental circuit was kept constant at 20°C and 45%RH.

[0069] Step S3.2: After the pressurization is stabilized, the discharge ultrasonic signal generated by the metallized film capacitor within one hour is collected by an oscilloscope, and the corresponding data is saved.

[0070] In a preferred embodiment of the method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics, step S4 includes:

[0071] Step S4.1: Perform Fourier transform analysis on the discharge ultrasonic signal to obtain a spectrum diagram.

[0072] Step S4.2: Classify the frequency range corresponding to the highest peak in the spectrum diagram, and divide the frequency domain into three frequency bands: 0-100kHz, 100-200kHz, and greater than 200kHz. The highest peak corresponds to 0-100kHz, 100-200kHz, and greater than 200kHz. Statistically calculate the proportion of discharge signals in these three frequency bands.

[0073] Step S4.3: Classify the number of peaks in the frequency domain into single peaks and multiple peaks, and count the respective proportions of single peaks and multiple peaks in the discharge signal within one hour.

[0074] In a preferred embodiment of the method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics, step S5 includes:

[0075] Step S5.1: Collect frequency domain feature proportion data under different health states of metallized film capacitors, the frequency domain feature proportion data includes the frequency range proportion corresponding to the highest peak and the peak quantity proportion, and calculate the relative change rate of the frequency domain feature proportion. The relative change rate formula is: 00%,

[0076] Step S5.2: Use the logistic regression model to establish a prediction model for capacitor failure, with the relative change rate as the input feature and the health status as the target variable. The prediction model outputs the probability P (Y=1) that each sample is in a failure state. The form of the prediction model is:

[0077]

[0078] Among them, Y is a binary response variable, failure status, 1 means failure, 0 means health, X1 and X2 are the relative change rates of the two frequency characteristics, β0 is the intercept term, β1 and β2 are the coefficients of the characteristics,

[0079] Step S5.3: Set a discrimination threshold. During the monitoring process, count the frequency domain feature proportion of the ultrasonic discharge signal and compare it with the historical data, calculate the corresponding change rate, and input the change rate as input data into the prediction model to obtain the probability P(Y=1). By comparing the probability with the set threshold, the health status of the capacitor is judged. If P(Y=1)>threshold, it is judged as a failure; otherwise, it is judged as healthy.

[0080] A health status assessment system for metallized film capacitors based on acoustic emission frequency domain characteristics includes:

[0081] Aging test circuit, which accelerates aging experiments to obtain metallized film capacitors of different health levels;

[0082] The discharge ultrasonic DC monitoring experimental circuit is connected to the metallized film capacitor in different health states;

[0083] An acoustic emission sensor is fixed to the metallized film capacitor, and a discharge ultrasonic monitoring experimental circuit collects a discharge ultrasonic signal of the metallized film capacitor within one hour;

[0084] A frequency domain analysis unit, which performs frequency domain analysis on the discharge ultrasonic signal, counts two frequency domain features, namely, the frequency range corresponding to the highest peak value and the number of peak values ​​in the spectrum graph, and performs classification within each frequency domain feature;

[0085] A calculation unit calculates the distribution of the frequency domain characteristics of the ultrasonic signal of the metallized film capacitor in different health states, obtains the relative change rate of the frequency domain characteristics, and uses a logistic regression model to predict the probability of capacitor failure, so as to realize online monitoring and evaluation of the health status of the metallized film capacitor.

[0086] In one embodiment, in the discharge ultrasonic DC monitoring experimental circuit, the DC source and the protective resistor with a resistance value of 25kΩ are connected in series to output a DC voltage, and the rated capacity of the metallized film capacitor is 18μF and the rated voltage is 2.0kV. During the experiment, the metallized film capacitor element is placed in a programmable high and low temperature test box, the equipment model is: EL-10JT, which can maintain constant temperature and humidity. The Tektronix high-voltage probe is placed in the high-voltage section of the constant temperature box to detect the capacitor voltage, and the equipment model is P6015A. The acoustic transmitter uses a PXR15 sensor, which is arranged at the low-voltage end of the component through a coupling agent and a fixing. The acoustic sensor has a sensitivity of more than 60dB between 120kHz and 300kHz, indicating that the ultrasonic sensor has good frequency response characteristics within this range. This frequency band can cover the main frequency band of the ultrasonic signal generated by the self-healing of the metallized film capacitor, and the ultrasonic signal can be better collected. The amplifier is connected to the acoustic sensor, which has the characteristics of high gain, wide dynamic range, low noise, and high input impedance. It can amplify the detected ultrasonic signal by 100 times, while filtering out the noise part and improving the signal-to-noise ratio. The maximum output signal of the amplifier under a 50Ω load can reach 20V, which can meet the needs of amplifying ultrasonic signals. The oscilloscope is used to record voltage signals and sound signals. Its model is Tektronix MDO34. The high-voltage probe and sound transmitter are connected to the oscilloscope through the amplifier through a signal transmission line.

[0087] In the accelerated aging test circuit, the DC source and the protection resistor with a resistance value of 25kΩ are connected in series to output a DC voltage. The rated capacity of the metallized film capacitor is 18μF and the rated voltage is 2.0kV. During the experiment, multiple metallized film capacitors are placed in parallel in a programmable high and low temperature test chamber, equipment model: EL-10JT, which can maintain constant temperature and humidity.

[0088] In a preferred embodiment of the system, the computing unit is a central processing unit.

[0089] A computer storage medium includes computer instructions, which, when executed on a computer, cause the computer to execute the method described.

[0090] An electronic device, comprising:

[0091] A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein:

[0092] When the processor executes the program, the method described is implemented.

[0093] In one embodiment, step S2: obtaining metallized film capacitors of different health levels through accelerated aging experiments, and connecting them to Figure 1 The circuit is shown and the acoustic emission sensor is fixed.

[0094] Step S2.1: The circuit of the accelerated aging experiment is as follows Figure 2 As shown, the experimental temperature is set to 65°C, the humidity is 45%RH, and the experimental voltage is 1.4U N After each 12-hour aging period, the capacitance of the capacitor is measured using a high-precision bridge to obtain the different health states of the metallized film capacitor elements.

[0095] Step S2.2: When the capacitor ages to different health levels, such as 99%C0, it is taken out and connected to the discharge ultrasonic DC monitoring experimental circuit.

[0096] Step S2.3: Use special ultrasonic coupling agent and fasteners to fix the acoustic emitter, which is arranged at the ground end of the capacitor element. The selected acoustic emitter should have good frequency response characteristics in the frequency band of 120kHz to 300kHz to ensure that the ultrasonic signal can be effectively collected. The amplifier uses a PXPA3 amplifier, which can amplify the detected ultrasonic signal by 100 times, filter out the noise part, and improve the signal-to-noise ratio.

[0097] Step S3: applying a DC voltage across the metallized film capacitor element, and collecting the discharge ultrasonic signal of the capacitor within one hour through an oscilloscope.

[0098] Step S3.1: Apply 1.4U across the metallized film capacitor element N The DC voltage was kept constant at 20°C and the humidity was 45%RH.

[0099] Step S3.2: After the pressure is stabilized, the discharge ultrasonic signal generated by the metallized film capacitor element within one hour is collected by an oscilloscope, and the corresponding data is saved.

[0100] Step S4: Perform frequency domain analysis on the discharge ultrasonic signal collected from the capacitor within one hour. Count the two frequency domain features of the frequency range corresponding to the highest peak in the spectrum graph and the number of peaks, and perform corresponding classification within each feature.

[0101] Step S4.1: Perform Fourier transform analysis on the discharge ultrasonic signal collected within one hour.

[0102] Step S4.2: Classify the frequency range corresponding to the highest peak in the spectrum graph and divide the frequency domain into three frequency bands: 0-100kHz, 100-200kHz and greater than 200kHz. Figure 3 , Figure 4 , Figure 5 As shown in the figure, the highest peaks correspond to 0-100kHz, 100-200kHz and greater than 200kHz. The proportion of discharge signals in these three frequency bands is counted to analyze their distribution within one hour.

[0103] Step S4.3: Classify the number of peaks in the frequency domain into single peak and multiple peaks, as shown in Figure 6 (a) and Figure 6 (b). Count the proportion of single peak and multiple peak in the discharge signal within one hour.

[0104] Step S5: Calculate the distribution of the frequency domain characteristics of the ultrasonic signal of the metallized film capacitor in different health states, so as to further obtain the relative change rate of the frequency domain characteristics. Use the logistic regression model to predict the probability of capacitor failure, so as to achieve online monitoring and evaluation of the health status of the capacitor.

[0105] Step S5.1: Collect the frequency domain feature proportion data of the metallized film capacitor under different health conditions, and calculate the relative change rate of the frequency domain feature proportion. Relative change rate formula: 00%.

[0106] Step S5.2: Use the logistic regression model to establish a prediction model for capacitor failure. The relative rate of change is used as the input feature and the health status is used as the target variable. The final model outputs the probability P (Y=1) that each sample is in a failure state. The basic form of logistic regression is:

[0107]

[0108] Where Y is a binary response variable (i.e., failure status, 1 for failure and 0 for health). X1 and X2 are the relative rates of change of the two frequency features. β0 is the intercept term, and β1 and β2 are the coefficients of the features.

[0109] Step S5.3: Set the discrimination threshold according to the actual application requirements. During the monitoring process, the frequency domain feature ratio of the ultrasonic discharge signal within one hour is counted and compared with the historical data to calculate the corresponding rate of change. The rate of change is input into the prediction model as input data to obtain the probability P(Y=1). By comparing the probability with the set threshold, the health status of the capacitor is judged. If P(Y=1)>threshold, it is judged as failure; otherwise, it is judged as healthy.

[0110] An example of using specific frequency domain characteristic parameters to determine the health status of metallized film capacitors:

[0111] Figure 7The figure shows the distribution of the frequency range corresponding to the highest peak of the acoustic emission signal in the frequency domain of the metallized film capacitor in different health states. After FFT analysis of the ultrasonic signals collected at different aging stages, it can be seen that when the capacitor is in a 100% C0 health state, the frequency corresponding to the highest peak appears in the frequency range of 100kHz-200kHz, accounting for 100%. When the capacitor is in a 96%-99% C0 state, the frequency corresponding to the highest peak in the range of 100kHz-200kHz accounts for more than 80%. When the capacitor is close to 95% C0, the frequency corresponding to the highest peak in the frequency range of 100kHz-200kHz accounts for a significant decrease to 60%. The relative rate of change jumps from 2% to 36%. The relative rate of change of the frequency range corresponding to the highest peak in the frequency domain jumps significantly, indicating that the capacitor is "degraded".

[0112] Figure 8 The figure shows the distribution of the number of peaks in the frequency domain of the acoustic emission signal of the metallized film capacitor in different health states. After FFT analysis of the ultrasonic signals collected at different aging stages, it can be seen that when the capacitor is in a healthy state of 96%-100% C0, within the frequency range of 20kHz-500kHz, about 80% of the ultrasonic signals have a single peak, and the number of signals with multiple peaks is not large. When the capacitor is in a healthy state close to 95% C0, the proportion of single peaks drops to 34%. Its relative rate of change jumps from 12% to 58%. The relative rate of change of the number of peaks in the frequency domain jumps significantly, indicating that the capacitor is "degraded".

[0113] Although the embodiments of the present invention are described above in conjunction with the accompanying drawings, the present invention is not limited to the above specific embodiments and application fields, and the above specific embodiments are only illustrative and instructive, rather than restrictive. A person of ordinary skill in the art can also make many forms under the guidance of this specification and without departing from the scope of protection of the claims of the present invention, all of which belong to the protection of the present invention.

Claims

1. A method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics, characterized in that: The steps include: Step S1, constructing a discharge ultrasonic DC monitoring experimental circuit and an aging test circuit; Step S2, obtaining metallized film capacitors with different health levels through accelerated aging experiments, disassembling the metallized film capacitors in different health states, connecting them to the discharge ultrasonic monitoring experimental circuit, and fixing the acoustic emission sensor; Step S3, in the discharge ultrasonic DC monitoring experimental circuit, a DC voltage is applied across the metallized film capacitor, and a discharge ultrasonic signal of the metallized film capacitor within one hour is collected by an oscilloscope; Step S4, performing frequency domain analysis on the discharge ultrasonic signal, counting two frequency domain features, namely, the frequency range corresponding to the highest peak value and the number of peak values ​​in the spectrum graph, and performing classification within each frequency domain feature; Step S5, calculate the distribution of the frequency domain characteristics of the ultrasonic signal of the metallized film capacitor in different health states, obtain the relative change rate of the frequency domain characteristics, and use the logistic regression model to predict the probability of capacitor failure, so as to realize online monitoring and evaluation of the health status of the metallized film capacitor.

2. The method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics according to claim 1 is characterized in that: Preferably, the step S2 comprises: Step S2.1: In the accelerated aging test loop, the test temperature is set to 65°C, the humidity is 45%RH, and the test voltage is 1.4U N , after each 12-hour aging cycle, the capacitance value of the metallized film capacitor is measured using a bridge to obtain metallized film capacitors in different health states; Step S2.2: After the metallized film capacitor has aged to different predetermined percentages of health, the capacitor is taken out and connected to a discharge ultrasonic DC monitoring experimental circuit; Step S2.3: Fix the acoustic emitter using ultrasonic coupling agent and fasteners, and arrange the acoustic emitter at the grounding end of the metallized film capacitor.

3. A method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics according to claim 2, characterized in that: The selected acoustic transmitter has a frequency response characteristic of 120kHz to 300kHz. The amplifier uses a PXPA3 amplifier to amplify the detected discharge ultrasonic signal by 100 times while filtering out noise to improve the signal-to-noise ratio.

4. The method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics according to claim 1, characterized in that: The step S3 comprises: Step S3.1: Apply 1.4U across the metallized film capacitor N The DC voltage of the discharge ultrasonic DC monitoring experimental circuit is kept constant at 20℃ and 45%RH. Step S3.2: After the pressurization is stabilized, the discharge ultrasonic signal generated by the metallized film capacitor within one hour is collected by an oscilloscope, and the corresponding data is saved.

5. The method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics according to claim 1, characterized in that: The step S4 comprises: Step S4.1: Perform Fourier transform analysis on the discharge ultrasonic signal to obtain a spectrum diagram. Step S4.2: Classify the frequency range corresponding to the highest peak in the spectrum diagram, and divide the frequency domain into three frequency bands: 0-100kHz, 100-200kHz, and greater than 200kHz. The highest peak corresponds to 0-100kHz, 100-200kHz, and greater than 200kHz. Statistically calculate the proportion of discharge signals in these three frequency bands. Step S4.3: Classify the number of peaks in the frequency domain into single peaks and multiple peaks, and count the respective proportions of single peaks and multiple peaks in the discharge signal within one hour.

6. The method for evaluating the health status of a metallized film capacitor based on acoustic emission frequency domain characteristics according to claim 1, characterized in that: The step S5 comprises: Step S5.1: Collect frequency domain feature proportion data under different health conditions of metallized film capacitors. The frequency domain feature proportion data includes the frequency range proportion corresponding to the highest peak value and the peak quantity proportion. Calculate the relative change rate of the frequency domain feature proportion. The relative change rate formula is: 00%, Step S5.2: Use the logistic regression model to establish a prediction model for capacitor failure, with the relative change rate as the input feature and the health status as the target variable. The prediction model outputs the probability P (Y=1) that each sample is in a failure state. The form of the prediction model is: ， Among them, Y is a binary response variable, failure status, 1 means failure, 0 means health, X1 and X2 are the relative change rates of the two frequency characteristics, β0 is the intercept term, β1 and β2 are the coefficients of the characteristics, Step S5.3: Set a discrimination threshold. During the monitoring process, count the frequency domain feature proportion of the ultrasonic discharge signal and compare it with the historical data, calculate the corresponding change rate, and input the change rate as input data into the prediction model to obtain the probability P(Y=1). By comparing the probability with the set threshold, the health status of the capacitor is judged. If P(Y=1)>threshold, it is judged as a failure; otherwise, it is judged as healthy.

7. A system for evaluating the health status of metallized film capacitors based on acoustic emission frequency domain characteristics, characterized in that: These include, Aging test circuit, which is used for: accelerated aging experiment to obtain metallized film capacitors with different health levels; The discharge ultrasonic DC monitoring experimental circuit is used to: connect to metallized film capacitors in different health states; The acoustic emission sensor is used to: be fixed to the metallized film capacitor, and the discharge ultrasonic monitoring experimental circuit collects the discharge ultrasonic signal of the metallized film capacitor within one hour; A frequency domain analysis unit, which is used to: perform frequency domain analysis on the discharge ultrasonic signal, count two frequency domain features, the frequency range corresponding to the highest peak value and the number of peak values ​​in the spectrum graph, and perform classification within each frequency domain feature; The calculation unit is used to calculate the distribution of the frequency domain characteristics of the ultrasonic signal of the metallized film capacitor in different health states, obtain the relative change rate of the frequency domain characteristics, and use the logistic regression model to predict the probability of capacitor failure, so as to realize online monitoring and evaluation of the health status of the metallized film capacitor.

8. The system according to claim 7, characterized in that The computing unit is the central processing unit.

9. A computer storage medium, characterized in that The storage medium includes computer instructions, which, when executed on a computer, enable the computer to execute the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: The electronic device comprises: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method according to any one of claims 1 to 7 is implemented.

Images