CN114441859A - High-precision capacitance measurement method based on BP neural network - Google Patents

High-precision capacitance measurement method based on BP neural network Download PDF

Info

Publication number
CN114441859A
CN114441859A CN202210108623.5A CN202210108623A CN114441859A CN 114441859 A CN114441859 A CN 114441859A CN 202210108623 A CN202210108623 A CN 202210108623A CN 114441859 A CN114441859 A CN 114441859A
Authority
CN
China
Prior art keywords
neural network
sample
sample set
capacitance
signal
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210108623.5A
Other languages
Chinese (zh)
Inventor
闭吕庆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yulin Normal University
Original Assignee
Yulin Normal University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Yulin Normal University filed Critical Yulin Normal University
Priority to CN202210108623.5A priority Critical patent/CN114441859A/en
Publication of CN114441859A publication Critical patent/CN114441859A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R27/00—Arrangements for measuring resistance, reactance, impedance, or electric characteristics derived therefrom
    • G01R27/02—Measuring real or complex resistance, reactance, impedance, or other two-pole characteristics derived therefrom, e.g. time constant
    • G01R27/26—Measuring inductance or capacitance; Measuring quality factor, e.g. by using the resonance method; Measuring loss factor; Measuring dielectric constants ; Measuring impedance or related variables
    • G01R27/2605—Measuring capacitance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Computational Mathematics (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Artificial Intelligence (AREA)
  • Health & Medical Sciences (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • General Health & Medical Sciences (AREA)
  • Mathematical Analysis (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Operations Research (AREA)
  • Probability & Statistics with Applications (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Measurement Of Resistance Or Impedance (AREA)

Abstract

The invention discloses a high-precision capacitance measuring method based on a BP (back propagation) neural network, which relates to the technical field of capacitance measurement and solves the technical problem of poor capacitance measuring precision, and the measuring method comprises the following steps: voltage data at two ends of a sample capacitor are obtained through a data acquisition circuit, and an original sample set is constructed; cleaning and denoising the original sample set to obtain a preprocessed sample set; performing feature selection on the preprocessed sample set; after normalization processing is carried out on the preprocessing sample set, the preprocessing sample set is input into a BP neural network for training, and a test model is obtained; and measuring the actual capacitance to be measured by using the test model. The capacitance measuring precision of the invention reaches 97 percent, and the measuring range is as follows: 1-470pF, the single measurement time is only 0.5s, the method has good generalization capability, and can be applied to a real-time online capacitive sensor application system.

Description

High-precision capacitance measurement method based on BP neural network
Technical Field
The invention relates to the technical field of capacitance measurement, in particular to a high-precision capacitance measurement method based on a BP neural network.
Background
At present, capacitance measurement methods commonly used include a capacitance meter method, a three-meter method, a bridge method, and a resonance method. The capacitance method uses a capacitance measuring instrument to measure a target capacitance value, the measurement accuracy of the capacitance method depends on the used measuring instrument, manual operation is required, and the capacitance method is not suitable for application occasions requiring on-line real-time measurement. The three-meter method is used in early stage of circuit design, and uses AC voltmeter, AC ammeter and power meter to obtain the voltage, current and power electric parameters of two ends of tested element, and then calculates them to obtain the required parameter values. The method is complicated in operation, the measurement error is influenced by 3 measuring instruments and is difficult to improve, and the method is generally only applied to experimental verification. The bridge method and the resonance method respectively connect the capacitance to be measured into the bridge circuit and the resonance circuit, then measure the output voltage of the bridge and the resonance frequency of the resonance circuit, and calculate the capacitance value of the element to be measured by using the output voltage and the resonance frequency. Although the method can be applied to a real-time system, the method is limited by nonlinearity of a circuit when the circuit works in a wide frequency band, and when the target capacitance variation range is large, the measurement error is large.
Disclosure of Invention
The invention aims to solve the technical problem of the prior art, and aims to provide a high-precision capacitance measuring method based on a BP neural network, which can improve the measuring precision.
The technical scheme of the invention is as follows: a high-precision capacitance measurement method based on a BP neural network comprises the following steps:
voltage data at two ends of a sample capacitor are obtained through a data acquisition circuit, and an original sample set is constructed;
cleaning and denoising the original sample set to obtain a preprocessed sample set;
performing feature selection on the preprocessed sample set;
after normalization processing is carried out on the preprocessing sample set, the preprocessing sample set is input into a BP neural network for training, and a test model is obtained;
and measuring the actual capacitance to be measured by using the test model.
As a further improvement, the data acquisition circuit comprises a main control board, a signal generator, an amplifier, a sampler, a resistor and a sample capacitor, wherein the main control board is sequentially connected with one end of the signal generator, the amplifier, the resistor and the sample capacitor in series, one end of the sample capacitor is connected with the main control board through the sampler, and the other end of the sample capacitor is grounded.
Further, the process of obtaining the voltage data across the sample capacitor is:
the main control board starts a signal generator to linearly output a sinusoidal test signal with the frequency range of 1MHz-8MHz by stepping at 100kHz, totaling 71 frequency points and maintaining time of each frequency point signal at 500 ms;
the amplifier carries out amplitude limiting amplification output on the sinusoidal test signal, so that the signal output amplitude of all frequency points is kept unchanged at 1V;
and performing effective value detection and analog-to-digital conversion on the voltage amplitudes at two ends of the sample capacitor through the sampler, and storing the obtained signal amplitudes as characteristic values of the data sample.
Further, the cleaning treatment comprises the following steps: respectively measuring the voltages of 50 sample capacitors for 6 times, wherein for the same sample capacitor, if the voltage amplitude of a certain measurement is smaller than half of the average value of other five measurements or larger than 1.5 times of the average value of other five measurements, the same sample capacitor is considered as an abnormal data point, and the abnormal data point is replaced by the average value of other five measurements;
the denoising treatment adopts a median filter method;
after removing the abnormal data points and filtering, the average of 6 measurements is taken as the new eigenvalue of the sample.
Further, in feature selection, 71 signal voltage amplitudes with frequency values greater than 1MHz are taken as features of the sample.
Further, the BP neural network is a 5-layer network structure of 71 × 16 × 32 × 16 × 1.
Further, the coefficient R is determined by mean square error MSE and2these two indices evaluate the performance of the test model.
Further, the training parameters of the BP neural network include: the data set partition ratio is train: evaluation: test is 70:15:15, the training function is Levenberg-Marquardt, the evaluation function is MSE, the maximum iteration number Epoch is 1000, the learning target is 1.0e-7, the learning rate is 0.001, the minimum Validation failure number Validation Checks is 6, the hidden layer function is tandig, and the output layer function is purelin.
Advantageous effects
Compared with the prior art, the invention has the advantages that:
the capacitance measuring precision of the invention reaches 97 percent, and the measuring range is as follows: 1-470pF, the single measurement time is only 0.5s, the method has good generalization capability, and can be applied to a real-time online capacitive sensor application system.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a circuit diagram of data acquisition in the present invention;
FIG. 3 is a diagram of a BP neural network structure according to the present invention;
FIG. 4 is a graph of the convergence of the BP neural network training in the present invention;
FIG. 5 is a graph of a regression fit of a training set in accordance with the present invention;
FIG. 6 is a comparison graph of the predicted result and the original value in the present invention.
Detailed Description
The invention will be further described with reference to specific embodiments shown in the drawings.
Referring to fig. 1 to 6, a high-precision capacitance measurement method based on a BP neural network includes:
voltage data at two ends of a sample capacitor C are obtained through a data acquisition circuit, and an original sample set is constructed;
cleaning and denoising an original sample set to obtain a preprocessed sample set;
selecting characteristics of the preprocessed sample set;
after normalization processing is carried out on the preprocessed sample set, the preprocessed sample set is input into a BP neural network for training, and a test model is obtained;
and measuring the actual capacitance to be measured by using the test model.
In this embodiment, the data acquisition circuit includes a main control board DSP, a signal generator DDS, an amplifier AGC, a sampler ADC, a resistor R, and a sample capacitor C, where the main control board DSP is sequentially connected in series with one end of the signal generator DDS, the amplifier AGC, the resistor R, and one end of the sample capacitor C is connected to the main control board DSP through the sampler ADC, and the other end of the sample capacitor C is grounded, as shown in fig. 2.
The DSP of the main control board adopts a 32-bit C2000 series high-performance real-time microcontroller TMS320F28379D manufactured by TI (Texas instruments); the operation rate of the system reaches 800MIPS, the capacity of the flash memory is 1MB, and the requirements of real-time performance and precision of a measuring system can be met. The signal generator DDS adopts a direct digital frequency synthesizer AD9910 produced by AD (analog devices), configured by a DSP controller and started, and linearly outputs sine wave frequency sweeping signals with the frequency range of 1MHz-8MHz and 71 frequency points in total. The amplifier AGC is an AD8367 amplifier of AD company, and the AD8367 amplifier can keep the amplitude of the sweep frequency signal at 1V in the whole frequency range. The sampler ADC is composed of an effective value detector and an AD converter, wherein the effective value detection is realized by an effective value measuring chip AD637 produced by AD company, the working frequency range of the sampler ADC is 0-8 MHz, and the voltage measuring range is-45 dBm- +30 dBm; the AD converter adopts a DSP with a built-in 16 bit-ADC.
The process of obtaining voltage data across the sample capacitor C is:
a main control board DSP starts a signal generator DDS to output a sine test signal with a frequency range of 1MHz-8MHz in a linear mode, wherein the step is 100kHz, the total number of frequency points is 71, and the signal maintaining time of each frequency point is 500 ms;
the amplifier AGC carries out amplitude limiting amplification output on the sinusoidal test signal, so that the signal output amplitude of all frequency points is kept unchanged at 1V;
and performing effective value detection and analog-to-digital conversion on the voltage amplitudes at two ends of the sample capacitor C through the sampler ADC, and storing the obtained signal amplitudes as characteristic values of the data sample.
The cleaning treatment comprises the following steps: respectively measuring the voltages of 50 sample capacitors C for 6 times, wherein the voltage amplitude of a certain measurement is smaller than half of the average value of other five measurements or larger than 1.5 times of the average value of other five measurements for the same sample capacitor C at intervals of 24 hours, and replacing the abnormal data point with the average value of other five measurements;
the denoising process adopts a median filter method, which is a high-efficiency nonlinear filtering method and can effectively filter random noise such as white noise in analog discrete signals, wherein in a discrete time domain Standard Median (SM) filtering algorithm, a window slides on a signal sequence X, and the median of samples in the window is used as the output of each position. The median filtering operation can be expressed as:
Y＝Med*X1,X2,…,XN}
in the formula, Y is a filtering output value; x is a discrete analog signal sequence; n is the size of the filtering window; x1,X2,…,XNAre samples within a filter window;
after removing the abnormal data points and filtering, the average of 6 measurements is taken as the new eigenvalue of the sample.
At low frequency (<1MHz), the capacitor is equivalent to an open circuit, the amplitudes of the voltages at two ends of the capacitor are input signal amplitudes at this time, and no distinction is made among all samples, so that 71 signal voltage amplitudes with frequency values greater than 1MHz are taken as the characteristics of the samples during characteristic selection.
The BP neural network has a 5-layer network structure of 71 × 16 × 32 × 16 × 1, and as shown in fig. 3, the number of input layer nodes is equal to the number of characteristics of a sample, and is 71; the number of output layer nodes is 1; the number of nodes of the three hidden layers is 16, 32 and 16 respectively.
And (3) converting the data into a [0,1] interval by adopting a minmax normalization method for the preprocessed sample set, wherein the conversion formula is as follows:
Figure BDA0003494655870000051
in the formula, xiFor initial data, min (x)i) And max (x)i) Maximum and minimum values of the initial data, xi' is normalized output data.
By mean square error MSE and decision coefficient R2These two indices evaluate the performance of the test model.
MSE is the mean of the squares of the differences between predicted and actual values, and is given by the formula:
Figure BDA0003494655870000061
in the formula, n is the number of samples; y isiAnd
Figure BDA0003494655870000062
the real value and the predicted value of the sample are respectively. MSE is the square amplification of the error, and more emphasis is placed on amplifying errors with larger deviations than other methods, so that the MSE can be used for evaluating the overall stability of the model. The smaller the MSE value, the better the model performance, and vice versa.
R2The proportion of variance in the explanation of the designed model is represented as a comparison dimension of the model relative to the mean model, and the calculation formula is as follows:
Figure BDA0003494655870000063
Figure BDA0003494655870000064
Figure BDA0003494655870000065
wherein SSE (sum of die to errors) is the sum of squares of the errors; SST (Total sum of squares) is the sum of the squares of the deviations of the samples from the mean. When R is2When the SSE is equal to 0, the predicted value is completely consistent with the true value, and the designed model perfectly explains the change of the dependent variable. When R is2When the value is 0, SSE is SST, which indicates that the prediction performance is the same as that of the mean model, and the model has no further interpretability relative to the mean model and is not available. Thus, R2In [0,1]]Intervals can be used to evaluate the performance of the model, with closer to 1 indicating better performance of the model.
The training parameters of the BP neural network comprise: the data set partition ratio is train: evaluation: test is 70:15:15, the training function is Levenberg-Marquardt, the evaluation function is MSE, the maximum iteration number Epoch is 1000, the learning target is 1.0e-7, the learning rate is 0.001, the minimum Validation failure number validationchecks is 6, the hidden layer function is tandig, and the output layer function is purelin. As shown in table 1.
Table 1: neural network model training parameters
Parameter name Set value
Data set partitioning ratio (train: evaluation: test) 70:15:15
Training function Levenberg-Marquardt
Evaluation function MSE
Maximum number of iterations (Epoch) 1000
Learning objective 1.0e-7
Learning rate 0.001
Minimum number of Validation failures (Validation Checks) 6
Function of hidden layer tansig
Output layer boxNumber of purelin
Fig. 4 is a training convergence curve of the BP neural network. As can be seen from fig. 4, the MSE performance curve decreases rapidly with the increase of the number of iterations, and at the 10 th time, the MSE value of the training set reaches the set learning target value (1.0 e)-7) And the training is finished, and the MSE values of the verification set and the test set tend to be stable after the 3 rd iteration. Fig. 5 is a regression fit curve of the BP neural network on the training set. Determining the coefficient R2Values on training, validation and test data are 1, 0.99747 and 0.99876 respectively, and 0.99709 is also achieved on the whole training set, which shows that the BP network model provided by the invention has high prediction accuracy.
In order to evaluate the prediction capability of the model on new data, the trained model is used for predicting the test set, and the prediction results are shown in the following table 2.
Table 2 test set test results
Figure BDA0003494655870000071
Figure BDA0003494655870000081
As can be seen from table 2, the error of the model prediction is only 0.417% at the minimum, 10.1073% at the maximum, and the average error is about 2.6291%, and there is a relatively large prediction error when the capacitance is less than 10 pF. The comparison of the predicted results with the original values is shown in fig. 6. As can be seen from the figure, the prediction result of the model has good consistency with the original value, and the coefficient of determination R on the test set20.99914 was also reached, indicating that the model has good generalization ability.
The invention discloses a high-precision capacitance measuring method based on a BP neural network. Acquiring voltage division amplitudes of 50 standard capacitors at 71 different frequency points by using an acquisition circuit, and constructing a 50 x 71C-U data set after data cleaning and median filtering denoising pretreatment; and then, training a BP neural network model with a hidden layer structure of 16 multiplied by 32 multiplied by 16 by using the data set to obtain a capacitance prediction model. Test results show that the average relative error between the model predicted value and the actual value is only 2.6291%, and the method has good generalization capability.
The above is only a preferred embodiment of the present invention, and it should be noted that it is obvious to those skilled in the art that several variations and modifications can be made without departing from the structure of the present invention, which will not affect the effect of the implementation of the present invention and the utility of the patent.

Claims (8)

1. A high-precision capacitance measuring method based on a BP neural network is characterized by comprising the following steps:
voltage data at two ends of a sample capacitor (C) are obtained through a data acquisition circuit, and an original sample set is constructed;
cleaning and denoising the original sample set to obtain a preprocessed sample set;
performing feature selection on the preprocessed sample set;
after normalization processing is carried out on the preprocessing sample set, the preprocessing sample set is input into a BP neural network for training, and a test model is obtained;
and measuring the actual capacitance to be measured by using the test model.
2. The BP neural network-based high-precision capacitance measuring method according to claim 1, wherein the data acquisition circuit comprises a main control board (DSP), a signal generator (DDS), an Amplifier (AGC), a sampler (ADC), a resistor (R) and a sample capacitor (C), the main control board (DSP) is connected with one end of the signal generator (DDS), the Amplifier (AGC), the resistor (R) and the one end of the sample capacitor (C) in series in sequence, one end of the sample capacitor (C) is connected with the main control board (DSP) through the sampler (ADC), and the other end of the sample capacitor (C) is grounded.
3. The BP neural network-based high-precision capacitance measurement method according to claim 2, wherein the process of obtaining the voltage data across the sample capacitance (C) is as follows:
the main control board (DSP) starts a signal generator (DDS) to output a sine test signal with the frequency range of 1MHz-8MHz in a linear mode, wherein the step is 100kHz, the total number of frequency points is 71, and the signal maintaining time of each frequency point is 500 ms;
the Amplifier (AGC) carries out amplitude limiting amplification output on the sinusoidal test signal, so that the signal output amplitude of all frequency points is kept unchanged at 1V;
and performing effective value detection and analog-to-digital conversion on the voltage amplitudes at the two ends of the sample capacitor (C) through the sampler (ADC), and storing the obtained signal amplitudes as the characteristic values of the data samples.
4. The BP neural network-based high-precision capacitance measuring method according to claim 1, wherein the cleaning process is: respectively measuring the voltages of 50 sample capacitors (C) 6 times, wherein for the same sample capacitor (C), if the voltage amplitude of a certain measurement is smaller than half of the average value of other five measurements or larger than 1.5 times of the average value of other five measurements, the same sample capacitor (C) is considered as an abnormal data point, and replacing the abnormal data point with the average value of other five measurements;
the denoising treatment adopts a median filter method;
after removing the abnormal data points and filtering, the average of 6 measurements is taken as the new eigenvalue of the sample.
5. The method of claim 1, wherein 71 signal voltage amplitudes with frequency values greater than 1MHz are used as the characteristics of the sample during the characteristic selection.
6. The high-precision capacitance measurement method based on the BP neural network as claimed in claim 1, wherein the BP neural network is a 5-layer network structure of 71 x 16 x 32 x 16 x 1.
7. The method of claim 1The high-precision capacitance measuring method based on the BP neural network is characterized by comprising the steps of determining a coefficient R through a mean square error MSE2These two indices evaluate the performance of the test model.
8. The high-precision capacitance measurement method based on the BP neural network as claimed in claim 1, wherein the training parameters of the BP neural network comprise: the data set partition ratio is train: evaluation: test is 70:15:15, the training function is Levenberg-Marquardt, the evaluation function is MSE, the maximum iteration number Epoch is 1000, the learning target is 1.0e-7, the learning rate is 0.001, the minimum Validation failure number validationchecks is 6, the hidden layer function is tandig, and the output layer function is purelin.
CN202210108623.5A 2022-01-28 2022-01-28 High-precision capacitance measurement method based on BP neural network Pending CN114441859A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210108623.5A CN114441859A (en) 2022-01-28 2022-01-28 High-precision capacitance measurement method based on BP neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210108623.5A CN114441859A (en) 2022-01-28 2022-01-28 High-precision capacitance measurement method based on BP neural network

Publications (1)

Publication Number Publication Date
CN114441859A true CN114441859A (en) 2022-05-06

Family

ID=81371453

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210108623.5A Pending CN114441859A (en) 2022-01-28 2022-01-28 High-precision capacitance measurement method based on BP neural network

Country Status (1)

Country Link
CN (1) CN114441859A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116976754A (en) * 2023-09-19 2023-10-31 北京柏瑞安电子技术有限公司 High-precision capacitance measurement method

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116976754A (en) * 2023-09-19 2023-10-31 北京柏瑞安电子技术有限公司 High-precision capacitance measurement method
CN116976754B (en) * 2023-09-19 2023-12-26 北京柏瑞安电子技术有限公司 High-precision capacitance measurement method

Similar Documents

Publication Publication Date Title
CN104090214B (en) A kind of Cable fault examination and aging analysis method
CN110161421B (en) Method for reconstructing battery impedance in set frequency range on line
CN111474510B (en) Error evaluation method and system for voltage transformer with non-stable output
Popkirov et al. Optimization of the perturbation signal for electrochemical impedance spectroscopy in the time domain
CN113484603B (en) Electric energy metering system and method
CN112540316B (en) Complicated battery impedance spectrum analysis method
CN114441859A (en) High-precision capacitance measurement method based on BP neural network
CN102043084B (en) Method for detecting resistive current of lightning arrester
Hoja et al. Portable analyzer for impedance spectroscopy
CN113100740A (en) Abdominal component analyzer and analysis method thereof
CN117150386B (en) Assessment method and device for measurement uncertainty of humidity sensor based on self-adaption
CN111896849B (en) Oil paper insulation performance analysis method and device considering non-uniform medium
CN105974278B (en) Oil clearance telegram in reply holds accelerated test method under low frequency mixed excitation based on Sine-Fitting
CN116256566A (en) Evaluation method for dielectric loss factor of power equipment under action of high voltage of oscillating wave
CN116359824A (en) Electric measurement correction method based on frequency domain
US7834641B1 (en) Phase-gain calibration of impedance/admittance meter
CN112904088B (en) Dielectric loss tangent calculation method in broadband dielectric loss detection device
CN107800434A (en) A kind of fast A/D C method of testings based on parameter extraction
CN113064013A (en) Lightning arrester state monitoring method, device and system
CN111398735A (en) Transformer substation grounding grid fault detection method based on information entropy
CN112957028B (en) Comprehensive abdominal health detection analyzer and analysis method thereof
CN110927504A (en) Analysis method, device and equipment of combined network based on LCR
Minda et al. Accurate frequency estimation using DFT and artificial neural networks
Ramos How signal processing is changing impedance spectroscopy
Baccigalupi et al. Least square procedures to improve the results of the three-parameter sine fitting algorithm

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination