CN115993567B - Calibration method and system for bidirectional feedback power supply data - Google Patents

Calibration method and system for bidirectional feedback power supply data Download PDF

Info

Publication number
CN115993567B
CN115993567B CN202310279409.0A CN202310279409A CN115993567B CN 115993567 B CN115993567 B CN 115993567B CN 202310279409 A CN202310279409 A CN 202310279409A CN 115993567 B CN115993567 B CN 115993567B
Authority
CN
China
Prior art keywords
calibration
trend
point
value
period
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.)
Active
Application number
CN202310279409.0A
Other languages
Chinese (zh)
Other versions
CN115993567A (en
Inventor
马磊
陆财宏
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.)
Shenzhen Beihan Technology Co ltd
Original Assignee
Shenzhen Beihan Technology Co ltd
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 Shenzhen Beihan Technology Co ltd filed Critical Shenzhen Beihan Technology Co ltd
Priority to CN202310279409.0A priority Critical patent/CN115993567B/en
Publication of CN115993567A publication Critical patent/CN115993567A/en
Application granted granted Critical
Publication of CN115993567B publication Critical patent/CN115993567B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Indication And Recording Devices For Special Purposes And Tariff Metering Devices (AREA)

Abstract

The invention relates to the technical field of data processing, in particular to a method and a system for calibrating bidirectional feedback power supply data, comprising the following steps: setting a calibration point, calibrating the calibration point, obtaining a calibration difference of the calibration point, forming a calibration difference sequence, obtaining a calibrated display current sequence, decomposing the calibrated display current sequence by utilizing different window sizes, further obtaining the window property of each window size, further decomposing a residual error item corresponding to the optimal window, obtaining a fourth trend item according to the result of the two decomposition, performing polynomial fitting on the fourth trend item to obtain a trend curve, obtaining the calibration difference of the interpolation point according to the trend value of the calibration point and the calibration difference, and further calibrating the interpolation point according to the calibration difference of the interpolation point. The calibration result of the invention is more accurate.

Description

Calibration method and system for bidirectional feedback power supply data
Technical Field
The invention relates to the technical field of data processing, in particular to a method and a system for calibrating bidirectional feedback power supply data.
Background
The existing method usually calibrates the display current of the bidirectional feedback digital power supply through the measurement value of a five-bit half digital multimeter (DMM), and because the calibration points of the full range of the digital power supply are limited, the display current between adjacent calibration points is often calculated through the nearest interpolation method so as to calibrate the full range of the digital power supply.
However, when the conductor is electrified, the conductor generates heat, the longer the electrified time or the larger the current, the more the heat is generated, and meanwhile, the longer the digital power supply is used, the larger the resistance is, the more the heat is generated, so that the error under different current values is different. Therefore, the display current between adjacent calibration points obtained by nearest neighbor interpolation has larger error.
Disclosure of Invention
The invention provides a calibration method and a calibration system for bidirectional feedback power supply data, which are used for solving the existing problems.
The invention relates to a calibration method of bidirectional feedback power supply data, which adopts the following technical scheme:
an embodiment of the present invention provides a method for calibrating bi-directional feedback power data, including the steps of:
setting calibration points, calibrating the calibration points, and obtaining the actual current after calibration and the display current after calibration of each calibration point; taking the absolute value of the difference value between the actual current after calibration and the display current after calibration corresponding to each calibration point as the calibration difference of each calibration point; forming a calibration difference sequence by the calibration differences of all the calibration points;
forming a calibrated display current sequence by the calibrated display currents of all the calibration points; acquiring a plurality of window sizes; STL decomposition is carried out on the calibrated display current sequence by adopting different window sizes respectively to obtain a trend item, a period item and a residual item corresponding to each window size;
acquiring differences between each element in the residual error item corresponding to each window size and the element in the corresponding position in the calibration difference sequence to form a first difference sequence of each window size; acquiring the period of each window size according to the period item corresponding to each window size; acquiring a period of a calibration difference sequence; acquiring the window property of each window size according to the first difference sequence and the period of each window size and the period of the calibration difference sequence; taking the window size with the largest window property as the optimal window;
respectively taking a trend item, a period item and a residual item corresponding to the optimal window as a first trend item, a first period item and a first residual item; STL decomposition is carried out on the first residual error item by adopting the optimal window size, so that a second trend item and a second residual error item are obtained; subtracting the second trend term from the first trend term to obtain a third trend term; subtracting the third trend term from the calibrated display current sequence to obtain an updated trend term; acquiring a third residual error item according to the updated trend removal item; subtracting the second residual term from the third residual term to obtain a residual-removed term, performing Fourier transform on the residual-removed term to obtain a fourth period term, and subtracting the fourth period term from the residual-removed term to obtain a fourth trend term;
performing polynomial fitting on the fourth trend term to obtain a trend curve; acquiring two reference points of an interpolation point, and acquiring a first ratio and a second ratio according to the trend value of the interpolation point on a trend curve and the trend values of the two reference points; and obtaining the calibration difference of the interpolation point according to the calibration difference of the two reference points, the first ratio and the second ratio, and calibrating the interpolation point according to the calibration difference of the interpolation point.
Preferably, the step of obtaining the period of each window size according to the period item corresponding to each window size includes the following specific steps:
and carrying out Fourier transform on the period item corresponding to each window size, converting the period item corresponding to each window size into a frequency domain space, and obtaining the reciprocal of the maximum frequency domain value of the period item corresponding to each window size in the frequency domain space as the period of each window size.
Preferably, the step of obtaining the windowability of each window size according to the first difference sequence, the period of each window size and the period of the calibration difference sequence includes the following specific steps:
Figure SMS_1
wherein the method comprises the steps of
Figure SMS_3
Window properties for the ith window size; />
Figure SMS_6
A period of the ith window size; f is the period of the calibration difference sequence; />
Figure SMS_9
A first difference sequence of an ith window size; />
Figure SMS_4
The smallest value in the first sequence of differences being the ith window size; />
Figure SMS_5
Is the maximum value in the first difference sequence of the ith window size; />
Figure SMS_7
A larger value for the period of the ith window size and the period of the calibration difference sequence; />
Figure SMS_8
For the period of the ith window size and the period of the calibration difference sequence; />
Figure SMS_2
Is an exponential function with a base of natural constant.
Preferably, the obtaining the two reference points of the interpolation point includes the following specific steps:
the reference points comprise a left reference point and a right reference point, the current value of the non-calibration point is taken as an interpolation point, and the nearest calibration point at the left side of the interpolation point on the trend curve is taken as the left reference point of the interpolation point; and taking the nearest calibration point on the right side of the interpolation point on the trend curve as the right reference point of the interpolation point.
Preferably, the first ratio and the second ratio are obtained according to the trend value of the interpolation point on the trend curve and the trend values of the two reference points, and the specific steps are as follows:
taking the trend value of the interpolation point on the trend curve as a fitting value of the interpolation point;
when the fitting value of the interpolation point is larger than or equal to the trend value of the left reference point on the trend curve, the first ratio is obtained by utilizing the trend value of the left reference point on the trend curve to the fitting value of the interpolation point; when the fitting value of the interpolation point is smaller than or equal to the trend value of the left reference point on the trend curve, the fitting value of the interpolation point is utilized to be compared with the trend value of the left reference point on the trend curve, so that a first ratio is obtained;
when the fitting value of the interpolation point is larger than or equal to the trend value of the right reference point on the trend curve, a second ratio is obtained by utilizing the trend value of the right reference point on the trend curve to the fitting value of the interpolation point; and when the fitting value of the interpolation point is smaller than or equal to the trend value of the right reference point on the trend curve, obtaining a second ratio by utilizing the fitting value of the interpolation point to the trend value of the right reference point on the trend curve.
Preferably, the obtaining the calibration difference of the interpolation point according to the calibration difference of the two reference points, the first ratio and the second ratio includes the following specific steps:
normalizing the first ratio and the second ratio, and taking the normalized first ratio and second ratio as trend weights of the left reference point and the right reference point respectively; taking the reciprocal of the absolute value of the difference between the interpolation point and the left reference point as a first reciprocal and taking the reciprocal of the absolute value of the difference between the interpolation point and the right reference point as a second reciprocal; normalizing the first reciprocal and the second reciprocal, and taking the normalized first reciprocal and the normalized second reciprocal as the distance weights of the left reference point and the right reference point respectively;
taking the average value of the trend weight and the distance weight of the left reference point as the overall weight of the left reference point, and taking the average value of the trend weight and the distance weight of the right reference point as the overall weight of the right reference point; and weighting and summing the calibration differences of the left reference point and the right reference point by using the overall weight of the left reference point and the overall weight of the right reference point to obtain the calibration difference of the interpolation point.
The embodiment of the invention provides a calibration system for bidirectional feedback power supply data, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes any one step of a calibration method based on the bidirectional feedback power supply data when executing the computer program.
The technical scheme of the invention has the beneficial effects that: the bidirectional feedback digital power supply has limited calibration points in the whole range, and interpolation operation is carried out between adjacent calibration points in the prior method to obtain the display current between the adjacent calibration points. However, the difference between the actual current after calibration and the display current after calibration corresponding to different current values is different, and the current value without actual test is subjected to corresponding display current after calibration by the nearest neighbor interpolation method in the prior method, so that larger error exists. According to the embodiment of the invention, the error of a few calibration points is calculated to obtain the calibration difference sequence, the variation trend of the error is obtained according to the calibration difference sequence, the calibration difference of a plurality of interpolation points is obtained, the current value with larger calibration difference is calibrated, a better integral calibration result can be obtained, and the reliability of the current display value of the bidirectional feedback digital power supply after calibration is improved. The existing nearest neighbor difference method only considers the distance from the interpolation point to the left reference point and the right reference point, and the accuracy of the interpolation result is not high. According to the method, the trend weight is obtained according to the fitting value of the difference point and the trend values of the left reference point and the right reference point on the trend curve, meanwhile, the interpolation result is obtained by combining the trend weight and the distance from the interpolation point to the left reference point and the right reference point, the obtained result is more accurate, and the reliability of the calibrated bidirectional feedback digital power supply current display value is further improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart showing the steps of a method for calibrating bi-directional feedback power data according to the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description refers to specific embodiments, structures, features and effects of a calibration method for bi-directional feedback power data according to the present invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of a calibration method for bidirectional feedback power supply data provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for calibrating bi-directional feedback power data according to an embodiment of the invention is shown, the method includes the following steps:
s001, setting a calibration point, and calibrating the current of the calibration point.
And setting a plurality of uniformly distributed calibration points in the full range of the bidirectional feedback digital power supply, wherein each calibration point corresponds to one current value.
And connecting a resistor with the DMM in series, and accessing a bidirectional feedback digital power supply. And setting the current of the bidirectional feedback digital power supply as a current value corresponding to the calibration point, and starting operation. The measured current of the DMM is recorded as the actual current before calibration, and the display current of the bi-directional feedback digital power supply is recorded as the display current before calibration. For example, when the calibration point is 5A, the current of the bidirectional feedback digital power supply is set to 5A, the measured current of the dmm is 5.1045A, the actual current before calibration is 5.1045A, the display current of the bidirectional feedback digital power supply is 5.0004A, and the display current before calibration is 5.0004A. The error at this time was 2%. Where a is the unit of current, representing amperes.
The measurement current (namely actual current before calibration) of the DMM is input into a bidirectional feedback digital power supply through a touch screen, and the bidirectional feedback digital power supply adjusts an output value through closed-loop control, so that the calibration of the current is realized. And recording the measured current of the calibrated DMM as the actual current after calibration, and recording the display current of the bidirectional feedback digital power supply after calibration as the display current after calibration. For example, when the actual current 5.1045A before calibration is input to the bi-directional feedback digital power supply, the measured current of the DMM is 5.0008A, the actual current after calibration is 5.0008A, the display current of the bi-directional feedback digital power supply is 5.0007a, and the display current after calibration is 5.0007a. At the moment, the error is 0.002%, the error is reduced, and the current value output by the bidirectional feedback digital power supply is more accurate.
Thus, the calibration of the calibration point current is completed.
S002, obtaining a calibration difference sequence.
It should be noted that, the calibration points of the full range of the bidirectional feedback digital power supply are limited, so that interpolation operation is performed between adjacent calibration points in the existing method to obtain the display current between the adjacent calibration points. However, the difference between the actual current after calibration and the display current after calibration corresponding to different current values is different, and the current value without actual test is subjected to corresponding display current after calibration by the nearest neighbor interpolation method in the prior method, so that larger error exists. Therefore, the embodiment of the invention calculates the errors of a few calibration points to obtain the calibration difference sequence, and then obtains the variation trend of the errors according to the calibration difference sequence, so as to obtain the prediction errors of a plurality of interpolation points, calibrate the current value with larger prediction errors, obtain a better integral calibration result, and improve the reliability of the current display value of the bidirectional feedback digital power supply after calibration.
In the embodiment of the invention, the specific steps for acquiring the calibration difference sequence are as follows:
and acquiring the actual current after calibration and the display current after calibration corresponding to each calibration point, calculating the absolute value of the difference value between the actual current after calibration and the display current after calibration corresponding to each calibration point, and taking the absolute value of the difference value as the calibration difference of each calibration point. And arranging the calibration differences of all the calibration points according to the sequence from small to large of the calibration points to obtain a calibration difference sequence.
Thus, a calibration difference sequence is obtained.
S003, obtaining an optimal window of STL decomposition.
It should be noted that, the current in the bidirectional feedback digital power supply is a sine wave, that is, the periodicity of the current sequence is strong, and the STL decomposition algorithm is a method for decomposing the sequence with the strong periodicity of the error and calculating the data trend, so the embodiment of the invention calculates the current variation trend value through the idea of STL decomposition. The STL decomposition algorithm can decompose an original sequence into a trend term, a period term and a residual, specifically, the trend term is calculated firstly through a moving average method, then the trend term is subtracted by the original sequence to obtain a trending term, the trending term is subjected to Fourier transformation to obtain the period term, and finally the period term is subtracted by the trending term to obtain the residual term.
It should be further noted that, the purpose of the embodiment of the present invention is to obtain the variation trend of the error according to the calibration difference sequence, further obtain the prediction errors of a plurality of interpolation points, and calibrate the current value with larger prediction error. If the calibrated bidirectional feedback digital power supply is regarded as a system
Figure SMS_10
The DMM instrument at calibration is considered as a system +.>
Figure SMS_13
Since the calibration difference sequence is obtained by taking the difference between the current values of the two systems, if the calculation of the trend term is directly performed on the calibration difference sequence, on the one hand, the residual term is difficult to obtain, and on the other hand, the final purpose is to +.>
Figure SMS_15
Performing calibration, wherein the system +.>
Figure SMS_11
The partial information in (3) results in a large error in the calculation of trend terms directly on the calibration difference sequence. Thus, the system is ++>
Figure SMS_14
Approximate simulation of the calibration difference sequence, taking into account that the display current after calibration is independent of the system
Figure SMS_16
Thus, by analyzing the calibrated display current, a system independent of +.>
Figure SMS_17
Residual terms of the display current of (2) to exclude the system +.>
Figure SMS_12
Is a function of (a) and (b).
The calibration difference sequence is equivalent to the residual terms of the calibrated display current, and the change rule of the trend terms of the calibration difference sequence is not known, namely, the trend terms can be obtained according to the moving average calculation of which window size is not known when the STL is decomposed.
In the embodiment of the invention, the display currents after calibration of all the calibration points are arranged according to the sequence from the small calibration points to the large calibration points, so as to obtain a display current sequence after calibration. The number of calibration points is denoted as n, and the length of the current sequence after calibration is denoted as n.
Will be
Figure SMS_18
Respectively, as a window size. Wherein->
Figure SMS_19
For a first preset threshold,/a>
Figure SMS_20
For a second preset threshold value, in the present embodiment +.>
Figure SMS_21
=10,/>
Figure SMS_22
In other embodiments, the practitioner can set as desired
Figure SMS_23
Is a value of (2).
STL decomposition is carried out on the calibrated display current sequence, and in the decomposition process, trend items, period items and residual items are respectively obtained by adopting different window sizes. It should be noted that the trend term, the period term, and the residual term are all sequences, and the sequence length is the same as the length of the calibrated display current sequence, and is n. Each window size gets a trend term, a period term, and a residual term.
It should be further noted that, residual terms corresponding to different window sizes have different degrees of approximation with the calibration difference sequence, when the difference between the residual terms and the calibration difference sequence is smaller, that is, the closer the residual terms are to the calibration difference sequence, the better the corresponding window sizes are; because the periodicity of the current value is strong, the periodicity of the calibrated difference sequence is similar to the periodicity of the calibrated display current sequence, and therefore, the closer the period length of the calibrated display current sequence is to the period length of the calibrated difference sequence, the better the corresponding window size is.
In the embodiment of the invention, the difference between each element in the residual error item corresponding to each window size and the element at the corresponding position in the calibration difference sequence is obtained to form a first difference sequence of each window size.
And carrying out Fourier transform on the period item corresponding to each window size, converting the period item corresponding to each window size into a frequency domain space, and obtaining the reciprocal of the maximum frequency domain value of the period item corresponding to each window size in the frequency domain space as the period of each window size.
And similarly, performing Fourier transformation on the calibration difference sequence, converting the calibration difference sequence into a frequency domain space, and obtaining the reciprocal of the maximum frequency domain value of the calibration difference sequence in the frequency domain space as the period of the calibration difference sequence.
Calculating windowability of each window size, e.g. windowability of ith window size
Figure SMS_24
The method comprises the following steps: />
Figure SMS_25
Wherein the method comprises the steps of
Figure SMS_27
Window properties for the ith window size; />
Figure SMS_29
A period of the ith window size; f is the period of the calibration difference sequence; />
Figure SMS_32
A first difference sequence of an ith window size; />
Figure SMS_26
The smallest value in the first sequence of differences being the ith window size; />
Figure SMS_31
Is the maximum value in the first difference sequence of the ith window size; />
Figure SMS_34
A larger value for the period of the ith window size and the period of the calibration difference sequence; />
Figure SMS_35
For the period of the ith window size and the period of the calibration difference sequence; />
Figure SMS_28
Is an exponential function with a natural constant as a base; when the maximum value and the minimum value in the first difference sequence of the ith window size are closer to 1, namely the difference value between the ratio and 1 is smaller, the trend difference between the residual item corresponding to the ith window size and the calibration difference sequence is smaller; when the minimum value in the first difference sequence of the ith window size is smaller, the difference between the residual error item corresponding to the ith window size and the calibration difference sequence is smaller; when->
Figure SMS_30
The larger the ratio is, the smaller the period difference between the residual error item corresponding to the size of the ith window and the calibration difference sequence is; thus, when the maximum value and the minimum value in the first difference sequence of the ith window size are closer to 1, the minimum value in the first difference sequence of the ith window size is smaller, and +.>
Figure SMS_33
The greater the ratioAnd when the residual error item corresponding to the ith window size is more similar to the calibration difference sequence, the window property of the ith window size is larger.
The window size with the largest windowability is taken as the optimal window.
So far, the optimal window is obtained.
S004, obtaining a trend curve of the calibration difference sequence.
And taking the trend item, the period item and the residual item corresponding to the optimal window as a first trend item, a first period item and a first residual item respectively. And carrying out STL decomposition on the first residual error item, and acquiring a trend item, a period item and a residual error item of the first residual error item by adopting the optimal window size in the decomposition process, wherein the trend item, the period item and the residual error item are respectively used as a second trend item, a second period item and a second residual error item.
It should be noted that, due to the calculation sequence of the STL, the trend information of the first residual term is originally included in the first trend term, so that the first residual term has no trend information, and thus, the change rule of the calibration difference sequence cannot be accurately represented. Therefore, the embodiment of the invention subtracts the second trend term from the first trend term, thereby obtaining the trend term which does not contain the variation trend of the first residual term. The method and the device for correcting the variation trend of the difference sequence aim to obtain the variation trend of the accurate first residual error item, update the first residual error item according to the trend item which does not contain the variation trend of the first residual error item to obtain a third residual error item, subtract the second residual error item and the corresponding period item from the third residual error item, remove the characteristics of the second residual error item and the corresponding period item, and obtain the accurate trend item which is high in accuracy and reliability and capable of accurately representing the variation rule of the difference sequence.
In the embodiment of the invention, the second trend term is subtracted from the first trend term to obtain the third trend term. And subtracting the third trend term from the calibrated display current sequence to obtain an updated trend term. And carrying out Fourier transformation on the updated trending item to obtain a third period item, and finally subtracting the third period item from the updated trending item to obtain a third residual item. It should be noted that, performing fourier transform on the updated trending term to obtain the period term is the prior art in the STL decomposition algorithm, and will not be described in detail herein.
Subtracting the second residual term from the third residual term to obtain a residual-removed term, performing Fourier transform on the residual-removed term to obtain a fourth period term, and subtracting the fourth period term from the residual-removed term to obtain a fourth trend term. The fourth trend term is the best trend term for calibrating the difference sequence.
And performing polynomial fitting on the fourth trend term by using a least square method, and taking the fitted curve as a trend curve of the calibration difference sequence. In the embodiment of the present invention, the degree of the polynomial is 5, and in other embodiments, the practitioner may set the degree of the polynomial according to the actual implementation situation.
Thus, a trend curve of the calibration difference sequence is obtained.
S005, obtaining a current value with larger calibration difference, and further calibrating.
According to the trend curve of the calibration difference sequence, the calibration difference of the current value of the non-calibration point is obtained, specifically:
and taking the current value of the non-calibrated point as an interpolation point, and taking the trend value corresponding to the interpolation point on the trend curve as a fitting value of the interpolation point. And acquiring the nearest calibration point on the left side of the interpolation point on the trend curve as a left reference point of the interpolation point, and acquiring the ratio of the fitting value of the interpolation point to the trend value of the left reference point on the trend curve as a first ratio. It should be noted that, the method for obtaining the ratio of the fitting value of the interpolation point to the trend value of the left reference point on the trend curve is as follows: when the fitting value of the interpolation point is larger than or equal to the trend value of the left reference point on the trend curve, the trend value of the left reference point on the trend curve is utilized to be compared with the fitting value of the interpolation point; when the fitting value of the interpolation point is smaller than or equal to the trend value of the left reference point on the trend curve, the fitting value of the interpolation point is utilized to be compared with the trend value of the left reference point on the trend curve.
And similarly, the nearest calibration point on the right side of the interpolation point on the trend curve of the calibration difference sequence is obtained and used as the right reference point of the interpolation point, and the ratio of the fitting value of the interpolation point to the trend value of the right reference point on the trend curve of the calibration difference sequence is obtained and used as the second ratio.
Normalizing the first ratio and the second ratio, and taking the normalized first ratio and second ratio as trend weights of the left reference point and the right reference point respectively.
When the calibration difference of the interpolation point is obtained by the existing nearest neighbor difference method, weights are set according to the distances from the interpolation point to the left reference point and the right reference point, and the calibration differences of the left reference point and the right reference point are weighted and summed to obtain the calibration difference of the interpolation point. However, the method only considers the distance from the interpolation point to the left reference point and the right reference point, and the accuracy of the interpolation result is not high. Therefore, the embodiment of the invention combines the trend weights of the left reference point and the right reference point to obtain the interpolation result.
In the embodiment of the invention, the reciprocal of the absolute value of the difference between the interpolation point and the left reference point is taken as a first reciprocal, and the reciprocal of the absolute value of the difference between the interpolation point and the right reference point is taken as a second reciprocal. And normalizing the first reciprocal and the second reciprocal, and taking the normalized first reciprocal and the normalized second reciprocal as the distance weights of the left reference point and the right reference point respectively.
Taking the average value of the trend weight and the distance weight of the left reference point as the integral weight of the left reference point, taking the average value of the trend weight and the distance weight of the right reference point as the integral weight of the right reference point, and carrying out weighted summation on the calibration difference of the left reference point and the right reference point by utilizing the integral weight of the left reference point and the integral weight of the right reference point to obtain the calibration difference of the interpolation point.
For example: when the calibration differences corresponding to the left reference point and the right reference point of the interpolation point 7A are respectively 5A and 10A, and the calibration differences corresponding to the 5A and 10A are respectively 0.5 and 0.6, the distance weights of the left reference point and the right reference point are respectively
Figure SMS_36
、/>
Figure SMS_37
The calibration difference of interpolation point 7A obtained by the traditional nearest neighbor interpolation method is that:/>
Figure SMS_38
. Assuming that the trend weights of the left reference point and the right reference point are 0.7 and 0.3 respectively, the overall weights of the left reference point and the right reference point are +.>
Figure SMS_39
、/>
Figure SMS_40
. The calibration difference of the interpolation point 7A obtained by combining the overall weights of the left reference point and the right reference point is as follows:
Figure SMS_41
thus, the calibration difference of the interpolation points is obtained.
Dividing the calibration difference of the interpolation point by the current value corresponding to the interpolation point to obtain the calibration error of the interpolation point, and when the calibration error is larger than a third preset threshold value
Figure SMS_42
In this case, it is explained that the current value corresponding to the interpolation point has a large error after calibration according to the calibration point, and the current value corresponding to the interpolation point is calibrated by the method in step S001. In the embodiment of the present invention, a third preset threshold +.>
Figure SMS_43
In other embodiments, the practitioner may be set according to the actual implementation.
Thus, the calibration of the bidirectional feedback digital power supply is completed.
It should be noted that, the conventional calibration method has fewer calibration points, and the error obtained by directly calibrating other non-calibration points according to the calibration points by using the nearest neighbor interpolation method is larger. According to the embodiment of the invention, the calibration difference of other non-calibration points is obtained according to the calibration difference of the calibration points, and the non-calibration points with larger calibration difference are calibrated again, so that the calibration difference of all final current values is smaller, and the whole calibration result of the bidirectional feedback digital power supply is more accurate.
The embodiment of the invention provides a calibration system for bidirectional feedback power supply data, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes any one step of a calibration method based on the bidirectional feedback power supply data when executing the computer program.
According to the embodiment of the invention, the error of a few calibration points is calculated to obtain a calibration difference sequence, the variation trend of the error is obtained according to the calibration difference sequence, the calibration difference of a plurality of interpolation points is obtained, the current value with larger calibration difference is calibrated, and compared with the current value which is not actually tested and has larger error after corresponding calibration through the nearest neighbor interpolation method in the prior art, the embodiment of the invention can obtain a better overall calibration result, and the reliability of the current display value of the bidirectional feedback digital power supply after calibration is improved; the existing nearest neighbor difference method only considers the distance from the interpolation point to the left reference point and the right reference point, the accuracy of the interpolation result is not high, the embodiment of the invention obtains the trend weight according to the fitting value of the difference point and the trend values of the left reference point and the right reference point on the trend curve, and simultaneously obtains the interpolation result by combining the trend weight and the distance from the interpolation point to the left reference point and the right reference point, the obtained result is more accurate, and the reliability of the current display value of the bidirectional feedback digital power supply after calibration is further improved.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.

Claims (7)

1. A method for calibrating bi-directional feedback power supply data, the method comprising the steps of:
setting calibration points, calibrating the calibration points, and obtaining the actual current after calibration and the display current after calibration of each calibration point; taking the absolute value of the difference value between the actual current after calibration and the display current after calibration corresponding to each calibration point as the calibration difference of each calibration point; forming a calibration difference sequence by the calibration differences of all the calibration points;
forming a calibrated display current sequence by the calibrated display currents of all the calibration points; acquiring a plurality of window sizes; STL decomposition is carried out on the calibrated display current sequence by adopting different window sizes respectively to obtain a trend item, a period item and a residual item corresponding to each window size;
acquiring differences between each element in the residual error item corresponding to each window size and the element in the corresponding position in the calibration difference sequence to form a first difference sequence of each window size; acquiring the period of each window size according to the period item corresponding to each window size; acquiring a period of a calibration difference sequence; acquiring the window property of each window size according to the first difference sequence and the period of each window size and the period of the calibration difference sequence; taking the window size with the largest window property as the optimal window;
respectively taking a trend item, a period item and a residual item corresponding to the optimal window as a first trend item, a first period item and a first residual item; STL decomposition is carried out on the first residual error item by adopting the optimal window size, so that a second trend item and a second residual error item are obtained; subtracting the second trend term from the first trend term to obtain a third trend term; subtracting the third trend term from the calibrated display current sequence to obtain an updated trend term; acquiring a third residual error item according to the updated trend removal item; subtracting the second residual term from the third residual term to obtain a residual-removed term, performing Fourier transform on the residual-removed term to obtain a fourth period term, and subtracting the fourth period term from the residual-removed term to obtain a fourth trend term;
performing polynomial fitting on the fourth trend term to obtain a trend curve; acquiring two reference points of an interpolation point, and acquiring a first ratio and a second ratio according to the trend value of the interpolation point on a trend curve and the trend values of the two reference points; and obtaining the calibration difference of the interpolation point according to the calibration difference of the two reference points, the first ratio and the second ratio, and calibrating the interpolation point according to the calibration difference of the interpolation point.
2. The method for calibrating bi-directional feedback power data according to claim 1, wherein the step of obtaining the period of each window according to the period item corresponding to each window comprises the following specific steps:
and carrying out Fourier transform on the period item corresponding to each window size, converting the period item corresponding to each window size into a frequency domain space, and obtaining the reciprocal of the maximum frequency domain value of the period item corresponding to each window size in the frequency domain space as the period of each window size.
3. The method for calibrating bi-directional feedback power data according to claim 1, wherein the step of obtaining the windowability of each window according to the first difference sequence, the period of each window and the period of the calibration difference sequence comprises the following specific steps:
Figure QLYQS_1
wherein the method comprises the steps of
Figure QLYQS_3
Window properties for the ith window size; />
Figure QLYQS_5
A period of the ith window size; f is the period of the calibration difference sequence; />
Figure QLYQS_7
A first difference sequence of an ith window size; />
Figure QLYQS_4
The smallest value in the first sequence of differences being the ith window size; />
Figure QLYQS_6
Is the maximum value in the first difference sequence of the ith window size; />
Figure QLYQS_8
A larger value for the period of the ith window size and the period of the calibration difference sequence; />
Figure QLYQS_9
For the period of the ith window size and the period of the calibration difference sequence; />
Figure QLYQS_2
Is an exponential function with a base of natural constant.
4. The method for calibrating bi-directional feedback power data according to claim 1, wherein the step of obtaining two reference points of the interpolation point comprises the following specific steps:
the reference points comprise a left reference point and a right reference point, the current value of the non-calibration point is taken as an interpolation point, and the nearest calibration point at the left side of the interpolation point on the trend curve is taken as the left reference point of the interpolation point; and taking the nearest calibration point on the right side of the interpolation point on the trend curve as the right reference point of the interpolation point.
5. The method for calibrating bi-directional feedback power data according to claim 1, wherein the step of obtaining the first ratio and the second ratio according to the trend values of the interpolation points on the trend curve and the trend values of the two reference points comprises the following specific steps:
taking the trend value of the interpolation point on the trend curve as a fitting value of the interpolation point;
when the fitting value of the interpolation point is larger than or equal to the trend value of the left reference point on the trend curve, the first ratio is obtained by utilizing the trend value of the left reference point on the trend curve to the fitting value of the interpolation point; when the fitting value of the interpolation point is smaller than or equal to the trend value of the left reference point on the trend curve, the fitting value of the interpolation point is utilized to be compared with the trend value of the left reference point on the trend curve, so that a first ratio is obtained;
when the fitting value of the interpolation point is larger than or equal to the trend value of the right reference point on the trend curve, a second ratio is obtained by utilizing the trend value of the right reference point on the trend curve to the fitting value of the interpolation point; and when the fitting value of the interpolation point is smaller than or equal to the trend value of the right reference point on the trend curve, obtaining a second ratio by utilizing the fitting value of the interpolation point to the trend value of the right reference point on the trend curve.
6. The method for calibrating bi-directional feedback power data according to claim 1, wherein the step of obtaining the calibration difference of the interpolation point according to the calibration difference of the two reference points, the first ratio and the second ratio comprises the following specific steps:
normalizing the first ratio and the second ratio, and taking the normalized first ratio and second ratio as trend weights of the left reference point and the right reference point respectively; taking the reciprocal of the absolute value of the difference between the interpolation point and the left reference point as a first reciprocal and taking the reciprocal of the absolute value of the difference between the interpolation point and the right reference point as a second reciprocal; normalizing the first reciprocal and the second reciprocal, and taking the normalized first reciprocal and the normalized second reciprocal as the distance weights of the left reference point and the right reference point respectively;
taking the average value of the trend weight and the distance weight of the left reference point as the overall weight of the left reference point, and taking the average value of the trend weight and the distance weight of the right reference point as the overall weight of the right reference point; and weighting and summing the calibration differences of the left reference point and the right reference point by using the overall weight of the left reference point and the overall weight of the right reference point to obtain the calibration difference of the interpolation point.
7. A calibration system for bi-directional feedback power supply data, comprising a memory, a processor and a computer program stored in said memory and running on said processor, characterized in that said processor, when executing said computer program, carries out the steps of a method for calibrating bi-directional feedback power supply data according to any of claims 1-6.
CN202310279409.0A 2023-03-22 2023-03-22 Calibration method and system for bidirectional feedback power supply data Active CN115993567B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310279409.0A CN115993567B (en) 2023-03-22 2023-03-22 Calibration method and system for bidirectional feedback power supply data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310279409.0A CN115993567B (en) 2023-03-22 2023-03-22 Calibration method and system for bidirectional feedback power supply data

Publications (2)

Publication Number Publication Date
CN115993567A CN115993567A (en) 2023-04-21
CN115993567B true CN115993567B (en) 2023-05-19

Family

ID=85992321

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310279409.0A Active CN115993567B (en) 2023-03-22 2023-03-22 Calibration method and system for bidirectional feedback power supply data

Country Status (1)

Country Link
CN (1) CN115993567B (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3584186A (en) * 1969-04-25 1971-06-08 Chemetron Corp Direct current power supply with adjustable inductance control
CN101765243A (en) * 2009-12-16 2010-06-30 中兴通讯股份有限公司 CDMA and GSM bimodule digital mobile communication terminal
CN103218475A (en) * 2013-03-14 2013-07-24 天津大学 In-process evaluation based complex spatial surface error feedback compensating method
CN103454497A (en) * 2013-09-10 2013-12-18 南京理工大学 Phase difference measuring method based on improved windowing discrete Fourier transform
CN106679777A (en) * 2017-01-04 2017-05-17 承德石油高等专科学校 Electronic scale temperature compensation device and compensation method thereof
CN108802654A (en) * 2018-07-06 2018-11-13 山东大学 A kind of the automatic Calibration acquisition system and method for forming and capacity dividing test power supply
CN110967661A (en) * 2019-12-20 2020-04-07 宁夏凯晨电气集团有限公司 Electrical data calibration method based on curve fitting
CN112036041A (en) * 2020-09-02 2020-12-04 河北工业大学 STL model permeation error comprehensive compensation method in 3DP process
CN112833913A (en) * 2020-12-31 2021-05-25 福建星海通信科技有限公司 Inertial navigation system and correction method thereof

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3584186A (en) * 1969-04-25 1971-06-08 Chemetron Corp Direct current power supply with adjustable inductance control
CN101765243A (en) * 2009-12-16 2010-06-30 中兴通讯股份有限公司 CDMA and GSM bimodule digital mobile communication terminal
CN103218475A (en) * 2013-03-14 2013-07-24 天津大学 In-process evaluation based complex spatial surface error feedback compensating method
CN103454497A (en) * 2013-09-10 2013-12-18 南京理工大学 Phase difference measuring method based on improved windowing discrete Fourier transform
CN106679777A (en) * 2017-01-04 2017-05-17 承德石油高等专科学校 Electronic scale temperature compensation device and compensation method thereof
CN108802654A (en) * 2018-07-06 2018-11-13 山东大学 A kind of the automatic Calibration acquisition system and method for forming and capacity dividing test power supply
CN110967661A (en) * 2019-12-20 2020-04-07 宁夏凯晨电气集团有限公司 Electrical data calibration method based on curve fitting
CN112036041A (en) * 2020-09-02 2020-12-04 河北工业大学 STL model permeation error comprehensive compensation method in 3DP process
CN112833913A (en) * 2020-12-31 2021-05-25 福建星海通信科技有限公司 Inertial navigation system and correction method thereof

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
传感器数据补偿算法的设计与实现;罗潇;中国优秀硕士学位论文全文数据库信息科技辑(第7期);I140-172 *
基于DSP全数字开关电源及功率因数校正器的设计;葛加林;中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑(第3期);C042-577 *

Also Published As

Publication number Publication date
CN115993567A (en) 2023-04-21

Similar Documents

Publication Publication Date Title
CN104716930B (en) System and method for the dutycycle adjustment with error resiliency
EP1655841A2 (en) Calibration system and method for a linearity corrector using filter products
CN102163972A (en) Correction of non-linearities in adcs
JP2011228815A (en) Frequency characteristic correction method for arbitrary waveform generator
CN110752844B (en) Time interleaved digital-to-analog converter correction
CN112414997A (en) ICP-AES overlapping spectral line separation method based on RBF neural network
US11329633B2 (en) Joint optimization of FIR filters in a non-linear compensator
CN115993567B (en) Calibration method and system for bidirectional feedback power supply data
Bluemm et al. Correcting nonlinearity and temperature influence of sensors through B-spline modeling
CN114280526B (en) Digital differential traceability system and method for electronic transformer calibrator
JP2014071115A (en) Temperature correction method and temperature correction device for received signal strength indicator
CN108650768B (en) Filament current control method and device
CN116994979B (en) Square resistance measuring method, electronic device and storage medium
CN116318449A (en) Temperature calibration method and device
CN116911356A (en) InSAR phase unwrapping method and device based on deep convolutional neural network optimization and storage medium
CN114567386B (en) High-precision channel group delay characteristic fitting and simulation implementation method, system, storage medium and communication system
CN111983312A (en) Method for determining noise parameters and terminal equipment
CN112485651B (en) Chebyshev-based analog circuit element fault parameter identification method
WO2015011496A1 (en) A system of providing calibration information to a frequency attenuator means
CN116596061B (en) Power amplifier harmonic prediction method based on transfer learning
Nikolova et al. Curve fitting for sensors’ analog behavioural modelling
CN115840182B (en) Frequency response curve correction method and device
CN117668301A (en) Asynchronous monitoring data screening method for interval harmonic state estimation
Diamanta et al. Using measurements with large round‐off errors for interval estimation of normal process variance
CN115979437A (en) Real aperture radiometer calibration method based on deep learning network

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
GR01 Patent grant
GR01 Patent grant