A sampling value switching method with variable threshold and sampling circuit

Abstract

The present invention discloses a sampling value switching method with a variable threshold. Sensor output signals are collected through two sampling channels with different amplification factors. The sampling values ​​of the two ADs are read by a DSP, and the sampling values ​​are calibrated, normalized, and switched. The sampling value switching threshold is calculated based on the sensor calibration coefficient K. When used with sensors of different specifications, a reasonable sampling value switching threshold can be adaptively calculated and determined to maximize the multi-channel AD measurement value switching effect, thereby improving measurement accuracy and engineering design adaptability.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grids, and in particular relates to a sampling value switching method with a variable threshold and a sampling circuit. Background Art

[0002] Electronic instrument transformers are increasingly used in AC and DC power transmission systems, particularly in my country's ultra-high voltage DC and flexible DC transmission projects in recent years. They are used to measure high currents and high voltages in DC fields. For active DC electronic instrument transformers, the primary sensor converts the measured value into a small analog voltage signal that is linearly related to the measured DC current or voltage. This signal is then processed by an on-site acquisition module and converted into a digital value. The acquisition module then sends the processed data to a merging unit, which ultimately combines multiple measurement points and outputs the result.

[0003] Measurement accuracy is a key specification for DC electronic transformers. DC systems have high requirements for measurement accuracy. For example, current measurement accuracy should reach Class 0.2 within a range of at least 6 times, and voltage measurement accuracy should reach Class 0.2 within a range of at least 1.5 times. Furthermore, high accuracy requirements are also imposed on the low-frequency point. In the sampling process, a single-channel AD is often unable to meet these requirements. Currently, a common practice is to connect the measured signal to multiple (at least 2) sampling channels with different amplification factors. These channels are then converted by the multiple AD channels, and the final sampled value is output after switching. This approach achieves a balance between a wide range and low-frequency accuracy.

[0004] Application Publication No. CN102323498A discloses a multi-stage segmented data sampling method that divides the input electrical quantity signal into five segments, each with a different amplification factor. By switching the analog switches corresponding to each channel, the sampled signal is amplified at different levels. The final output is the switching of the five-stage segmented sampled data, improving sampling accuracy. However, this application suffers from the following issues: The switching threshold is a fixed threshold, which is suitable for applications where the sensor output signal is fixed and accurate. However, when the front-end sensor specifications are changed or the output ratio is adjusted, the AD measurement range is fixed, which can lead to frequent switching or even failure to switch at full scale. For example, if the rated output of the sensor is 20% lower than before, if switching is still performed at the original 120% and 80% points, frequent switching will occur at the lower measurement points, causing fluctuations in the sampled value. If the rated output of the sensor is 20% higher than before, if switching is still performed at the original 120% point, the AD may saturate due to the increased signal and fail to switch properly.

[0005] The rated primary current and rated primary voltage of DC electronic transformers are not completely standardized. Different engineering parameters vary, and primary sensors of different specifications cannot guarantee completely consistent outputs. For example, the high-voltage arm of the 800kV, 660kV, and 75kV DC voltage dividers consists of eight, six, and one 100MΩ resistor tubes connected in series, respectively, while the parameters of the low-voltage arm and secondary voltage divider unit are fixed, with rated secondary outputs of 0.66V, 0.726V, and 0.495V, respectively. When used with primary sensors of different specifications in actual projects, the rated input of the sampling circuit varies, making the selection of a fixed switching threshold difficult. A threshold that is too large or too small can be problematic, making it difficult to fully and effectively utilize the advantages of multi-range switching. This results in poor project adaptability and may even require the development of different hardware. Summary of the Invention

[0006] The purpose of the present invention is to provide a sampling value switching method with a variable threshold. When used with sensors of different specifications, it can adaptively calculate a reasonable sampling switching threshold, maximize the multi-channel AD measurement value switching effect, and improve measurement accuracy and engineering design adaptability.

[0007] In order to achieve the above-mentioned invention objectives, the technical solutions adopted in this application are:

[0008] As a first aspect of the present application, a sampling value switching method with a variable threshold is proposed, comprising:

[0009] The original analog sampling value signal output by the sensor is processed by a first operational amplifier circuit with an amplification factor of K1 and a second operational amplifier circuit with an amplification factor of K2 to obtain a first analog sampling value signal and a second analog sampling value signal, wherein K1>K2;

[0010] The first analog sample value signal is converted into a first digital sample value signal by a first AD module, and the second analog sample value signal is converted into a second digital sample value signal by a second AD module.

[0011] The DSP receives the first digital sample value signal and the second digital sample value signal output by the first AD module and the second AD module, calibrates and normalizes the first digital sample value signal and the second digital sample value signal to obtain a first sample value signal V1 and a second sample value signal V2 respectively; and obtains a final sample value V after switching the first sample value signal V1 and the second sample value signal V2, specifically including:

[0012] The value of the first sampling value switching threshold X1 is calculated according to the sensor calibration coefficient K: X1 = K*X1N;

[0013] Among them, the output value of the first AD module under the rated signal is V1N, and the value of X1N is between 1.05*V1N and the full code value of the first AD module;

[0014] The second sampling value switching threshold X2 = m*X1, where m is a constant less than 1;

[0015] When V1≤X1, the final sampling value takes the V1 data; when V1>X1, it switches to output the V2 data; once switched to the V2 data, it only switches back to the V1 data when V2<X2 is satisfied.

[0016] Preferably, the sensor calibration coefficient K is related to the sensor and is obtained through actual tests; when the sensor output is the ideal value, the calibration coefficient K is 1; when the actual output value of the sensor is greater than the ideal value, the calibration coefficient K is less than 1; when the actual output value of the sensor is less than the ideal value, the calibration coefficient K is greater than 1.

[0017] Preferably, the processing method of the calibration is: multiplying the sampled value signal received from the AD module by the calibration coefficient K.

[0018] Preferably, the processing method of the normalization is to normalize the second digital quantity sampled value signal to the first digital quantity sampled value signal.

[0019] Preferably, a switching flag f is introduced. When the previous final sampled value output is the V1 data, f is 0; when the previous final sampled value output is the V2 data, f is 1, and the initial state f is 0; when V1>X1, it switches to the V2 data, that is, outputs V = V2, and at the same time the switching flag f is set to 1, otherwise V = V1 and f is set to 0; once switched to the V2 data, the switching return judgment condition is V2<X2, and if it is satisfied, it switches to V = V1 and f is set to 0, otherwise V = V2 and f is set to 1.

[0020] As the second aspect of the present application, a sampling circuit with a variable threshold is proposed, including:

[0021] The first operational amplifier circuit is used to receive the original analog quantity sampled value signal output by the sensor, amplify it by K1 times, and then output the first analog quantity sampled value signal;

[0022] The second operational amplifier circuit is used to receive the original analog quantity sampled value signal output by the sensor, amplify it by K2 times, and then output the second analog quantity sampled value signal; where K1>K2;

[0023] The first AD module is used to receive the first analog quantity sampled value signal and perform analog-to-digital conversion to obtain the first digital quantity sampled value signal;

[0024] The second AD module is used to receive the second analog quantity sampled value signal and perform analog-to-digital conversion to obtain the second digital quantity sampled value signal;

[0025] A DSP, which is used to receive the first digital quantity sampled value signal and the second digital quantity sampled value signal output by the first AD module and the second AD module, and calibrate and normalize the first digital quantity sampled value signal and the second digital quantity sampled value signal to obtain a first sampled value signal V1 and a second sampled value signal V2 respectively; the final sampled value V is obtained after the first sampled value signal V1 and the second sampled value signal V2 are processed by switching, specifically including:

[0026] The value of the first sampled value switching threshold X1 is calculated according to the sensor calibration coefficient K: X1 = K * X1N;

[0027] Wherein, the output value of the first AD module under the rated signal is V1N, and the value of X1N is between 1.05 * V1N and the full code value of the first AD module;

[0028] The second sampled value switching threshold X2 = m * X1, where m is a constant less than 1;

[0029] When V1 ≤ X1, the final sampled value takes the V1 data; when V1 > X1, it switches to output the V2 data; once it switches to the V2 data, it only switches back to the V1 data when V2 < X2 is satisfied.

[0030] Preferably, the sensor calibration coefficient K is related to the sensor and is obtained through actual tests; when the sensor output is the ideal value, the calibration coefficient K is 1; when the actual output value of the sensor is greater than the ideal value, the calibration coefficient K is less than 1; when the actual output value of the sensor is less than the ideal value, the calibration coefficient K is greater than 1.

[0031] Preferably, the processing method of the calibration is: multiplying the sampled value signal received from the AD module by the calibration coefficient K.

[0032] Preferably, the processing method of the normalization is to normalize the second digital quantity sampled value signal to the first digital quantity sampled value signal.

[0033] Preferably, the DSP introduces a switching flag f in the switching process. When the previous final sampled value output is V1 data, f is 0; when the previous final sampled value output is V2 data, f is 1, and the initial state f is 0; when V1 > X1, it switches to V'2 data, that is, outputs V = V2, and at the same time the switching flag f is set to 1, otherwise V = V1 and f is set to 0; once it switches to V2 data, the switching return judgment condition is V2 < X2. If it is satisfied, it switches to V = V1 and f is set to 0, otherwise V = V2 and f is set to 1.

[0034] The beneficial effects of the present invention are as follows: This application simultaneously performs digital conversion of analog signals through two sampling channels with different amplification factors. The software completes normalization processing and coefficient calibration of the two channels of data, and calculates and determines a reasonable sampling value switching threshold based on the calibration coefficient, and finally outputs the sampled data after multi-channel switching. This improves the multi-channel sampling value switching technology and can adaptively calculate and determine a reasonable sampling value switching threshold, solving the problem of difficult selection of fixed thresholds for traditional sampling value switching, maximizing the effect of multi-channel AD measurement value switching, and improving measurement accuracy and engineering design adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a schematic diagram of a sampling circuit with a variable threshold according to an embodiment of the present application.

[0036] Figure 2 This is a flowchart of a sampling value switching method with a variable threshold according to an embodiment of the present application.

[0037] Figure 3 This is another switching flow chart of an embodiment of the present application. DETAILED DESCRIPTION

[0038] The following description of the embodiments of the present invention is provided in more detail with reference to the accompanying drawings and examples to provide a better understanding of the present invention and its advantages in various aspects. However, the embodiments and examples described below are for illustrative purposes only and are not intended to limit the present invention.

[0039] Figure 1 The present invention shows a sampling circuit with a variable threshold according to an embodiment of the present invention, comprising: a first operational amplifier circuit, a second operational amplifier circuit, a first AD module, a second AD module, and a DSP, wherein:

[0040] The first operational amplifier circuit is used for receiving the original analog sampling value signal output by the sensor, amplifying it by K1 times, and then outputting the first analog sampling value signal.

[0041] The second operational amplifier circuit is used for receiving the original analog sampling value signal output by the sensor, amplifying it by K2 times, and then outputting a second analog sampling value signal; wherein K1>K2.

[0042] The first AD module is used to receive the first analog sampling value signal and obtain a first digital sampling value signal through analog-to-digital conversion.

[0043] The second AD module is used to receive the second analog sampling value signal and obtain a second digital sampling value signal through analog-to-digital conversion.

[0044] In the sampling circuit of this application, there are two sampling channels with different amplification factors. The first operational amplifier circuit and the first AD module form the first sampling channel, and the second operational amplifier circuit and the second AD module form the second sampling channel. The amplification factor of the first sampling channel is larger, so the first sampling channel has a smaller measurement range; the amplification factor of the second sampling channel is smaller, so the second sampling channel has a larger measurement range. The two sampling values have a fixed proportional relationship.

[0045] The DSP is used to receive the first digital quantity sampling value signal and the second digital quantity sampling value signal output by the first AD module and the second AD module, and calibrate and normalize the first digital quantity sampling value signal and the second digital quantity sampling value signal to obtain the first sampling value signal V1 and the second sampling value signal V2 respectively; the first sampling value signal V1 and the second sampling value signal V2 are processed by switching to obtain the final sampling value V.

[0046] Because the calibration coefficient reflects the degree to which the sensor deviates from the ideal value. For example, if the coefficient K>1, it means that the actual output of the sensor is smaller. At this time, the original sampling value of the AD is smaller, and there is a larger margin from saturation. Therefore, to make full use of the advantage that the first AD module has higher accuracy within the measurable range, the switching value X1 can be made larger. On the contrary, if the AD saturation problem needs to be considered, the switching value X1 should be made smaller.

[0047] Based on the above considerations, in this embodiment, the process that the first sampling value signal V1 and the second sampling value signal V2 are processed by switching to obtain the final sampling value V specifically includes:

[0048] The value of the first sampling value switching threshold X1 is calculated according to the sensor calibration coefficient K: X1 = K * X1N;

[0049] Among them, the output value of the first AD module under the rated signal is V1N, and the value of X1N is between 1.05 * V1N and the full code value of the first AD module;

[0050] The second sampling value switching threshold X2 = m * X1, where m is a constant less than 1;

[0051] When V1 ≤ X1, the final sampling value takes the V1 data; when V1 > X1, it switches to output the V2 data; once it switches to the V2 data, it only switches back to the V1 data when V2 < X2 is satisfied.

[0052] Based on Figure 1 the sampling circuit shown, a sampling switching method with a variable threshold in this application is as Figure 2 shown, including the following steps:

[0053] S1: The original analog sampled value signals output by the sensors are processed by a first operational amplifier circuit with an amplification factor of K1 and a second operational amplifier circuit with an amplification factor of K2 respectively to obtain a first analog sampled value signal and a second analog sampled value signal, where K1 > K2.

[0054] S2: The first analog sampled value signal is subjected to analog-to-digital conversion by a first AD module to obtain a first digital sampled value signal, and the second analog sampled value signal is subjected to analog-to-digital conversion by a second AD module to obtain a second digital sampled value signal.

[0055] S3: The DSP receives the first digital sampled value signal and the second digital sampled value signal output by the first AD module and the second AD module, and after calibrating and normalizing the first digital sampled value signal and the second digital sampled value signal, obtains a first sampled value signal V1 and a second sampled value signal V2 respectively; the first sampled value signal V1 and the second sampled value signal V2 are processed by switching to obtain a final sampled value V, which specifically includes:

[0056] The value of the first sampled value switching threshold X1 is calculated according to the sensor calibration coefficient K: X1 = K * X1N;

[0057] Among them, the output value of the first AD module under the rated signal is V1N, and the value of X1N is the value between 1.05 * V1N and the full code value of the first AD module;

[0058] The second sampled value switching threshold X2 = m * X1, where m is a constant less than 1;

[0059] When V1 ≤ X1, the final sampled value takes the V1 data; when V1 > X1, it switches to output the V2 data; once it switches to the V2 data, it only switches back to the V1 data when V2 < X2 is satisfied.

[0060] Due to the influence of different specifications of the sensors or the errors of the sensors themselves, there is a deviation between the actual output and the ideal output. Therefore, the sensor calibration coefficient K is obtained through testing and coefficient compensation is performed by the DSP. The sensor calibration coefficient K is related to the sensor and is obtained through actual testing; when the sensor output is the ideal value, the calibration coefficient K is 1; when the actual output value of the sensor is greater than the ideal value, the calibration coefficient K is less than 1; when the actual output value of the sensor is less than the ideal value, the calibration coefficient K is greater than 1. The calibration processing method is: multiplying the sampled value signal received from the AD module by the calibration coefficient K.

[0061] Because the two sampled values are in a fixed proportional relationship, therefore, to make a comparison, it is necessary to perform a normalization process to normalize the AD sampled value with a larger measurement range to the AD sampled value with a smaller measurement range, that is, to normalize the second digital sampled value signal to the first digital sampled value signal.

[0062] In some embodiments, when performing the switching process, the following processing method is adopted: Figure 3 Introduce a switching flag f. When the output of the previous final sampling value is V1 data, f is 0; when the output of the previous final sampling value is V2 data, f is 1, and the initial state f is 0. When V1 > X1, switch to V2 data, that is, output V = V2, and at the same time set the switching flag f to 1; otherwise, V = V1 and f is set to 0. Once switched to V2 data, the switching return judgment condition is V2 < X2. If it is satisfied, switch to V = V1 and f is set to 0; otherwise, V = V2 and f is set to 1.

[0063] Taking the aforementioned 800 kV, 660 kV, and 75 kV DC voltage dividers as examples, the switching scheme of the present application will be specifically introduced below. The rated secondary outputs of the sensors of the 800 kV, 660 kV, and 75 kV DC voltage dividers are 0.66 V, 0.726 V, and 0.495 V respectively. Considering different design parameters, errors and other factors in practice, their rated secondary outputs are within a certain range, which can be considered in the range of 0.4 V to 0.8 V. The sampling circuit is designed according to the ideal output of 0.66 V. After being converted by two 16-bit ADs (AD1 and AD2), the original sampling code values are designed as 20,000 and 5,000 respectively. At this time, K = 1, and the switching threshold value can be selected as 20,000 × 120% = 24,000, which is between 20,000 * 1.05 and the full code value of 32,767, ensuring high measurement accuracy and the AD is not saturated at the maximum of 0.8 V. For the 660 kV and 75 kV DC voltage dividers, the rated secondary outputs of the sensors are 0.726 V and 0.495 V respectively. At this time, the original sampling code values after AD1 conversion are 22,000 and 15,000 respectively, and the sensor correction coefficients K are 0.9091 and 1.3333 respectively. The selection of the switching value X1 should consider improving measurement accuracy, avoiding not switching when the AD is saturated, and avoiding frequent switching at and below the rated point. Therefore, it is more reasonable to choose 24,000 * K, that is, the sampling switching values X1 of the 660 kV and 75 kV DC voltage dividers after coefficient calibration and normalization are 21,818 and 32,000 respectively. The sampling switching value X2 is mainly to avoid frequent switching and should take a value smaller than X1. X2 = m * X1, and m can preferably be a constant between 0.9 and 1. For example, if m = 0.95, the value of X2 can be determined.

[0064] After starting sampling, read the original sampling values, normalize and calibrate the sampling values of AD1 and AD2. Normalization specifically means normalizing the sampling value AD2 with a larger measurement range to the sampling value AD1 with a smaller measurement range. Calibration means multiplying both sampling values by the calibration coefficient K, and then obtaining the sampling values V1 and V2 of the two channels respectively.

[0065] On this basis, the method for judging and selecting the final sampled output is as follows: introduce a switching flag f, where f is 0 when the previous output is V1 and f is 1 when the previous output is V2, and the initial state of f is 0. When V1 > X1, switch to the V2 data, that is, output V = V2, and at the same time set the switching flag f to 1; otherwise, V = V1 and f is set to 0. Once switched to the V2 data, the switching return judgment condition is V2 < X2, and if it is satisfied, then switch to V = V1 and f is set to 0; otherwise, V = V2 and f is set to 1.

[0066] After the switching judgment, the final sampled value V is output. In actual applications, the sampling system collects, processes, and transmits the above-mentioned sampled values at a certain frequency.

[0067] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A sampling value switching method with a variable threshold, characterized in that: Comprising: The original analog sampled value signals output by the sensor are respectively processed by a first operational amplifier circuit with an amplification factor of K1 and a second operational amplifier circuit with an amplification factor of K2 to obtain a first analog sampled value signal and a second analog sampled value signal, where K1 > K2; The first analog sampled value signal is subjected to analog-to-digital conversion through a first AD module to obtain a first digital sampled value signal, and the second analog sampled value signal is subjected to analog-to-digital conversion through a second AD module to obtain a second digital sampled value signal; The DSP receives the first digital sampled value signal and the second digital sampled value signal output by the first AD module and the second AD module, and after calibrating and normalizing the first digital sampled value signal and the second digital sampled value signal, respectively obtains a first sampled value signal V1 and a second sampled value signal V2; The first sampled value signal V1 and the second sampled value signal V2 are subjected to switching processing to obtain a final sampled value V, specifically including: The value of the first sampled value switching threshold X1 is calculated according to the sensor calibration coefficient K: X1 = K * X1N; Wherein, the output value of the first AD module under the rated signal is V1N, and the value of X1N is between 1.05 * V1N and the full code value of the first AD module; The second sampled value switching threshold X2 = m * X1, where m is a constant less than 1; When V1 ≤ X1, the final sampled value takes the V1 data; when V1 > X1, it switches to output the V2 data; once it switches to the V2 data, only when V2 < X2 is satisfied, it switches back to the V1 data.

2. The switching method according to claim 1, wherein: The sensor calibration coefficient K is related to the sensor and is obtained through actual testing; when the sensor output is the ideal value, the calibration coefficient K is 1; when the actual output value of the sensor is greater than the ideal value, the calibration coefficient K is less than 1; when the actual output value of the sensor is less than the ideal value, the calibration coefficient K is greater than 1.

3. The switching method according to claim 1, wherein: The processing method of the calibration is: multiplying the sampled value signal received from the AD module by the calibration coefficient K.

4. The switching method according to claim 1, wherein: The processing method of the normalization is to normalize the second digital sampled value signal to the first digital sampled value signal.

5. The switching method according to claim 1, wherein: Introduce a switching flag f. When the previous final sampled value output is V1 data, f is 0, and when the previous final sampled value output is V2 data, f is 1. The initial state f is 0; when V1 > X1, it switches to V2 data, that is, outputs V = V2, and at the same time the switching flag f is set to 1, otherwise V = V1 and f is set to 0; once it switches to V2 data, the switching return judgment condition is V2 < Xx2. If it is satisfied, it switches to V = V1 and f is set to 0, otherwise V = V2 and f is set to 1.

6. A sampling circuit with a variable threshold, characterized in that: Comprising: A first operational amplifier circuit for receiving the original analog sampled value signal output by the sensor, amplifying it by K1 times, and then outputting a first analog sampled value signal; A second operational amplifier circuit for receiving the original analog sampled value signal output by the sensor, amplifying it by K2 times, and then outputting a second analog sampled value signal; Where K1 > K2; A first AD module for receiving the first analog sampled value signal and performing analog-to-digital conversion to obtain a first digital sampled value signal; ​ A DSP, which is used to receive the first digital quantity sampled value signal and the second digital quantity sampled value signal output by the first AD module and the second AD module, and calibrate and normalize the first digital quantity sampled value signal and the second digital quantity sampled value signal to obtain a first sampled value signal V1 and a second sampled value signal V2 respectively; the first sampled value signal V1 and the second sampled value signal V2 are processed by switching to obtain a final sampled value V, specifically including: The value of the first sampled value switching threshold X1 is calculated according to the sensor calibration coefficient K: X1 = K * X1N; Wherein, the output value of the first AD module under the rated signal is V1N, and the value of X1N is the value between 1.05 * V1N and the full code value of the first AD module; The second sampled value switching threshold X2 = m * X1, where m is a constant less than 1; When V1 ≤ X1, the final sampled value takes the V1 data; when V1 > X1, it switches to output the V2 data; once it switches to the V2 data, it only switches back to the V1 data when V2 < X2 is satisfied.

7. The sampling circuit according to claim 6, wherein: The sensor calibration coefficient K is related to the sensor and is obtained through actual tests; when the sensor output is the ideal value, the calibration coefficient K is 1; when the actual output value of the sensor is greater than the ideal value, the calibration coefficient K is less than 1; when the actual output value of the sensor is less than the ideal value, the calibration coefficient K is greater than 1.

8. The sampling circuit according to claim 6, wherein: The processing method of the calibration is: multiplying the sampled value signal received from the AD module by the calibration coefficient K.

9. The sampling circuit according to claim 6, wherein: The processing method of the normalization is to normalize the second digital quantity sampled value signal to the first digital quantity sampled value signal.

10. The sampling circuit according to claim 6, wherein: The DSP introduces a switching flag f in the switching process. When the previous final sampled value output is V1 data, f is 0; when the previous final sampled value output is V2 data, f is 1, and the initial state f is 0; when V1 > X1, it switches to V2 data, that is, outputs V = V2, and at the same time the switching flag f is set to 1, otherwise V = V1 and f is set to 0; once it switches to V2 data, the switching return judgment condition is V2 < X2. If it is satisfied, it switches to V = V1 and f is set to 0, otherwise V = V2 and f is set to 1.

Images