Displacement Detection Method Based on a Combined Code-Track Eddy Current Grid Displacement Sensor

Abstract

The present invention belongs to the field of absolute displacement positioning and is a new displacement detection algorithm based on the structure of a combined code track eddy current grating displacement sensor. It adopts a measurement method combining an encoding code track and a measurement code track. The encoding code track bits are pre-encoded and grouped, and a forward and backward encoding is used for the uncertain encoding bits to ensure the certainty of the encoding process. The displacement of the measurement code track is corrected to compensate for the displacement of the encoding code track, ensuring that the total output displacement changes continuously without large-scale jumps. The output displacement undergoes a 4-point moving average to reduce the deviation caused by installation and manufacturing processes, and basically eliminates the jitter of the percentile of the measured displacement. This algorithm has achieved good results in the modeling test of the combined code track eddy current grating displacement sensor, verifying the practicality of the algorithm.

Description

Technical Field

[0001] The present invention belongs to the field of absolute displacement positioning, and particularly relates to a displacement detection method based on a combined code track eddy current grating displacement sensor. Background Art

[0002] The eddy current grating displacement sensor is a displacement sensor in which China has core intellectual property rights in the field of digital display waterproof vernier calipers. Based on the inductive sensor and the eddy current effect, it has the function of waterproof and dustproof by nature. Since the first eddy current grating displacement sensor came out, various structures have emerged, including sensors that use two code tracks with different measurement wavelengths for measurement, combined code track sensors, etc.

[0003] The combined code track eddy current grating displacement sensor is a frequency modulation type absolute displacement sensor that uses multiple code tracks. It has a simple structure and high precision, and has a wide range of application fields. The combined code track eddy current grating displacement sensor includes a measurement code track and a coding code track. The metal reflection conductors of the measurement code track are evenly distributed and are used for precise displacement measurement. The coding code track is distributed with metal reflection conductors of different lengths and widths, and the length is an integer multiple of the measurement wavelength. The coding index displacement method is used for rough measurement. The current displacement correction and calibration algorithm for the combined code track eddy current grating displacement sensor has a small error in small-range precise measurement and high accuracy. The main existing problems include: determining the coding value through the magnitude of the precise displacement and the coil coupling relationship. When the coil and the reflection conductor just have a half-coupling situation, it will lead to the uncertainty of the coding value, resulting in large jumps in displacement and non-unique displacement during the displacement calibration process; in addition, even when ensuring that both the measurement code track and the coding code track can correctly measure the displacements of each code track, integrating the displacements of the two different code tracks into the total displacement, without a unified displacement reference will also cause jumps in the output total displacement at some displacement nodes; due to the installation of hardware devices and the deviation of data reading, there will be offsets and jumps in the displayed output displacement at fixed positions. Summary of the Invention

[0004] To overcome the deficiencies of the above-mentioned prior art, the present invention provides a displacement detection method based on a combined code track eddy current grating displacement sensor. By using the displacement of the measurement code track as a reference and combining the coding method of carry and borrow with the displacement of the measurement code track, the uncertainty in the coding process is eliminated, and at the same time, the problem of large jumps in the displacement measurement process is solved. The operation of pre-coding grouping is adopted to effectively reduce the complexity of coding and improve the readability of the algorithm. The sliding average is performed on the output displacement to effectively reduce the problem of percentile jumps in displacement measurement caused by installation deviation.

[0005] The technical solution adopted by the present invention is as follows:

[0006] A new displacement detection algorithm based on a combined code track eddy current grating displacement sensor, comprising the following steps:

[0007] 1) Selectively energize each code track coil;

[0008] 2) Read the frequency of the measurement code track sensing coil to measure the precise displacement: Subtract the frequencies of two coils with a center position separation of λ1 / 2 in the measurement code track to obtain two difference frequency signals f 12 、f 34 , Substitute the two frequency difference signals into the formula to obtain the phase value where λ1 is the periodic displacement length of the reflective conductor placed in the measurement code track. The atan2() function is used in the program to output the phase Calculate the actual output displacement of the measurement code track x ∈ (0, λ1]. Correct and compensate the output displacement of the measurement code track so that the output displacement range is between ±λ1 / 2 for rough displacement compensation of the coding code track;

[0009] 3) Read the frequency values of the coding code track sensing coils for pre-coding, grouping, and performing carry coding or borrow coding in combination with the precise displacement of the measurement code track to complete the rough displacement measurement of the final coding index:

[0010] The pre-coding is used to determine the uncertain bits in the coding process. Read the frequencies of each coil of two coding code tracks. Subtract the frequencies of the corresponding coils of the two code tracks and then perform frequency compensation to complete the normalization of the frequency signal, and save the normalized frequency difference signal. Encode each group of normalized frequency difference signals. Set two decision thresholds a and b (a > b). If the normalized frequency difference signal is greater than a, set the corresponding coding bit to 0. If it is less than b, set the corresponding bit to 1, and set the coding bit between a and b to 2 (uncertain coding bit).

[0011] The grouping is used for subsequent carry or borrow coding of the uncertain coding bits. Traverse the pre-coding, find the coding bits where the uncertain coding is located and assign this bit to the corresponding group number. The grouping rule is: For a six-bit coding, scan from the highest bit to the lowest bit. Consecutive and adjacent uncertain coding bits are grouped into one group, and a single uncertain coding bit forms one group. The group number of the determined coding bits (coding value not equal to 2) is 0.

[0012] The carry or borrow coding is used to calibrate the unique coding value at a certain displacement. Use the actual displacement measured by the measurement code track as the reference to determine whether the actual measured displacement of the measurement code track is greater than λ1 / 2 (when the measurement code track moves λ1 / 2, the coupling between the coding code track coil and the reflective conductor is half, and at this time there will be coding uncertainty and it can be encoded into two different binary codings). If the actual measured displacement of the measurement code track is less than or equal to λ1 / 2, use the borrow coding method. Otherwise, use the carry coding method.

[0013] The coding rule of the carry - down coding is as follows: find the max - th bit with the same group number. If the coding of the (max + 1)-th bit is 0, then the coding of the max - th bit is 0, the coding of the (max - 1)-th bit is 1, and the coding alternates in turn until the min - th bit. If the max - th bit is the left - most bit of the coding, then find the min - th bit. If the (min - 1)-th bit is 0, then the coding of the min - th bit is 1, the coding of the (min + 1)-th bit is 0, and the coding alternates in turn to the max - th bit.

[0014] The coding rule of the carry - up coding is as follows: find the max - th bit with the same group number. If the coding of the (max + 1)-th bit is 0, then the coding of the max - th bit is 1, the coding of the (max - 1)-th bit is 0, and the coding alternates in turn until the min - th bit. If the max - th bit is the left - most bit of the coding, then find the min - th bit. If the (min - 1)-th bit is 0, then the coding of the min - th bit is 0, the coding of the (min + 1)-th bit is 1, and the coding alternates in turn to the max - th bit.

[0015] 4) Make an accurate compensation for the index displacement of the coding track. For the displacement compensation of the coding track, use the measured displacement of the corrected compensation coding track, add it to the coding index displacement of the coding track, and output the total displacement;

[0016] 5) Perform a 4 - point moving average on the total displacement, and use the algorithm to reduce the error caused by hardware installation and frequency perception to eliminate the problem of displacement jitter at fixed positions.

[0017] Using Matlab software to build an algorithm model for the combined eddy current grating displacement sensor, compared with the existing absolute displacement positioning technology, the present invention has the following advantages:

[0018] 1. By means of pre - coding, grouping, and carry - up / down coding, group the uncertain coding bits in the pre - coding. Combining with the accurate displacement size measured by the measurement track, perform carry - down or carry - up coding on the uncertain coding bits in the same group respectively. Compared with the existing coding track coding recognition algorithm, it reduces the coding uncertainty generated by directly coding according to the complementary relationship without grouping, and will not have a coding value jump when the coil and the reflection conductor are just half - coupled, has strong coding anti - interference ability, and at the same time reduces the algorithm difficulty.

[0019] 2. Take the displacement of the measurement track as the measurement reference to compensate for the displacement of the coding track. Through the correction and compensation of the displacement of the measurement track, eliminate the installation deviation at zero displacement, and at the same time compensate for the displacement of the coding track to eliminate the jump of the total output displacement.

[0020] 3. Introduce a moving average algorithm, perform a 4 - point moving average on the output total displacement, effectively reduce the data jitter caused by hardware installation and recognition, and solve the problem of displacement jitter of the eddy current grating displacement sensor at fixed positions. Description of the Drawings

[0021] Figure 1 is the track layout diagram of the eddy current grid displacement sensor in the prior art

[0022] Figure 2 is Figure 1 schematic diagram of partial track magnification

[0023] Figure 3 is the displacement measurement flow chart of the measurement track

[0024] Figure 4 is the displacement measurement flow chart of the coding track

[0025] Figure 5 is the flow chart of the borrowing algorithm

[0026] Figure 6 is the flow chart of the carrying algorithm

[0027] Figure 7 is the schematic diagram of 4-point moving average

[0028] Figure 8 is the displacement measurement effect diagram after modeling of the measurement track

[0029] Figure 9 is the displacement measurement effect diagram after modeling of the coding track

[0030] Figure 10 is the total displacement measurement effect diagram

[0031] Figure 11 is the displacement index diagram of the coding track Specific embodiments

[0032] The present invention will be further described below in conjunction with embodiments and the accompanying drawings, but the protection scope of the present invention should not be limited thereby.

[0033] The track layout structure of the combined track eddy current grid displacement sensor (see Figure 1 and Figure 2) There are a total of 3 track channels, including one measurement track channel and two symmetric coding track channels. The measurement track channel is placed between the two coding track channels, and the metal reflection conductors are evenly distributed. The lateral distribution period of the reflection conductors is 5.08 mm, and the lateral width of the reflection conductors is 2.54 mm. The lateral width of the coil is 2.54 mm, and the center spacing between the 1st and 2nd coils and between the 3rd and 4th coils is 2.54 mm. The center distance between the 2nd and 3rd coils is 3 / 4 of the measurement wavelength, which is 3.81 mm. The coding track channels are located on both sides of the measurement track channel, and the reflection conductor distributions of the two track channels are in a complementary structure, symmetrically distributed on both sides of the measurement track channel. During the movement of the coil, the coil change trends of the two track channels are opposite. The track channel layout structure of the combined track channel eddy current grating displacement sensor will effectively reduce the influence of factors such as power signal fluctuations, frequency fluctuations of the oscillation circuit, and temperature, and improve the reliability of coding.

[0034] The novel displacement detection method based on the structure of the combined track channel eddy current grating displacement sensor in this embodiment includes the following steps:

[0035] 1) Select and activate the coils of each track channel;

[0036] 2) The process of measuring the displacement of the measurement track channel is shown in Figure 3 :

[0037] ① Read the frequencies f1, f2, f3, and f4 of the four groups of coils of the measurement track channel;

[0038] ② Subtract f1 from f2 and f3 from f4 to obtain two groups of frequency difference signals f 12 、f 34 ;

[0039] ③ Calculate the phase of the measurement track channel. The formula is as follows:

[0040]

[0041] In the formula, λ1 is the displacement length of the placement period of the reflection conductors of the measurement track channel. The atan2() function is used in the program to output the phase

[0042] ④ Calculate the actual output displacement x of the measurement track channel. The formula is as follows:

[0043]

[0044] ⑤ Due to installation errors, the displacement measurement of the measurement track channel at zero displacement is not zero. In addition, it is necessary to measure the displacement compensation of the subsequent coding track channels by the measurement track channel. Therefore, for the displacement correction of the measurement track channel, the compensation displacement of the measurement track channel should ensure that the displacement at zero displacement is zero, and the compensation displacement changes periodically from -λ1 / 2 to λ1 / 2 (see Figure 8));

[0045] 3) The process of encoding the code track displacement is as follows (see Figure 4 ));

[0046] ① Read the frequencies of the coils of two complementary code tracks (each code track includes six coils).

[0047] ② Subtract the frequencies of the corresponding coils of the two complementary code tracks to obtain a difference frequency signal;

[0048] ③ Normalize the frequency difference signal to obtain a normalized difference frequency signal;

[0049] ④ Set two decision thresholds (set to 0.9 and 0.1 here), and pre - encode the normalized difference frequency signal. If a certain normalized frequency difference signal is greater than 0.9, set the corresponding coding bit to 0. If it is less than 0.1, set the corresponding bit to 1. The coding bits between 0.1 and 0.9 are set to 2 (uncertain coding bits).

[0050] ⑤ Traverse the six - bit pre - coding, and assign corresponding group numbers to the coding bits where the uncertain coding is located. The grouping rule is: Scan the six - bit coding from the highest bit to the lowest bit. Consecutive and adjacent uncertain coding bits are grouped into one group, and a single uncertain coding bit is grouped into one group. The group number of the determined coding bits (coding value not 2) is 0. (For example, the group number of 220021 can be grouped as 110020). During the movement of the moving grating coil, it is continuously coupled with the fixed grating reflection conductor. At most 3 group numbers will appear in this sensor structure. Next, the 3 groups will be encoded separately.

[0051] ⑥ Determine whether the displacement of the measurement code track is greater than 2.54 mm. If the displacement measured by the measurement code track is less than or equal to 2.54 mm, adopt a retreating coding method (see Figure 5 ). Otherwise, adopt an advancing coding method (see Figure 6); Retiring encoding step: Scan the six - digit group number encoding. If there is a group number 1, encode the encoding bits with group number 1, and find the highest bit (the left - most bit) max and the lowest bit (the right - most bit) min of group number 1. According to the distribution of the sensor's reflected conductors, during the movement, the difference between the max bit and the min bit can range from 0 to 3, which means that at least one encoding bit has group number 1 and at most four encoding bits have group number 1. The encoding rule is: Find the max bit of group number 1. If the encoding of the max + 1 bit is 0, then the encoding of the max bit is 0, and the encoding of the max - 1 bit is 1, and the encoding alternates in turn until the min bit is encoded. If the max bit is the left - most encoding bit, then find the min bit. If the encoding of the min - 1 bit is 0, then the encoding of the min bit is 1, and the encoding of the min + 1 bit is 0, and the encoding alternates in turn to the max bit. If there is a group number 2, encode the encoding bits with group number 2 (the encoding rule is the same as that of group number 1). According to the distribution of the fixed grating's reflected conductors, during the movement of the sensor, the difference between max and min can range from 0 to 2, which means that at least one encoding bit has group number 2 and at most three bits have group number 1. If there is a group number 3, encode the encoding bits with group number 3, and find the highest bit max and the lowest bit min of group number 3. According to Figure 1 the distribution of the reflected conductors, during the movement of the moving grating, the difference between max and min can only be 0, that is, at most one encoding with group number 3 can appear, and the position where group number 3 is located can only appear at the right - most bit of the coil encoding. Therefore, only need to find the max bit and check the encoding value of the max + 1 bit, which is the encoding of group number 3. Carrying encoding step: Scan the six - digit group number encoding. If there is a group number 1, encode the encoding bits with group number 1, and find the highest bit (the left - most bit) max and the lowest bit (the right - most bit) min of group number 1. According to the distribution of the sensor's reflected conductors, during the movement, the difference between the max bit and the min bit can range from 0 to 3, which means that at least one encoding bit has group number 1 and at most four encoding bits have group number 1. The encoding rule is: Find the max bit of group number 1. If the encoding of the max + 1 bit is 0, then the encoding of the max bit is 1, and the encoding of the max - 1 bit is 0, and the encoding alternates in turn until the min bit is encoded. If the max bit is the left - most encoding bit, then find the min bit. If the encoding of the min - 1 bit is 0, then the encoding of the min bit is 0, and the encoding of the min + 1 bit is 1, and the encoding alternates in turn to the max bit. If there is a group number 2, encode the encoding bits with group number 2 (the encoding rule is the same as that of group number 1). According to the distribution of the fixed grating's reflected conductors, during the movement of the sensor, the difference between max and min can range from 0 to 2, which means that at least one encoding bit has group number 2 and at most three bits have group number 1. If there is a group number 3, encode the encoding bits with group number 3, and find the highest bit max and the lowest bit min of group number 3. According to Figure 1Reflective conductor distribution. During the movement of the moving grating, the difference between max and min can only be 0, that is, there can be at most one code with a group number of 3, and the position where the group number is 3 can only appear in the rightmost bit of the coil code. Therefore, only need to find the max bit and check that the inverted value of the code value at the (max + 1)-th bit is the code with a group number of 3.

[0052] ⑦ Make 33 code values that will appear when the 6-bit coil moves into a coded track displacement index diagram (see Figure 11 ), and output the coarse displacement measured by the coded track (see the Matlab coded track model output in Figure 9 ).

[0053] ⑧ The total displacement output is the sum of the coded track displacement and the compensation of the measured track displacement (see the Matlab total displacement output in Figure 10 ).

[0054] ⑨ Perform a 4-point moving average on the total displacement (see Figure 7 ). While ensuring good displacement following characteristics during the sliding process of the sensor, the problem of displacement percentile jitter at fixed positions is eliminated.

Claims

1. A displacement detection method for a combined code track eddy current grid displacement sensor, characterized in that, The method includes the following steps: Step 1) Selectively activate each track coil: During the displacement, the coupling area between the four groups of measurement track coils and the reflective conductor continuously changes. According to the change in the coupling area of each measurement track coil, the coil frequency signal varies between 4.55 MHz and 5.00 MHz. Among them, in the state of complete coupling with the reflective conductor, the frequency is 5 MHz, and in the state of complete non - coupling, the frequency signal is 4.55 MHz; During the displacement, the coupling area between the six groups of coding track coils and the reflective conductor continuously changes. According to the change in the coupling area of each coding track coil, the coil frequency signal varies between 7.00 MHz and 8.00 MHz. Among them, in the state of complete coupling with the reflective conductor, the frequency is 8.00 MHz, and in the state of complete non - coupling, the frequency signal is 7.00 MHz; Step 2) Measure the displacement of each track: Measurement track: ① Read the frequencies f1, f2, f3, f4 of the four groups of coils of the measurement track; ②Subtract f1 from f2 and subtract f3 from f4 to obtain two groups of frequency difference signals f 12 、f 34 ; ③ Calculate the phase of the measured track The formula is as follows: Wherein, λ1 is the periodic displacement length of the reflection conductor of the measurement track. In the present invention, 5.08 mm is adopted, and the atan2() function is used to output the phase ④ Calculate the actual output displacement x of the measurement track, and the formula is as follows: ⑤ Perform displacement compensation on the measurement track to ensure that the displacement of the measurement track at zero displacement is zero, and the displacement changes periodically between -λ1 / 2 and λ1 / 2; Coding track; ① Read the frequencies of the coils of each of the two complementary coding tracks. Each coding track includes six coils; ② Subtract the frequencies of the corresponding coils of the two complementary coding tracks to obtain six groups of differential frequency signals; ③ Normalize the differential frequency signals to obtain normalized differential frequency signals; ④ Set two decision thresholds, set as 0.9 and 0.1, and perform pre - coding on the normalized differential frequency signals: if a certain normalized frequency difference signal is greater than 0.9, set the corresponding coding bit to 0; if it is less than 0.1, set the corresponding bit to 1; and set the coding bit between 0.1 and 0.9 to 2, that is, an uncertain coding bit; ⑤ Traverse the six - bit pre - coding, and assign corresponding group numbers to the coding bits where the uncertainty lies; The grouping rule is: scan the six - bit coding from the highest bit to the lowest bit. Consecutive and adjacent uncertain coding bits are divided into one group, and a single uncertain coding bit is a group. For the determined coding bits, the group number with a coding value not equal to 2 is 0; ⑥ Judge whether the displacement of the measurement track is greater than 2.54 mm. If the displacement measured by the measurement track is less than or equal to 2.54 mm, use the carry - down coding method, otherwise use the carry - up coding method; Make a coding track displacement index table for the 33 coding values that appear when the 6 - bit coils move, and output the coarse displacement measured by the coding track; The total displacement output is the sum of the coding track displacement and the measurement track displacement compensation; ⑨ Put the output total displacement values into an array storing four - bit data in sequence and perform 4 - point sliding.

2. The displacement detection method of the combined code track eddy current grid displacement sensor according to claim 1, characterized in that, The carry - down coding steps are as follows: Scan the six - digit group number encoding. If group number 1 exists, encode the encoding bits with group number 1. Find the most significant bit, i.e., the left - most bit max, and the least significant bit, i.e., the right - most bit min, of group number 1. According to the distribution of the sensor's reflected conductors, during the movement, the difference between the max bit and the min bit can range from 0 to 3, which means that at least one encoding bit has group number 1 and at most four encoding bits have group number 1. The encoding rule is as follows: find the max bit of group number 1. If the encoding of the max + 1 bit is 0, then the encoding of the max bit is 0 and the encoding of the max - 1 bit is 1, and the encoding alternates in turn until the min bit is encoded. If the max bit is the left - most encoding bit, then find the min bit. If the min - 1 bit is 0, then the encoding of the min bit is 1 and the encoding of the min + 1 bit is 0, and the encoding alternates in turn until the max bit. If group number 2 exists, encode the encoding bits with group number 2. The encoding rule is the same as that of group number 1. According to the distribution of the fixed grating's reflected conductors, during the movement of the sensor, the difference between max and min can range from 0 to 2, with at least one encoding bit having group number 2 and at most three bits having group number 1. If group number 3 exists, encode the encoding bits with group number 3. Find the most significant bit max and the least significant bit min of group number 3. According to the reflected conductor distribution in Figure 1, during the movement of the moving grating, the difference between max and min can only be 0, that is, at most one encoding with group number 3 can appear, and the position where group number 3 is located can only appear at the right - most bit of the coil encoding. Therefore, only need to find the max bit and check the encoding value of the max + 1 bit, which is the encoding of group number 3.

3. The displacement detection method of the combined code track eddy current grid displacement sensor according to claim 1, characterized in that, The carry encoding steps are as follows: Scan the six - digit group number encoding. If group number 1 exists, encode the encoding bits with group number 1. Find the most significant bit (the left - most bit) max and the least significant bit (the right - most bit) min of group number 1. According to the distribution of the sensor - reflected conductors, during the movement, the difference between the max bit and the min bit can range from 0 to 3, meaning that at least one encoding bit has group number 1 and at most four encoding bits have group number 1. The encoding rule is: find the max bit of group number 1. If the encoding of the max + 1 bit is 0, then the encoding of the max bit is 1, and the encoding of the max - 1 bit is 0. The encoding alternates in turn until the min bit is encoded. If the max bit is the left - most encoded bit, then find the min bit. If the min - 1 bit is 0, then the encoding of the min bit is 0, and the encoding of the min + 1 bit is 1. The encoding alternates in turn until the max bit. If group number 2 exists, encode the encoding bits with group number 2. The encoding rule is the same as that of group number 1. According to the distribution of the fixed grating - reflected conductors, during the movement of the sensor, the difference between max and min can range from 0 to 2. At least one encoding bit has group number 2 and at most three bits have group number 1. If group number 3 exists, encode the encoding bits with group number 3. Find the most significant bit max and the least significant bit min of group number 3. According to the distribution of the reflected conductors, during the movement of the moving grating, the difference between max and min can only be 0, that is, at most one encoding bit with group number 3 can appear, and the position where group number 3 is located can only appear at the right - most bit of the coil encoding. Therefore, only need to find the max bit and check the inverted encoding value of the max + 1 bit, which is the encoding of group number 3.

Images