A multi-point linear displacement device detection method based on serial polling sampling

Abstract

The present invention relates to the field of magnetic levitation technology, and specifically to a multi-point linear displacement device detection method using serial polling sampling, which solves the problems of low accuracy and small measurement range of the existing two-point Hall linear displacement detection method, and high cost and difficult maintenance of the parallel sampling multi-point absolute linear displacement measurement method. A multi-point linear displacement device detection method using serial polling sampling, comprising the following steps: S1: When performing linear displacement detection on a suspended body in a magnetic levitation system, twenty-four magnetic sensors are provided in the detection equipment to obtain voltage analog sine and cosine signals, and the phases of the sine and cosine signals on each magnetic sensor differ by ninety degrees. The present invention uses a high-precision linear displacement sensor to achieve accurate detection of the linear displacement of a mover or a suspended body in a magnetic levitation system, thereby improving the operating stability and control accuracy of the magnetic levitation system.

Description

Technical Field

[0001] The invention relates to the field of magnetic suspension technology, in particular to a multi-point linear displacement equipment detection method for serial polling sampling. Background Art

[0002] As an advanced non-contact support and guidance technology, magnetic levitation technology has significant advantages such as high speed, low noise, low wear, and no mechanical contact friction. Among them, magnetic levitation motor, as a new type of electric motor, has been widely used in industrial automation, aerospace, transportation and other fields because of its advantages such as no contact, low friction loss, low vibration, long life and high efficiency.

[0003] The stable operation of the magnetic levitation motor depends on high-precision position recognition technology to achieve accurate monitoring and control of the position of the mover. The position detection methods of the magnetic levitation motor mainly include two-point AMR Hall linear displacement ruler, multi-point parallel displacement ruler, photoelectric, electromagnetic induction and other technologies.

[0004] The existing two-point Hall linear displacement detection method cannot directly detect the position of the mover when power is turned on. The mover needs to be slid manually, which has low accuracy and small measurement range. The parallel sampling multi-point absolute linear displacement measurement method is costly and difficult to maintain. When the mover runs at high speed, errors are prone to occur due to the large amount of data processed. The program occupies more resources and has high requirements on devices. Therefore, it does not meet the existing needs. In this regard, we propose a multi-point linear displacement equipment detection method with serial polling sampling. Summary of the invention

[0005] The purpose of the present invention is to provide a multi-point linear displacement device detection method with serial polling sampling, so as to solve the problems that the existing two-point Hall linear displacement detection method proposed in the above background technology cannot directly detect the position of the mover when power is turned on, and the mover needs to be manually slid, the accuracy is low, the measurement range is small, and the parallel sampling multi-point absolute linear displacement measurement method is costly and difficult to maintain. When the mover runs at high speed, errors are prone to occur due to the large amount of data processed, the program occupies more resources, and has high requirements on devices.

[0006] To achieve the above object, the present invention provides the following technical solution: a multi-point linear displacement device detection method using serial polling sampling, comprising the following steps:

[0007] S1: When performing linear displacement detection on the suspended body in the magnetic suspension system, twenty-four magnetic sensors are set in the detection equipment to obtain voltage analog sine and cosine signals. The sine and cosine signals on each magnetic sensor are ninety degrees out of phase, and the data obtained by the twenty-four magnetic sensors are collected through six multiplexing switches. The switching of the six multiplexing switches is controlled by a state machine. The multiplexing switch outputs an ADC sampling completion signal every four switchings. Each multiplexing switch outputs two signals, namely, the sine and cosine signals of a magnetic sensor;

[0008] S2: Perform static position calculation while the multiplex switch is switched, and fix the multiplex switch when the mover is running stably. The voltage analog sine and cosine signals output by the six multiplex switches are connected to two ADC chips, namely AD7606 chips. Each ADC uses six input ports to perform analog-to-digital conversion and outputs two digital signals. The data output module is designed according to the output characteristics of the AD7606 chip, and the converted digital signals are output as six channels, and each channel is a 16-bit sine and cosine digital signal;

[0009] S3: Use the NSI1306 chip to collect the voltage of the magnetic sensor and convert it into a digital value, which is used as a parameter for sine and cosine amplitude compensation. The obtained sine and cosine amplitude data is stored in the EEPROM, and the sine and cosine amplitude is compensated with the collected real-time voltage of the magnetic sensor;

[0010] S4: converting the sine and cosine signals sampled in step S2 into positive and negative alternating data, the conversion method is to subtract the offset from the current sine and cosine values, the offset is equal to half of the sum of the maximum value and the minimum value, the sine and cosine signal conversion process involves a sine and cosine signal maximum and minimum value update module, a standard sine and cosine generation module, etc., designing an inverse tangent angle calculation module in the FPGA, subtracting the offset from the current sine and cosine values ​​into the inverse tangent angle calculation module to calculate the angle, and using the Cordic algorithm to calculate the inverse tangent angle in the FPGA;

[0011] S5: The serial multi-point absolute encoder is divided into two axes, each axis is composed of twelve magnetic sensors, three multiplex switches, and one ADC. In the process of the multiplex switch being switched, the static positions of the two axes are independently calculated through the twenty-four inverse tangent angles, linear area and other data output by the preceding module. At the same time, the static position number of turns, position and other information calculated by the static position calculation module are transmitted to the subsequent dynamic mode single magnetic sensor coarse position calculation module. The number of static position turns is used as the initial number of turns of the single magnetic sensor. The number of magnetic pole turns of the current coarse position is counted through the jump number of the inverse tangent angle output by the inverse tangent angle calculation module, and the linear area judgment module is used to judge whether the angle is the angle required for calculating the position, and the direction of the calculated coarse position also needs to be considered.

[0012] S6: Use a single magnetic sensor in dynamic mode to perform fine position calculation. The calculation process of the fine position is to add the coarse position calculated in step S5 to the current angle. For two-point magnetic sensor position fitting, when the multiplexing switch is switched, the static position calculated by the static position calculation module is transferred to the position fitting module for fitting; after the multiplexing switch is fixed, the fine position calculated by the single magnetic sensor in dynamic mode is transferred to the position fitting module for fitting;

[0013] S7: The linear region signals, angle signals, position signals, etc. calculated in all steps are passed to the error detection module to detect error types such as position mutation, mover overlap, static position calculation error, etc. After an error occurs in the serial sampling multi-point absolute encoder, a manual error reset signal needs to be input to allow the fixed multiplexer switch to switch back to calculate the static position, and then switch to the dynamic position calculation.

[0014] Preferably, the magnetic sensor adopts the magnetoresistive chip ADA4571BRZ. The magnetic sensor detects the magnetic field signal generated by the movable magnetic plate and forms a sine and cosine voltage signal with a phase difference of ninety degrees. The alternating transformation process of the magnetic field NS of the movable magnetic plate generates a pair of complete sine and cosine signals of 360 degrees, and the amplitude is basically fixed.

[0015] Preferably, the multiplexing switch is TPW4052, and the multiplexing switch is connected to two groups of signal sources, namely signal source group A and signal source group B. Signal source group A includes four signals a0~a3, and signal source group B includes four signals b0~b3. The multiplexing switch controls output ports a and b through logic ports S11 and S12, and each TPW4052 outputs a sine and cosine signal of a magnetic sensor at the same time.

[0016] Preferably, the sine and cosine voltage signals output by the multiplexer are converted into digital signals through AD7606, the AD7606 chip outputs two 64-bit digital signals, the two digital signals input by the AD7606 chip are divided into six 16-bit sine and cosine signals through the driving module provided therein, and each of the ADC chips is connected to three multiplexers and occupies six input ports.

[0017] Preferably, the ADC chip collects original sine and cosine signals, and the ADC chip uses a sliding mean filter module with a window of N to filter the signal. The sliding mean filter module formula is new average value = (result of this conversion + sum of past N-1 values) / N, and the N value is the threshold of the sliding window.

[0018] Preferably, the sine and cosine signals of the twenty-four magnetic sensors are transmitted alternately through a multiplexing switch and converted into digital signals through an AD7606 chip;

[0019] The multiplex switch switching process outputs the original signals of twenty-four magnetic sensors, the multiplex switch fixing process outputs the original signal of one magnetic sensor, and each of the AD7606 chips is connected to three multiplex switches and outputs three original signals.

[0020] Preferably, the signals output by the AD7606 chip are all positive values, and the process of converting the sine and cosine signals into standard sine and cosine signals is standard Sin=current Sin-(SinMax+SinMin) / 2, the standard Sin is the converted standard sine value, the current Sin is the original sine value, the SinMax is the original sine maximum value, the SinMin is the original sine minimum value, and the standard Sin is accurately calculated by the sine and cosine maximum and minimum value real-time update module.

[0021] Preferably, the inverse tangent angle is calculated by a linear region calculation module, and the effective radius calculated in the linear region module is:

[0022] Rrad²=(StdSin-(SinMax+SinMin) / 2)²+(StdCos-(CosMax+CosMin) / 2)²;

[0023] Rref²=((SinMax-SinMin) / 2)²;

[0024] Ddelta = Rref² - Rrad²;

[0025] Wherein Rrad is the actual measured radius value, Rref is the theoretical radius value, Ddelta is the effective radius, SinMax is the standard sine maximum value after the signal output by the AD7606 chip is converted, SinMin is the standard sine minimum value after the signal output by the AD7606 chip is converted, StdSin is the standard sine value after the signal output by the AD7606 chip is converted, CosMax is the standard cosine maximum value after the signal output by the AD7606 chip is converted, CosMin is the standard cosine minimum value after the signal output by the AD7606 chip is converted, and StdCos is the standard cosine value after the signal output by the AD7606 chip is converted;

[0026] A 14-stage pipeline is designed in the FPGA, and the method for calculating the inverse tangent angle by the Cordic algorithm is a coordinate rotation method, and the coordinate rotation method is X2=X1-Y1*2 -i and Y2=Y1+X1*2 -i .

[0027] Preferably, the position calculation of the single magnetic sensor includes coarse position calculation and fine position calculation, the coarse position calculation is the accumulation of turns according to the change of angle, and the fine position calculation is the sum of the coarse position and the current angle, and the fine position of the single magnetic sensor calculated by the static position and the dynamic position algorithm is the same.

[0028] Preferably, the position fitting module adopts a dynamic position fitting algorithm, and the position fitting module performs dynamic fitting through a state machine, and the dynamic fitting process is sequentially dynamic fitting range judgment state, dynamic fitting difference update state, dynamic fitting difference validity judgment state, actual selection of fitting difference calculation state, actual start of fitting position calculation judgment state and final position output calculation state.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] 1. The present invention uses a magnetic sensor array to detect position information based on the Hall effect and the change of the magnetic field. It is divided into a static position calculation process and a dynamic position calculation process. After the static position calculation is stable, it switches to the dynamic position calculation. It combines the advantages of the two-point Hall ruler position detection method and the parallel sampling multi-point absolute position detection method, thereby achieving high-precision and high-stability position detection. At the same time, it has the function of error detection during the movement of the mover, and also has the advantages of fast response, simple structure, low cost, and no need for manual intervention during use.

[0031] 2. The present invention realizes accurate detection of the position of the mover of the magnetic levitation motor by optimizing the arrangement of magnetic sensors and signal processing algorithms, uses a magnetic sensor array to detect changes in the magnetic field generated by the permanent magnet on the mover, and extracts position information through a signal processing algorithm. Without manual intervention, the mover position can be directly detected on a linear displacement detection scale, and serial sampling can better save the use of devices, with the advantages of high precision and good stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a hardware schematic diagram of the present invention as a whole;

[0033] Figure 2 A flow chart of the present invention as a whole;

[0034] Figure 3 It is a structural schematic diagram of the original signal module of the present invention;

[0035] Figure 4 The waveform diagram of the standard sine and cosine of the present invention;

[0036] Figure 5 It is a schematic diagram of the angle and linear region of the present invention;

[0037] Figure 6 It is a data schematic diagram of the linear region and angle of the present invention;

[0038] Figure 7 A schematic diagram of calculating the number of turns and position of the first magnetic sensor of the present invention;

[0039] Figure 8 It is a schematic diagram of position calculation when the mover of the present invention is in the linear region;

[0040] Fig. 9 A schematic diagram showing the calculation of the position and number of turns of the ninth magnetic sensor of the present invention;

[0041] Fig.10 It is a structural schematic diagram of a single magnetic sensor position calculation module of the present invention;

[0042] Fig.11 Position maps for static and dynamic position calculations of the present invention;

[0043] Fig.12 It is a structural schematic diagram of the position fitting module of the present invention. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0045] See also Figures 1 to 12 , an embodiment of the present invention provides: a multi-point linear displacement device detection method for serial polling sampling, comprising the following steps:

[0046] S1: When performing linear displacement detection on the suspended body in the magnetic suspension system, twenty-four magnetic sensors are set in the detection equipment to obtain voltage analog sine and cosine signals. The sine and cosine signals on each magnetic sensor are ninety degrees out of phase, and the data obtained by the twenty-four magnetic sensors are collected through six multiplexing switches. The switching of the six multiplexing switches is controlled by a state machine. The multiplexing switch outputs an ADC sampling completion signal every four switchings. Each multiplexing switch outputs two signals, namely, the sine and cosine signals of a magnetic sensor;

[0047] S2: Perform static position calculation while the multiplex switch is switched, and fix the multiplex switch when the mover is running stably. The voltage analog sine and cosine signals output by the six multiplex switches are connected to two ADC chips, namely AD7606 chips. Each ADC uses six input ports to perform analog-to-digital conversion and outputs two digital signals. The data output module is designed according to the output characteristics of the AD7606 chip, and the converted digital signals are output as six channels, and each channel is a 16-bit sine and cosine digital signal;

[0048] S3: Use the NSI1306 chip to collect the voltage of the magnetic sensor and convert it into a digital value, which is used as a parameter for sine and cosine amplitude compensation. The obtained sine and cosine amplitude data is stored in the EEPROM, and the sine and cosine amplitude is compensated with the collected real-time voltage of the magnetic sensor;

[0049] S4: converting the sine and cosine signals sampled in step S2 into positive and negative alternating data, the conversion method is to subtract the offset from the current sine and cosine values, the offset is equal to half of the sum of the maximum value and the minimum value, the sine and cosine signal conversion process involves a sine and cosine signal maximum and minimum value update module, a standard sine and cosine generation module, etc., designing an inverse tangent angle calculation module in the FPGA, subtracting the offset from the current sine and cosine values ​​into the inverse tangent angle calculation module to calculate the angle, and using the Cordic algorithm to calculate the inverse tangent angle in the FPGA;

[0050] S5: The serial multi-point absolute encoder is divided into two axes, each axis is composed of twelve magnetic sensors, three multiplex switches, and one ADC. In the process of the multiplex switch being switched, the static positions of the two axes are independently calculated through the twenty-four inverse tangent angles, linear area and other data output by the preceding module. At the same time, the static position number of turns, position and other information calculated by the static position calculation module are transmitted to the subsequent dynamic mode single magnetic sensor coarse position calculation module. The number of static position turns is used as the initial number of turns of the single magnetic sensor. The number of magnetic pole turns of the current coarse position is counted through the jump number of the inverse tangent angle output by the inverse tangent angle calculation module, and the linear area judgment module is used to judge whether the angle is the angle required for calculating the position, and the direction of the calculated coarse position also needs to be considered.

[0051] S6: Use a single magnetic sensor in dynamic mode to perform fine position calculation. The calculation process of the fine position is to add the coarse position calculated in step S5 to the current angle. For two-point magnetic sensor position fitting, when the multiplexing switch is switched, the static position calculated by the static position calculation module is transferred to the position fitting module for fitting; after the multiplexing switch is fixed, the fine position calculated by the single magnetic sensor in dynamic mode is transferred to the position fitting module for fitting;

[0052] S7: The linear region signals, angle signals, position signals, etc. calculated in all steps are passed to the error detection module to detect error types such as position mutation, mover overlap, static position calculation error, etc. After an error occurs in the serial sampling multi-point absolute encoder, a manual error reset signal needs to be input to allow the fixed multiplexer switch to switch back to calculate the static position, and then switch to the dynamic position calculation.

[0053] Among them, the magnetic sensor adopts the magnetoresistive chip ADA4571BRZ. The magnetic sensor detects the magnetic field signal generated by the mover magnetic plate and forms a sine and cosine voltage signal with a phase difference of ninety degrees. The magnetic field NS alternating transformation process of the mover magnetic plate generates a pair of 360-degree complete sine and cosine signals, and the amplitude is basically fixed. The multiplexer switch is TPW4052. The multiplexer switch is connected to two groups of signal sources, namely signal source group A and signal source group B. Signal source group A includes four signals from a0 to a3, and signal source group B includes four signals from b0 to b3. The multiplexer controls the output ports a and b through the logic ports S11 and S12. Each TPW4052 outputs a sine and cosine signal of a magnetic sensor at the same time.

[0054] Among them, the sine and cosine voltage signals output by the multiplexed switch are converted into digital signals through AD7606. The AD7606 chip outputs two 64-bit digital signals. The two digital signals input by the AD7606 chip are divided into six 16-bit sine and cosine signals through its driving module. Each ADC chip is connected to three multiplexed switches and occupies six input ports. The ADC chip collects the original sine and cosine signals. The ADC chip uses a sliding mean filter module with a window of N to filter the signal. The sliding mean filter module formula is new average value = (the result of this conversion + the sum of the values ​​of the past N-1 times) / N, and the N value is the threshold of the sliding window. The signal is filtered by the sliding mean filter module to achieve smoothing of the sine and cosine signals.

[0055] Furthermore, the sine and cosine signals of the twenty-four magnetic sensors are transmitted alternately through a multiplexing switch and converted into digital signals through an AD7606 chip;

[0056] The multiplex switch switching process outputs the original signals of twenty-four magnetic sensors, and the multiplex switch fixing process outputs the original signal of one magnetic sensor. Each AD7606 chip is connected to three multiplex switches and outputs three original signals. The signals output by the AD7606 chip are all positive values. The process of converting the sine and cosine signals to standard sine and cosine signals is standard Sin=current Sin-(SinMax+SinMin) / 2. Standard Sin is the converted standard sine value. The current Sin is the original sine value, SinMax is the original sine maximum value, and SinMin is the original sine minimum value. The standard Sin is accurately calculated by the sine and cosine maximum and minimum real-time update module.

[0057] See also Figure 4 The arc tangent angle is calculated by the linear region calculation module to determine whether a pair of sine and cosine values ​​currently obtained from the magnetic sensor is valid;

[0058] The effective radius calculated in the linear region module is:

[0059] Rrad²=(StdSin-(SinMax+SinMin) / 2)²+(StdCos-(CosMax+CosMin) / 2)²;

[0060] Rref²=((SinMax-SinMin) / 2)²;

[0061] Ddelta = Rref² - Rrad²;

[0062] Where Rrad is the actual measured radius value, Rref is the theoretical radius value, Ddelta is the effective radius, SinMax is the standard sine maximum value after the signal output by the AD7606 chip is converted, SinMin is the standard sine minimum value after the signal output by the AD7606 chip is converted, StdSin is the standard sine value after the signal output by the AD7606 chip is converted, CosMax is the standard cosine maximum value after the signal output by the AD7606 chip is converted, CosMin is the standard cosine minimum value after the signal output by the AD7606 chip is converted, and StdCos is the standard cosine value after the signal output by the AD7606 chip is converted.

[0063] See also Figure 5 , a 14-stage pipeline is designed in FPGA. The method used by Cordic algorithm to calculate the inverse tangent angle is the coordinate rotation method. The coordinate rotation method is X2=X1-Y1*2-i and Y2=Y1+X1*2-i. In interval 1, the change of the inverse tangent angle is linear; in interval 2, the change of the inverse tangent angle is nonlinear, and the effective radius of the magnetic sensor is larger than that of interval 1. When calculating the position, the inverse tangent angle value of interval 2 is discarded due to the nonlinear change of interval 2; interval 3 is an invalid interval.

[0064] As attached Figure 6 As shown, when the magnetic suspension mover passes through the encoder, the angle data and linear zone data output by the twelve magnetic sensors of one axis of the encoder can be divided into three parts, C1, C2 and C3. Part C1 is when the mover gradually enters the linear zone, and the linear zone mark gradually changes from 0000_0000_0000 to 0111_1111_1111; Part C2 is when the mover is in the linear zone, and the linear zone mark is 1111_1111_1111; Part C3 is when the mover gradually leaves the linear zone, and the change in the linear zone is from 1111_1111_1110 to 0000_0000_0000. When calculating the static position of the encoder, it is also divided into the above three parts. Parts C1 and C2 calculate the position and number of turns of the first magnetic sensor, and parts C2 and C3 calculate the position and number of turns of the ninth magnetic sensor.

[0065] When the magnetic suspension mover slides forward, the angle signal output by a single magnetic sensor increases to 1440 and then jumps to 0, corresponding to the actual magnetic field angle jumping from 360 degrees to zero. At this time, one magnetic pole of the magnetic suspension mover completely passes the current magnetic sensor, and the adjacent Figure 8 There are a total of twelve angle data from top to bottom, corresponding to the twelve magnetic sensors on the single-axis encoder. Each angle data has twelve jumps from 1440 to 0, corresponding to the twelve magnetic poles of the magnetic levitation mover.

[0066] See also Fig.10and Fig.11 The position calculation of a single magnetic sensor includes coarse position calculation and fine position calculation. The coarse position calculation is the cumulative number of turns according to the change of angle, and the fine position calculation is the sum of the coarse position and the current angle. The fine position of a single magnetic sensor through static position calculation and dynamic position algorithm is the same. According to the set resolution, the coarse position is increased or decreased by a certain value every turn. The method for judging whether the position increases or decreases is to calculate the difference between the current angle and the angle at the previous moment. If the difference is positive, it indicates that the mover is running in the reverse direction, and the coarse position needs to be subtracted by one turn; if the difference is negative, it indicates that the mover is running in the forward direction, and the position needs to be increased by one turn. The initial number of turns of the coarse position is the data passed in by the number of turns calculated by the static position algorithm.

[0067] See also Fig.12 The position fitting module adopts a dynamic position fitting algorithm. The position fitting module performs dynamic fitting through a state machine. The dynamic fitting process is sequentially as follows: dynamic fitting range judgment state, dynamic fitting difference update state, dynamic fitting difference validity judgment state, actual selection of fitting difference calculation state, actual start of fitting position calculation judgment state and final position output calculation state.

[0068] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A multi-point linear displacement device detection method using serial polling sampling, characterized in that: The following steps are involved: S1: When performing linear displacement detection on the suspended body in the magnetic suspension system, twenty-four magnetic sensors are set in the detection equipment to obtain voltage analog sine and cosine signals. The sine and cosine signals on each magnetic sensor are ninety degrees out of phase, and the data obtained by the twenty-four magnetic sensors are collected through six multiplexing switches. The switching of the six multiplexing switches is controlled by a state machine. The multiplexing switch outputs an ADC sampling completion signal every four switchings. Each multiplexing switch outputs two signals, namely, the sine and cosine signals of a magnetic sensor; S2: Perform static position calculation while the multiplex switch is switched, and fix the multiplex switch when the mover is running stably. The voltage analog sine and cosine signals output by the six multiplex switches are connected to two ADC chips, namely AD7606 chips. Each ADC uses six input ports to perform analog-to-digital conversion and outputs two digital signals. The data output module is designed according to the output characteristics of the AD7606 chip, and the converted digital signals are output as six channels, and each channel is a 16-bit sine and cosine digital signal; S3: Use the NSI1306 chip to collect the voltage of the magnetic sensor and convert it into a digital value, which is used as a parameter for sine and cosine amplitude compensation. The obtained sine and cosine amplitude data is stored in the EEPROM, and the sine and cosine amplitude is compensated with the collected real-time voltage of the magnetic sensor; S4: converting the sine and cosine signals sampled in step S2 into positive and negative alternating data, the conversion method is to subtract the offset from the current sine and cosine values, the offset is equal to half of the sum of the maximum value and the minimum value, the sine and cosine signal conversion process involves a sine and cosine signal maximum and minimum value update module and a standard sine and cosine generation module, designing an inverse tangent angle calculation module in the FPGA, subtracting the offset from the current sine and cosine values ​​into the inverse tangent angle calculation module to calculate the angle, and using the Cordic algorithm to calculate the inverse tangent angle in the FPGA; S5: The serial multi-point absolute encoder is divided into two axes, each axis is composed of twelve magnetic sensors, three multiplex switches, and one ADC. In the process of the multiplex switch being switched, the static positions of the two axes are independently calculated through the twenty-four inverse tangent angles and linear region data output by the preceding module. At the same time, the number of static position turns and position information calculated by the static position calculation module are transmitted to the subsequent dynamic mode single magnetic sensor coarse position calculation module. The number of static position turns is used as the initial number of turns of the single magnetic sensor. The number of magnetic pole turns of the current coarse position is counted through the jump number of the inverse tangent angle output by the inverse tangent angle calculation module, and the linear region judgment module is used to judge whether the angle is the angle required for calculating the position, and the direction of the calculated coarse position also needs to be considered. S6: Use a single magnetic sensor in dynamic mode to perform fine position calculation. The calculation process of the fine position is to add the coarse position calculated in step S5 to the current angle. For two-point magnetic sensor position fitting, when the multiplexing switch is switched, the static position calculated by the static position calculation module is transferred to the position fitting module for fitting; after the multiplexing switch is fixed, the fine position calculated by the single magnetic sensor in dynamic mode is transferred to the position fitting module for fitting; S7: The linear region signals, angle signals, and position signals calculated in all steps are passed to the error detection module to detect the types of position mutations, mover overlaps, and static position calculation errors. After an error occurs in the serial sampling multi-point absolute encoder, a manual error reset signal needs to be input to allow the fixed multiplexer switch to switch back to calculate the static position, and then switch to the dynamic position calculation.

2. The method for detecting a multi-point linear displacement device by serial polling sampling according to claim 1, characterized in that: The magnetic sensor adopts the magnetoresistive chip ADA4571BRZ. The magnetic sensor detects the magnetic field signal generated by the movable magnetic plate and forms a sine and cosine voltage signal with a phase difference of ninety degrees. The alternating transformation process of the magnetic field NS of the movable magnetic plate generates a pair of complete sine and cosine signals of 360 degrees, and the amplitude is basically fixed.

3. The method for detecting a multi-point linear displacement device by serial polling sampling according to claim 1, characterized in that: The multiplexing switch is TPW4052, and the multiplexing switch is connected to two groups of signal sources, namely signal source group A and signal source group B. Signal source group A includes four signals a0~a3, and signal source group B includes four signals b0~b3. The multiplexing switch controls output ports a and b through logic ports S11 and S12, and each TPW4052 outputs a sine and cosine signal of a magnetic sensor at the same time.

4. The method for detecting a multi-point linear displacement device by serial polling sampling according to claim 3 is characterized in that: The sine and cosine voltage signals output by the multiplex switch are converted into digital signals through AD7606. The AD7606 chip outputs two 64-bit digital signals. The two digital signals input by the AD7606 chip are divided into six 16-bit sine and cosine signals through the driving module provided therein. Each of the ADC chips is connected to three multiplex switches and occupies six input ports.

5. The method for detecting a multi-point linear displacement device by serial polling sampling according to claim 4, characterized in that: The ADC chip collects the original sine and cosine signals, and uses a sliding mean filter module with a window of N to filter the signals. The sliding mean filter module formula is new average value = (result of this conversion + sum of values ​​of past N-1 times) / N, and the N value is the threshold of the sliding window.

6. The method for detecting a multi-point linear displacement device by serial polling sampling according to claim 1, characterized in that: The sine and cosine signals of the twenty-four magnetic sensors are transmitted in turn through a multiplexing switch and converted into digital signals through an AD7606 chip; The multiplex switch switching process outputs the original signals of twenty-four magnetic sensors, the multiplex switch fixing process outputs the original signal of one magnetic sensor, and each of the AD7606 chips is connected to three multiplex switches and outputs three original signals.

7. The method for detecting a multi-point linear displacement device by serial polling sampling according to claim 1, characterized in that: The signals output by the AD7606 chip are all positive values. The process of converting the sine and cosine signals into standard sine and cosine signals is standard Sin=current Sin-(SinMax+SinMin) / 2. The standard Sin is the converted standard sine value, the current Sin is the original sine value, the SinMax is the original sine maximum value, and the SinMin is the original sine minimum value. The standard Sin is accurately calculated by the sine and cosine maximum and minimum value real-time update module.

8. The method for detecting a multi-point linear displacement device by serial polling sampling according to claim 1, characterized in that: The inverse tangent angle is calculated by a linear region calculation module, and the effective radius calculated in the linear region module is: Rrad²=(StdSin-(SinMax+SinMin) / 2)²+(StdCos-(CosMax+CosMin) / 2)²; Rref²=((SinMax-SinMin) / 2)²; Ddelta = Rref² - Rrad²; Wherein Rrad is the actual measured radius value, Rref is the theoretical radius value, Ddelta is the effective radius, SinMax is the standard sine maximum value after the signal output by the AD7606 chip is converted, SinMin is the standard sine minimum value after the signal output by the AD7606 chip is converted, StdSin is the standard sine value after the signal output by the AD7606 chip is converted, CosMax is the standard cosine maximum value after the signal output by the AD7606 chip is converted, CosMin is the standard cosine minimum value after the signal output by the AD7606 chip is converted, and StdCos is the standard cosine value after the signal output by the AD7606 chip is converted; A 14-stage pipeline is designed in the FPGA, and the method for calculating the inverse tangent angle by the Cordic algorithm is a coordinate rotation method, and the coordinate rotation method is X2=X1-Y1*2 -i and Y2=Y1+X1*2 -i .

9. The method for detecting a multi-point linear displacement device by serial polling sampling according to claim 1, characterized in that: The position calculation of the single magnetic sensor includes coarse position calculation and fine position calculation. The coarse position calculation is to accumulate the number of turns according to the change of angle, and the fine position calculation is the sum of the coarse position and the current angle. The fine position of the single magnetic sensor calculated by static position and dynamic position algorithm is the same.

10. The method for detecting a multi-point linear displacement device by serial polling sampling according to claim 1, characterized in that: The position fitting module adopts a dynamic position fitting algorithm. The position fitting module performs dynamic fitting through a state machine. The dynamic fitting process is sequentially a dynamic fitting range judgment state, a dynamic fitting difference update state, a dynamic fitting difference validity judgment state, an actual selection of fitting difference calculation state, an actual start fitting position calculation judgment state and a final position output calculation state.

Images