Signal processing device

Abstract

The present disclosure relates to a signal processing device that detects a phase difference between two signals by applying a Lissajous figure, compensates for an error between the signals by easily detecting the phase difference between the signals by recognizing a pattern, improves a control efficiency in processing the signals, such as an efficiency of controlling a resolver sensor or a motor, by improving a precision of the signals through the compensated signals, improves a utilization by applying the device to various devices, such as a vehicle, and improves a performance of the device.

Description

Technical Field

[0001] Exemplary embodiments of this disclosure relate to a signal processing apparatus for detecting and compensating for errors between a plurality of signals having a phase difference. Background Technology

[0002] Electric vehicles are equipped with batteries, which are charged by supplying charging current to the batteries and operated by receiving operating power from the batteries.

[0003] Typically, electric vehicles use permanent magnet synchronous motors (PMSMs) and require position sensors (such as resolver sensors) to operate the motor.

[0004] The rotary transformer sensor detects the angle of the motor's rotor. Because the stator is configured as an inductor, there is a problem that when a sinusoidal excitation signal with several kilohertz is applied as an input signal, a delay occurs between the excitation signal and the two output signals, and the two output signals change in amplitude and phase.

[0005] Korean Patent Application Publication No. 2020-0053772, entitled "A device for identifying the angle of a rotary transformer, a motor drive device having the device, and a vehicle," proposes a method for detecting and processing the rotor angle of a rotary transformer sensor.

[0006] The performance of the motor output signal and the resolver sensor may be degraded due to changes in the phase of the signal.

[0007] Therefore, a method is needed to detect the phase difference between signals and correct errors during the processing of signals with phase differences.

[0008] [Related Technical Documents]

[0009] [Patent Documents]

[0010] (Patent document 1) Korean Patent Application Publication No. 2020-0053772. Summary of the Invention

[0011] Various embodiments relate to a signal processing apparatus that improves the performance of a control rotary transformer sensor or motor by using the phase difference between two input signals to detect and compensate for errors.

[0012] In one embodiment, a signal processing apparatus includes: a signal input unit configured to receive and digitize two signals having a phase difference; a filter unit configured to filter the two digitized signals and output a filtered first signal and a second signal; a pattern recognition unit configured to convert the first signal and the second signal into coordinates and create a signal trajectory with a predetermined shape; a signal processing unit configured to create a converted sensor signal by normalizing the signal trajectory, calculate an error by comparing the converted sensor signal and a reference signal, and detect anomalies in the first signal and the second signal; a signal compensation unit configured to compensate the first signal and the second signal based on the error between the converted sensor signal and the reference signal when an anomaly is detected; and a control unit configured to verify whether the first signal and the second signal are properly compensated by applying the first signal and the second signal compensated by the signal compensation unit to the pattern recognition unit and the signal processing unit, the control unit being configured to calculate an angular velocity from the compensated first signal and the second signal based on the verification result.

[0013] The filter unit may include a high-pass filter and a low-pass filter. The high-pass filter and the low-pass filter may be connected in series or in parallel to form a notch filter or a band-pass filter to remove noise from the signal and separate the signal in a specific frequency domain.

[0014] The signal processing unit can call up a reference signal that has the same phase difference as the phase difference between the first signal and the second signal, and compare the reference signal with the first signal and the second signal.

[0015] The signal processing unit can normalize the signal trajectory created by the pattern recognition unit so that the ratio between the amplitudes of the first signal and the second signal is 1:1.

[0016] The signal processing unit can perform normalization to eliminate variations caused by the amplitudes of the first and second signals, and compare variations caused by the phase difference.

[0017] When the result of comparing the reference signal and the converted sensor signal indicates that the error between the reference signal and the converted sensor signal is 0, the signal processing unit can determine that the first signal and the second signal are normal, and when the error between them is not 0, the signal processing unit can determine that the first signal and the second signal are abnormal.

[0018] The pattern recognition unit can convert the signal value of the first signal into X-axis coordinates, convert the signal value of the second signal into Y-axis coordinates, and create a signal trajectory that connects the converted coordinates.

[0019] The pattern recognition unit can create signal trajectories based on Lissajous figures and recognize the signal trajectories as patterns.

[0020] When the first and second compensated signals are input, the pattern recognition unit can create a signal trajectory based on the first and second compensated signals, and the signal processing unit can calculate the error by comparing the signal trajectory created based on the first and second compensated signals with the reference signal again, thereby verifying the compensated signal.

[0021] When the error is 0, the control unit can calculate the angular velocity based on the first and second signals without verifying the signals, and when the error is not 0, the control unit can allow the signal compensation unit to compensate for the error.

[0022] According to one aspect, the signal processing apparatus of this disclosure can improve the accuracy of the signal by using a Lissajous figure to compensate for the error between two signals having a phase difference and verifying the compensated signal.

[0023] According to one aspect of this disclosure, the signal processing apparatus can be applied to processes of processing various signals, such as processes of controlling rotary transformer sensors or motors, to improve control efficiency and enhance the performance of the apparatus. Attached Figure Description

[0024] Figure 1 This is a diagram illustrating the configuration of a signal processing apparatus according to an embodiment of the present disclosure.

[0025] Figure 2 This is a diagram illustrating the hardware configuration of a signal processing apparatus according to an embodiment of the present disclosure.

[0026] Figure 3 This is a diagram illustrating the configuration of a filter in a signal processing apparatus according to an embodiment of the present disclosure.

[0027] Figure 4 and Figure 5 This is a flowchart illustrating a method of operating a signal processing apparatus according to an embodiment of the present disclosure.

[0028] Figure 6A and Figure 6B This is a diagram illustrating the operation of a filter in a signal processing apparatus according to an embodiment of the present disclosure.

[0029] Figure 7 It is shown Figure 6A and Figure 6B A diagram showing the filtered signal in the image.

[0030] Figures 8A to 8C This is a diagram illustrating a method for generating a Lissajous figure using a signal processing apparatus according to an embodiment of the present disclosure.

[0031] Figures 9A to 9C It shows from Figures 8A to 8CThe Lissajous signal detected in the figure is shown in the diagram.

[0032] Figure 10 This is an exemplary diagram showing a Lissajous figure based on the phase difference and ratio between signal amplitudes.

[0033] Figure 11A and Figure 11B This is a diagram illustrating anomaly detection based on Lissajous figures for a signal processing apparatus according to embodiments of the present disclosure. Detailed Implementation

[0034] This disclosure will be described below with reference to the accompanying drawings.

[0035] Here, for clarity and convenience, the thickness of lines or the dimensions of components shown in the accompanying drawings may be exaggerated. Furthermore, the terminology used below is defined in consideration of its function within this disclosure and may be changed according to the intent of the user or operator or common practice. Therefore, these terms should be defined based on the entirety of this specification.

[0036] Figure 1 This is a diagram illustrating the configuration of a signal processing apparatus according to an embodiment of the present disclosure.

[0037] like Figure 1 As shown, the signal processing device includes a signal input unit 120, a filter unit 130, a pattern recognition unit 140, a signal processing unit 150, a signal compensation unit 160, a signal output unit 190, a data unit 180, and a control unit 110.

[0038] The signal input unit 120 receives two signals with a phase difference and converts the analog signal into a digital signal. The signal input unit 120 also receives a signal detected by a sensor, such as a resolver sensor, and converts that signal into a digital signal.

[0039] The signal input unit 120 may include multiple input terminals and an analog-to-digital converter (ADC).

[0040] The filter unit 130 removes noise by filtering two signals with a phase difference and separates the signals in the frequency domain to be actually used. The filter unit 130 stores the separated signals in the data unit 180.

[0041] The filter unit 130 includes a high-pass filter and a low-pass filter. The high-pass filter and the low-pass filter are connected in series or in parallel to form a band-pass filter or a notch filter.

[0042] The pattern recognition unit 140 creates a signal trajectory based on a Lissajous figure from two filtered signals with a phase difference. The pattern recognition unit 140 identifies the shape of the created signal trajectory as a pattern.

[0043] The signal processing unit 150 creates a sensor signal based on the Lissajous figure (i.e., the trajectory of the signal) of the pattern recognition unit 140. The sensor signal is normalized and converted for signal comparison relative to the pattern.

[0044] Signal processing unit 150 creates a reference signal (which is a standard for signal processing) and stores the reference signal in data unit 180. Signal processing unit 150 can create a reference signal based on a Lissajous figure relative to two signals with signal amplitudes having a specific integer ratio and a specified phase difference. Signal processing unit 150 can create and store multiple reference signals based on the phase difference.

[0045] The signal processing unit 150 retrieves a reference signal corresponding to the phase difference between the sensor signal from the data unit 180 and performs signal processing.

[0046] The signal processing unit 150 compares the converted sensor signal and the reference signal, and calculates the error between the reference signal and the converted sensor signal.

[0047] When the reference signal equals the converted sensor signal, the signal processing unit 150 calculates the error to be 0 and determines that both filtered signals are normal. Conversely, when the error is not 0, the signal processing unit 150 determines that both filtered signals are abnormal.

[0048] Furthermore, the pattern recognition unit 140 and the signal processing unit 150 verify, based on the Lissajous figure, whether the phase difference between the two compensated signals is normal relative to the signal compensated by the signal compensation unit 160.

[0049] The signal compensation unit 160 compensates for the error between the converted sensor signal and the reference signal output from the signal processing unit 150. The signal compensation unit 160 performs proportional-integral (PI) control on the two filtered signals until the error is reduced.

[0050] The signal output unit 190 outputs the signal finally compensated by the signal compensation unit 160 and the angular velocity calculated by the control unit 110.

[0051] The control unit 110 controls the input and output of the control signals and stores the data created for each step in the data unit 180. The control unit 110 detects and stores the offset value of the conversion signal.

[0052] Based on the signal comparison result between the reference signal and the converted sensor signal of the signal processing unit 150, when the error between the converted sensor signal and the reference signal is 0, since the converted sensor signal and the reference signal are equal to each other, the control unit 110 calculates the angular velocity based on the two filtered signals without performing separate verification.

[0053] When the error calculated by the signal processing unit 150 is not zero, that is, when the converted sensor signal and the reference signal are different from each other, the control unit 110 allows the signal compensation unit 160 to compensate for the error. In addition, the control unit 110 applies the compensated signal to the pattern recognition unit 140 and the signal processing unit 150, and performs control to verify the compensated signal.

[0054] When the signal is normally compensated based on the verification results, the control unit 110 calculates the angular velocity based on the compensated signal and outputs the angular velocity.

[0055] Figure 2 This is a diagram illustrating the hardware configuration of a signal processing apparatus according to an embodiment of the present disclosure.

[0056] like Figure 2 As shown, a sensor board on which sensor 10 is mounted outputs two signals 20 with different phases, and inputs these two signals 20 into a signal processing device (electronic control unit (ECU)) 30. For example, the sensor may be a resolver sensor and is used in sensors or devices that utilize phase differences.

[0057] The signal processing device (ECU) 30 may include an analog-to-digital converter (ADC) and a main control unit (MCU).

[0058] The ADC is included in the signal input unit 120, and the MCU includes a signal processing unit 150 and a control unit 110.

[0059] The signal processing device 30 monitors the signal consistently by using a signal HW2 with a phase difference and a signal HW1 generated by feedback, and compensates for errors related to the phase difference.

[0060] Figure 3 This is a diagram illustrating the configuration of a filter in a signal processing apparatus according to an embodiment of the present disclosure.

[0061] like Figure 3 As shown, the signal input unit 120 receives two signals with a phase difference and converts these two signals into digital signals. The signal input unit 120 converts the input analog sensor signal HM1 into a digital signal SW1 using an ADC.

[0062] The filter unit 130 includes a high-pass filter 131 and a low-pass filter 132.

[0063] Filter unit 130 is connected in series with high-pass filter 131 and low-pass filter 132 to form band-pass filter 134. Alternatively, filter unit 130 can be connected in parallel with high-pass filter 131 and low-pass filter 132 to form notch filter 133.

[0064] The filter unit 130 filters the signal by using a notch filter 133 or a bandpass filter 134 composed of a high-pass filter 131 and a low-pass filter 132.

[0065] Therefore, the filter unit 130 removes noise included in the signal by using multiple filters and creates a signal in the desired specific frequency domain.

[0066] Figure 4 and Figure 5 This is a flowchart illustrating a method of operating a signal processing apparatus according to an embodiment of the present disclosure.

[0067] like Figure 4 As shown, the signal processing device 30 receives two signals with a phase difference detected by the sensor 10 (S310), and converts the two signals into digital signals by using the analog-to-digital converter of the signal input unit 120 (S320).

[0068] The control unit 110 collects information about the raw data by means of the analog-to-digital converter (ADC) of the signal input unit 120, identifies the offset by finding the zero-cross point based on the converted signal, and then stores the offset value in the data unit 180 (S330).

[0069] The filter unit 130 removes noise and separates the signals in a predetermined frequency domain by filtering the two signals using a bandpass filter or a notch filter (S340).

[0070] The filter unit 130 configures the notch filter 133 or the bandpass filter 134 by connecting the high-pass filter 131 and the low-pass filter 132 in parallel or in series, and processes the signal.

[0071] The control unit 110 creates the absolute value of the filtered signal by applying the stored offset value and stores the absolute value.

[0072] The pattern recognition unit 140 converts two signals with a phase difference, filtered by the filter unit 130, into x-axis and y-axis coordinate values ​​respectively, and creates a signal trajectory with a predetermined shape based on a Lissajous figure. The pattern recognition unit 140 calculates the coordinates by setting the first filtered signal as the x-axis and the second filtered signal as the y-axis, and identifies the signal trajectory created by connecting the coordinates based on a Lissajous figure.

[0073] The signal processing unit 150 creates a sensor signal that is converted by normalizing the signal trajectory to a predetermined size (S350).

[0074] The converted sensor signal is generated by normalizing the trajectory of a signal created by digitizing and filtering the detected sensor signal and then converting the sensor signal into a vector dimension.

[0075] like Figure 5 As shown, the signal processing unit 150 calls a reference signal that is used as a stored signal (S370). The signal processing unit 150 calls a reference signal that has the same phase difference as the sensor signal.

[0076] The reference signal is generated by normalizing the trajectory of the signal relative to a signal having a predetermined phase difference in the idle state, and the reference signal is stored in the data unit 180. For example, the reference signal is a circular signal trajectory created by combining two signals with a 90-degree phase difference into a vector dimension.

[0077] The signal processing unit 150 compares the converted sensor signal and the reference signal and determines whether the signals are the same as each other (S380).

[0078] When the comparison result indicates that the reference signal and the converted sensor signal are equal to each other, the control unit 110 outputs the angular velocity based on the converted sensor signal (S450).

[0079] Simultaneously, when the converted sensor signal differs from the reference signal, the signal processing unit 150 calculates the error between the filtered signal and the reference signal. The signal processing unit 150 calculates the error as the value obtained by subtracting the reference signal from the converted sensor signal.

[0080] The signal compensation unit 160 reduces the error by repeatedly executing PI control using the calculated error. The signal compensation unit 160 performs phase difference compensation (S410) on the signal processed by the signal processing unit 150.

[0081] The control unit 110 applies the compensated signal to the pattern recognition unit 140 and the signal processing unit 150.

[0082] The pattern recognition unit 140 and the signal processing unit 150 use a Lissajous figure to verify whether the compensation has been performed correctly (S420). The signal processing unit 150 normalizes the trajectory of the signal created by the pattern recognition unit 140 based on the compensated signal, and then calculates the error by comparing the compensated signal with a reference signal. When the error is 0, the control unit determines that the compensated signal is normal, and the control unit completes the verification.

[0083] When the verification result indicates that the compensation was not performed normally (S430), the control unit 110 repeatedly performs the processing of calculation error, the processing of reducing error by performing PI control, and the processing of compensating for phase difference (S400, S410 and S420).

[0084] When the verification is completed normally, the control unit 110 calculates the angular velocity based on the compensated signal and outputs the angular velocity (S450).

[0085] Therefore, according to this disclosure, the error between two signals with a phase difference can be compensated, and the angular velocity can be calculated by compensating for the phase difference.

[0086] Figure 6A and Figure 6B This is a diagram illustrating the operation of a filter in a signal processing apparatus according to an embodiment of the present disclosure. Figure 7 It is shown Figure 6A and Figure 6B A diagram showing the filtered signal in the image.

[0087] like Figure 6A and Figure 6B As shown, the signal input unit 120 detects a first sensor signal S1 and a second sensor signal S2 from the sensor 10. The first sensor signal S1 and the second sensor signal S2 are sensor signals generated by synthesizing a frequency and have a phase difference with a predetermined amplitude.

[0088] The signal input unit 120 converts the first sensor signal S1 and the second sensor signal S2 input from the sensor 10 into digital signals by using an analog-to-digital converter (ADC).

[0089] The filter unit 130 removes noise by filtering the digitized first sensor signal S1 and second sensor signal S2, and outputs a first signal S11 and a second signal S12 by filtering the first sensor signal S1 and the second sensor signal S2 with a specific frequency band. The filter unit 130 outputs the first signal S11 by filtering the first sensor signal S1, and outputs the second signal S12 by filtering the second sensor signal S2.

[0090] The filter unit 130 filters the first sensor signal S1 and the second sensor signal S2 using a bandpass filter 134. A bandpass filter is used because the delay caused by the integrator characteristics when using a regular low-pass filter would affect the actual operating performance of the regular low-pass filter.

[0091] like Figure 7 As shown, the signals filtered by filter unit 130, namely the first signal S11(A) and the second signal S12(B), have a phase difference with a predetermined amplitude. For example, the first signal S11(A) and the second signal S12(B) have a phase difference of 90 degrees.

[0092] Control unit 110 calculates the average value of the first signal S11(A) and the second signal S12(B) over one cycle, and performs absoluteification by calculating the offset at the zero crossing point of the average value. Control unit 110 compensates for negative values ​​by calculating and performing absoluteification of the offset based on the average value and the zero crossing point.

[0093] The control unit 110 controls the motor by calculating the angular velocity using filtered signals, namely the first signal S11(A) and the second signal S12(B).

[0094] Before calculating the angular velocity, the control unit 110 identifies whether the sensor signal input from the sensor 10 is normal. When the sensor signal input from the sensor 10 is abnormal, the control unit 110 compensates for the abnormality and then calculates the angular velocity.

[0095] The control unit 110 allows the pattern recognition unit 140 and the signal processing unit 150 to identify whether the signal is abnormal, and allows the signal compensation unit 160 to compensate for errors caused by the abnormality. Furthermore, the control unit 110 verifies the compensated signal by applying the compensated signal to the pattern recognition unit 140 and the signal processing unit 150, and then calculates the angular velocity based on the compensated signal.

[0096] Figures 8A to 8C This is a diagram illustrating a method for creating a Lissajous figure using a signal processing apparatus according to embodiments of the present disclosure, and Figures 9A to 9C It shows from Figures 8A to 8C The Lissajous signal detected in the figure is shown in the diagram.

[0097] like Figure 8A and Figure 8B As shown, the pattern recognition unit 140 converts the signal values ​​of the first signal S11 and the second signal S12, which have a phase difference, into X-axis and Y-axis coordinate vector values ​​over time.

[0098] like Figure 8AAs shown, the pattern recognition unit 140 converts the signal value of the first signal S11 at the first time t1 into an x-axis coordinate value and the signal value of the second signal S12 into a y-axis coordinate value.

[0099] Therefore, as Figure 9A As shown, the first signal S11 and the second signal S12 at the first time t1 are converted into the first point P11.

[0100] like Figure 8B As shown, the pattern recognition unit 140 converts the signal value of the first signal S11 at the second time t2 into an x-axis coordinate value and the signal value of the second signal S12 into a y-axis coordinate value.

[0101] Therefore, as Figure 9B As shown, the first signal S11 and the second signal S12 at the second time t2 are converted into the second point P12.

[0102] In addition, such as Figure 8C As shown, the pattern recognition unit 140 converts the signal value of the first signal S11 at the third time t3 into an x-axis coordinate value and the signal value of the second signal S12 into a y-axis coordinate value.

[0103] Therefore, as Figure 9C As shown, the first signal S11 and the second signal S12 at the third time t3 are converted into the third point P13.

[0104] Therefore, the pattern recognition unit 140 accumulates coordinate values ​​over time by converting a first signal S11 and a second signal S12 with a phase difference, and creates a signal trajectory with a predetermined shape connecting the coordinate values. The pattern recognition unit 140 recognizes the shape of the signal trajectory as a pattern. The created signal trajectory is based on a Lissajous figure.

[0105] like Figure 9C As shown, when the ratio between the amplitudes of the first signal S11 and the second signal S12 is 1:1 and the phase difference between them is 90 degrees, a circular signal trajectory is created. When the phase difference between the signals deviates from 90 degrees or the amplitude of the signals changes, the pattern recognition unit 140 creates a signal trajectory with an elliptical or butterfly shape and identifies the signal trajectory as a pattern.

[0106] Therefore, when the pattern recognition unit 140 recognizes a circular signal pattern with respect to the created signal trajectory, the pattern recognition unit 140 can recognize that the ratio between the signal amplitudes of the two signals S11 and S12 is 1:1 and the phase difference between them is 90 degrees.

[0107] Figure 10 This is an exemplary diagram illustrating a Lissajous figure based on the ratio between the phase difference and amplitude of a signal.

[0108] like Figure 10 As shown, even with the same phase difference, the Lissajous figure creates trajectories of different shapes based on the ratio between the amplitudes of the two signals.

[0109] For example, when the phase difference between the filtered signals (first signal S11 and second signal S12) is 0, a signal trajectory with a straight line shape is created when the ratio between the amplitudes of the first signal S11 and the second signal S12 is 1:1; a trajectory with a parabolic shape is created when the ratio between the amplitudes of the first signal S11 and the second signal S12 is 1:2; a signal trajectory with a waveform shape is created when the ratio between the amplitudes of the first signal S11 and the second signal S12 is 1:3; and a signal trajectory with an alpha shape is created when the ratio between the amplitudes of the first signal S11 and the second signal S12 is 2:3.

[0110] Furthermore, when the phase difference between the first signal S11 and the second signal S12 is 45 degrees, a signal trajectory with an elliptical shape is created when the ratio between the amplitudes of the two signals is 1:1, and a signal trajectory with a butterfly shape that is horizontally asymmetrical and vertically symmetrical is created when the ratio between the amplitudes of the first signal S11 and the second signal S12 is 1:2.

[0111] When the phase difference between the first signal S11 and the second signal S12 is 90 degrees, a signal trajectory with a circular shape is created when the ratio between the amplitudes of the first signal S11 and the second signal S12 is 1:1, and a signal trajectory with a butterfly shape that is horizontally and vertically symmetrical is created when the ratio between the amplitudes of the first signal S11 and the second signal S12 is 1:2.

[0112] Furthermore, when the phase difference between the first signal S11 and the second signal S12 is 135 degrees, a basic shape with the same ratio between signal amplitudes as when the phase difference between them is 45 degrees is created, but with a vertically reversed shape. When the phase difference between the first signal S11 and the second signal S12 is 180 degrees, a different shape is created with the same ratio between signal amplitudes as when the phase difference between them is 0 degrees, but with a vertically reversed shape.

[0113] Because even if the phase differences are the same, the trajectory changes according to the signal amplitude, the signal processing unit 150 normalizes the signal amplitude so that the ratio between the amplitudes of the first signal S11 and the second signal S12 is 1:1. The signal processing unit 150 can change the ratio between the signal amplitudes to a desired ratio by using a normalization module (not shown). For example, the signal processing unit 150 performs normalization so that the signal amplitude is 1.

[0114] The signal processing unit 150 creates a converted sensor signal (third signal) by normalizing the trajectory of the signal created by the pattern recognition unit 140.

[0115] The signal processing unit 150 can normalize the signal relative to the trajectory of the signal created based on the principle of Lissajous figures, thereby eliminating variations in the signal trajectory based on the amplitude of the signal and calculating the shape changes caused by the phase difference.

[0116] The signal processing unit 150 stores data in a sorted list, identifies the trajectory (circle) of the signal formed by synthesizing two signals (first signal S1 and second signal S2) that are actually measured and processed, and identifies the error between the trajectories. In this case, the error is not the phase difference between the first signal and the second signal, but rather the error associated with whether the current phase difference between the first signal and the second signal, which have a phase difference, is normal.

[0117] Therefore, according to this disclosure, by excluding the amplitude of the signal and emphasizing the portion associated with the phase difference, the efficiency of detecting errors based on the phase difference can be improved.

[0118] Figure 11A and Figure 11B This is a diagram illustrating anomaly detection based on Lissajous figures for a signal processing apparatus according to embodiments of the present disclosure.

[0119] like Figure 11A As shown, the pattern recognition unit 140 creates a signal trajectory (circle) by converting the filtered and processed first signal S11 (A) and second signal S12 (B).

[0120] The signal processing unit 150 normalizes the trajectory of the created signal. By normalizing the signal, the signal processing unit 150 excludes signal variations based on signal amplitude from the signal trajectory created by the pattern recognition unit 140, leaving only signal variations based on phase difference.

[0121] The signal processing unit 150 detects errors by comparing the trajectory of the detected signal (the transformed sensor signal, the third signal) with a reference signal. The reference signal is a Lissajous figure signal with a predetermined shape, created by two reference signals having a phase difference equal to the phase difference between the two input sensor signals S1 and S2.

[0122] When the phase difference between the two sensor signals S1 and S2 input by the signal input unit 120 is 90 degrees, the signal processing unit 150 calls a reference signal with a 90-degree phase difference. When the phase difference between the two input sensor signals is 45 degrees, the signal processing unit 150 calls a reference signal with a 45-degree phase difference.

[0123] The signal processing unit 150 creates signals based on the first signal and the second signal. Figure 11A The signal trajectory in the middle, that is, the converted sensor signal (third signal) and the reference signal. Figure 11B The error is calculated by comparing the trajectory 51 of the signal in the signal processing unit 150. That is, the signal processing unit 150 compares the converted sensor signal created based on the filtered first and second signals with a reference signal having the same phase difference and calculates the error based on its radian.

[0124] When the converted sensor signal (third signal) is the same as the reference signal, the signal processing unit 150 determines that the phase difference is the same and the error is 0.

[0125] When the converted sensor signal (third signal) differs from the reference signal, the signal processing unit 150 calculates the error based on the difference between them.

[0126] The signal compensation unit 160 compensates for the error between the filtered signals (i.e., the first signal S11 and the second signal S12) based on the error between the converted sensor signal and the reference signal.

[0127] When the error calculated by the signal processing unit 150 is not zero, the signal compensation unit 160 determines that the phase difference between the first signal S11 and the second signal S12 is abnormal, and the signal compensation unit 160 compensates for the error between the two signals based on the calculated error.

[0128] Therefore, the signal compensation unit 160 compensates for the phase difference between the first signal S11 and the second signal S12.

[0129] For example, when the error between the first signal S11 and the second signal S12, which have a 90-degree phase difference calculated by the signal processing unit 150, is not zero, the actual phase difference is not 90 degrees. Therefore, the signal compensation unit 160 compensates for the signal by measuring the error.

[0130] The signal compensation unit 160 compensates for signal errors by performing PI control using the calculated error. The signal compensation unit 160 performs PI control and thus additionally compensates for values ​​normalized by using feedforward.

[0131] The signal compensation unit 160 can more easily and quickly compensate for the phase difference by using the calculated error between the two signals. The signal compensation unit 160 outputs the compensated sensor signals (the fifth signal and the sixth signal).

[0132] When the error calculated by the signal processing unit 150 is 0, the converted sensor signal (third signal) and the reference signal are identical. Therefore, the control unit 110 determines that the filtered signals, namely the first signal S11 and the second signal S12, are normal signals.

[0133] When the first signal S11 and the second signal S12 are normal signals, the control unit 110 calculates the angular velocity based on the first signal S11 and the second signal S12 without performing separate compensation and verification.

[0134] Meanwhile, when the error calculated by the signal processing unit 150 is not zero, the control unit 110 determines that the first signal S11 and the second signal S12 are abnormal.

[0135] When the signal is abnormal, the control unit 110 allows the signal compensation unit 160 to compensate for the signal.

[0136] The control unit 110 verifies whether the signal is properly compensated by applying the compensation sensor signals (the fifth signal and the sixth signal) output from the signal compensation unit 160 to the pattern recognition unit 140 and the signal processing unit 150.

[0137] When the verification is complete, the control unit 110 calculates the angular velocity based on the compensated sensor signals (the fifth signal and the sixth signal).

[0138] Therefore, according to this disclosure, anomalies in sensor signals can be detected using Lissajous figures, and the motor can be controlled by easily and quickly detecting phase differences.

[0139] Although this disclosure has been described with reference to embodiments shown in the accompanying drawings, the embodiments are for illustrative purposes only, and those skilled in the art will understand that various modifications to the embodiments and any other equivalent embodiments are available. Therefore, the true scope of protection of this disclosure should be determined by the appended claims.

Claims

1. A signal processing apparatus, comprising: The signal input unit is configured to receive and digitize two signals with a phase difference; The filter unit is configured to filter two digitized signals and output the filtered first and second signals; The pattern recognition unit is configured to convert the first signal and the second signal into coordinates and create a signal trajectory with a predetermined shape; A signal processing unit is configured to create a transformed sensor signal by normalizing the signal trajectory, calculate an error by comparing the transformed sensor signal with a reference signal, and detect anomalies in the first signal and the second signal, wherein the signal processing unit invokes a reference signal having a phase difference equal to the phase difference between the first signal and the second signal; A signal compensation unit is configured to compensate for the first signal and the second signal based on the error between the converted sensor signal and the reference signal when an anomaly is detected; and The control unit is configured to verify whether the first signal and the second signal are properly compensated by applying the first signal and the second signal compensated by the signal compensation unit to the pattern recognition unit and the signal processing unit, and the control unit is configured to calculate the angular velocity from the compensated first signal and the second signal based on the verification result.

2. The signal processing apparatus according to claim 1, wherein, The filter unit includes a high-pass filter and a low-pass filter, which are connected in series or in parallel to form a notch filter or a band-pass filter to remove noise from the signal and separate the signal in a specific frequency domain.

3. The signal processing apparatus according to claim 1, wherein, The signal processing unit normalizes the signal trajectory created by the pattern recognition unit, such that the ratio between the signal amplitudes of the first signal and the second signal is 1:

1.

4. The signal processing apparatus according to claim 1, wherein, The signal processing unit performs normalization to eliminate variations caused by the signal amplitudes of the first and second signals, and compares variations caused by the phase difference.

5. The signal processing apparatus according to claim 1, wherein, When the comparison between the reference signal and the converted sensor signal indicates that the error between the reference signal and the converted sensor signal is 0, the signal processing unit determines that the first signal and the second signal are normal. Specifically, when the error between the first signal and the second signal is not zero, the signal processing unit determines that the first signal and the second signal are abnormal.

6. The signal processing apparatus according to claim 1, wherein, The pattern recognition unit converts the signal value of the first signal into X-axis coordinates, converts the signal value of the second signal into Y-axis coordinates, and creates the signal trajectory connecting the converted coordinates.

7. The signal processing apparatus according to claim 1, wherein, The pattern recognition unit creates the signal trajectory based on the Lissajous figure and identifies the signal trajectory as a pattern.

8. The signal processing apparatus according to claim 1, wherein, When a compensated first signal and a second signal are input, the pattern recognition unit creates a signal trajectory based on the compensated first signal and the second signal. The signal processing unit verifies the compensated signal by calculating the error again by comparing the signal trajectory created based on the compensated first and second signals with the reference signal.

9. The signal processing apparatus according to claim 1, wherein, When the error is 0, the control unit calculates the angular velocity based on the first signal and the second signal, without verifying the signals. When the error is not zero, the control unit allows the signal compensation unit to compensate for the error.

Images