Method for suppressing influence of air disturbance on collimated light beam in laser collimation measurement

Abstract

The invention discloses a method for suppressing influence of air disturbance on collimated light beams in laser collimation measurement, which has the characteristic of adaptive decomposition, does not need to understand signal characteristics in advance, does not need to specify a fixed primary function and a window function in the decomposition process, and is suitable for analyzing and representing non-periodic and non-stable laser collimation signals. Compared with a traditional air disturbance influence suppression method, the empirical mode decomposition and least square combined laser collimation air disturbance influence suppression method provided by the invention has remarkable advantages.

Description

Technical Field

[0001] The invention relates to the field of optical and electronic manufacturing precision measurement, and in particular to a method for suppressing the influence of air disturbance on a collimated light beam in laser collimation measurement. Background Art

[0002] Laser collimated beams are widely used in precision measurement fields such as straightness and angles. Their stability is the premise and guarantee for achieving stable measurement based on laser collimation. However, in addition to the laser drift caused by the thermal deformation of the laser itself, the random disturbance of the air in the laser propagation environment is also an important factor affecting the stability of the laser collimated straight line reference.

[0003] In view of the influence of air disturbance on the laser straight line reference, the conventional methods currently include optical path shielding method, dual beam method, piezoelectric drive compensation method, digital filtering method, etc. The optical path shielding method is to protect the propagation optical path by using a shielding tube to isolate the interference of the environment on the laser beam, but this method has poor flexibility and is only suitable for measurement in fixed scenes. The dual beam method mainly monitors the jitter of the laser alignment straight line reference caused by air disturbance in real time, and uses the monitoring results of the reference optical path to effectively compensate the measurement optical path, thereby eliminating the influence of air disturbance on the straight line reference. However, although the measurement optical path and the compensation optical path are very close, there is still a certain error, so its compensation accuracy is limited to a certain extent. Piezoelectric drive control compensation can effectively eliminate the influence of air disturbance on the stability of the laser alignment straight line reference. However, this method has high requirements for piezoelectric control and requires fast tracking response speed and accuracy, otherwise it will affect the laser drift compensation effect, thereby increasing the control difficulty and device cost. The digital filtering method uses sliding mean filtering, Kalman filtering and other methods to filter the laser alignment signal from the perspective of the laser alignment signal. However, in reality, the laser collimation signal obtained by the detector under air disturbance is very complex. When there is air disturbance, the laser collimation signal will contain random noise interference and non-stationary noise interference, and the collimation signal components are complex, so simple digital filtering is difficult to effectively filter out the interference of air disturbance on the collimation signal.

[0004] Therefore, a method for suppressing the influence of air disturbance on the collimated light beam in laser collimation measurement solves the above problems. Summary of the invention

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A method for suppressing the influence of air disturbance on a collimated light beam in laser collimation measurement comprises the following steps:

[0007] 1. Use a photosensitive detector to obtain the laser alignment signal;

[0008] 2. Decompose the laser collimation signal using empirical mode decomposition to obtain intrinsic mode function components and trend term components;

[0009] 3. For the decomposed intrinsic modal components and trend term components, use the correlation calculation method to find the first local minimum value;

[0010] 4. The component before the first local minimum can be considered as a noise signal, so the threshold is set to 0, and the signal reconstruction is performed on the component and trend item after the local minimum;

[0011] 5. Divide the reconstructed signal into three parts, and extend the middle part of the signal to both sides to cover the original reconstructed signal;

[0012] 6. Perform least square fitting on the reconstructed signal and finally obtain the laser alignment signal without air disturbance.

[0013] In step 1, the laser alignment signal obtained by the detector is recorded as:

[0014] f(t)=f DC (t)+f1(t)+f2(t)+f3(t) (1)

[0015] Where t is the signal sampling time, f(t) is the laser alignment signal detected by the detector under the influence of air disturbance in the linear motion geometric error measurement system; f DC (t) is the ideal laser collimation signal, which is a constant and represents the collimation position of the light beam; f1(t) and f2(t) represent the random and non-periodic disturbance signals caused by air disturbances, respectively, with random amplitude and frequency changes; f3(t) is a fixed-frequency interference signal, which is caused by mechanical vibration, air disturbance, environmental vibration or other factors.

[0016] In step 2, the maximum value e of the laser alignment measurement signal obtained in step 1 is obtained respectively. + (t) and the minimum value e - (t), and use cubic spline interpolation fitting to fit the maximum and minimum curves respectively, obtain the upper and lower envelopes and calculate their mean m(t), the formula is:

[0017]

[0018] After finding the mean of the upper and lower envelopes, the difference between the original signal and the mean is further calculated.

[0019] h(t)=f(t)-m(t) (3)

[0020] According to the IMF judgment criteria, if the conditions are met, the EMD decomposition of the laser collimation signal is completed, otherwise the above steps are repeated.

[0021] Furthermore, the constraints on the IMF results are as follows:

[0022]

[0023] Where T is the signal length, ε is 0.2-0.3, and when SD is less than this value, the screening is stopped.

[0024] Furthermore, based on the first IMF component separated from the IMF, the first IMF component is subtracted from the original signal to obtain the trend term component, which is recorded as r1(t). The calculation formula is:

[0025] r1(t)=f(t)-h1(t) (5)

[0026] The trend term component is used as a new signal, and the above calculation method and steps are repeated for separation and judgment to obtain a series of IMFs until the last trend term component r n If (t) exhibits a constant term or a monotonic function, the calculation can be stopped.

[0027] Furthermore, the laser collimation signal can be decomposed into a finite IMF component from high frequency to low frequency and a trend term with a frequency of 0 Hz. The calculation formula is:

[0028]

[0029] Furthermore, in step 3, the laser collimation signal is decomposed based on EMD, and the random and non-periodic laser collimation signal is processed by EMD to obtain a series of IMFs components and r n By comparing the correlation between the IMF component and the original signal, the minimum extreme point between the two IMF components dominated by noise and the effective signal can be found; the IMF component before this extreme point is considered to be the main component of the noise. The effective component and the noise component in the laser collimation signal are separated, and the formula for calculating the correlation coefficient is as follows:

[0030]

[0031] In the formula,

[0032] According to the above formula (7), the first minimum value of R corresponds to IMF k , then IMF k Defined as the dividing point between high-frequency components and low-frequency components; high-frequency components are considered to be noise, and hard threshold processing is used, that is, IMF1 to IMF k Components are set to zero.

[0033] Furthermore, in step 4, the IMF k+1 To IMF n The signal is reconstructed by the following formula:

[0034]

[0035] Where N is the signal length.

[0036] Furthermore, in step 5, in order to eliminate the endpoint fluctuation effect of the reconstructed signal f'(t), the mirror extension method is used to process the reconstructed signal. This method takes the center of the signal length as the extension point and performs mirror extension to the left and right sides to eliminate the endpoint effect problem caused by the EMD method and improve the accuracy of signal reconstruction. In this way, the reconstructed signal f"(t) after mirror extension processing can be obtained.

[0037] Furthermore, in step 6, the baseline correction is performed on the signal reconstructed by the image extension, that is, the laser alignment signal is estimated by the least squares fitting method, so as to obtain a more reliable laser alignment signal. The final processed laser alignment signal can be expressed as:

[0038]

[0039] Due to the adoption of the above technical solution, the present invention has the following technical advances compared with the prior art:

[0040] This method has the characteristics of adaptive decomposition, does not require prior knowledge of signal characteristics, and does not require the specification of fixed basis functions and window functions during the decomposition process. It is suitable for analyzing and representing non-periodic and non-stationary laser collimation signals.

[0041] Compared with the traditional method for suppressing the influence of air disturbance, the laser alignment air disturbance suppression method combining empirical mode decomposition and least squares proposed in the present invention has significant advantages.

[0042] Adaptability: As an adaptive data decomposition method, EMD can automatically adjust the number of decomposed modes and the frequency range according to the characteristics of the data, and can better adapt to different types of signals and disturbances.

[0043] Accuracy: EMD can decompose the signal into multiple intrinsic mode functions (IMFs). These IMFs represent components of different frequencies and amplitudes in the signal, which can more accurately describe the characteristics of the signal and separate disturbances of different frequencies.

[0044] Noise resistance: EMD has a certain degree of noise resistance when processing signals. It can effectively suppress the impact of noise on signal decomposition and reconstruction and reduce the interference of noise on the system.

[0045] Least squares optimization: Through the least squares method, the reconstruction of the signal can be optimized, making the reconstructed signal more consistent with the actual situation, reducing errors and distortions, and improving the accuracy and reliability of data processing.

[0046] Comprehensive performance: Combining EMD and the least squares method can fully utilize the advantages of both to achieve more effective signal processing and reconstruction, improve the laser alignment system's ability to suppress air disturbances, and thus improve the system's stability and performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0048] Figure 1 A schematic diagram of a method for suppressing the influence of air disturbance on the collimated light beam in laser collimation measurement according to the present invention;

[0049] Figure 2 Schematic diagram of the effect of air disturbance on laser alignment signal in the present invention;

[0050] Figure 3 It is a decomposition flow chart of the EMD laser collimation signal in the present invention;

[0051] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0052] The following embodiments may help those skilled in the art to more fully understand the present invention, but shall not limit the present invention in any way.

[0053] A method for suppressing the influence of air disturbance on a collimated light beam in laser collimation measurement comprises the following steps:

[0054] 1. Use a photosensitive detector to obtain the laser alignment signal;

[0055] 2. Decompose the laser collimation signal using empirical mode decomposition to obtain a set of intrinsic mode function components and trend term components;

[0056] 3. For the decomposed intrinsic modal components and trend term components, use the correlation calculation method to find the first local minimum value;

[0057] 4. The component before the first local minimum can be considered as a noise signal, so the threshold is set to 0, and the signal reconstruction is performed on the component and trend item after the local minimum;

[0058] 5. Divide the reconstructed signal into three parts, and extend the middle part of the signal to both sides to cover the original reconstructed signal;

[0059] 6. Perform least square fitting on the reconstructed signal and finally obtain the laser alignment signal without air disturbance.

[0060] The photoelectric signal obtained by the photosensitive detector includes: the ideal laser alignment signal (frequency is 0Hz), the random disturbance signal caused by air disturbance and the non-stationary disturbance signal, and some natural frequency vibration interference in the environment will also affect the detector's measurement of the aligned beam signal. Therefore, under the influence of air disturbance and environmental vibration, in step 1, the laser alignment signal obtained by the detector is recorded as:

[0061] f(t)=f DC (t)+f1(t)+f2(t)+f3(t) (1)

[0062] Where t is the signal sampling time, f(t) is the laser alignment signal detected by the detector under the influence of air disturbance in the linear motion geometric error measurement system; f DC (t) is the ideal laser collimation signal, which is a constant and represents the collimation position of the light beam; f1(t) and f2(t) represent the random and non-periodic disturbance signals caused by air disturbances, respectively, with random amplitude and frequency changes; f3(t) is a fixed-frequency interference signal, which is caused by mechanical vibration, air disturbance, environmental vibration or other factors.

[0063] Based on the above analysis of the characteristics of the laser alignment signal, the laser alignment signal obtained by the detector is random and non-stationary. Conventional digital filtering methods are difficult to suppress the influence of air disturbance on the laser alignment straight line reference. The empirical mode decomposition (EMD) method does not require prior knowledge of the signal characteristics, and does not require the specification of fixed basis functions and window functions during the decomposition process. It is suitable for analyzing and representing non-stationary signals.

[0064] IMF is a sequence whose instantaneous frequency has physical meaning and satisfies the following conditions:

[0065] (a) The number of zero-crossing points and extreme value points of the sequence on the time axis is the same, or differs by at most one;

[0066] (b) The sequence is locally symmetric about the horizontal axis, and the mean of the upper and lower fitting envelopes determined by the maximum and minimum points is zero.

[0067] For the laser alignment measurement signal obtained in step 1, calculate its maximum value e + (t) and the minimum value e - (t), and use cubic spline interpolation fitting to fit the maximum and minimum curves respectively, obtain the upper and lower envelopes and calculate their mean m(t), the formula is:

[0068]

[0069] After finding the mean of the upper and lower envelopes, the difference between the original signal and the mean is further calculated.

[0070] h(t)=f(t)-m(t) (3)

[0071] According to the IMF judgment criteria, if the conditions are met, the EMD decomposition of the laser collimation signal is completed, otherwise the above steps are repeated.

[0072] When EMD decomposes the laser collimation signal, the IMF does not meet the definition ideally. In order to make the IMF have physical meaning in amplitude and frequency, the number of IMF iterations needs to be limited to prevent excessive decomposition from causing unstable IMF results. The limit criteria are as follows:

[0073]

[0074] Where T is the signal length, ε is 0.2-0.3, and when SD is less than this value, the screening is stopped.

[0075] Based on the first IMF component separated from the IMF, the trend term component is obtained by subtracting the first IMF component from the original signal, which is denoted as r1(t). The calculation formula is:

[0076] r1(t)=f(t)-h1(t) (5)

[0077] The trend term component is used as a new signal, and the above calculation method and steps are repeated for separation and judgment to obtain a series of IMFs until the last trend term component r n If (t) exhibits a constant term or a monotonic function, the calculation can be stopped.

[0078] The laser collimation signal can be decomposed into a finite IMF component from high frequency to low frequency and a trend term with a frequency of 0 Hz. The calculation formula is:

[0079]

[0080] Based on EMD, the laser collimation signal is decomposed and the random and non-periodic laser collimation signal is processed by EMD to obtain a series of IMFs components and r nBy comparing the correlation between the IMF component and the original signal, the minimum extreme point between the two IMF components dominated by noise and the effective signal can be found; the IMF component before this extreme point is considered to be the main component of the noise. The effective component and the noise component in the laser collimation signal are separated, and the formula for calculating the correlation coefficient is as follows:

[0081]

[0082] In the formula,

[0083] According to the above formula (7), the first minimum value of R corresponds to IMF k , then IMF k Defined as the dividing point between high-frequency components and low-frequency components; high-frequency components are considered to be noise, and hard threshold processing is used, that is, IMF1 to IMF k The components are set to zero. k+1 To IMF n The signal is reconstructed by the following formula:

[0084]

[0085] Where N is the signal length.

[0086] In order to eliminate the endpoint fluctuation effect of the reconstructed signal f'(t), the mirror extension method is used to process the reconstructed signal. This method takes the center of the signal length as the extension point and performs mirror extension to the left and right sides to eliminate the endpoint effect problem caused by the EMD method and improve the accuracy of signal reconstruction. In this way, the reconstructed signal f"(t) after mirror extension processing can be obtained. Finally, the baseline correction is performed on the signal reconstructed by mirror extension, that is, the laser alignment signal is estimated by the least squares fitting method to obtain a more reliable laser alignment signal. The final processed laser alignment signal can be expressed as:

[0087]

[0088] The above generally describes the present invention in detail, but it is obvious to a person skilled in the art that some modifications or improvements can be made to the present invention. Therefore, modifications or improvements that do not depart from the spirit of the present invention are within the scope of protection of the present invention.

Claims

1. A method for suppressing the influence of air disturbance on the collimated light beam in laser collimation measurement, characterized in that: The following steps are involved: (1) Obtaining laser alignment signals using a photosensitive detector; (2) Decomposing the laser collimation signal using empirical mode decomposition to obtain intrinsic mode function components and trend term components; (3) Using the correlation calculation method, find the first local minimum value for the decomposed intrinsic modal components and trend term components; (4) The component before the first local minimum is considered to be a noise signal, so the threshold is set to 0, and the signal reconstruction is performed on the component after the local minimum and the trend term; (5) Divide the reconstructed signal into three parts, and extend the middle part of the signal to both sides to cover the original reconstructed signal; (6) Perform least square fitting on the reconstructed signal and finally obtain the laser alignment signal without air disturbance.

2. The method for suppressing the influence of air disturbance on the collimated light beam in laser collimation measurement according to claim 1, characterized in that: In step 1, the laser alignment signal obtained by the detector is recorded as: f(t)=f DC (t)+f1(t)+f2(t)+f3(t) Where t is the signal sampling time, f(t) is the laser alignment signal detected by the detector under the influence of air disturbance in the linear motion geometric error measurement system; f DC (t) is the ideal laser collimation signal, which is a constant and represents the collimation position of the light beam; f1(t) and f2(t) represent the random and non-periodic disturbance signals caused by air disturbances, respectively, with random amplitude and frequency changes; f3(t) is a fixed frequency interference signal.

3. The method for suppressing the influence of air disturbance on the collimated light beam in laser collimation measurement according to claim 2, characterized in that: In step 2, the maximum value e of the laser alignment measurement signal obtained in step 1 is obtained respectively. + (t) and the minimum value e - (t), and use cubic spline interpolation fitting to fit the maximum and minimum curves respectively, obtain the upper and lower envelopes and calculate their mean m(t), the formula is: After finding the mean of the upper and lower envelopes, further find the difference between the original signal and the mean; h(t)=f(t)-m(t) According to the IMF judgment criteria, if the conditions are met, the EMD decomposition of the laser collimation signal is completed, otherwise the above steps are repeated.

4. The method for suppressing the influence of air disturbance on the collimated light beam in laser collimation measurement according to claim 3, characterized in that: The constraints on the IMF results are as follows: In the formula, T is the signal length, ε is 0.2-0.3, and when SD is less than the value of ε, the screening is stopped; Based on the first IMF component separated from the IMF, the trend term component is obtained by subtracting the first IMF component from the original signal, which is denoted as r1(t). The calculation formula is: r1(t)=f(t)-h1(t) The trend term component is used as a new signal, and the above calculation method and steps are repeated for separation and judgment to obtain a series of IMFs until the last trend term component r n If (t) exhibits a constant term or a monotonic function, the calculation can be stopped.

5. The method for suppressing the influence of air disturbance on the collimated light beam in laser collimation measurement according to claim 4, characterized in that: The laser collimation signal can be decomposed into a finite IMF component from high frequency to low frequency and a trend term with a frequency of 0 Hz. The calculation formula is:

6. The method for suppressing the influence of air disturbance on the collimated light beam in laser collimation measurement according to claim 5, characterized in that: In step 3, the laser collimation signal is decomposed based on EMD. After the random and non-periodic laser collimation signal is processed by EMD, a series of IMFs components and r are obtained. n By comparing the correlation between the IMF component and the original signal, the minimum extreme point between the two IMF components dominated by noise and the effective signal can be found; the IMF component before the extreme point is considered to be the main component of the noise; the effective component and the noise component in the laser collimation signal are separated, and the formula for calculating the correlation coefficient is as follows: In the formula, According to the first minimum value of R, it corresponds to IMF k , then IMF k Defined as the dividing point between high-frequency components and low-frequency components; high-frequency components are considered to be noise, and hard threshold processing is used, that is, IMF1 to IMF k Components are set to zero.

7. The method for suppressing the influence of air disturbance on the collimated light beam in laser collimation measurement according to claim 6, characterized in that: Step 4: IMF k+1 To IMF n The signal is reconstructed by the following formula: Where N is the signal length.

8. The method for suppressing the influence of air disturbance on the collimated light beam in laser collimation measurement according to claim 7, characterized in that: In step 5, in order to eliminate the endpoint fluctuation effect of the reconstructed signal f'(t), the mirror extension method is used to process the reconstructed signal. The center of the signal length is taken as the extension point, and mirror extension is performed to the left and right sides to eliminate the endpoint effect problem caused by the EMD method, and the reconstructed signal f"(t) after mirror extension processing is obtained.

9. The method for suppressing the influence of air disturbance on the collimated light beam in laser collimation measurement according to claim 8, characterized in that: In step 6, the baseline correction is performed on the signal reconstructed by the image extension, that is, the laser alignment signal is estimated by the least squares fitting method, so as to obtain a more reliable laser alignment signal. The final processed laser alignment signal can be expressed as:

Images