Large rotary equipment axial profile measurement method and device based on multi-level median mixed filtering, computer and storage medium

Abstract

A method, apparatus, computer, and storage medium for measuring the axial profile of large rotating equipment based on multi-level median hybrid filtering are disclosed, relating to the field of surface profile measurement. This method solves the problem of insufficient accuracy in end-face measurement of large rotating equipment. The method includes: establishing an end-face measurement model; fixing the large rotating equipment on a self-aligning and tilting table; adjusting the self-aligning and tilting table, using inductive sensors to detect the eccentricity and eccentricity angle, tilt and tilt angle, and the profile data of the reference surface end face and the measurement surface end face relative to the rotating spindle; when the eccentricity and tilt of the reference surface are less than 5μm, the self-aligning and tilting is terminated, and the eccentricity and eccentricity angle of the reference surface are obtained; error separation is performed based on the end-face measurement model and the eccentricity and eccentricity angle of the reference surface, obtaining the error-separated profile measurement data of the reference surface end face and the measurement surface end face; and multi-level median hybrid filtering is used to process the data to obtain the equipment end-face profile data. This method is applied to the field of accurate measurement of the axial profile of the assembly surface of large rotating equipment.

Description

Technical Field

[0001] This invention relates to the field of surface profile measurement, and more particularly to a method for measuring the axial profile of large rotating equipment based on multi-level median hybrid filtering. Background Technology

[0002] Large rotating equipment, including aerospace engines, aircraft engines, missile engines, and gas turbines, are core equipment in defense industries and other fields. Precision measurement of large rotating equipment is fundamental to ensuring accurate assembly. Because large rotating equipment operates at high speeds, assembly deviations are rapidly amplified, potentially causing significant equipment damage. Therefore, improving the measurement system for large rotating equipment is crucial for ensuring the overall operation of the machine. Currently, the main measurement indicators for large rotating equipment are the axial perpendicularity and radial coaxiality of its assembly surfaces. In particular, the end faces of the assembly surfaces will progressively transmit rotor eccentricity and tilting errors, ultimately affecting the overall performance of the large rotating equipment. Therefore, a measurement method is urgently needed to solve the problem of accurate measurement of the assembly surfaces of large rotating equipment. Summary of the Invention

[0003] This invention proposes an axial measurement model that considers five errors: eccentricity error, tilt error, sensor probe radius error, probe offset error, and probe support rod tilt error of large rotating equipment. At the same time, it uses a hybrid multi-level median filtering to purify the final measurement data, thus solving the problem of insufficient accuracy in end face measurement of large rotating equipment.

[0004] This invention provides a method for measuring the axial profile of large rotating equipment based on multi-level median hybrid filtering, the measurement method comprising:

[0005] An axial measurement model is established based on the geometric coupling characteristics of the eccentricity error, tilt error, sensor probe radius error, probe offset error, and probe support rod tilt error of the large rotating equipment to be measured.

[0006] The large rotary equipment is fixed to the self-aligning and tilting table;

[0007] Adjust the centering and tilting table, and use inductive sensors to detect the eccentricity and eccentricity angle, tilt and tilt angle, and the contour data of the reference surface and the measuring surface of the large rotary equipment relative to the rotary spindle. When the eccentricity and tilt of the reference surface are less than 5μm, the centering and tilting is stopped, and the eccentricity and eccentricity angle of the reference surface of the large rotary equipment are obtained.

[0008] Based on the eccentricity and eccentricity angle of the axial measurement model and the reference surface of the large rotary equipment, error separation is performed on the contour data of the reference surface end face and the measurement surface end face, and the measurement data of the reference surface end face and the measurement surface end face after error separation are obtained.

[0009] Multi-level median hybrid filtering is used to process the contour measurement data of the reference surface end face and the measurement surface end face after error separation, so as to obtain pure axial contour data of large rotary equipment.

[0010] Furthermore, a preferred embodiment is also provided, wherein obtaining the eccentricity and eccentricity angle of the reference plane of the large rotary equipment specifically involves:

[0011]

[0012] Where e0 is the eccentricity of the reference plane of the large rotary equipment, e m It is an axial end face eccentricity, z m The height of the end face is α0, and the eccentricity angle of the reference plane is α. m It is the eccentric angle, γ m The angle of inclination of the geometric axis of the large rotary equipment relative to the main shaft of rotation is given by (l', m', n').

[0013] Furthermore, a preferred embodiment is also provided, wherein the acquisition of the contour measurement data of the reference surface end face and the measurement surface end face after error separation specifically includes:

[0014]

[0015] Where r0 is the end-face sampling radius, θ' i β is the actual sampling angle of the end face of the large rotating equipment, where β is the angle between the tilt direction and the measurement direction, and Δz is the actual sampling angle. i It refers to the end face runout, and r is the deviation caused by the sensor's probe radius error. It is the tilt angle of the probe support rod, θ i For the ideal sampling angle of the end face of large rotating equipment, d m It is the probe offset error, θ i ′ is the actual sampling angle.

[0016] Furthermore, a preferred embodiment is provided, wherein the multi-level median hybrid filtering process for the error-separated reference surface end face and measurement surface end face contour measurement data specifically includes:

[0017] Normalize the measurement data of the reference surface end face and the measurement surface end face after error separation to obtain graphical sampled contour data;

[0018] The graphical sampled contour data is processed using a multi-stage median blending filter, which has four sampling windows:

[0019]

[0020] stn∈[-d,d]

[0021] Where K(X) represents a sequence centered at X, K1(X) represents a sequence of (X(x1-d,x2), X(x1-d+1,x2),...,X(x1+d,x2)), K2(X) represents a sequence of (X(x1,x2-d), X(x1,x2-d+1),...,X(x1,x2+d)), and K3(X) represents a sequence of (X(x1-d,x2-d), X(x1-d+1),...,X ... Let K4(X) represent the sequence (X(x1+d,x2+d),X(x1+d-1,x2+d-1),...,X(x1-d,x2-d)), where X(x1,x2) represents the coordinates of a point on the image after the contour measurement data is graphically represented, n represents an integer in the range from -d to d, and d represents the size of the filter window.

[0022] This invention also provides an axial profile measuring device for large rotating equipment based on multi-level median hybrid filtering. The measuring device includes a self-aligning and tilting platform, multiple inductive sensors, and a computer. The computer controls the adjustment actions of the self-aligning and tilting platform. The multiple sensors are fixed on the self-aligning and tilting platform to detect the physical parameters of the large rotating equipment to be measured and send the detection results to the computer. The computer has an embedded measurement module implemented with computer software. The measurement module includes:

[0023] The end-face measurement model establishment unit is used to establish an end-face measurement model based on the geometric coupling characteristics of the eccentricity error, tilt error, sensor probe radius error, probe offset error, and probe support rod tilt error of large rotating equipment.

[0024] The self-aligning and tilting platform adjustment unit is used to send control signals to the self-aligning and tilting platform, and to control the adjustment of the eccentricity and eccentricity angle, tilt and tilt angle of the reference surface of the large rotary equipment relative to the rotary spindle. When the eccentricity and tilt of the reference surface are less than 5μm, the self-aligning and tilting is terminated.

[0025] The data acquisition unit is used to acquire data on the eccentricity and eccentricity angle, tilt and tilt angle, and the contour data of the reference plane and the measuring plane relative to the main shaft of the large rotary equipment, as well as the eccentricity and eccentricity angle of the reference plane.

[0026] The data processing unit is used to separate the error of the reference surface end face and the measurement surface end face contour data based on the eccentricity and eccentricity angle of the end face measurement model and the reference surface of the large rotary equipment, and to obtain the reference surface end face and measurement surface end face contour measurement data after error separation.

[0027] The end face contour data acquisition unit is used for multi-level median mixed filtering of the reference surface end face and measurement surface end face contour measurement data after error separation to obtain clean end face contour data of large rotating equipment.

[0028] Furthermore, a preferred embodiment is also provided, wherein the data acquisition unit specifically comprises:

[0029]

[0030] Where e0 is the eccentricity of the reference plane of the large rotary equipment, e m It is an axial end face eccentricity, z m The height of the end face is α0, and the eccentricity angle of the reference plane is α. m It is the eccentric angle, γ m The angle of inclination of the geometric axis of the large rotary equipment relative to the main shaft of rotation is given by (l', m', n').

[0031] Furthermore, a preferred embodiment is also provided, wherein the data processing unit specifically comprises:

[0032]

[0033] Where r0 is the end-face sampling radius, θ' i β is the actual sampling angle of the end face of the large rotating equipment, where β is the angle between the tilt direction and the measurement direction, and Δz is the actual sampling angle. i It refers to the end face runout, and r is the deviation caused by the sensor's probe radius error. It is the tilt angle of the probe support rod, θ i For the ideal sampling angle of the end face of large rotating equipment, d m It is the probe offset error, θ i ′ is the actual sampling angle.

[0034] Furthermore, a preferred embodiment is also provided, wherein the end face contour data acquisition unit specifically comprises:

[0035] The graphical sampling contour data module is used to normalize the contour measurement data of the reference surface end face and the measurement surface end face after error separation, and to obtain graphical sampling contour data.

[0036] A multi-level median blending filter module is used to process graphical sampled contour data using multi-level median blending filters. The multi-level median blending filter has four sampling windows:

[0037]

[0038] stn∈[-d,d]

[0039] Where K(X) represents a sequence centered at X, K1(X) represents a sequence of (X(x1-d,x2), X(x1-d+1,x2),...,X(x1+d,x2)), K2(X) represents a sequence of (X(x1,x2-d), X(x1,x2-d+1),...,X(x1,x2+d)), and K3(X) represents a sequence of (X(x1-d,x2-d), X(x1-d+1),...,X ... Let K4(X) represent the sequence (X(x1+d,x2+d),X(x1+d-1,x2+d-1),...,X(x1-d,x2-d)), where X(x1,x2) represents the coordinates of a point on the image after the contour measurement data is graphically represented, n represents an integer in the range from -d to d, and d represents the size of the filter window.

[0040] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the axial profile measurement method for large rotating equipment based on multi-level median hybrid filtering as described in any of the preceding claims.

[0041] The present invention also provides a computer-readable storage medium for storing a computer program that executes the method for measuring the axial profile of a large rotating equipment based on multi-level median hybrid filtering as described in any one of the claims.

[0042] The advantages of this invention are:

[0043] The measurement method described in this invention considers an axial measurement model that addresses five errors in large rotating equipment: eccentricity error, tilt error, probe radius error, probe offset error, and probe support rod tilt error. Simultaneously, it utilizes a multi-level median hybrid filter, combining median filtering, mean filtering, and extreme value filtering, to purify the final measurement data, ensuring the measurement accuracy of large rotating equipment. This solves the problems of inaccurate perpendicularity measurement of the axial end face profile of large rotating equipment and measurement data noise.

[0044] Using a coaxiality standard (standard value of 12μm) as the measurement target, the measurement method of the present invention can improve the accuracy of end face perpendicularity assessment by 2 to 4μm compared with the end face perpendicularity assessment of the prior art.

[0045] The measurement method described in this invention, which involves first planarization, then mean filtering, and finally median filtering, offers the advantage of preserving the inherent characteristics of the contour data while visualizing the distribution of contour features through planarization. Utilizing multi-level median filtering to process the visualized data maximizes the correlation between data points during filtering, ensuring effective filtering. The optimal filtering sequence—mean filtering followed by median filtering—was determined through multiple experimental analyses, providing the most effective filtering of the contour data, achieving data "purification," and guaranteeing the final assessment accuracy.

[0046] The measuring device described in this invention is based on the aforementioned measuring method. It also considers an axial measurement model that addresses five errors in large rotating equipment: eccentricity error, tilt error, probe radius error, probe offset error, and probe support rod tilt error. Furthermore, it utilizes a multi-level median hybrid filter, combining median filtering, mean filtering, and extreme value filtering, to purify the final measurement data and ensure the measurement accuracy of large rotating equipment. This solves the problems of inaccurate perpendicularity measurement of the axial end face profile of large rotating equipment and measurement data noise.

[0047] This invention is applicable to the field of precise measurement of the axial profile of the assembly surface of large rotating equipment. Attached Figure Description

[0048] Figure 1 This is a diagram showing the axial error coupling relationship of the assembly surface of the large rotary equipment as described in Embodiment Eleven of the present invention.

[0049] Figure 2 This is a grayscale image corresponding to different contour sampling data of the large rotary equipment described in Embodiment 11 of the present invention, wherein... Figure 2 (a) is a data diagram of a large single-tilt slewing machine. Figure 2 (b) Data diagram of large saddle surface rotary equipment;

[0050] Figure 3 This is a flowchart of the multi-stage median hybrid filtering process described in Embodiment Eleven of the present invention. Detailed Implementation

[0051] To make the technical solutions and advantages of the present invention clearer, several embodiments of the present invention will now be described in further detail with reference to the accompanying drawings. However, the embodiments described below are only a few preferred embodiments of the present invention and are not intended to limit the invention.

[0052] Implementation Method 1: The axial profile measurement method for large rotating equipment based on multi-level median hybrid filtering described in this implementation method includes:

[0053] An end-face measurement model is established based on the geometric coupling characteristics of the eccentricity error, tilt error, sensor probe radius error, probe offset error, and probe support rod tilt error of the large rotating equipment to be measured.

[0054] The large rotary equipment is fixed to the self-aligning and tilting table;

[0055] Adjust the centering and tilting table, and use inductive sensors to detect the eccentricity and eccentricity angle, tilt and tilt angle, and the contour data of the reference surface and the measuring surface of the large rotary equipment relative to the rotary spindle. When the eccentricity and tilt of the reference surface are less than 5μm, the centering and tilting is stopped, and the eccentricity and eccentricity angle of the reference surface of the large rotary equipment are obtained.

[0056] Based on the end face measurement model and the eccentricity and eccentricity angle of the reference surface of the large rotary equipment, the error separation of the reference surface end face and the measurement surface end face contour data is performed, and the error-separated reference surface end face and measurement surface end face contour measurement data are obtained.

[0057] Multi-level median hybrid filtering is used to process the measurement data of the reference surface end face and the measurement surface end face after error separation, so as to obtain clean end face contour data of large rotating equipment.

[0058] In this embodiment, the eccentricity error, tilt error, sensor probe radius error, and probe offset error of the large rotating equipment are the measurement errors of the large rotating equipment. The centering and tilting table is mainly a posture adjustment mechanism for the rotating components, which makes the geometric center of the reference plane of the rotating equipment as close as possible to the rotation axis of the rotating spindle. It has four adjustment knobs, two for adjusting the eccentricity of the rotating equipment and two for adjusting the tilt of the rotating components. A tripod chuck is used to fix the large rotating equipment to ensure its stability; and inductive sensors are used to collect data. In practical applications, the reference plane cross-section and the measurement plane cross-section of the large rotating equipment are measured using lever-type inductive sensors, while the reference plane end face and the measurement plane end face are measured using telescopic inductive sensors. The measurement range of both the lever-type and telescopic inductive sensors is ±0.5mm, and the measurement resolution is 0.1μm.

[0059] Implementation Method Two: This implementation method further defines the axial profile measurement method for large rotating equipment based on multi-level median hybrid filtering described in Implementation Method One. Specifically, obtaining the eccentricity and eccentricity angle of the reference plane of the large rotating equipment involves:

[0060]

[0061] Where e0 is the eccentricity of the reference plane of the large rotary equipment, e m It is an axial end face eccentricity, zm The height of the end face is α0, and the eccentricity angle of the reference plane is α. m It is the eccentric angle, γ m The angle of inclination of the geometric axis of the large rotary equipment relative to the main shaft of rotation is given by (l', m', n').

[0062] Specifically, the eccentricity e0 of the reference plane of large slewing equipment is mainly caused by the misalignment of the measurement center of the reference plane and the measurement rotation center. The above calculation can accurately obtain the eccentricity and eccentricity angle of the reference plane of large slewing equipment.

[0063] Implementation Method 3: This implementation method further defines the axial profile measurement method for large rotating equipment based on multi-level median hybrid filtering described in Implementation Method 1. Specifically, the acquisition of the profile measurement data of the reference surface end face and the measurement surface end face after error separation is as follows:

[0064]

[0065] Where r0 is the end-face sampling radius, θ' i β is the actual sampling angle of the end face of the large rotating equipment, where β is the angle between the tilt direction and the measurement direction, and Δz is the actual sampling angle. i It refers to the end face runout, and r is the deviation caused by the sensor's probe radius error. It is the tilt angle of the probe support rod, θ i For the ideal sampling angle of the end face of large rotating equipment, d m It is the probe offset error, θ i ′ is the actual sampling angle.

[0066] In practical operation, the axial end face runout of large rotating equipment will be affected by the tilt angle at different sampling angles, resulting in deviations. The errors caused by the measuring device itself mainly include the probe radius error of the measuring sensor, the probe support rod tilt error, and the probe position offset error. The deviation caused by the sensor's probe radius error is its radius *r*, which is coupled with the probe support rod tilt error, and the tilt angle of the probe support rod... This will lead to amplified axial end face runout. The probe position offset error is caused by the probe's measurement direction not being collinear with the line connecting the measuring point to the center of rotation; the probe offset error d... m It is usually estimated using a step-by-step estimation method.

[0067] Implementation Method Four: This implementation method further defines the axial profile measurement method for large rotating equipment based on multi-level median hybrid filtering described in Implementation Method One. Specifically, the multi-level median hybrid filtering processes the profile measurement data of the reference surface end face and the measurement surface end face after error separation.

[0068] Normalize the measurement data of the reference surface end face and the measurement surface end face after error separation to obtain graphical sampled contour data;

[0069] The graphical sampled contour data is processed using a multi-stage median blending filter, which has four sampling windows:

[0070]

[0071] stn∈[-d,d]

[0072] Where K(X) represents a sequence centered at X, K1(X) represents a sequence of (X(x1-d,x2), X(x1-d+1,x2),...,X(x1+d,x2)), K2(X) represents a sequence of (X(x1,x2-d), X(x1,x2-d+1),...,X(x1,x2+d)), and K3(X) represents a sequence of (X(x1-d,x2-d), X(x1-d+1),...,X ... Let K4(X) represent the sequence (X(x1+d,x2+d),X(x1+d-1,x2+d-1),...,X(x1-d,x2-d)), where X(x1,x2) represents the coordinates of a point on the image after the contour measurement data is graphically represented, n represents an integer in the range from -d to d, and d represents the size of the filter window.

[0073] This implementation method solves the problem of noise generated by environmental interference in measured data through filtering.

[0074] Implementation Method 5: This implementation method describes an axial profile measuring device for large rotating equipment based on multi-level median hybrid filtering. The measuring device includes a self-aligning and tilting platform, multiple inductive sensors, and a computer. The computer controls the adjustment actions of the self-aligning and tilting platform. The multiple sensors are fixed on the self-aligning and tilting platform to detect the physical parameters of the large rotating equipment to be measured and send the detection results to the computer. The computer has an embedded measurement module implemented with computer software. The measurement module includes:

[0075] The end-face measurement model establishment unit is used to establish an axial measurement model based on the geometric coupling characteristics of the eccentricity error, tilt error, sensor probe radius error, probe offset error, and probe support rod tilt error of large rotating equipment.

[0076] The self-aligning and tilting platform adjustment unit is used to send control signals to the self-aligning and tilting platform, and to control the adjustment of the eccentricity and eccentricity angle, tilt and tilt angle of the reference surface of the large rotary equipment relative to the rotary spindle. When the eccentricity and tilt of the reference surface are less than 5μm, the self-aligning and tilting is terminated.

[0077] The data acquisition unit is used to acquire data on the eccentricity and eccentricity angle, tilt and tilt angle, and the contour data of the reference plane and the measuring plane relative to the main shaft of the large rotary equipment, as well as the eccentricity and eccentricity angle of the reference plane.

[0078] The data processing unit is used to separate the error of the reference surface end face and the measurement surface end face contour data based on the eccentricity and eccentricity angle of the axial measurement model and the reference surface of the large rotary equipment, and to obtain the reference surface end face and measurement surface end face contour measurement data after error separation.

[0079] The end face contour data acquisition unit is used for multi-level median mixed filtering of the reference surface end face and measurement surface end face contour measurement data after error separation to obtain clean end face contour data of large rotating equipment.

[0080] Implementation Method Six: This implementation method further defines the axial profile measuring device for large rotating equipment based on multi-level median hybrid filtering described in Implementation Method Five. Specifically, the data acquisition unit is:

[0081]

[0082] Where e0 is the eccentricity of the reference plane of the large rotary equipment, e m It is an axial end face eccentricity, z m The height of the end face is α0, and the eccentricity angle of the reference plane is α. m It is the eccentric angle, γ m The angle of inclination of the geometric axis of the large rotary equipment relative to the main shaft of rotation is given by (l', m', n').

[0083] Implementation Method Seven: This implementation method further defines the axial profile measuring device for large rotating equipment based on multi-level median hybrid filtering described in Implementation Method Five. The data processing unit specifically comprises:

[0084]

[0085] Where r0 is the end-face sampling radius, θ' i β is the actual sampling angle of the end face of the large rotating equipment, where β is the angle between the tilt direction and the measurement direction, and Δz is the actual sampling angle. i It refers to the end face runout, and r is the deviation caused by the sensor's probe radius error. It is the tilt angle of the probe support rod, θ i For the ideal sampling angle of the end face of large rotating equipment, d m It is the probe offset error, θ i ′ is the actual sampling angle.

[0086] Implementation Method Eight: This implementation method further defines the axial profile measuring device for large rotating equipment based on multi-level median hybrid filtering described in Implementation Method Five. Specifically, the end face profile data acquisition unit is:

[0087] The graphical sampling contour data module is used to normalize the contour measurement data of the reference surface end face and the measurement surface end face after error separation, and to obtain graphical sampling contour data.

[0088] A multi-level median blending filter module is used to process graphical sampled contour data using multi-level median blending filters. The multi-level median blending filter has four sampling windows:

[0089]

[0090] stn∈[-d,d]

[0091] Where K(X) represents a sequence centered at X, K1(X) represents a sequence of (X(x1-d,x2), X(x1-d+1,x2),...,X(x1+d,x2)), K2(X) represents a sequence of (X(x1,x2-d), X(x1,x2-d+1),...,X(x1,x2+d)), and K3(X) represents a sequence of (X(x1-d,x2-d), X(x1-d+1),...,X ... Let K4(X) represent the sequence (X(x1+d,x2+d),X(x1+d-1,x2+d-1),...,X(x1-d,x2-d)), where X(x1,x2) represents the coordinates of a point on the image after the contour measurement data is graphically represented, n represents an integer in the range from -d to d, and d represents the size of the filter window.

[0092] Implementation Method Nine: A computer device according to this implementation method includes a memory and a processor. The memory stores a computer program. When the processor runs the computer program stored in the memory, the processor executes a method for measuring the axial profile of a large rotating equipment based on multi-level median hybrid filtering, as described in any one of Implementation Methods One to Four.

[0093] Implementation Method 10: A computer-readable storage medium according to this implementation method, the computer-readable storage medium being used to store a computer program, the computer program executing any one of Implementation Methods 1 to 4, a method for measuring the axial profile of a large rotating equipment based on multi-level median hybrid filtering.

[0094] Implementation Method 11, see below Figure 1 , Figure 2 and Figure 3This embodiment describes a specific implementation of the axial profile measurement method for large rotating equipment based on multi-level median hybrid filtering provided in Embodiment 1, and is used to explain the axial profile measurement method for large rotating equipment based on multi-level median hybrid filtering provided in any one of Embodiments 1 to 4. Specifically:

[0095] Step 1: Establish an axial measurement model based on the geometric coupling characteristics of the eccentricity error, tilt error, sensor probe radius error, probe offset error, and probe support rod tilt error of the large rotating equipment to be measured;

[0096] Step 2: Place the large rotating equipment to be tested on the self-aligning and tilting platform and fix it with the tripod chuck on the self-aligning and tilting platform.

[0097] Step 3: Place four sensors at the reference surface section, the reference surface end face, the measurement surface section, and the measurement surface end face of the large rotating equipment, respectively, to collect the eccentricity and eccentricity angle, tilt and tilt angle, and the contour data of the reference surface end face and the measurement surface end face of the large rotating equipment relative to the rotating spindle.

[0098] Based on the sensor measurement, the centering and tilting table is controlled to adjust the attitude of the large rotating equipment under test, so that the eccentricity and tilt of the reference plane of the large rotating equipment are less than 5μm, and the eccentricity and eccentricity angle of the reference plane of the large rotating equipment are obtained.

[0099] Step 4: Control the rotation of the air-bearing turntable in the self-aligning and tilting platform. During the rotation, the sensor collects the contour data of the reference surface end face and the measurement surface end face of the large rotary equipment, and separates the measurement error of the large rotary equipment based on the axial measurement model, thereby obtaining the measurement data of the reference surface end face and the measurement surface end face contour after error separation.

[0100] Step 5: After error separation, the contour measurement data of the reference surface end face and the measurement surface end face are subjected to multi-level median mixing filtering in the host computer system to obtain clean axial contour data of the large rotary equipment. Based on this axial contour data, the verticality of the large rotary equipment is evaluated according to ISO standards.

[0101] For large rotating equipment, the measurement error of the axial assembly surface of the large rotating equipment should be considered first. For example... Figure 1 As shown, the large rotary equipment is placed on a turntable plane using a tripod chuck. The axial reference plane of the large rotary equipment serves as a reference. The axial measuring plane of the large rotary equipment will couple because machining errors in the large rotary equipment cause eccentricity errors. For example... Figure 1 As shown in the left figure, the elliptical dashed line is the ideal outline, while the elliptical solid line and the shaded area are the actual outline; Figure 1The right-hand figure is a top view of the axial error coupling relationship of the assembly surface of a large rotating equipment. As shown in the figure, the axial end face eccentricity of the assembly surface of the large rotating equipment is e. m The eccentricity is α m This causes a cosine error in the sampling angle of the end face of the large rotating equipment; the tilt caused by the tilt of the large rotating equipment itself and the machining error will also lead to the tilt error of the large rotating equipment. Let γ m The tilt angle of the geometric axis of a large rotating equipment relative to the main rotating shaft causes deviations in the axial end face runout of the equipment at different sampling angles due to the influence of the tilt angle. Errors caused by the measuring device itself mainly include the probe radius error of the measuring sensor, the probe support rod tilt error, and the probe position offset error. Existing technology typically obtains measurement data of the axial assembly surface of the large rotating equipment by having a measuring sensor in contact with it. The deviation caused by the sensor's probe radius error is its radius *r*, which is coupled with the probe support rod tilt error, and the tilt angle of the probe support rod... This will lead to amplified axial end face runout. The probe position offset error is caused by the probe's measurement direction not being collinear with the line connecting the measuring point to the center of rotation; the probe offset error d... m It is usually estimated using a step-by-step separation estimation method.

[0102] based on Figure 1 Based on the geometric relationship, the axial measurement equation is obtained. According to the equation, the contour measurement data of the reference surface end face and the measurement surface end face after error separation can be obtained as follows:

[0103]

[0104] Where r0 represents the end-face sampling radius, γ m θ' represents the angle of inclination of the geometric axis of a large rotary equipment relative to the main shaft of rotation. i β represents the actual sampling angle of the end face of the large rotating equipment, β represents the angle between the tilt direction and the measurement direction, and Δz represents the angle between the tilt direction and the measurement direction. i This indicates end face runout, and r represents the deviation caused by the sensor's probe radius error. θ represents the inclination angle of the probe support rod. i d represents the ideal sampling angle of the end face of a large rotating equipment. m θ represents the probe offset error. i ′ represents the actual sampling angle.

[0105] The influence of tilting error on the end face eccentricity of large rotating equipment is related to the eccentricity of the reference plane and the tilting angle of the large rotating equipment. The eccentricity e0 of the reference plane of the large rotating equipment is mainly caused by the misalignment of the measurement center of the reference plane and the rotation center of the measurement transposition. The geometric relationship is as follows:

[0106]

[0107] Where e0 represents the eccentricity of the reference plane of the large rotary equipment, α0 represents the eccentricity angle of the reference plane, r0 represents the sampling radius, and z m Indicates the height of the end face, γ m The angle of inclination of the geometric axis is represented by (l',m',n'), and the direction vector corresponding to the geometric axis is represented by (l',m',n').

[0108] Based on the measurement equation, the measured contour data of the assembly surface of large rotary equipment can be obtained by using the step-by-step error separation method. However, since the measured data is also subject to environmental interference and noise, it is necessary to further filter the measured data.

[0109] First, the 1×1000 sampled data is normalized to the range of 0-255 pixel values ​​and then transformed into a 40×25 matrix structure. Figure 2 Grayscale images corresponding to different contour sampling data of large rotating equipment. Figure 2 (a) is a data diagram of a large single-tilt slewing machine. Figure 2 (b) Data diagram of large saddle-face rotating equipment. For the graphical sampled contour data, multi-level median blending filtering was used for processing. For example... Figure 3 As shown, the multi-stage median blending filter has four sampling windows. Let the point coordinates be X(x1, x2), and the four sub-windows can be represented as follows:

[0110]

[0111] stn∈[-d,d]

[0112] Here, K(X) represents a sequence centered at X. The mean of the values ​​in the four windows is calculated, and the maximum and minimum values ​​of the four means are found. The median of the results and the values ​​themselves is then calculated to achieve numerical filtering.

[0113] The output of a multi-stage median mixing filter is defined as:

[0114] Y(X)=median{Y1(X),Y2(X),X(x1,x2)}

[0115]

[0116] Here, `median()` represents ordinary median filtering, and `average()` represents mean filtering. Multi-level median blending filtering is directional, thus preserving detailed information relatively well. Combining multi-level median blending filtering with error separation based on an error model allows the final measurement data to accurately reflect the contour characteristics of the assembly surfaces of large rotating equipment.

[0117] In this embodiment, planarization of contour data preserves the inherent characteristics of the data while visualizing the distribution of contour features. Utilizing multi-level median hybrid filtering to process the visualized data maximizes the correlation between data points during filtering, ensuring effective filtering and ultimately obtaining accurate axial contour measurement data for large rotating equipment.

[0118] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0119] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products implemented according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0120] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0121] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0122] The present application has been described in detail above through specific embodiments. However, the above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, combinations of embodiments, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included within the protection scope of the present application.

Claims

1. A large-scale rotary equipment axial profile measurement method based on multi-stage median hybrid filtering, the indicators of the measurement are the axial perpendicularity and the radial coaxiality of its assembly surface, characterized in that, The measurement method includes: An axial measurement model is established based on the geometric coupling characteristics of the eccentricity error, tilt error, sensor probe radius error, probe offset error, and probe support rod tilt error of the large rotating equipment to be measured. The large rotary equipment is fixed to the self-aligning and tilting table; Adjust the centering and tilting table, and use inductive sensors to detect the eccentricity and eccentricity angle, tilt and tilt angle, and the contour data of the reference surface and the measuring surface of the large rotary equipment relative to the rotary spindle. When the eccentricity and tilt of the reference surface are less than 5μm, the centering and tilting is stopped, and the eccentricity and eccentricity angle of the reference surface of the large rotary equipment are obtained. Based on the eccentricity and eccentricity angle of the axial measurement model and the reference surface of the large rotary equipment, error separation is performed on the contour data of the reference surface end face and the measurement surface end face, and the measurement data of the reference surface end face and the measurement surface end face after error separation are obtained. Multi-level median hybrid filtering is used to process the contour measurement data of the reference surface end face and the measurement surface end face after error separation to obtain pure axial contour data of large rotary equipment. The eccentricity error, tilt error, sensor probe radius error, and probe offset error of the large rotating equipment are the measurement errors of the large rotating equipment. The eccentricity and eccentricity angle of the reference plane for obtaining the large rotary equipment are specifically as follows: ， in, e 0 represents the eccentricity of the reference surface for large rotary equipment. e m It is an axial end face eccentricity. z m The height of the end face. α 0 is the initial eccentricity angle. α m It is an off-center angle. γ m The tilt angle of the geometric axis of the large rotary equipment relative to the main shaft of rotation is given by the direction vector corresponding to the geometric axis. l' , m' , n' ); The acquisition of the contour measurement data of the reference surface end face and the measurement surface end face after error separation is specifically as follows: ， in, r 0 is the end-face sampling radius. θ' i This refers to the actual sampling angle of the end face of a large rotating equipment. It is the angle between the tilt direction and the measurement direction. It is end face runout. The deviation is caused by the error in the sensor's probe radius. It is the tilt angle of the probe support rod. θ i For the ideal sampling angle of the end face of large rotating equipment, It is the probe offset error. That is the actual sampling angle.

2. The method for measuring the axial profile of large rotating equipment based on multi-level median hybrid filtering according to claim 1, characterized in that, The multi-level median hybrid filtering process for the error-separated contour measurement data of the reference surface end face and the measurement surface end face is specifically as follows: Normalize the measurement data of the reference surface end face and the measurement surface end face after error separation to obtain graphical sampled contour data; The graphical sampled contour data is processed using a multi-stage median blending filter, which has four sampling windows: ， in, K ( X ) indicates with X The sequence centered on, K 1( X )express( X ( x 1- d , x 2), X ( x 1- d +1, x 2),..., X ( x 1+ d , x 2)) sequence, K 2( X )express( X ( x 1, x 2- d ), X ( x 1, x 2- d +1),..., X ( x 1, x 2+ d The sequence of numbers, K 3( X )express( X ( x 1- d , x 2- d ), X ( x 1- d +1, x 2- d +1),..., X ( x 1+ d , x 2+ d The sequence of numbers, K 4( X )express( X ( x 1+ d , x 2+ d ), X ( x 1+ d -1, x 2+ d -1),..., X ( x 1- d , x 2- d The sequence of numbers, X ( x 1, x 2) Represents the coordinates of a point on the image after the contour measurement data has been graphically represented. The range is - d arrive d The integer d represents the size of the filter window.

3. A axial profile measuring device for large rotary equipment based on multi-level median hybrid filtering, the measuring device comprising a self-aligning and tilting platform, multiple inductive sensors, and a computer, wherein the computer controls the adjustment action of the self-aligning and tilting platform, the multiple sensors are fixed on the self-aligning and tilting platform to detect the physical parameters of the large rotary equipment to be measured, and send the detection results to the computer; the measured indicators are the axial perpendicularity and radial coaxiality of its assembly surface, characterized in that... The computer has an embedded measurement module implemented in computer software, and the measurement module includes: The end-face measurement model establishment unit is used to establish an end-face measurement model based on the geometric coupling characteristics of the eccentricity error, tilt error, sensor probe radius error, probe offset error, and probe support rod tilt error of large rotating equipment. The self-aligning and tilting platform adjustment unit is used to send control signals to the self-aligning and tilting platform, and to control the adjustment of the eccentricity and eccentricity angle, tilt and tilt angle of the reference surface of the large rotary equipment relative to the rotary spindle. When the eccentricity and tilt of the reference surface are less than 5μm, the self-aligning and tilting is terminated. The data acquisition unit is used to acquire data on the eccentricity and eccentricity angle, tilt and tilt angle, and the contour data of the reference plane and the measuring plane relative to the main shaft of the large rotary equipment, as well as the eccentricity and eccentricity angle of the reference plane. The data processing unit is used to separate the error of the reference surface end face and the measurement surface end face contour data based on the eccentricity and eccentricity angle of the axial measurement model and the reference surface of the large rotary equipment, and to obtain the reference surface end face and measurement surface end face contour measurement data after error separation. The axial profile data acquisition unit is used for multi-level median mixed filtering of the reference surface end face and measurement surface end face profile measurement data after error separation to obtain pure axial profile data of large rotary equipment. The eccentricity error, tilt error, sensor probe radius error, and probe offset error of the large rotating equipment are the measurement errors of the large rotating equipment. The data acquisition unit is specifically: ， in, e 0 represents the eccentricity of the reference surface for large rotary equipment. e m It is an axial end face eccentricity. z m The height of the end face. α 0 is the initial eccentricity angle. α m It is an off-center angle. γ m The tilt angle of the geometric axis of the large rotary equipment relative to the main shaft of rotation is given by the direction vector corresponding to the geometric axis. l' , m' , n' ); The data processing unit specifically comprises: ， Among them, r 0 is the end-face sampling radius. θ' i This refers to the actual sampling angle of the end face of a large rotating equipment. It is the angle between the tilt direction and the measurement direction. It is end face runout. The deviation is caused by the error in the sensor's probe radius. It is the tilt angle of the probe support rod. θ i For the ideal sampling angle of the end face of large rotating equipment, It is the probe offset error. That is the actual sampling angle.

4. The axial profile measuring device for large rotating equipment based on multi-level median hybrid filtering according to claim 3, characterized in that, The end face contour data acquisition unit specifically comprises: The graphical sampling contour data module is used to normalize the contour measurement data of the reference surface end face and the measurement surface end face after error separation, and to obtain graphical sampling contour data. A multi-level median blending filter module is used to process graphical sampled contour data using multi-level median blending filters. The multi-level median blending filter has four sampling windows: ， in, K ( X ) indicates with X The sequence centered on, K 1( X )express( X ( x 1- d , x 2), X ( x 1- d +1, x 2),..., X ( x 1+ d , x 2)) sequence, K 2( X )express( X ( x 1, x 2- d ), X ( x 1, x 2- d +1),..., X ( x 1, x 2+ d The sequence of numbers, K 3( X )express( X ( x 1- d , x 2- d ), X ( x 1- d +1, x 2- d +1),..., X ( x 1+ d , x 2+ d The sequence of numbers, K 4( X )express( X ( x 1+ d , x 2+ d ), X ( x 1+ d -1, x 2+ d -1),..., X ( x 1- d , x 2- d The sequence of numbers, X ( x 1, x 2) Represents the coordinates of a point on the image after the contour measurement data has been graphically represented. The range is - d arrive d The integer d represents the size of the filter window.

5. A computer device, characterized in that: It includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes a method for measuring the axial profile of a large rotating equipment based on multi-level median hybrid filtering as described in any one of claims 1-2.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program that executes the axial profile measurement method for large rotating equipment based on multi-level median hybrid filtering as described in any one of claims 1-2.

Images