High-precision measuring device for thermal error of ball screw under variable working conditions and use method
By installing temperature sensors and displacement sensors on the ball screws, combined with the improved fuzzy clustering FCM algorithm, the high-precision measurement problem of ball screw thermal error under complex temperature distribution is solved, real-time accurate measurement and compensation under variable working conditions is achieved, and the needs of high-precision motion control are met.
Patent Information
- Application Number
- CN202510422250.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-07
- Publication Date
- 2025-08-01
AI Technical Summary
The existing ball screw thermal error measurement methods are insufficient in complex temperature distribution and are difficult to accurately measure and compensate in real time. The traditional methods are costly or sensitive to vibration, and cannot meet the high-precision requirements under varying working conditions.
Using a measuring device including a temperature sensor, a displacement sensor and a grating scale, combined with an improved fuzzy clustering FCM algorithm, the thermal error and temperature changes of the ball screw are measured in real time through the precise positioning of temperature sensitive points and data processing, and automated control is achieved using a programmable logic controller and driver.
It realizes high-precision measurement of thermal error of ball screw under variable working conditions, simplifies the operation process, improves the accuracy and real-time measurement, and adapts to the needs of precision motion control under complex working conditions.
Smart Images

Figure CN120404131A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a ball screw measuring device, and in particular to a high-precision measuring device for thermal errors of a ball screw under variable working conditions and a method for using the device. Background Art
[0002] The feed system is a major component in modern high-precision machine tools and precision motion control systems. Its main function is to convert rotational motion into linear motion. Heat is generated due to factors such as friction and load, and heat is also transferred due to factors such as heat conduction, heat radiation and heat convection. The combined effect leads to uneven temperature changes on the ball screw and thermal expansion, which causes thermal errors in the workbench. When the ball screw works under different loads and speeds, the temperature changes caused by the generation and dissipation of heat will also be different, which in turn leads to complex thermal expansion behavior. This makes thermal error a dynamic error that changes with working conditions. In order to achieve high-precision motion control in practical applications, it is necessary to accurately measure the temperature conditions and predict and compensate for these errors caused by temperature changes in real time.
[0003] Existing methods for measuring ball screw thermal errors involve installing sensors such as thermocouples or thermistors on the ball screw to measure temperature changes and calculating the length change of the screw using thermal expansion formulas. This method is simple and easy to implement, but its accuracy is limited by the assumed uniformity of the temperature field and is difficult to handle with complex, non-uniform temperature distributions. Blindly determining the location of temperature measurement points reduces the reliability of the overall temperature distribution of the screw. Another high-precision measurement method is laser interferometry, which uses a laser interferometer to measure minute displacement changes in the ball screw and calculate its thermal error. The advantage of laser interferometry lies in its extremely high accuracy, capable of detecting submicron expansion changes. However, this method is complex and expensive, and is highly sensitive to external vibrations, making it difficult to obtain the required temperature change data. Alternatively, thermal field simulation and error compensation methods are widely used. Finite element analysis and thermal field simulation techniques are combined with actual temperature data to estimate the thermal error of the ball screw and perform real-time error compensation through a control system. While this method can model complex temperature distributions and nonlinear thermal effects, model accuracy and computational complexity remain major challenges. Summary of the Invention
[0004] In order to solve the problems existing in the background technology, the present invention provides a high-precision measurement device for thermal error of a ball screw under variable working conditions and a method for using the device.
[0005] The technical solution adopted in the present invention is:
[0006] 1. A high-precision measurement device for ball screw thermal errors under variable operating conditions, comprising:
[0007] The control box is electrically connected to the ball screw platform and is used to control the operation of the ball screw platform under variable working conditions.
[0008] The data acquisition and display system is installed on the ball screw platform and is used to obtain the sensing data of the ball screw platform. The data acquisition and display system includes several temperature sensors which are respectively installed at various temperature-sensitive point positions of the ball screw platform.
[0009] The described ball screw platform includes a vibration isolation platform, a servo motor, a coupling, two bearings, two guide rail sliders, a nut pair, a panel, two guide rails, a ball screw and a bottom plate. The bottom plate is horizontally installed on the vibration isolation platform. The two guide rails are installed on the bottom plate at intervals along the length direction of the bottom plate. The ball screw is installed on the bottom plate along the length direction of the bottom plate through two bearings and is located between the two guide rails. The body of the servo motor is installed on the bottom plate through a motor mounting bracket and is close to one end of the ball screw. The output shaft of the servo motor is synchronously connected to one end of the ball screw through a coupling. The panel is installed directly above the ball screw through the nut pair. The nut pair and the ball screw adopt a double-nut pre-tightening method. The two sides of the bottom surface of the panel are respectively slidably installed on the two guide rails through two guide rail sliders. Each temperature sensor is installed on the bottom plate at intervals and is located directly below the ball screw. The control box includes a programmable logic controller PLC (Programmable Logic Controller) and a driver connected in sequence. The programmable logic controller PLC is electrically connected to the data acquisition and display system. The driver is electrically connected to the servo motor through an encoder wire.
[0010] The described data acquisition and display system includes a displacement sensor, a display and a grating scale. The displacement sensor is installed on the bottom plate and is directly opposite to the other end of the ball screw. The display is placed on the vibration isolation platform. The grating scale is installed on the side of the panel. Each temperature sensor, displacement sensor, grating scale and the programmable logic controller PLC of the control box are all connected to the display.
[0011] It also includes three limit switches for panel limit. The three limit switches are installed on the bottom plate at intervals through a limit switch bracket and are located on one side of one guide rail away from the ball screw. Two of the limit switches are located on the side close to the servo motor and are respectively opposite to one end position and the preset zero position within the preset movement range during the movement of the panel. The other limit switch is located on the side away from the servo motor and is opposite to the other end position within the preset movement range during the movement of the panel. When the ball screw platform is not operating, the control panel moves to the preset zero position. When the ball screw platform is operating, the movement range does not exceed the preset movement range. The three limit switches are respectively connected to the programmable logic controller PLC of the control box.
[0012] II. According to a usage method of a high-precision measuring device for ball screw thermal error, it includes:
[0013] Step 1: Control the driver through a programmable logic controller (PLC) to drive the servo motor to run, so that the panel completes a preset action within a preset moving range on the ball screw.
[0014] Step 2: Collect the temperature data of the ball screw during operation through each temperature sensor, use one end of the ball screw as the origin, use the distance of each temperature sensor from the origin as the temperature measurement point position, and process the temperature data according to each temperature measurement point position and the collected temperature data using an improved fuzzy c-means (FCM) algorithm to obtain several temperature-sensitive point positions, retain the temperature sensors at each temperature-sensitive point position, and remove other temperature sensors.
[0015] Step 3: During each operation of the panel, obtain the thermal elongation of the ball screw through a displacement sensor, obtain the temperature of the ball screw through the temperature sensors at each temperature-sensitive point position, and obtain the thermal error from the movement trajectory data of the panel obtained by the driver and grating ruler in the control box to measure the positioning error, and display and save the data on the display.
[0016] Step 4: After each operation is completed, control the panel to move to the preset zero position for the next thermal error and temperature measurement.
[0017] In the above Step 2, the improved fuzzy c-means (FCM) algorithm is used to process the temperature data according to each temperature measurement point position, specifically as follows:
[0018] Step 2.1: Use the fuzzy c-means (FCM) algorithm to establish a set X of clustering prototype vectors for each temperature measurement point position.
[0019] Step 2.2: Establish the objective function of the improved fuzzy c-means (FCM) algorithm, input the set X of clustering prototype vectors into the objective function, and output several initial measurement positions and their membership matrix U after processing. Take each initial measurement position corresponding to the maximum membership value in the membership matrix U as the final temperature-sensitive point position.
[0020] In the above Step 2.2, the objective function of the improved fuzzy c-means (FCM) algorithm is specifically as follows:
[0021]
[0022] where U represents the membership matrix; V represents the cluster center matrix; λ represents the Lagrange multiplier matrix, and λ k represents the k-th Lagrange multiplier in the Lagrange multiplier matrix; J FCM() represents the objective function of the fuzzy c-means (FCM) algorithm; X represents the set of cluster prototype vectors; N represents the total number of measurement points; c represents the number of clusters;
[0023] When the objective function reaches its minimum value, the iteration ends.
[0024] In step 3, the ideal movement trajectory of the panel is obtained through the driver of the control box, and the actual movement trajectory of the panel is obtained through the grating ruler. The thermal error is obtained by subtracting the ideal movement trajectory from the actual movement trajectory.
[0025] The beneficial effects of the present invention are as follows:
[0026] The measurement of the thermal error, temperature, and overall thermal deformation elongation of the method of the present invention can be completed at one time, with high measurement accuracy, strong readability, and simple usage method. For tests under different working conditions, only the control program needs to be changed to realize the motion settings of the feed system under multiple working conditions, which is convenient for operators to use and the measurement is rapid. It solves the problems that it is difficult to measure the temperature and thermal deformation of the ball screw, the measurement accuracy is low, it is difficult to display in real time, and it is difficult to determine the position and quantity of the temperature measurement points. At the same time, it can complete data communication and data storage, providing strong support for data analysis. The present invention is modified on the basis of the traditional machine tool feed system, can realize the real-time and accurate measurement of the thermal error and temperature with high precision under variable working conditions, the data display effect is intuitive, and it can meet the precise motion control requirements under complex working conditions. Description of the Drawings
[0027] Figure 1 is a schematic diagram of the measuring device of the present invention;
[0028] Figure 2 is a schematic diagram of the layout of the sensing system of the measuring device of the present invention;
[0029] Figure 3 is a partial schematic diagram of the layout of the limit switches of the measuring device of the present invention;
[0030] Figure 4 is a partial schematic diagram of the layout of the thermal error measurement sensors of the measuring device of the present invention;
[0031] Figure 5 is a schematic diagram of the installation interval of the temperature sensors of the measuring device of the present invention;
[0032] Figure 6 is a flowchart of the method for screening temperature-sensitive points based on the improved fuzzy c-means (FCM) method of the present invention;
[0033] In the figure: 1. Vibration isolation platform; 2. Control box; 3. Temperature sensor; 4. Motor mounting bracket 5; 5. Servo motor; 6. Coupling; 7. Bearing; 8. Guide rail slider; 9. Nut pair; 10. Panel; 11. Guide rail; 12. Ball screw; 13. Limit switch bracket; 14. Limit switch; 15. Displacement sensor; 16. Display; 17. Base plate; 18. Grating scale. Detailed implementation mode
[0034] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0035] As Figure 1 shown, the high-precision measurement device for the thermal error of the ball screw under variable working conditions of the present invention includes a control box 2 and a data acquisition and display system. The control box 2 is electrically connected to the ball screw platform and is used to control the operation of the ball screw platform under variable working conditions; the data acquisition and display system is installed on the ball screw platform and is used to obtain the sensing data of the ball screw platform. The data acquisition and display system includes a number of temperature sensors 3 which are respectively installed at various temperature-sensitive point positions of the ball screw platform.
[0036] As Figure 2 and Figure 3As shown in the figure, the ball screw platform includes a vibration isolation platform 1, a servo motor 5, a coupling 6, two bearings 7, two guide rail sliders 8, a nut pair 9, a panel 10, two guide rails 11, a ball screw 12 and a bottom plate 17. The bottom plate 17 is horizontally installed on the vibration isolation platform 1. The two guide rails 11 are installed on the bottom plate 17 at intervals along the length direction of the bottom plate 17. The ball screw 12 is installed on the bottom plate 17 along the length direction of the bottom plate 17 through two bearings 7 and is located between the two guide rails 11. The body of the servo motor 5 is installed on the bottom plate 17 through a motor mounting bracket 4 and is close to one end of the ball screw 12. The output shaft of the servo motor 5 is synchronously connected to one end of the ball screw 12 through the coupling 6. The panel 10 is installed directly above the ball screw 12 through the nut pair 9. The nut pair 9 and the ball screw 12 adopt a double-nut pre-tightening method. The two sides of the bottom surface of the panel 10 are respectively slidably installed on the two guide rails 11 through two guide rail sliders 8. Each temperature sensor 3 is installed on the bottom plate 17 at intervals and is located directly below the ball screw 12. The control box 2 includes a programmable logic controller PLC and a driver connected in sequence. The programmable logic controller PLC is electrically connected to the data acquisition and display system. The driver is electrically connected to the servo motor 5 through an encoder cable. The device also includes three limit switches 14 for limiting the panel 10. The three limit switches 14 are installed on the bottom plate 17 at intervals through a limit switch bracket 13 and are located on one side of one of the guide rails 11 away from the ball screw 12. Two of the limit switches 14 are located on the side close to the servo motor 5 and are respectively opposite to one end position and the preset zero position within the preset movement range during the movement of the panel 10. The other limit switch 14 is located on the side away from the servo motor 5 and is opposite to the other end position within the preset movement range during the movement of the panel 10. When the ball screw platform is not running, the control panel 10 is moved to the preset zero position. When the ball screw platform is running, the movement range does not exceed the preset movement range. The three limit switches 14 are respectively connected to the programmable logic controller PLC of the control box 2.
[0037] As Figure 1 , Figure 2 and Figure 4 shown, the data acquisition and display system includes a displacement sensor 15, a display 16 and a grating scale 18. The displacement sensor 15 is installed on the bottom plate 17 and is opposite to the other end of the ball screw 12. The display 16 is placed on the vibration isolation platform 1. The grating scale 18 is installed on the side of the panel 10. Each temperature sensor 3, displacement sensor 15, grating scale 18 and the programmable logic controller PLC of the control box 2 are all connected to the display 16. The displacement sensor 26 is installed 3 - 5 cm away from the other end of the ball screw 12.
[0038] The usage method of the high-precision ball screw thermal error measurement device of the present invention is specifically as follows:
[0039] Step 1: Write a control driver program in the programmable logic controller (PLC) of the control box 2 to control the driver to drive the output shaft of the servo motor 5 to rotate, and the rotation drives the panel 10 to complete a preset action within the preset movement range on the ball screw 12.
[0040] Step 2: Collect the temperature data of the ball screw 12 during operation through each temperature sensor 3. Take one end of the ball screw 12 as the origin, and take the distance of each temperature sensor 3 from the origin as the temperature measurement point position. Process the temperature measurement point positions and the collected temperature data using the improved fuzzy c-means (FCM) algorithm to obtain several temperature sensitive point positions. Retain the temperature sensors 3 at each temperature sensitive point position and remove other temperature sensors 3.
[0041] As Figure 5 and Figure 6 shown, process the temperature measurement point positions and the collected temperature data using the improved fuzzy c-means (FCM) algorithm as follows:
[0042] Step 2.1: Use the fuzzy c-means (FCM) algorithm to establish a set of clustering prototype vectors X for each temperature measurement point position; in order to optimize N measurement points into c clusters that can represent the overall data distribution, it is necessary to establish a corresponding set of clustering prototype vectors X, where X = {x1, x2, x3,... x k ..., x N}, and each data vector x k has n characteristic indicators at the kth measurement point, that is, corresponding to n data acquisitions, and can be expressed as x k = (x 1k , x 2k ,..., x nk ). Divide the set of clustering prototype vectors X into c clusters: X1, X2... X T i ... X c , where c represents the number of clusters, which is an integer greater than 1; the device of the present invention hopes to arrange four temperature measurement points, so the number of clusters c = 4 is set, that is, it is hoped to cluster N temperature measurement points into c clusters. Among them, X1 = {x1, x2, x3}, X2 = {x4, x5, x6}, X3 = {x7, x8, x9}, X4 = {x 10 , x 11 , x 12 , x 13}, and use the fuzzy membership function u ik to represent the membership degree of the sample x k to the cluster X i . Then each classification result of X, that is, the membership degree matrix U calculated by selecting a prototype vector from each cluster, can correspond to a c×N-order Boolean fuzzy function matrix U = [u ik c×N , where \(k = 1, 2, \cdots, N\), representing the membership probability of each measurement point to \(c\) clusters, and satisfying:
[0043]
[0044] Set the weighted index, initialize the cluster center matrix, and construct the objective function for the first round. To select a measurement point from each cluster as a representative, the nearest neighbor method is used to construct the objective function. It is stipulated that the cluster center matrix of \(c\) classes is \(V = [v_1, v_2, \cdots, v i \cdots, v c \), where \(v_1, v_2, \cdots, v i \cdots, v c correspond to the central positions of each corresponding cluster in the \(n\) characteristic indicators, i.e., the \(n\)-dimensional space. Referring to the required experimental conditions, the clustering results of the temperature measurement points are obtained. The specific objective function is as follows:
[0045]
[0046] where \(d(x k , v i )=\|x k - v i \|\), representing the distortion degree of the measurement point \(x k corresponding to the cluster center \(v i \), and is measured by the Euclidean distance between these two vectors. The smaller \(J FCM \) is, the better the classification effect; the weighted index \(m\) is also called the smoothing index, and its value lacks effective proof or theoretical support. Generally, the best value range is determined according to practical experience as \([1.5, 3]\), and the weighted index \(m = 2\) is set.
[0047] Step 2.2: Establish the objective function of the improved fuzzy c-means (FCM) algorithm Specifically as follows:
[0048]
[0049] where \(U\) represents the membership matrix; \(V\) represents the cluster center matrix; \(\lambda\) represents the Lagrange multiplier matrix, and \(\lambda k represents the \(k\)-th Lagrange multiplier in the Lagrange multiplier matrix; \(J FCM ()\) represents the objective function of the fuzzy c-means (FCM) algorithm; \(X\) represents the set of clustering prototype vectors; \(N\) represents the total number of measurement points, with an actual size of 13; \(c\) represents the number of clusters, with an actual size of 4;
[0050] Input the set of clustering prototype vectors \(X\) into the objective function. When the objective function When a minimum value lower than the preset threshold is obtained, the iteration ends. After processing, several initial measurement positions and their membership matrix U are output. Each initial measurement position corresponding to the maximum membership degree in the membership matrix U is used as the final temperature-sensitive point position.
[0051] According to the length of the ball screw 12 being 1 m and the maximum one-way movement distance being 0.917 m, several evenly distributed temperature measurement point positions on it are selected as the objects to be classified. Initially, 13 position points are selected. After clustering, the optimal measurement points are obtained, namely x1 in cluster X1, x5 in cluster X2, x7 in cluster X3, and x in cluster X4 13 , and finally, four positions with distances from the origin position of T1 = 0.02 m, T5 = 0.244 m, T7 = 0.351 m, and T13 = 0.809 m are selected to measure the temperatures of the four temperature-sensitive point positions on the ball screw 12 respectively. Thus, the problem of multicollinearity that may be caused by too many temperature measurement points is eliminated.
[0052] Step 3: Each time the panel 10 runs, the thermal elongation at the end of the ball screw 12 is obtained through the displacement sensor 15, the temperature of the ball screw 12 is obtained through the temperature sensors 3 at each temperature-sensitive point position, the ideal movement trajectory of the panel 10 is obtained through the driver of the control box 2, and the actual movement trajectory of the panel 10 is obtained through the grating scale 18. The actual movement trajectory is subtracted from the ideal movement trajectory to obtain the thermal error for measuring the positioning error, and the data is both displayed and saved on the display 16.
[0053] Step 4: After each operation is completed, the control program drives the ball screw 12 to move to drive the nut pair 9 to return to the origin, and the control panel 10 is moved to the preset zero position for the next measurement of thermal error and temperature.
[0054] The above are only the preferred embodiments of the present application and are not used to limit the present application. For those skilled in the art, it should be understood that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the various embodiments of the present invention, and they should all be covered by the scope of the claims and the description of the present invention.
Claims
1. A high-precision measuring device for the thermal error of a ball screw under variable operating conditions, characterized in that Comprising: A control box (2), electrically connected to the ball screw platform and used to control the operation of the ball screw platform under variable working conditions; A data acquisition and display system, installed on the ball screw platform and used to obtain the sensing data of the ball screw platform. The data acquisition and display system includes several temperature sensors (3) which are respectively installed at each temperature-sensitive point position of the ball screw platform.
2. The high-precision measuring device for the thermal error of a ball screw under variable operating conditions according to claim 1, characterized in that: The described ball screw platform includes a vibration isolation platform (1), a servo motor (5), a coupling (6), two bearings (7), two guide rail sliders (8), a nut pair (9), a panel (10), two guide rails (11), a ball screw (12) and a bottom plate (17). The bottom plate (17) is horizontally installed on the vibration isolation platform (1). The two guide rails (11) are installed on the bottom plate (17) at intervals along the length direction of the bottom plate (17). The ball screw (12) is installed on the bottom plate (17) along the length direction of the bottom plate (17) through two bearings (7) and is located between the two guide rails (11). The body of the servo motor (5) is installed on the bottom plate (17) through a motor mounting bracket (4) and is close to one end of the ball screw (12). The output shaft of the servo motor (5) is synchronously connected to one end of the ball screw (12) through a coupling (6). The panel (10) is installed directly above the ball screw (12) through a nut pair (9). The two sides of the bottom surface of the panel (10) are respectively slidably installed on the two guide rails (11) through two guide rail sliders (8). Each temperature sensor (3) is installed on the bottom plate (17) at intervals and is located directly below the ball screw (12). The control box (2) includes a programmable logic controller PLC and a driver connected in sequence. The programmable logic controller PLC is electrically connected to the data acquisition and display system. The driver is electrically connected to the servo motor (5) through a coding wire.
3. The high-precision measurement device for the thermal error of a ball screw under variable operating conditions according to claim 2, characterized in that: The described data acquisition and display system includes a displacement sensor (15), a display (16) and a grating scale (18). The displacement sensor (15) is installed on the bottom plate (17) and faces the other end of the ball screw (12). The display (16) is placed on the vibration isolation platform (1). The grating scale (18) is installed on the side of the panel (10). Each temperature sensor (3), displacement sensor (15), grating scale (18) and the programmable logic controller PLC of the control box (2) are all connected to the display (16).
4. The high-precision measurement device for the thermal error of a ball screw under variable operating conditions according to claim 2, characterized in that: It further includes three limit switches (14) for limiting the panel (10). The three limit switches (14) are installed on the bottom plate (17) at intervals and are located on one side of one of the guide rails (11) away from the ball screw (12). Two of the limit switches (14) are located on the side close to the servo motor (5) and respectively face one end position and the preset zero position within the preset moving range during the movement of the panel (10). The other limit switch (14) is located on the side away from the servo motor (5) and faces the other end position within the preset moving range during the movement of the panel (10). The three limit switches (14) are respectively connected to the programmable logic controller PLC of the control box (2).
5. The method of using the high-precision measuring device for the thermal error of the ball screw according to any one of claims 1-4, characterized in that, Comprising: Step 1: Control the driver through a programmable logic controller (PLC) to drive the servo motor (5) to operate, so that the panel (10) completes a preset action within a preset movement range on the ball screw (12). Step 2: Collect the temperature data of the ball screw (12) during operation through each temperature sensor (3). Take one end of the ball screw (12) as the origin, and take the distance of each temperature sensor (3) from the origin as the temperature measurement point position. Process the temperature measurement point positions and the collected temperature data using an improved fuzzy c-means (FCM) algorithm to obtain several temperature-sensitive point positions. Retain the temperature sensors (3) at each temperature-sensitive point position and remove other temperature sensors (3). Step 3: During each operation of the panel (10), obtain the thermal elongation of the ball screw (12) through the displacement sensor (15), obtain the temperature of the ball screw (12) through the temperature sensors (3) at each temperature-sensitive point position, and obtain the thermal error from the movement trajectory data of the panel (10) obtained by the driver and the grating scale (18) of the control box (2) to measure the positioning error, and display them all on the display (16). Step 4: After each operation is completed, control the panel (10) to move to the preset zero position.
6. The method for using the high-precision measuring device for the thermal error of the ball screw according to claim 5, characterized in that: In the said Step 2, the temperature measurement point positions and the collected temperature data are processed using an improved fuzzy c-means (FCM) algorithm as follows: Step 2.1: Use the fuzzy c-means (FCM) algorithm to establish a set of clustering prototype vectors X for each temperature measurement point position. Step 2.2: Establish the objective function of the improved fuzzy c-means (FCM) algorithm. Input the set of clustering prototype vectors X into the objective function. After processing, output several initial measurement positions and their membership matrix U. Take each initial measurement position corresponding to the maximum membership degree in the membership matrix U as the final temperature-sensitive point position.
7. The method for using the high-precision measuring device for the thermal error of the ball screw according to claim 6, characterized in that: In step 2.2 described above, the objective function of the improved fuzzy c-means (FCM) algorithm is as follows: Among them, U represents the membership matrix; V represents the cluster center matrix; V represents the Lagrange multiplier matrix, and λ k represents the k-th Lagrange multiplier in the Lagrange multiplier matrix; J FCM ( ) represents the objective function of the fuzzy c-means (FCM) algorithm; X represents the set of cluster prototype vectors; N represents the total number of measurement points; c represents the number of clusters; When the objective function reaches the minimum value, the iteration ends.
8. The method of using the high-precision measuring device for the thermal error of the ball screw according to claim 5, characterized in that: In the said Step 3, obtain the ideal movement trajectory of the panel (10) through the driver of the control box (2), obtain the actual movement trajectory of the panel (10) through the grating scale (18), and subtract the ideal movement trajectory from the actual movement trajectory to obtain the thermal error.