A planetary roller screw matching grinding online evaluation method and device

Abstract

The present invention provides an online evaluation method and device for planetary roller screw matching grinding, which belongs to the field of screw processing technology and solves the problem that the processing accuracy and assembly quality cannot be monitored online in the prior art. It includes a screw nut to be tested fixed on a machine tool at one end, and a standard screw is threadedly connected to the other end of the screw nut to be tested. The end of the standard screw away from the screw nut to be tested is connected to a torque sensor, and the torque sensor is connected to a mounting plate, and a laser rangefinder for monitoring the displacement distance of the standard screw is provided on the mounting plate. In the process of rotating the screw, the torque sensor monitors the torque applied to the standard screw in real time, and the laser rangefinder monitors the displacement distance of the standard screw in real time. The amount of machine tool to be modified is calculated by a computer. After using the device of the present application, the processing accuracy of the planetary roller screw can be improved, and its processing and assembly quality can be quantitatively evaluated, reducing the requirements for operators.

Description

Technical Field

[0001] The invention belongs to the technical field of screw processing, and in particular to an online evaluation method and device for matching grinding of a planetary roller screw. Background Art

[0002] The planetary roller screw is a transmission mechanism that can realize the mutual conversion between rotational motion and linear motion. It is mainly composed of parts such as screw rods, nuts, rollers, etc. It is significantly superior to traditional ball screws in core key performance indicators such as accuracy, life, and load. It has been used in new energy vehicles, high-end precision machine tools, humanoid robots and other fields.

[0003] Planetary roller screws are high-end precision products, and thread grinding is currently widely used to meet their high-precision manufacturing needs. Due to factors such as temperature changes, grinding wheel wear, tool setting, and grinding rod deformation, high-precision internal thread grinding is difficult to process in one go. However, the cumulative repeated positioning errors caused by repeated clamping will ultimately determine the product accuracy of the planetary roller screw. In the actual processing process, matching grinding is often used to avoid repeated clamping. The matching process requires skilled operators to make judgments based on their own experience, and it is impossible to form a standardized operating process. However, the assembly clearance is a key indicator parameter of the planetary roller screw.

[0004] Therefore, it is urgent to develop a device that can monitor processing accuracy and assembly quality online, form a set of product qualification evaluation methods, realize standardized and normalized operation procedures in the planetary roller screw processing process, and improve product performance. Summary of the invention

[0005] In view of the above problems, the purpose of the present invention is to provide an online evaluation method and device for planetary roller screw matching grinding. During the rotation of the screw, the torque sensor monitors the torque applied to the standard screw in real time, and the laser rangefinder monitors the displacement distance of the standard screw in real time. The torque and displacement distance data are transmitted to the computer, and the machine tool modification amount is calculated by the deep neural network algorithm model trained in the computer. The machine tool parameters are adjusted according to the machine tool modification amount, and the screw nut to be tested is processed again until the design requirements are met. After using the device of the present application, the processing accuracy of the planetary roller screw can be improved, the processing and assembly quality can be quantitatively evaluated, and the requirements for operators can be reduced, which is conducive to improving the product quality of the planetary roller screw and realizing standardized and standardized operation procedures in the processing process.

[0006] The technical solution adopted by the present invention is as follows:

[0007] An online evaluation device for planetary roller screw matching grinding comprises a screw nut to be tested with one end fixed on a machine tool, the other end of the screw nut to be tested is threadedly connected to a standard screw, the end of the standard screw away from the screw nut to be tested is connected to a torque sensor, the torque sensor is connected to a mounting plate, and a laser rangefinder for monitoring the displacement distance of the standard screw is arranged on the mounting plate.

[0008] Preferably, a sleeve is sleeved on the standard screw rod, a block is provided on the inner wall of the sleeve, a slot cooperating with the block is opened on the outer wall of the standard screw rod along the length direction, the sleeve is connected to a first flange, and second flanges are provided at both ends of the torque sensor, and the torque sensor is detachably connected to the first flange and the mounting plate through the second flange.

[0009] Preferably, the laser rangefinder is connected to a processor disposed on a mounting disk.

[0010] Preferably, the processor and the torque sensor are wirelessly connected to a computer.

[0011] Preferably, a machine headstock is provided on a side of the lead screw nut to be tested away from the standard lead screw, and a chuck for clamping the lead screw nut to be tested is provided on the machine headstock.

[0012] Preferably, a handle is provided on the mounting plate.

[0013] A planetary roller screw matching grinding online evaluation method is provided, wherein the planetary roller screw matching grinding online evaluation device is used for evaluation, and the evaluation method comprises the following steps:

[0014] S1. Establish a deep neural network algorithm model based on the model and database on the computer, construct a mapping relationship between the torque of the standard screw rod, the displacement distance parameter of the standard screw rod and the amount of the machine tool to be modified, and realize automatic calculation of the next step of the machine tool to be modified according to the measured torque and displacement distance parameters. In the early stage, the deep neural network algorithm model is trained through the database and model;

[0015] S2. The mounting plate is rotated by the handle, and the standard lead screw rotates synchronously with the mounting plate and is screwed into the lead screw nut to be tested. The torque sensor monitors the torque of the standard lead screw in real time, and the laser rangefinder monitors the displacement distance of the standard lead screw in real time.

[0016] S3. The torque and displacement distance data are transmitted to the computer. The machine tool modification amount is calculated by the deep neural network algorithm model trained in the computer. The machine tool parameters are adjusted according to the machine tool modification amount to process the screw nut to be tested again until the design requirements are met.

[0017] Preferably, in step S1, the machine tool quantities to be modified include taper and feed rate.

[0018] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0019] During the process of rotating the screw, the torque sensor monitors the torque applied to the standard screw in real time, and the laser rangefinder monitors the displacement distance of the standard screw in real time. The torque and displacement distance data are transmitted to the computer, and the deep neural network algorithm model trained in the computer calculates the amount of machine tool to be modified. The machine tool parameters are adjusted according to the amount of machine tool to be modified, and the screw nut to be tested is processed again until the design requirements are met. After using the device of the present application, the processing accuracy of the planetary roller screw can be improved, the processing and assembly quality can be quantitatively evaluated, and the requirements for operators can be reduced, which is conducive to improving the product quality of the planetary roller screw and realizing standardized and standardized operation procedures in the processing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 A schematic diagram of a three-dimensional structure provided by an embodiment of the present invention;

[0022] Figure 2 for Figure 1 The enlarged structural diagram at A in the middle;

[0023] Figure 3 A schematic diagram of the structure of the installation disk provided in an embodiment of the present invention;

[0024] Figure 4 A schematic diagram of a sleeve structure provided in an embodiment of the present invention;

[0025] Figure 5 A schematic diagram of the structure of a torque sensor provided in an embodiment of the present invention.

[0026] Figure markings: 1-machine headstock; 2-chuck; 3-screw nut to be tested; 4-standard screw; 5-sleeve; 6-first flange; 7-second flange; 8-laser rangefinder; 9-processor; 10-mounting plate; 11-handle; 12-torque sensor; 13-slot; 14-block. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0028] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] In the description of the present invention, it should be noted that if the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside" and the like appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the accompanying drawings, or is the orientation or position relationship in which the product of the application is usually placed when used. It is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0030] Combine the following Figure 1-Figure 5 The present invention is described in detail.

[0031] Example:

[0032] A planetary roller screw grinding online evaluation device, such as Figure 1 and 2 As shown, it includes a screw nut 3 to be tested, one end of which is fixed on the machine tool, the other end of the screw nut 3 to be tested is threadedly connected to a standard screw 4, the end of the standard screw 4 away from the screw nut 3 to be tested is connected to a torque sensor 12, the torque sensor 12 is connected to a mounting plate 10, and the mounting plate 10 is provided with a laser rangefinder 8 for monitoring the displacement distance of the standard screw 4.

[0033] The operator rotates the mounting disk 10, and the mounting disk 10 drives the standard screw 4 to rotate together through the torque sensor 12. During the rotation process, the torque sensor 12 monitors the torque applied to the standard screw 4 in real time, and the laser rangefinder 8 monitors the displacement distance of the standard screw 4 in real time. The machine tool modification amount is calculated based on the torque and displacement distance data, and the machine tool parameters are adjusted based on the machine tool modification amount to perform processing on the screw nut 3 to be tested again until the design requirements are met.

[0034] like Figure 2 and 4As shown, a sleeve 5 is sleeved on the standard screw rod 4, a block 14 is arranged on the inner wall of the sleeve 5, a slot 13 cooperating with the block 14 is opened on the outer wall of the standard screw rod 4 along the length direction, and the sleeve 5 is connected to a first flange 6, as shown in FIG. Figure 3 , 4 As shown in FIG. 5 , second flanges 7 are provided at both ends of the torque sensor 12 , and the torque sensor 12 is detachably connected to the first flange 6 and the mounting plate 10 via the second flange 7 .

[0035] The card slot 13 is connected to the end face of the standard screw 4, which facilitates the card block 14 to be inserted into the card slot 13. The sleeve 5 is limited to rotate on the standard screw 4 through the card block 14 and the card slot 13, ensuring that the mounting plate 10, the torque sensor 12 and the standard screw 4 rotate synchronously. The sleeve 5 and the torque sensor 12, and the torque sensor 12 and the mounting plate 10 are all removable, which is convenient for subsequent replacement of parts. Among them, the second flange 7 and the first flange 6 are detachably connected by a screw and a nut, and the second flange 7 and the mounting plate 10 are detachably connected by a screw; multiple card blocks 14 and card slots 13 can be set.

[0036] The sleeve 5 can be fixed on the standard lead screw 4 by a screw (not shown in the figure) to avoid axial sliding. Specifically, the screw penetrates the sleeve 5 along the radial thread of the sleeve 5 and abuts against the standard lead screw 4. If the sleeve 5 is not fixed, the sleeve 5 needs to be pressed axially during the rotation process to avoid axial displacement of the sleeve 5 during the synchronous rotation process with the standard lead screw 4, which affects the measurement data of the laser rangefinder 8.

[0037] The laser rangefinder 8 is connected to a processor 9 disposed on a mounting plate 10; the processor 9 and the torque sensor 12 are wirelessly connected to a computer. The processor 9 can process the data collected by the laser rangefinder 8 and transmit it to the computer, and the data collected by the torque sensor 12 is directly transmitted to the computer, and the computer automatically calculates the amount to be modified of the machine tool according to the torque and displacement distance parameters. Among them, the laser rangefinder 8 and the processor 9 are powered by batteries (not shown in the figure), and the batteries are disposed in the mounting plate 10; the wireless torque sensor 12 is also powered by batteries (not shown in the figure).

[0038] A machine headstock 1 is provided on the side of the screw nut 3 to be tested away from the standard screw 4, and a chuck 2 for clamping the screw nut 3 to be tested is provided on the machine headstock 1. The chuck 2 is a four-jaw chuck, which can stably clamp the standard screw 4. The laser emitted by the laser rangefinder 8 can hit the end face of the machine headstock 1 or the screw nut 3 to be tested.

[0039] like Figure 3 As shown, a handle 11 is provided on the mounting plate 10 , so that an operator can use the handle to rotate the mounting plate 10 conveniently.

[0040] An online evaluation method for planetary roller screw matching grinding is provided, wherein the online evaluation device for planetary roller screw matching grinding is used for evaluation, and the evaluation method comprises the following steps:

[0041] S1. Establish a deep neural network algorithm model based on the model and database on the computer, construct a mapping relationship between the torque of the standard screw rod 4, the displacement distance parameter of the standard screw rod 4 and the amount of the machine tool to be modified, and realize automatic calculation of the amount of the machine tool to be modified in the next step according to the measured torque and displacement distance parameters. In the early stage, the deep neural network algorithm model is trained through the database and model;

[0042] S2. The mounting plate 10 is rotated by the handle 11, and the standard screw 4 rotates synchronously with the mounting plate 10 and is screwed into the screw nut 3 to be tested. The torque sensor 12 monitors the torque of the standard screw 4 in real time, and the laser rangefinder 8 monitors the displacement distance of the standard screw 4 in real time;

[0043] S3, the torque and displacement distance data are transmitted to the computer, and the machine tool modification amount is calculated by the deep neural network algorithm model trained in the computer. The machine tool parameters are adjusted according to the machine tool modification amount to process the screw nut 3 to be tested again until the design requirements are met.

[0044] When the maximum torque measured over the full stroke of the standard screw 4 screwed into the screw nut 3 to be tested is greater than a set value, the standard screw 4 is screwed out and the grinding process of the screw nut 3 to be tested is continued until the maximum torque is less than the set value to meet the design requirements.

[0045] In step S1, the machine tool quantities to be modified include taper and feed rate. The taper is modified by calculation, and the feed rate is modified by manually adjusting process parameters.

[0046] About the calculation process of taper:

[0047] 1. Assume that the median diameter of the screw nut to be tested is According to the friction torque data tested by the device, the normal force between the contact interface of the screw nut 3 to be tested and the standard screw 4 is calculated by formulas (1) and (2);

[0048] The relationship between the friction torque and the axial force between the standard screw 4 and the screw nut 3 to be tested is:

[0049] (1)

[0050] In the formula, is the friction torque, is the median diameter of the screw nut 3 to be tested, It is the major diameter of the bearing surface at the initial end of the standard screw 4 thread. The minor diameter of the bearing surface at the initial end of the standard screw 4 thread. is the thread half angle, is the friction coefficient, is the thread pitch, The axial force on the screw nut 3 to be tested and the standard screw 4 during the rotation process;

[0051] The normal force at the contact interface between the screw nut 3 to be measured and the standard screw 4 during the screwing process of the standard screw 4 for:

[0052] (2)

[0053] In the formula, is the normal force between the contact interfaces, is the contact angle between the two interfaces;

[0054] 2. According to the normal force calculated in step 1, the number of asperity contacts n is calculated by formula (5), and then the total number of asperities N is calculated by formula (4). Finally, the nominal contact area An between the interface of the standard screw 4 and the screw nut 3 to be tested is calculated by formula (3);

[0055] The contact between two rough surfaces is equivalent to the contact between a smooth rigid plane and a rough surface. Assume that the radius of curvature of each micro-convex body is R1, z is the height of the micro-convex body, and d is the separation distance between the average height of the micro-convex body and the rigid plane. The above-mentioned interface contact characteristic parameters can be obtained by optical profilometer testing;

[0056] set up is the height distribution function of the micro-convex body, and it obeys Gaussian distribution. According to Hertz contact theory, the total number of micro-convex bodies N is:

[0057] (3)

[0058] In the formula, is the surface density function of the microconvex body; is the nominal contact area;

[0059] When the deformation of a single microconvex body is greater than 0, that is, When , the asperities will come into contact, and the number of asperities in contact n is:

[0060] (4)

[0061] The total contact load F is obtained from the Hertz contact theory:

[0062] (5)

[0063] In the formula, is the equivalent elastic modulus of the two contact bodies at the interface;

[0064] 3. Based on the nominal contact area An calculated in step 2, calculate the median diameter of the nut under this contact area using formula (8): ;

[0065] When the standard screw 4 is screwed into the screw nut 3 to be tested by self-rotation, the major diameter of the thread will increase linearly with the rotation angle, and the relationship is as follows:

[0066] (6)

[0067] In the formula, is the major diameter of the thread; It is the major diameter of the bearing surface at the initial end of the standard screw 4 thread. The standard lead screw has 4 tapers (i.e. the radius change per unit length). is the thread lead. Usually, the number of thread starts of the standard screw 4 and the screw nut 3 to be tested is the same, that is, the lead of the standard screw 4 is the product of the single-thread pitch of the screw nut 3 to be tested and the number of starts; It is a standard screw rod with 4 turns; It is the total rotation angle of the standard screw 4 during its rotation.

[0068] The relationship between the screw-in distance of the standard screw 4 and the nominal contact area An is as follows:

[0069] (7)

[0070] Formula (7) can be changed to:

[0071] (8)

[0072] 4. Comparison , If the error value between the two is less than the given error value, then is the median diameter of the nut at this position; if it is greater than the given error value, adjust Resolve until the final error is less than the given error value;

[0073] 5. Calculate the nut median diameter , the taper change is calculated using the variation law of the median diameter and the total length.

[0074] The deep neural network algorithm model of this application adopts the back propagation neural network (BP neural network). The BP neural network is a multi-layer feedforward neural network. It learns the parameters of the neural network based on the BP algorithm and has a strong ability to handle nonlinear problems. This application constructs a mapping relationship between parameters such as torque and displacement and the amount of machine tool to be trimmed (taper, feed rate) through the BP neural network.

[0075] The deep neural network algorithm model is mainly divided into the following processes:

[0076] 1. Forward propagation process: Let L be the number of layers of the neural network, that is, the overall output layer of the neural network is L, and the hidden layer is L-1 (the input layer is generally defined as the 0th layer of the network); is the connection weight between the kth neuron in the l-1th layer and the jth neuron in the lth layer, where l=1,2...,L, k=1,2...,K, j=1,2...,J, K is the number of neurons in the l-1th layer, and J is the number of neurons in the lth layer; is the threshold of the jth neuron in the lth layer; is the activation function of the lth layer of neurons; the activation value output of the jth neuron in the lth layer is for:

[0077] (9)

[0078] In the formula, is the output value of the l-1th layer of neurons. The input value of the input layer (i.e., the 0-layer network) in the present invention is the torque information M(x) at each position. The output value of each layer can be calculated through the BP neural network. The output value of the last layer is the machine tool taper Y(x1) and the feed amount adjustment amount Y(x2), thereby completing the forward transmission process of the input signal from layer to layer to the output layer output.

[0079] 2. Definition of loss function: During the BP neural network training process, the loss function measures the gap between the network output value and the expected value. It is an important indicator for evaluating the training effect of hyperparameters such as neural network connection weights and thresholds. The loss function in the form of mean square error is used in this invention:

[0080] (10)

[0081] In the above formula, is the expected value of machine tool taper, Y(x1) is the taper output value of the neural network, is the expected value of the feed amount, and Y(x2) is the output value of the neural network feed amount. The above formula can be used to calculate the error between the output value of the neural network and the expected value, so that the BP algorithm can be used to back propagate the error and update the parameters of each layer of neurons.

[0082] 3. Error back propagation process: The update of the parameters of each layer in the network is actually to regard the weights and thresholds of each layer of the neural network as the independent variables of the loss function, and regard each sample in the training set (including input values ​​and expected output values) as the known quantity of the loss function. The parameter values ​​are adjusted with the goal of minimizing the loss function. The gradient descent method is generally used to implement this process. The specific expression is as follows:

[0083] (11)

[0084] (12)

[0085] In the formula, is the learning rate, and its value is generally between 0 and 1. The smaller the parameter adjustment, the faster the model converges by selecting an appropriate learning rate; t is the number of iterations; is the connection weight between the kth neuron in the l-1 layer and the jth neuron in the l layer at the tth iteration, is the connection weight between the kth neuron in the l-1 layer and the jth neuron in the l layer at the t+1th iteration; is the threshold of the jth neuron in layer l at the tth iteration, is the threshold of the jth neuron in layer l at the t+1th iteration; is the error loss function at the tth iteration.

[0086] This application first collects a large amount of actual processing data and model data as a training set, and uses the training set data to train the core parameters of the BP neural network such as weights and thresholds. Finally, based on the BP neural network algorithm, a mapping relationship between parameters such as torque and displacement and the amount of machine tool to be trimmed (taper, feed rate) is constructed to realize the online processing and evaluation method of the internal thread of the planetary roller screw.

[0087] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A planetary roller screw matching grinding online evaluation method, characterized in that: An online evaluation device for planetary roller screw matching grinding is used for evaluation, the online evaluation device for planetary roller screw matching grinding comprising a screw nut (3) to be tested having one end fixed on a machine tool, the other end of the screw nut (3) to be tested being threadedly connected to a standard screw (4), the end of the standard screw (4) away from the screw nut (3) to be tested being connected to a torque sensor (12), the torque sensor (12) being connected to a mounting plate (10), a laser rangefinder (8) for monitoring the displacement distance of the standard screw (4) being arranged on the mounting plate (10), the laser rangefinder (8) being connected to a processor (9) arranged on the mounting plate (10), the processor (9) and the torque sensor (12) being wirelessly connected to a computer, and a handle (11) being arranged on the mounting plate (10); The evaluation method comprises the following steps: S1. Establishing a deep neural network algorithm model based on a model and a database on a computer, constructing a mapping relationship between the torque applied to the standard screw rod (4), the displacement distance parameter of the standard screw rod (4) and the amount of machine tool to be modified, and realizing automatic calculation of the amount of machine tool to be modified in the next step according to the measured torque and displacement distance parameters. In the early stage, the deep neural network algorithm model is first trained through the database and the model; S2, the mounting plate (10) is rotated by the handle (11), the standard screw (4) and the mounting plate (10) are rotated synchronously and screwed into the screw nut (3) to be tested, the torque sensor (12) monitors the torque applied to the standard screw (4) in real time, and the laser rangefinder (8) monitors the displacement distance of the standard screw (4) in real time; S3, the torque and displacement distance data are transmitted to the computer, and the machine tool modification amount is calculated by the deep neural network algorithm model trained in the computer. The machine tool parameters are adjusted according to the machine tool modification amount, and the screw nut (3) to be tested is processed again until the design requirements are met.

2. The on-line evaluation method for planetary roller screw matching grinding according to claim 1, characterized in that: In step S1, the machine tool quantities to be modified include taper and feed rate.

3. The on-line evaluation method for planetary roller screw matching grinding according to claim 1, characterized in that: The standard screw rod (4) is sleeved with a sleeve (5), the inner wall of the sleeve (5) is provided with a clamping block (14), the outer wall of the standard screw rod (4) is provided with a clamping groove (13) cooperating with the clamping block (14) along the length direction, the sleeve (5) is connected to a first flange (6), and second flanges (7) are provided at both ends of the torque sensor (12), and the torque sensor (12) is detachably connected to the first flange (6) and the mounting plate (10) via the second flange (7).

4. The on-line evaluation method for planetary roller screw matching grinding according to claim 1, characterized in that: A machine headstock (1) is arranged on the side of the screw nut (3) to be tested away from the standard screw (4), and a chuck (2) for clamping the screw nut (3) to be tested is arranged on the machine headstock (1).

Images