A method and system for regulating the rotational speed of a swaging spindle

Abstract

The present application relates to the technical field of spindle speed regulation, and particularly relates to a kind of rotary division processing spindle speed regulation method and system, the method steps include: based on workpiece tooth number / edge number and cutter disc cutter tooth number, calculate target speed ratio;Based on workpiece tooth number / edge number and workpiece shaft preset speed or cutter disc cutter tooth number and cutter disc preset speed, calculate the contact interval of workpiece and cutter disc cutter under target speed ratio, defined as theoretical contact interval;In the process of rotary division processing, the contact interval of workpiece and cutter disc cutter is collected, defined as actual contact interval;Based on theoretical contact interval and actual contact interval, calculate the matching degree of spindle speed, based on the matching degree of spindle speed, judge whether to carry out spindle speed regulation;Calculate the spindle speed regulation amount, and regulate the spindle speed through the driving end of spindle.The present application can effectively reduce the delay of spindle speed regulation and the deviation of matching degree of spindle speed and workpiece speed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of spindle speed regulation, and particularly relates to a method and system for regulating the speed of a main shaft in rotary dividing machining. BACKGROUND

[0002] The core of rotary dividing machining is to realize synchronous cutting and indexing through the fixed speed ratio of the cutter head and the workpiece, and the essential feature is that the contact interval of the cutter tooth and the workpiece is fixed, which is determined by the target speed ratio, the number of cutter teeth and the number of workpiece teeth.

[0003] In the prior art, the matching of the spindle speed and the workpiece speed is monitored by relying on a speed sensor, and the speed ratio is calculated after the speed of the cutter head and the speed of the workpiece are collected. The matching degree of the spindle speed and the workpiece speed is determined by the speed signal. This method often needs to accumulate a certain speed deviation to discover the mismatch, which can easily lead to abnormal cutting of multiple cutter teeth before triggering the speed regulation, and has a monitoring and regulation lag. In addition, the speed sensor is usually arranged at the driving end of the main shaft, for example, an absolute optical encoder is used as the speed sensor and is integrated in the main shaft to monitor the rotor speed of the main shaft and the position of the cutter head. However, the actual speed signal of the cutter head at the cutting end of the main shaft is easily affected by the transmission gap, sensor error and mechanical vibration interference, resulting in deviation in the matching degree of the spindle speed and the workpiece speed.

[0004] Based on the above problems, a spindle speed regulation technology suitable for rotary dividing machining is proposed. SUMMARY

[0005] To solve the above problems, the present application provides a method and system for regulating the speed of a main shaft in rotary dividing machining, which can reduce the delay of spindle speed regulation and the deviation in the matching degree of the spindle speed and the workpiece speed.

[0006] To achieve the above purpose, the technical scheme of the present application is as follows:

[0007] On the one hand, a method for regulating the speed of a main shaft in rotary dividing machining is provided, which comprises,

[0008] calculating a target speed ratio based on the number of workpiece teeth / edges and the number of cutter teeth;

[0009] calculating the contact interval of the workpiece and the cutter tooth at the target speed ratio based on the number of workpiece teeth / edges and the preset speed of the workpiece shaft or the number of cutter teeth and the preset speed of the cutter head, which is defined as the theoretical contact interval;

[0010] During the rotary dividing machining process, the contact interval of the workpiece and the cutter tooth is collected, which is defined as the actual contact interval;

[0011] calculating the matching degree of the spindle speed based on the theoretical contact interval and the actual contact interval, and determining whether to regulate the spindle speed based on the matching degree of the spindle speed.

[0012] The main shaft rotation speed regulation amount is calculated, and the main shaft rotation speed is regulated through the driving end of the main shaft.

[0013] Further, the target rotation speed ratio calculation formula is as follows:

[0014]

[0015] Wherein, is the number of teeth / edges of the workpiece; is the number of teeth of the cutter head.

[0016] Further, the theoretical contact interval calculation formula is as follows:

[0017]

[0018] Wherein, is the theoretical contact interval, used to represent the fixed contact period of the cutter tooth and the workpiece under the target rotation speed ratio; is the number of teeth / edges of the workpiece; is the number of teeth of the cutter head, is the preset rotation speed of the main shaft; is the preset rotation speed of the workpiece shaft.

[0019] Further, the actual contact interval acquisition method is one of the following:

[0020] (1) The driving end current signal is acquired through the current sensor electrically connected with the driving end of the main shaft, and when the current signal fluctuation reaches the preset threshold, it is determined that contact occurs;

[0021] (2) The pressure signal is acquired through the pressure sensor arranged between the cutter head and the cutter tooth, and when the pressure signal appears effective fluctuation, it is determined that contact occurs;

[0022] The calculation formula of the actual contact interval is:

[0023]

[0024] Wherein, is the actual contact interval; k is the contact number sequence number of the workpiece and the cutter tooth of the cutter head, k=1, 2, …, n.

[0025] Further, the main shaft rotation speed matching degree calculation formula is as follows:

[0026]

[0027] Wherein, is the theoretical contact interval; is the actual contact interval;

[0028] The matching degree threshold is set , the matching degree threshold is proportional to the workpiece machining precision; when , it is determined that the spindle speed is mismatched, triggering the spindle speed regulation.

[0029] Further, when judging whether to perform spindle speed regulation, a deviation prediction step is further included:

[0030] First, calculate the adjacent contact interval deviation:

[0031]

[0032] wherein, is the theoretical contact interval; is the actual contact interval;

[0033] Then construct the time series deviation dataset ; and perform enhancement processing and normalization processing on the dataset:

[0034]

[0035] wherein, is the maximum value of the historical deviation, is the minimum value of the historical deviation;

[0036] Then divide the input deviation sequence using a sliding window to divide the training set, validation set and test set;

[0037] Finally, based on the lightweight LSTM model, construct and train the deviation prediction model, input the deviation sequence into the trained deviation prediction model, and output the predicted deviation .

[0038] The deviation prediction model training uses the mean square error loss function:

[0039]

[0040] wherein, is the kth predicted deviation, is the kth actual deviation.

[0041] Further, the spindle speed regulation amount calculation formula is:

[0042]

[0043] wherein:

[0044] the proportional term , is the proportional coefficient;

[0045] the integral term , is the integral coefficient;

[0046] prediction compensation term , prediction compensation coefficient

[0047] further, the prediction compensation coefficient is a proportional coefficient derived by the following formula:

[0048]

[0049] wherein, selected based on load and processing accuracy requirements;

[0050] the integral coefficient is derived by the following formula:

[0051]

[0052] wherein, based on prediction deviation selected.

[0053] further, the prediction compensation coefficient the value range is 0.8~1.2; the value is inversely proportional to the matching degree of spindle speed.

[0054] In another aspect, a rotary dividing machining spindle speed control system is provided, comprising:

[0055] a data acquisition module for acquiring workpiece tooth number / edge number, cutter disc cutter tooth number, spindle preset speed and workpiece shaft preset speed, and cutter tooth and workpiece contact signal;

[0056] a data processing module for calculating target speed ratio, theoretical contact interval and actual contact interval;

[0057] a data output module for calculating speed matching degree, judging whether to perform spindle speed control, calculating spindle speed control amount, and sending the spindle speed control amount to the driving end.

[0058] The above scheme has the following beneficial effects:

[0059] 1. The prior art relies on speed sensor to collect speed and then indirectly calculates speed ratio, and needs to accumulate multiple sets of speed data to find mismatch; compared with the prior art, the present scheme directly collects the contact interval of cutter tooth and workpiece, calculates the actual contact interval and judges the speed matching degree in a single contact, shortens the time from mismatch occurrence to monitoring and identification, reduces the delay of spindle speed control, can trigger control quickly, avoids multi-cutter abnormal cutting, and significantly reduces the scrap rate.

[0060] 2、The prior art is that the speed sensor is installed on the driving end of the main shaft, there is a transmission gap between the cutting end where the cutter head is located, and the signal is easily affected by mechanical vibration and electromagnetic interference, and the matching degree of the judged speed is easy to deviate; compared with the prior art, the contact interval characteristics of the rotary division processing are taken as the judgment basis, the contact signal directly reflects the cutting division synchronization, and is not affected by the transmission gap and sensor error interference; through main shaft speed matching degree calculation and deviation prediction, the speed matching degree can be accurately judged.

[0061] 3、The prior art is that the deviation is adjusted after exceeding the threshold value, and frequent mismatching will seriously affect the processing stability; compared with the prior art, the future contact interval deviation is predicted through the deviation prediction model, early prediction and active regulation are realized, the speed mismatching frequency is reduced, and the processing stability is improved. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 It is a whole flow schematic diagram of the rotary division processing main shaft speed regulation method embodiment of the application;

[0063] Figure 2 It is a deviation prediction model architecture schematic diagram of the rotary division processing main shaft speed regulation method embodiment of the application;

[0064] Figure 3 It is a working flow schematic diagram of the rotary division processing main shaft speed regulation method embodiment of the application;

[0065] Figure 4 It is a structure schematic diagram of the rotary division processing main shaft speed regulation system embodiment of the application. DETAILED DESCRIPTION

[0066] The technical solutions of the application will be described clearly and completely below with reference to the drawings. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0067] In the description of the application, it should be noted that the terms “center”, “upper”, “lower”, “left”, “right”, “vertical”, “horizontal”, “inner”, “outer” and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the application. In addition, the terms “first”, “second”, “third” are only for the purpose of description, and cannot be understood as indicating or implying relative importance.

[0068] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connecting" should be understood in a broad sense, for example, it can be fixed connection, or detachable connection, or integrally connected; it can be mechanical connection, or electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or the internal communication of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0069] The following will be further described in detail through specific embodiments:

[0070] The hardware level of the rotary dividing machine includes the following improvements: first, a hardware for monitoring the contact between the workpiece and the cutter head teeth is arranged on the electric spindle; in the present embodiment, the hardware for monitoring the contact between the workpiece and the cutter head teeth is a current sensor, which is electrically connected with the driving end of the spindle. The current sensor is used to collect the current signal of the driving end, and the contact between the workpiece and the cutter head teeth is judged based on the fluctuation of the current signal reaching the fluctuation threshold. The fluctuation frequency of the current signal is used as the contact interval of the workpiece and the cutter head teeth. In some other embodiments, the hardware for monitoring the contact between the workpiece and the cutter head teeth is a pressure sensor, which is arranged between the cutter head and the cutter teeth. The pressure sensor is used to collect the pressure signal between the cutter head and the cutter teeth, and the contact between the workpiece and the cutter head teeth is judged based on the pressure signal between them. The pressure signal fluctuation frequency is used as the contact interval of the workpiece and the cutter head teeth.

[0071] Secondly, an additional data processing unit is added to judge and record the contact interval between the workpiece teeth and the cutter head teeth; in the present embodiment, the data processing unit can be a control component such as a microprocessor, a controller, etc., for example, it can be a general central processing unit (CPU), a general processor, a digital signal processing (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The data processing unit is used to record the time stamp of the contact time when the cutter head teeth and the workpiece occur cutting contact each time . In some other embodiments, a data processing module is integrated in the numerical control system, which is used to record the time stamp of the contact time when the cutter head teeth and the workpiece occur cutting contact each time .

[0072] On the basis of the above hardware level improvement of the rotary dividing machine,Figure 1 The spindle speed regulation method includes the following steps as shown in the figure:

[0073] S101: Obtain the workpiece tooth number / edge number and cutter disc cutter tooth number from the numerical control system of the rotary dividing machine tool, and calculate the target speed ratio based on the workpiece tooth number and the cutter disc cutter tooth number. Specifically, the workpiece tooth number / edge number is the tooth number / edge number of the workpiece to be machined, which can be a tooth to be machined workpiece, a chamfering workpiece or an edge cutting workpiece, etc. When the workpiece to be machined is a tooth to be machined workpiece and a chamfering workpiece, the workpiece tooth number is used for target speed ratio calculation, and when the workpiece to be machined is an edge cutting workpiece, the workpiece edge number is used for target speed ratio calculation. The specific calculation formula of the target speed ratio is as follows:

[0074]

[0075] Wherein, is the workpiece tooth number / edge number; is the cutter disc cutter tooth number; is the spindle speed (preset by the numerical control system); is the workpiece shaft speed (preset by the numerical control system); the spindle speed and the workpiece shaft speed can also be directly obtained from the numerical control system of the rotary dividing machine tool.

[0076] S102: Calculate the contact interval of the workpiece tooth and the cutter disc cutter under the target speed ratio based on the workpiece tooth number and the workpiece speed or the cutter disc cutter tooth number and the cutter disc speed, defined as the theoretical contact interval. The specific calculation formula of the theoretical contact interval is as follows:

[0077]

[0078] S103: In the rotary dividing machining process, the contact interval of the workpiece tooth and the cutter disc cutter is collected, defined as the actual contact interval; in this embodiment, the actual contact interval is calculated based on the driving end current fluctuation frequency of the spindle collected by the current sensor. Specifically, the time stamp of the contact time recorded by the data processing unit is extracted, and the time difference of the contact of adjacent workpiece and cutter disc cutter, i.e. the actual contact interval, is calculated:

[0079]

[0080] Wherein, k is the contact number sequence of the workpiece and the cutter disc cutter, k=1, 2, …, n.

[0081] S104: Calculate the spindle speed matching degree based on the theoretical contact interval and the actual contact interval, and judge whether to perform spindle speed regulation based on the spindle speed matching degree. Specifically, first, based on the machining precision requirement of the workpiece, set the matching degree threshold , the matching degree threshold is proportional to the machining precision of the workpiece, and when the spindle speed matching degree is lower than the matching degree threshold When the ratio is less than the threshold value, it is determined that the spindle speed is mismatched, and spindle speed regulation is required.

[0082]

[0083] Preferably, the matching degree threshold value is set to 99.5%, and when the ratio is less than the threshold value, it is determined that the spindle speed is mismatched, and spindle speed regulation is required.

[0084] S105: When the spindle speed is mismatched, the spindle speed regulation amount is calculated, and the spindle speed is regulated through the driving end of the spindle. Specifically, the regulation principle is to maintain the target speed ratio i, and by adjusting the spindle speed, the actual contact interval is returned to the theoretical contact interval, and the speed ratio of the cutting end of the cutter head to the workpiece is returned to the target speed ratio.

[0085] In this embodiment, in order to more efficiently and timely regulate the spindle speed to reduce the probability of abnormal cutting, it is predicted whether spindle speed regulation is required, and the advance regulation is triggered based on the prediction result. Specifically, first, the theoretical contact interval is obtained, and the actual contact interval is obtained in real time; then the adjacent contact interval deviation is calculated.

[0086]

[0087] wherein, the adjacent contact interval deviation, is the actual contact interval at the kth contact, is the theoretical contact interval; k is the contact number sequence of the workpiece and the cutter head, k = 1, 2, …, n.

[0088] Then, based on the sliding window, the first N contact interval deviations are selected, (N is a positive integer), and the prediction deviation of the future contact interval is output through the deviation prediction model .

[0089] The deviation prediction model of this embodiment is integrated in the data processing unit and is constructed based on the LSTM model, mainly including the following steps: obtaining the above-mentioned adjacent contact interval deviation, based on the time stamp of the time when the cutter head and the workpiece are in cutting contact , constructing a time series deviation data set ​​where m is the total number of contacts between the cutter and the workpiece. Preferably, a timing deviation dataset with m > 1000 is constructed for bias prediction model training to ensure the accuracy of the bias prediction model training. The data is enhanced by adding common interference scene data (such as material hard points, tool slight wear, load fluctuations) in the rotary division processing to generate simulated deviation data to avoid overfitting of the bias prediction model; for example: artificially adding random deviation of ± 0.0001 ~ ± 0.0005s to simulate slight speed fluctuations. The sliding window is used to divide the input deviation sequence (the deviation time sequence of the previous n cutter and workpiece contact intervals), and the window size N is preferably 5-10, i.e. each input is the previous 5-10 deviations ; the corresponding output is the k+1 deviation , i.e. In addition, the input deviation sequence also needs to be normalized:

[0090]

[0091] where, is the maximum value of the historical deviation, is the minimum value of the historical deviation; normalizing the input deviation sequence can avoid the influence of the order of magnitude difference on the bias prediction model training.

[0092] Considering the low delay control requirement of the spindle speed in rotary division processing, a lightweight LSTM model architecture is adopted to avoid the delay caused by complex networks. Preferably, the lightweight LSTM model architecture is as shown in Figure 2 , wherein the input dimension of the input layer is the size of the sliding window; the LSTM hidden layer contains 1-2 layers, each layer has 16-32 hidden units, and the activation function = tanh; the output dimension of the fully connected layer is 1; the activation function of the output layer = linear. The hyperparameters of the lightweight LSTM model: batch size = 8-16; learning rate = 0.001-0.005; training rounds = 50-100.

[0093] The timing deviation dataset is divided into training set (70%), validation set (20%), and test set (10%) according to the ratio of 7:2:1, the training set is used for bias prediction model parameter update, the validation set is used for overfitting monitoring, and the test set is used for final prediction accuracy verification. The mean square error is used to adapt to the bias regression task, and the formula is as follows:

[0094]

[0095] where, is the kth predicted deviation, is the kth actual deviation.

[0096] In some other embodiments, the bias prediction model is also optimized, for example, minimizing the loss function, dynamically adjusting the learning rate, pruning the network weights after training, reducing the computational load during inference, and ensuring low latency.

[0097] Based on the bias prediction model obtained above, the PID feedback regulation is integrated to realize the spindle speed regulation requirement prediction and control amount calculation. Specifically, the predicted bias is output by the bias prediction model before step S104 , and after the predicted bias is obtained in step S104, the spindle speed mismatch is determined in combination with the predicted bias : when and , it is determined that the spindle speed is not mismatched, and the current spindle speed is maintained. When or , it is determined that the spindle speed is mismatched, triggering step S105 to calculate the spindle speed control amount and regulating the spindle speed through the driving end of the spindle; in step S105, the PID regulation term is constructed:

[0098] the proportional term , wherein is the proportional coefficient;

[0099] the integral term , wherein is the integral coefficient;

[0100] the prediction compensation term , wherein is the prediction compensation coefficient.

[0101] The spindle speed control amount is calculated by combining the above-mentioned proportional term, integral term and prediction compensation term: , and output to the driving end of the spindle by the data processing unit, and the driving end of the spindle control amount is + , so that the actual contact interval returns to the theoretical contact interval .

[0102] In this embodiment, the proportional coefficient and the integral coefficient are coordinated with the bias prediction model, and the core principle is that the proportional coefficient increases with the increase of the cutting load, and the integral coefficient decreases with the increase of the cutting load; when the prediction result of the bias prediction model is too large, the proportional coefficient takes the upper limit to realize the rapid regulation of the spindle speed, and the corresponding integral coefficient takes the lower limit to avoid integral saturation; when the prediction result of the bias prediction model is too small, the proportional coefficient takes the lower limit.

[0103] Specifically, the value strategy of the proportional coefficient may be automatically selected based on the convergence of the adjacent contact interval deviation : if the deviation converges slowly (>5 contacts still do not return to the allowed range), the proportional coefficient is gradually increased by 20 r / min each time; if the spindle speed appears to be shaking (the contact interval deviation is large and small), the proportional coefficient is gradually reduced by 20 r / min each time. In some other embodiments, the proportional coefficient can also be directly calculated by the following formula :

[0104]

[0105] wherein, the proportional coefficient is selected based on the load and the machining precision requirement.

[0106] Specifically, the value strategy of the integral coefficient may be automatically selected based on the average of several contact interval deviations: if the average of several contact interval deviations , the integral coefficient is gradually increased by 0.05 r / (min·s) each time; if integral saturation occurs, i.e., the speed continues to be adjusted in one direction, the integral coefficient is gradually reduced by 0.05 r / (min·s) each time. In some other embodiments, the integral coefficient can also be calculated by the following formula :

[0107]

[0108] wherein, the integral coefficient is selected based on the predicted deviation .

[0109] In the present embodiment, the predicted compensation coefficient has a value range of 0.8-1.2, and is dynamically corrected according to the speed matching degree M: taking the matching degree threshold of 99.5% as an example, when , ; when , ; when , .

[0110] As shown in Figure 3 , the working process of the above method is as follows:

[0111] Initialization: collect the number of workpiece teeth, the number of cutter teeth, and the target speed ratio, automatically calculate the theoretical contact interval and the matching degree threshold; ​​

[0112] Signal acquisition: record the time stamp of each tooth cutting contact with the workpiece , and collect the spindle speed and workpiece shaft speed;

[0113] Matching degree and prediction calculation: real-time calculation of actual contact interval , spindle speed matching degree And the predicted deviation ; if Or , trigger the spindle speed regulation;

[0114] Dynamic regulation: calculate the spindle speed regulation amount , the drive end of the spindle executes the regulation; repeat signal acquisition-matching degree and prediction calculation-dynamic regulation until And .

[0115] Based on the above spindle speed regulation method, a spindle speed regulation system adapted to it is constructed, as shown in Figure 4 , mainly composed of data acquisition module, data processing module and data output module connected in turn. The data acquisition module is connected with the numerical control system and the current sensor signal, and the data output module is connected with the drive end of the spindle. Among them, the data acquisition module is used to obtain the workpiece tooth number / edge number and cutter tooth number from the numerical control system, and synchronously obtain the current signal collected by the current sensor; the data processing module is used to calculate the target speed ratio based on the workpiece tooth number / edge number and the cutter tooth number; at the same time, based on the workpiece tooth number / edge number and the workpiece speed or the cutter tooth number and the cutter speed, the contact interval of the workpiece and the cutter tooth under the target speed ratio is calculated, which is defined as the theoretical contact interval; the contact interval of the workpiece and the cutter tooth is calculated based on the current signal collected by the current sensor, which is defined as the actual contact interval. The data output module is used to calculate the spindle speed matching degree based on the theoretical contact interval and the actual contact interval, and to judge whether to perform spindle speed regulation based on the spindle speed matching degree; when it is judged that spindle speed regulation is needed, the spindle speed regulation amount is calculated and sent to the drive end of the spindle.

[0116] Obviously, the above embodiments are only examples for clear illustration, and not limitation of the embodiments. For ordinary skilled in the art, other different forms of changes or variations can be made on the basis of the above description. Here, it is not necessary and impossible to enumerate all the embodiments. The obvious changes or variations derived therefrom are still within the protection scope of the present application.

Claims

1. A method for regulating the rotational speed of a swaging spindle, characterized in that, The application relates to a method for controlling the rotation speed of a main shaft during the rotation dividing process of a workpiece, and belongs to the technical field of rotation dividing. The target rotation speed ratio is calculated based on the number of workpiece teeth / edges and the number of cutter teeth; The contact interval between the workpiece and the cutter teeth under the target rotation speed ratio is calculated based on the number of workpiece teeth / edges and the preset rotation speed of the workpiece shaft or the number of cutter teeth and the preset rotation speed of the cutter, and is defined as the theoretical contact interval; During the rotation dividing process, the contact interval between the workpiece and the cutter teeth is collected, and is defined as the actual contact interval; The main shaft rotation speed matching degree is calculated based on the theoretical contact interval and the actual contact interval, and whether the main shaft rotation speed control is performed is judged based on the main shaft rotation speed matching degree; When judging whether the main shaft rotation speed control is performed, the bias prediction step is further included: Firstly, the adjacent contact interval bias is calculated: ； wherein is the theoretical contact interval; is the actual contact interval; Reconstructing a timing bias dataset ; and performing augmentation and normalization on the dataset: ； wherein is a maximum value of the historical deviation, is a minimum value of the historical deviation; Then, the input bias sequence is divided into a training set, a verification set and a test set by using a sliding window; Finally, a bias prediction model is constructed and trained based on the lightweight LSTM model. By inputting the bias sequence into the trained bias prediction model, the predicted bias is output ; The bias prediction model training adopts a mean square error loss function: ； wherein, is the kth predicted deviation, is the kth actual deviation; The main shaft rotation speed control amount is calculated, and the main shaft rotation speed is controlled through the driving end of the main shaft. The main shaft rotation speed control amount calculation formula is as follows: ； Wherein: proportional term , is a proportionality factor; integral term , is an integral coefficient; prediction compensation term , is a prediction compensation coefficient.

2. The method of claim 1, wherein the rotational speed of the main shaft of the swaging machine is controlled by a control unit. The target rotation speed ratio calculation formula is as follows: ； wherein, is the number of teeth / edges of the workpiece; is the number of teeth of the cutter head.

3. The method of claim 1, wherein the rotational speed of the main shaft of the swaging machine is controlled by a control unit. The theoretical contact interval calculation formula is as follows: ； wherein, is the theoretical contact interval, used to characterize the fixed contact period of the tool teeth with the workpiece at the target rotational speed ratio; is the number of workpiece teeth / edges; is the number of tool teeth of the tool holder, is the preset rotational speed of the main shaft; is the preset rotational speed of the workpiece shaft.

4. The method of claim 1, wherein the rotational speed of the main shaft of the swaging machine is controlled by a control unit. The collection mode of the actual contact interval is one of the following: (1) The driving end current signal is collected through the current sensor electrically connected with the driving end of the main shaft, and the contact is determined to occur when the current signal fluctuation reaches a preset threshold; (2) The pressure signal is collected through the pressure sensor arranged between the cutter and the cutter teeth, and the contact is determined to occur when the pressure signal appears effective fluctuation; The actual contact interval calculation formula is as follows: ； wherein, is the actual contact interval; k is the contact number of the workpiece and the cutter head tooth, k = 1, 2, …, n.

5. The method of claim 1, wherein the rotational speed of the main shaft of the swaging machine is controlled by a controller. The main shaft rotation speed matching degree calculation formula is as follows: ； wherein is the theoretical contact interval; is the actual contact interval; Setting a matching degree threshold , the matching degree threshold is proportional to the workpiece machining accuracy; when , it is determined that the spindle speed is mismatched, triggering spindle speed regulation.

6. The method of claim 1, wherein the rotational speed of the main shaft of the swaging machine is controlled by a controller. The is a proportionality factor derived by the following equation: ； wherein, Based on load and processing precision requirements selection; The The integral coefficient is derived by the following equation: ； wherein based on the prediction bias selection.

7. The method of claim 1, wherein the rotational speed of the main shaft of the swaging machine is controlled by a controller. The prediction compensation coefficient The value range is 0.8-1.

2. The value is inversely proportional to the matching degree of the spindle speed.

8. A system for controlling the rotational speed of a machining spindle, particularly for controlling the rotational speed of a machining spindle according to any one of claims 1 to 7, characterized in that The application relates to a method for controlling the rotation speed of a main shaft during the rotation dividing process of a workpiece, and belongs to the technical field of rotation dividing. The data acquisition module is used for acquiring the number of workpiece teeth / edges, the number of cutter teeth, the preset rotation speed of the main shaft and the preset rotation speed of the workpiece shaft, and the contact signal of the cutter teeth and the workpiece; The data processing module is used for calculating the target rotation speed ratio, the theoretical contact interval and the actual contact interval; The data output module is used for calculating the rotation speed matching degree, judging whether the main shaft rotation speed control is performed, and further including the bias prediction step when judging whether the main shaft rotation speed control is performed: Firstly, the adjacent contact interval bias is calculated: ； wherein is the theoretical contact interval; is the actual contact interval; Reconstructing a timing skew dataset ; and performing augmentation and normalization on the dataset: ； wherein is a maximum value of the historical deviation, is a minimum value of the historical deviation; Then, the input bias sequence is divided into a training set, a verification set and a test set by using a sliding window; Finally, a bias prediction model is constructed and trained based on the lightweight LSTM model. By inputting the bias sequence into the trained bias prediction model, the predicted bias is output ; The bias prediction model training adopts a mean square error loss function: ； wherein, is the kth predicted deviation, is the kth actual deviation; The main shaft rotation speed control amount is calculated, and the main shaft rotation speed is controlled through the driving end of the main shaft. The main shaft rotation speed control amount calculation formula is as follows: ； Wherein: proportional term , is a proportionality factor; integral term , is an integral coefficient; prediction compensation term , is a prediction compensation coefficient.

Images