A worm wheel grinding surface waviness prediction method, device and equipment

Abstract

This invention relates to the fields of machine learning and tooth surface waviness prediction. It discloses a method, apparatus, and device for predicting the waviness of tooth surfaces ground by worm gear grinding. The method includes: acquiring vibration signals from the grinding wheel spindle in three directions during worm gear grinding of helical gears and annotating them with semantic information to obtain an annotated dataset; constructing a tooth surface waviness prediction model based on the Goose optimization algorithm and optimizing an Extreme Learning Machine (ELM); inputting the vibration signals from the annotated dataset into the tooth surface waviness prediction model, and iteratively optimizing the input weights and biases of the ELM using a parameter optimization module to obtain a trained tooth surface waviness prediction model; and real-time acquiring the vibration signals of the workpiece being processed, extracting features from the real-time vibration signals using a feature extraction module, and then inputting the extracted features into the trained tooth surface waviness prediction model to obtain the ghost-order amplitude of the tooth surface waviness. This invention can improve the accuracy and efficiency of waviness prediction for worm gear grinding.

Description

Technical Field

[0001] This invention belongs to the field of machine learning and tooth surface waviness prediction technology, and particularly relates to a method, device and equipment for predicting the waviness of tooth surfaces ground by worm gear grinding. Background Technology

[0002] With the technological iteration in high-precision transmission fields such as new energy vehicles and aerospace, gears, as core components of power transmission, have their tooth surface morphology machining quality directly affecting equipment service performance and lifespan. Existing research has demonstrated that tooth surface morphology has a significant impact on contact fatigue life, gear noise, and other performance parameters. There will inevitably be a difference between the machined tooth surface morphology and the ideal tooth surface morphology. The deviation between the two surfaces consists of tooth profile / tooth direction error, tooth surface waviness, and tooth surface roughness. These deviations will lead to transmission errors, which is one of the main sources of gear noise.

[0003] Noise caused by tooth surface waviness can be assigned to the gear meshing frequency or intermediate frequencies in spectrum analysis. The number of wavinesses per workpiece rotation corresponds to the workpiece's order. If the number of wavinesses does not correspond to the gear meshing order or its multiples, the waviness is called "ghost order." When the ghost order amplitude is too large, the gear tooth surface waviness will produce abnormal noise, which is considered a machining defect. However, tooth surface waviness is affected by many factors such as worm wheel wear, grinding process parameters, worm wheel dressing defects, machine tool vibration, and gear installation errors. In actual machining, it is difficult to accurately control tooth surface waviness. Existing technologies also use tooth surface waviness measuring instruments to detect the ghost order of tooth surface waviness, helping gear manufacturers to detect and control gear quality. However, these devices are not only expensive, but can only be measured after the workpiece is machined, and each measurement takes a lot of time. The above-mentioned offline detection methods have significant limitations in terms of cost, timeliness, and efficiency. Summary of the Invention

[0004] This invention provides a method, apparatus, and equipment for predicting the waviness of worm gear grinding surfaces, which can improve the accuracy and efficiency of waviness prediction for worm gear grinding surfaces.

[0005] To achieve the above objectives, the present invention provides a method for predicting the waviness of the tooth surface of a worm gear grinding wheel, comprising: Vibration signals of the grinding wheel spindle in three directions during worm gear grinding of helical gears were acquired and semantically labeled to obtain a labeled dataset; A tooth surface waviness prediction model based on the goose optimization algorithm to optimize the extreme learning machine is constructed. The tooth surface waviness prediction model includes a feature extraction module, a parameter optimization module, and a waviness prediction module. The vibration signals from the labeled dataset are input into the tooth surface waviness prediction model. The parameter optimization module is used to iteratively optimize the input weights and biases of the extreme learning machine to obtain the trained tooth surface waviness prediction model. The real-time vibration signal of the workpiece is acquired in real time, and the feature extraction module is used to extract the features of the real-time vibration signal. Then, the signal is input into the trained tooth surface waviness prediction model to obtain the ghost order amplitude of tooth surface waviness.

[0006] To address the aforementioned problems, the present invention also provides a device for predicting the waviness of the tooth surface ground by a worm gear grinding wheel, the device comprising: The data acquisition module is used to acquire vibration signals of the grinding wheel spindle in three directions during worm grinding of helical gears and to annotate them with semantic information to obtain an annotated dataset. The model building module is used to build a tooth surface waviness prediction model based on the goose optimization algorithm to optimize the extreme learning machine. The tooth surface waviness prediction model includes a feature extraction module, a parameter optimization module, and a waviness prediction module. The vibration signal in the labeled dataset is input into the tooth surface waviness prediction model. The parameter optimization module is used to iteratively optimize the input weights and biases of the extreme learning machine to obtain the trained tooth surface waviness prediction model. The tooth surface waviness ghost order amplitude prediction module is used to collect real-time vibration signals of the machined workpiece, and then input the real-time vibration signals into the trained tooth surface waviness prediction model after feature extraction using the feature extraction module to obtain the tooth surface waviness ghost order amplitude.

[0007] To address the above problems, the present invention also provides an electronic device, the electronic device comprising: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the waviness prediction method for worm gear grinding surfaces described above.

[0008] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one computer program, which is executed by a processor in an electronic device to implement the above-described method for predicting the waviness of the tooth surface of a worm gear grinding wheel.

[0009] This invention acquires vibration signals from the grinding wheel spindle in three directions during the grinding of helical gears using a worm gear grinding wheel and annotates them with semantic information to obtain an annotated dataset. This dataset captures the most direct and comprehensive physical characteristics of the grinding process. The semantic annotation establishes a quantitative mapping relationship between the abstract vibration waveform and the actual waviness quality. Furthermore, a tooth surface waviness prediction model based on the goose optimization algorithm and optimized extreme learning machine is constructed. This model includes a feature extraction module, a parameter optimization module, and a waviness prediction module, which can leverage machine learning models to improve the accuracy and generalization ability of the prediction model. Finally, the vibration signals from the annotated dataset are input into the tooth surface waviness... The prediction model utilizes a parameter optimization module to iteratively optimize the input weights and biases of the extreme learning machine, resulting in a trained tooth surface waviness prediction model. This model can adapt to signal variations under different gear specifications and grinding conditions, ensuring high recognition accuracy even when dealing with minute ghost-order features. Finally, real-time vibration signals of the machined workpiece are acquired and extracted using a feature extraction module before being input into the trained tooth surface waviness prediction model to obtain the ghost-order amplitude of tooth surface waviness. This improves the accuracy and efficiency of waviness prediction for worm wheel grinding. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating a method for predicting the waviness of a worm gear grinding tooth surface according to an embodiment of the present invention. Figure 2 A functional block diagram of a worm gear grinding tooth surface waviness prediction device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device for implementing the waviness prediction method for worm gear grinding tooth surfaces, as provided in an embodiment of the present invention.

[0011] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0012] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0013] This application provides a method for predicting the waviness of the tooth surface of a worm gear grinding wheel. The execution subject of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for predicting the waviness of the tooth surface of a worm gear grinding wheel can be executed by software or hardware installed on a terminal device or a server device. The software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0014] Reference Figure 1 The diagram shown is a flowchart illustrating a method for predicting the waviness of worm gear grinding tooth surfaces according to an embodiment of the present invention. In this embodiment, the method for predicting the waviness of worm gear grinding tooth surfaces includes: S1. Obtain the vibration signals of the grinding wheel spindle in three directions during the grinding of helical gears by worm gear grinding and annotate them with semantic information to obtain an annotated dataset.

[0015] Understandably, worm wheel grinding is a high-efficiency gear finishing process. Its principle is similar to worm gear transmission. The continuously rotating worm-shaped grinding wheel and the gear workpiece generate motion according to a specific transmission ratio, thereby grinding the tooth surface.

[0016] Understandably, semantic information annotation refers to the process of associating the collected raw vibration signals with their actual physical meanings during the machine learning data preprocessing stage. The annotated data becomes "label data," which is used to train the model to learn the mapping relationship between "vibration signal" and "ripple ghost amplitude."

[0017] Specifically, vibration signals of the grinding wheel spindle in three directions during worm gear grinding of helical gears are acquired and semantically annotated to obtain an annotated dataset, including: Vibration signals of the grinding wheel spindle in three directions are collected by an accelerometer installed on the grinding wheel spindle. Fourier transform is performed on the processed helical gear to obtain the waviness ghost order amplitude. The waviness ghost order amplitude is then used to annotate the collected vibration signal with semantic information to obtain an annotated dataset.

[0018] It is understandable that the three directions of the grinding wheel spindle refer to the X, Y, and Z directions of the grinding wheel spindle.

[0019] Understandably, the grinding wheel spindle is the core actuator of a machine tool, which drives the worm wheel to rotate at high speed. During the grinding process, the grinding wheel spindle carries the cutting force and can directly transmit dynamic vibrations caused by grinding wheel imbalance, machine tool transmission errors, or improper process parameters.

[0020] Understandably, the waviness ghost magnitude refers to the severity of a specific waviness formed on the tooth surface due to machining abnormalities.

[0021] Understandably, Fourier transform is a mathematical analysis method that converts the time-domain signal of tooth surface morphology to the frequency domain. By performing Fourier transform on the full tooth surface morphology error obtained by the gear inspection instrument, the complex error can be decomposed into simple harmonic waves of different frequencies, which can accurately identify periodic fluctuations of a specific order.

[0022] S2. Construct a tooth surface waviness prediction model based on the goose optimization algorithm to optimize the extreme learning machine. The tooth surface waviness prediction model includes a feature extraction module, a parameter optimization module, and a waviness prediction module.

[0023] Understandably, the Goose Optimization Algorithm is a metaheuristic algorithm that simulates the intelligent behavior of a flock of geese. It is inspired by the behavior of geese during migration and defense, such as standing on one leg to guard. It imitates the migratory behavior of geese and finds the optimal solution to the problem by simulating the dynamic adjustment strategy of the flock during the migration process.

[0024] For example, the specific process of the goose optimization algorithm is as follows: Step 1: First, detect the initial input weights and biases of ELM and calculate the Mean Absolute Percentage Error (MAPE). Set the maximum number of iterations and the number of geese in the flock. Step 2: Calculate the objective function value and update the optimal individual information X for each iteration of the population. The objective function value is MAPE, and the individual information X consists of the input weights and biases. If the value of the random parameter rnd is greater than or equal to 0.5, then proceed to the development phase.

[0025] Phase 1: Development phase. Using the current "optimal vibration feature-ripple mapping" relationship, local fine optimization of hyperparameters is performed, with the goal of quickly converging to a hyperparameter combination that accurately predicts ripple.

[0026] Step 3: If the stable random number SRN is greater than or equal to 12 and the value of the local convergence tendency parameter pro is greater than 0.2, then update the input weights and biases using equation (1).

[0027] (1); In the formula:x i For optimal input weights and biases; SLR This serves as the "step size reference" for local searches. DF This refers to the "distance feature" between the current optimal solution and the individual. TS The "time scale" refers to the time scale in the iterative process.

[0028] Step 4: If the stability random number is less than 12 and the value of pro is less than 0.2, then update the input weights and biases using equation (2).

[0029] (2); In the formula: IW The "influence weight" of group information; RC The "randomness control" representing the exploration can be calculated using the following formula: (3); (4); Phase 2: Exploration phase. Breaking free from the constraints of the "current best vibration characteristics", we will try new hyperparameters on a large scale and aggressively, with the goal of discovering the potential mapping relationship between "new vibration signal characteristics and ripple".

[0030] Step 5: rand The value is less than 0.5. Set the parameter. MT and exploration intensity coefficient EIC Used to improve the searchability of hyperparameters.

[0031] (5); In the formula: t This represents the current iteration number; MT The maximum number of iterations is given. The input weights and biases are updated using Equation (6). Through "random exploration + elite guidance", a new "vibration signal feature-ripple" mapping is discovered, which improves the generalization ability of the model.

[0032] (6); In the formula: randn(1, dim) represents the "randomness" of global exploration; X Best Preserve strategies for elites, explore based on historical best solutions, and avoid complete randomness.

[0033] Step Six: Update complete. Save the optimal input weights and bias assignments from each iteration. Step 7: Determine whether the mean absolute percentage error (MAPE) value of this iteration is smaller than that of the previous iteration. If so, assign the input weights and biases of this iteration to X; otherwise, retain the previous X value.

[0034] Step 8: Determine if the number of iterations has reached the maximum value. If not, return to step 2. If it has, output the X value, which is the optimal input weight and bias. Assign it to the Extreme Learning Machine (ELM) model to achieve the best model performance.

[0035] Furthermore, hyperparameters determine the optimization depth and global exploration capability. Based on the required hyperparameter settings, a grid search method is used to optimize the number of iterations (range [100, 200, 300, 400, 500, 600]) and population size (range [10, 20, 30, 40, 50, 60]), resulting in 36 parameter combinations. Each parameter combination is input into the GOOSE-ELM model and trained. After training, a validation set is used for verification. After traversing all 36 parameter combinations in the grid, the mean absolute percentage error (MAPE) scores of all combinations are compared. The parameter combination with the lowest MAPE is determined as the optimal hyperparameter combination. Experimental results show that when the number of iterations is 400 and the population size is 40, the model achieves the lowest MAPE on the test set, reaching optimal performance.

[0036] Specifically, the feature extraction module is used to perform noise reduction and feature extraction on the collected vibration signal, and to reduce the dimensionality of the extracted features to generate an optimized feature set.

[0037] Understandably, the feature parameters extracted by the feature extraction module include: peak value, valley value, peak-to-peak value, mean, root mean square value, standard deviation, skewness, kurtosis, waveform factor, average absolute differential value, maximum spectral amplitude, average spectral amplitude, spectral energy, dominant frequency, wavelet energy features, average peak value, peak standard deviation, average valley value, valley standard deviation, average peak interval, average valley interval, peak density, and valley density.

[0038] Furthermore, the noise reduction and feature extraction of the acquired vibration signals, and the dimensionality reduction of the extracted features to generate an optimized feature set, include: A fast threshold denoising method based on fast Fourier transform is used to denoise vibration signals; Extract the characteristic parameters of the noise-reduced vibration signal; Kernel principal component analysis was used to reduce the dimensionality of the extracted feature parameters, and features whose cumulative contribution rate reached the preset contribution rate threshold were selected to form an optimized feature set.

[0039] Understandably, Kernel Principal Component Analysis (KPCA) is a nonlinear extension of Principal Component Analysis (PCA). KPCA uses a "kernel function" to map the original features to a high-dimensional space, thereby extracting the principal feature components that are orthogonal (uncorrelated) to each other in that space. It can more effectively preserve the nonlinear structural features in the original data than ordinary PCA.

[0040] Understandably, the preset contribution rate threshold refers to a dividing line for the cumulative contribution rate, which is set at 90%.

[0041] Furthermore, the waviness prediction module is an extreme learning machine of a single hidden layer feedforward neural network, used to receive the optimized feature set and combine it with the input weights and biases optimized by the parameter optimization module, and output the predicted ghost order amplitude of the tooth surface waviness.

[0042] Understandably, Extreme Learning Machine (ELM) refers to a training algorithm for feedforward neural networks with a single hidden layer, which has extremely fast learning speed, strong generalization ability, and the ability to avoid getting trapped in local optima.

[0043] For example, the Extreme Learning Machine (ELM) is an intelligent mapping tool for "features → ripple". The input layer receives vibration features, the hidden layer extracts core patterns, and the output layer generates predicted values. The input layer has 18 nodes and is responsible for receiving an input feature vector of length 18. The hidden layer has 9 nodes, and this layer performs a fixed transformation on the input using GOOSE-optimized input weights, biases, and nonlinear activation functions. The output layer has 1 node, and the predicted ghost magnitude is obtained by linear weighted summation.

[0044] Understandably, a single-hidden-layer feedforward neural network refers to a basic and powerful fully connected artificial neural network topology, in which the data flow is unidirectional (from the input layer to the output layer).

[0045] S3. Input the vibration signal from the labeled dataset into the tooth surface waviness prediction model, and use the parameter optimization module to iteratively optimize the input weights and biases of the extreme learning machine to obtain the trained tooth surface waviness prediction model.

[0046] Specifically, the parameter optimization module iteratively optimizes the input weights and biases of the extreme learning machine, including: Initialize the input weights and biases of the extreme learning machine, and set the maximum number of iterations and the number of geese in the flock; Calculate the objective function value of each individual in the population in each iteration and generate random numbers, wherein the objective function value is the mean absolute percentage error; The value of the random number determines whether to enter the development or exploration phase, and the input weights and biases of the individual are updated accordingly. If the resulting prediction error is less than the error of the previous iteration, the updated input weights and biases are retained. The iterative update process is repeated until the maximum number of iterations is reached, and the optimal input weights and biases that achieve the minimum mean absolute percentage error are assigned to the extreme learning machine.

[0047] Furthermore, updating the individual's input weights and biases based on whether it has entered the development or exploration phase includes: When the random number is greater than or equal to the threshold, the development phase begins. Based on the relationship between the random number and the local convergence tendency parameter, a preset adjustment formula is selected to perform local fine-tuning of the input weights and biases. When the random number is less than the threshold, the exploration phase begins. Based on the randomness of global exploration and the elite retention strategy, the input weights and biases are updated using the exploration intensity coefficient to obtain a new feature-ripple mapping relationship.

[0048] Understandably, the development phase refers to the process by which the algorithm performs a local search in the vicinity of a known potential optimal solution.

[0049] Understandably, the exploration phase refers to the process by which the algorithm performs a wide-ranging search across all possible combinations of input weights and biases.

[0050] S4. Real-time vibration signal of the workpiece is acquired, and the real-time vibration signal is extracted using the feature extraction module and then input into the trained tooth surface waviness prediction model to obtain the ghost order amplitude of tooth surface waviness.

[0051] This invention acquires vibration signals from the grinding wheel spindle in three directions during the grinding of helical gears using a worm gear grinding wheel and annotates them with semantic information to obtain an annotated dataset. This dataset captures the most direct and comprehensive physical characteristics of the grinding process. The semantic annotation establishes a quantitative mapping relationship between the abstract vibration waveform and the actual waviness quality. Furthermore, a tooth surface waviness prediction model based on the goose optimization algorithm and optimized extreme learning machine is constructed. This model includes a feature extraction module, a parameter optimization module, and a waviness prediction module, which can leverage machine learning models to improve the accuracy and generalization ability of the prediction model. Finally, the vibration signals from the annotated dataset are input into the tooth surface waviness... The prediction model utilizes a parameter optimization module to iteratively optimize the input weights and biases of the extreme learning machine, resulting in a trained tooth surface waviness prediction model. This model can adapt to signal variations under different gear specifications and grinding conditions, ensuring high recognition accuracy even when dealing with minute ghost-order features. Finally, real-time vibration signals of the machined workpiece are acquired and extracted using a feature extraction module before being input into the trained tooth surface waviness prediction model to obtain the ghost-order amplitude of tooth surface waviness. This improves the accuracy and efficiency of waviness prediction for worm wheel grinding.

[0052] like Figure 2 The diagram shown is a functional block diagram of a worm gear grinding tooth surface waviness prediction device provided in an embodiment of the present invention.

[0053] The waviness prediction device 100 for worm gear grinding tooth surfaces described in this invention can be installed in an electronic device. Depending on the functions implemented, the waviness prediction device 100 for worm gear grinding tooth surfaces may include a data acquisition module 101, a model construction module 102, and a tooth surface waviness ghost-order amplitude prediction module 103.

[0054] The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0055] In this embodiment, the functions of each module / unit are as follows: The data acquisition module 101 is used to acquire vibration signals of the grinding wheel spindle in three directions during worm gear grinding of helical gears and to annotate them with semantic information to obtain an annotated dataset.

[0056] The model building module 102 is used to build a tooth surface waviness prediction model based on the goose optimization algorithm to optimize the extreme learning machine. The tooth surface waviness prediction model includes a feature extraction module, a parameter optimization module, and a waviness prediction module. The vibration signal in the labeled dataset is input into the tooth surface waviness prediction model, and the parameter optimization module is used to iteratively optimize the input weights and biases of the extreme learning machine to obtain the trained tooth surface waviness prediction model.

[0057] The tooth surface waviness ghost order amplitude prediction module 103 is used to collect the real-time vibration signal of the processed workpiece in real time, and after extracting the features of the real-time vibration signal using the feature extraction module, input it into the trained tooth surface waviness prediction model to obtain the tooth surface waviness ghost order amplitude.

[0058] like Figure 3 The diagram shown is a schematic representation of an electronic device for implementing a method for predicting the waviness of a worm gear grinding tooth surface, according to an embodiment of the present invention.

[0059] The electronic device may include a processor 10, a memory 11, a communication bus 12, and a communication interface 13. It may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a method program for predicting the waviness of the tooth surface of a worm gear grinding wheel.

[0060] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a method for predicting the waviness of tooth surfaces in worm gear grinding), and calls data stored in the memory 11 to perform various functions of the electronic device and process data.

[0061] The memory 11 includes at least one type of readable storage medium, including flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of an electronic device, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device, such as a plug-in portable hard drive, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc. Furthermore, the memory 11 can include both internal and external storage units of the electronic device. The memory 11 can be used not only to store application software and various types of data installed on the electronic device, such as the code of a method for predicting the waviness of a worm gear grinding tooth surface, but also to temporarily store data that has been output or will be output.

[0062] The communication bus 12 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable communication between the memory 11 and at least one processor 10, etc.

[0063] The communication interface 13 is used for communication between the aforementioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, Bluetooth interface, etc.), typically used to establish communication connections between the electronic device and other electronic devices. The user interface may be a display, an input unit (such as a keyboard), or optionally, a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device and to display a visual user interface.

[0064] Figure 3 Only electronic devices with components are shown; those skilled in the art will understand that... Figure 3The structure shown does not constitute a limitation on the electronic device and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0065] For example, although not shown, the electronic device may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0066] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0067] The memory 11 in the electronic device stores a program for predicting the waviness of the tooth surface of a worm gear grinding wheel. This program is a combination of multiple instructions, which, when run in the processor 10, can achieve the following: Vibration signals of the grinding wheel spindle in three directions during worm gear grinding of helical gears were acquired and semantically labeled to obtain a labeled dataset; A tooth surface waviness prediction model based on the goose optimization algorithm to optimize the extreme learning machine is constructed. The tooth surface waviness prediction model includes a feature extraction module, a parameter optimization module, and a waviness prediction module. The vibration signals from the labeled dataset are input into the tooth surface waviness prediction model. The parameter optimization module is used to iteratively optimize the input weights and biases of the extreme learning machine to obtain the trained tooth surface waviness prediction model. The real-time vibration signal of the workpiece is acquired in real time, and the feature extraction module is used to extract the features of the real-time vibration signal. Then, the signal is input into the trained tooth surface waviness prediction model to obtain the ghost order amplitude of tooth surface waviness.

[0068] Specifically, the specific implementation method of the processor 10 for the above instructions can be referred to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, and will not be repeated here.

[0069] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0070] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following: Vibration signals of the grinding wheel spindle in three directions during worm gear grinding of helical gears were acquired and semantically labeled to obtain a labeled dataset; A tooth surface waviness prediction model based on the goose optimization algorithm to optimize the extreme learning machine is constructed. The tooth surface waviness prediction model includes a feature extraction module, a parameter optimization module, and a waviness prediction module. The vibration signals from the labeled dataset are input into the tooth surface waviness prediction model. The parameter optimization module is used to iteratively optimize the input weights and biases of the extreme learning machine to obtain the trained tooth surface waviness prediction model. The real-time vibration signal of the workpiece is acquired in real time, and the feature extraction module is used to extract the features of the real-time vibration signal. Then, the signal is input into the trained tooth surface waviness prediction model to obtain the ghost order amplitude of tooth surface waviness.

[0071] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0072] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0073] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0074] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0075] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0076] The blockchain referred to in this invention is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.

[0077] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0078] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a system claim may also be implemented by a single unit or device through software or hardware. The terms "first," "second," etc., are used to indicate names and do not indicate any specific order.

[0079] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for predicting the waviness of the tooth surface ground by a worm gear grinding wheel, characterized in that, The method includes: Vibration signals of the grinding wheel spindle in three directions during worm gear grinding of helical gears were acquired and semantically labeled to obtain a labeled dataset; A tooth surface waviness prediction model based on the goose optimization algorithm to optimize the extreme learning machine is constructed. The tooth surface waviness prediction model includes a feature extraction module, a parameter optimization module, and a waviness prediction module. The vibration signals from the labeled dataset are input into the tooth surface waviness prediction model. The parameter optimization module is used to iteratively optimize the input weights and biases of the extreme learning machine to obtain the trained tooth surface waviness prediction model. The real-time vibration signal of the workpiece is acquired in real time, and the feature extraction module is used to extract the features of the real-time vibration signal. Then, the signal is input into the trained tooth surface waviness prediction model to obtain the ghost order amplitude of tooth surface waviness.

2. The method for predicting the waviness of the tooth surface ground by a worm gear grinding wheel as described in claim 1, characterized in that, The process involves acquiring vibration signals from the grinding wheel spindle in three directions during worm gear grinding of helical gears and annotating them with semantic information to obtain an annotated dataset, including: Vibration signals of the grinding wheel spindle in three directions are collected by an accelerometer installed on the grinding wheel spindle. Fourier transform is performed on the processed helical gear to obtain the waviness ghost order amplitude. The waviness ghost order amplitude is then used to annotate the collected vibration signal with semantic information to obtain an annotated dataset.

3. The method for predicting the waviness of the tooth surface ground by a worm gear grinding wheel as described in claim 1, characterized in that, The feature extraction module is used to perform noise reduction and feature extraction on the collected vibration signals, and to reduce the dimensionality of the extracted features to generate an optimized feature set.

4. The method for predicting the waviness of the tooth surface ground by a worm gear grinding wheel as described in claim 3, characterized in that, The process of denoising and extracting features from the acquired vibration signals, and then reducing the dimensionality of the extracted features to generate an optimized feature set, includes: A fast threshold denoising method based on fast Fourier transform is used to denoise vibration signals; Extract the characteristic parameters of the noise-reduced vibration signal; Kernel principal component analysis was used to reduce the dimensionality of the extracted feature parameters, and features whose cumulative contribution rate reached the preset contribution rate threshold were selected to form an optimized feature set.

5. The method for predicting the waviness of the tooth surface ground by a worm gear grinding wheel as described in claim 1, characterized in that, The waviness prediction module is an extreme learning machine of a single hidden layer feedforward neural network, which receives the optimized feature set and combines the input weights and biases optimized by the parameter optimization module to output the predicted ghost order amplitude of tooth surface waviness.

6. The method for predicting the waviness of the tooth surface ground by a worm gear grinding wheel as described in claim 1, characterized in that, The iterative optimization of the input weights and biases of the extreme learning machine using the parameter optimization module includes: Initialize the input weights and biases of the extreme learning machine, and set the maximum number of iterations and the number of geese in the flock; Calculate the objective function value of each individual in the population in each iteration and generate random numbers, wherein the objective function value is the mean absolute percentage error; The value of the random number determines whether to enter the development or exploration phase, and the input weights and biases of the individual are updated accordingly. If the resulting prediction error is less than the error of the previous iteration, the updated input weights and biases are retained. The iterative update process is repeated until the maximum number of iterations is reached, and the optimal input weights and biases that achieve the minimum mean absolute percentage error are assigned to the extreme learning machine.

7. The method for predicting the waviness of the tooth surface ground by a worm gear grinding wheel as described in claim 6, characterized in that, The process of updating the input weights and biases of individuals based on whether they have entered the development or exploration phase includes: When the random number is greater than or equal to the threshold, the development phase begins. Based on the relationship between the random number and the local convergence tendency parameter, a preset adjustment formula is selected to perform local fine-tuning of the input weights and biases. When the random number is less than the threshold, the exploration phase begins. Based on the randomness of global exploration and the elite retention strategy, the input weights and biases are updated using the exploration intensity coefficient to obtain a new feature-ripple mapping relationship.

8. A device for predicting the waviness of tooth surfaces ground by a worm gear grinding wheel, characterized in that, The apparatus is used to implement the waviness prediction method for worm gear grinding tooth surfaces as described in any one of claims 1 to 7, and the apparatus comprises: The data acquisition module is used to acquire vibration signals of the grinding wheel spindle in three directions during worm grinding of helical gears and to annotate them with semantic information to obtain an annotated dataset. The model building module is used to build a tooth surface waviness prediction model based on the goose optimization algorithm to optimize the extreme learning machine. The tooth surface waviness prediction model includes a feature extraction module, a parameter optimization module, and a waviness prediction module. The vibration signal in the labeled dataset is input into the tooth surface waviness prediction model. The parameter optimization module is used to iteratively optimize the input weights and biases of the extreme learning machine to obtain the trained tooth surface waviness prediction model. The tooth surface waviness ghost order amplitude prediction module is used to collect real-time vibration signals of the machined workpiece, and then input the real-time vibration signals into the trained tooth surface waviness prediction model after feature extraction using the feature extraction module to obtain the tooth surface waviness ghost order amplitude.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the waviness prediction method for worm gear grinding tooth surfaces as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the waviness prediction method for the tooth surface of a worm gear grinding wheel as described in any one of claims 1 to 7.

Images