Noise level prediction device

The noise level prediction device calculates gear tooth surface characteristics to forecast noise levels using machine learning, addressing the need for post-assembly measurement and improving gear quality evaluation efficiency.

JP2024168967AInactive Publication Date: 2024-12-05TOYOTA JIDOSHA KK
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2023086064
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-25
Publication Date
2024-12-05
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing gear manufacturing processes require post-assembly noise level measurement, complicating the evaluation of gear quality due to potential noise level exceedance despite shape errors within acceptable ranges.

Method used

A noise level prediction device that calculates the waviness of gear tooth surfaces and uses a learned prediction model to forecast noise levels based on variables like tooth profile roundness, crowning, pressure angle error, and helix angle error, utilizing a machine learning approach.

Benefits of technology

Enables early prediction of gear noise levels, simplifying quality assessment and allowing for timely adjustments in manufacturing conditions or tool replacement, thereby enhancing efficiency and accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024168967000001_ABST
    Figure 2024168967000001_ABST
Patent Text Reader

Abstract

To provide a noise level prediction device capable of predicting the noise level of a gear.SOLUTION: A noise level prediction device is provided, comprising a computation unit for computing the amount of undulation of a tooth surface based on the shape of the tooth surface of a gear, and a prediction unit configured to predict the noise level of the gear based on the computed amount of undulation using a machine-trained prediction model employing explanatory variables including the amount of undulation of the tooth surface of the gear and using the noise level of the gear as an objective variable.SELECTED DRAWING: Figure 2
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a noise level prediction device. [Background technology]

[0002] There is a device that evaluates the quality of gears based on the error in the shape of the gear tooth flanks (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2005-195360 A Summary of the Invention [Problem to be solved by the invention]

[0004] Even if such errors are within the appropriate range, when the gear is incorporated into a product and used, the noise level generated may exceed the specified value. In such cases, the manufacturing conditions of the gear must be reviewed. In this way, it is necessary to measure the noise level of the gear after incorporating the manufactured gear into the product, which makes the work complicated.

[0005] SUMMARY OF THE PRESENT EMBODIMENT An object of the present invention is to provide a noise level prediction device capable of predicting the noise level of a gear. [Means for solving the problem]

[0006] The above object can be achieved by a noise level prediction device including: a calculation unit that calculates an amount of waviness on a tooth flank of a gear based on a shape of the tooth flank; and a prediction unit that predicts the noise level of the gear based on the calculated amount of waviness using a machine-learned prediction model that includes the amount of waviness on the tooth flank of the gear as an explanatory variable and has the noise level of the gear as an objective variable. Effect of the Invention

[0007] According to the present invention, a noise level prediction device capable of predicting the noise level of a gear can be provided. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a schematic diagram of a noise level prediction system. [Diagram 2] 4 is a flowchart illustrating a noise level prediction control. [Diagram 3] 11 is a diagram illustrating an example of predicted values ​​of a noise level displayed on a display device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] 1 is a schematic diagram of a noise level prediction system 1. The noise level prediction system 1 includes a prediction device 10, a measuring device 20, an input device 30, and a display device 40. The measuring device 20, the input device 30, and the display device 40 are connected to the prediction device 10. The prediction device 10 is a computer including an arithmetic processing circuit that performs various arithmetic processing related to the control of the measuring device 20 and the display device 40, and a memory that stores programs and data for control. The prediction device 10 functionally realizes a calculation unit and a prediction unit, which will be described later, and also executes noise level prediction control.

[0010] The measuring device 20 is a stylus-type measuring device that measures the shape of the tooth flank of the gear G. The input device 30 is a device for inputting information to the prediction device 10, and includes, for example, a keyboard, a mouse, and a touch panel. The display device 40 is a device for displaying predetermined information by the prediction device 10, and includes, for example, a liquid crystal display.

[0011] 2 is a flowchart illustrating the noise level predictive control. The predicting device 10 executes the noise level predictive control as follows. The predicting device 10 measures the shape of the tooth flank of the gear G in the tooth profile direction and the tooth trace direction by the measuring device 20 (step S1).

[0012] The prediction device 10 calculates the amount of waviness, etc. of the tooth flank of the gear G based on the measurement results (step S2). Specifically, the prediction device 10 calculates the tooth profile roundness, crowning, pressure angle error, helix angle error, and amount of waviness of the tooth flank of the gear G. The pressure angle error and tooth profile roundness are calculated based on the measurement results of the tooth profile direction. Step S2 is an example of a process executed by the calculation unit.

[0013] The helix angle error and crowning are calculated based on the measurement results in the tooth trace direction. The tooth profile roundness is the maximum deviation between a straight line connecting one end and the other end of the inspection range on the tooth surface in the tooth profile direction and a regression curve between the one end and the other end, when viewed from the tooth trace direction. The regression curve is a quadratic curve obtained by measuring multiple points within the inspection range on the tooth surface and using the least squares method from these multiple points. The crowning is the maximum deviation between a straight line connecting one end and the other end of the inspection range on the tooth surface in the tooth trace direction and a regression curve between the one end and the other end, when viewed from the tooth profile direction. The pressure angle error is the difference between the actual pressure angle and the desired pressure angle. The helix angle error is the difference between the actual helix angle and the desired helix angle. The waviness is the error in the height direction (in other words, the depth direction) due to unevenness with respect to the ideal tooth surface from the root to the tip in the tooth profile direction. The amount of waviness is calculated by subtracting the waviness component and the twist angle error from the shape of the tooth flank.

[0014] Next, the prediction device 10 predicts the noise level of the gear G based on the above calculation results using a prediction model (step S3). The prediction model is a model that has been machine-learned by supervised learning. The explanatory variables of this prediction model are the tooth profile roundness, crowning, pressure angle error, helix angle error, and waviness amount. The objective variable of the prediction model is the noise level of the gear G. More specifically, the noise level of the gear G is the magnitude of noise relative to the vibration frequency of the gear G when the gear G is incorporated into a specific product and used. As the supervised learning, for example, any one or a plurality of logistic regression, random forest, support vector machine, k-nearest neighbor method, and neural network may be used. The prediction device 10 inputs the calculated tooth profile roundness, crowning, pressure angle error, helix angle error, and waviness amount into the prediction model. As a result, the prediction model outputs the noise level of the gear G. The prediction device 10 displays the predicted value of the noise level on the display device 40. Step S3 is an example of a process executed by the prediction unit.

[0015] 3 is a diagram showing an example of the predicted noise level values ​​displayed on the display device 40. The vertical axis indicates the predicted noise level values, and the horizontal axis indicates the vibration frequency. The display device 40 also displays a standard line. In this manner, the display device 40 displays whether the predicted noise level values ​​exceed the specified values. This allows the operator to easily determine whether the gear G is acceptable or not.

[0016] If the predicted value of the noise level exceeds the specified value, the worker can adjust the manufacturing conditions, for example, the processing conditions or replace the processing tool. As described above, the prediction device 10 can predict the noise level of the gear G in advance before the gear G is incorporated into a target product and used. This allows the worker to grasp the noise level of the gear G at an early stage, making the work easier. Furthermore, the prediction device 10 can accurately predict the noise level of the gear G by using the above-mentioned prediction model.

[0017] In the above embodiment, the explanatory variables of the prediction model are, but are not limited to, the tooth profile roundness, the crowning, the pressure angle error, the helix angle error, and the amount of waviness. The explanatory variables may include the amount of waviness.

[0018] In the above embodiment, the objective variable of the prediction model is the noise level for each vibration frequency, but is not limited to this. For example, the objective variable of the prediction model may be the maximum value of the noise level.

[0019] Although the embodiments of the present invention have been described in detail above, the present invention is not limited to such specific embodiments, and various modifications and variations are possible within the scope of the gist of the present invention described in the claims. [Explanation of symbols]

[0020] 1. Noise level prediction system 10 Prediction device (calculation unit, prediction unit) 20 Measuring instruments 30 Input Devices 40 Display device G Gear

Claims

[Claim 1] a calculation unit that calculates an amount of waviness of the tooth flank based on a shape of the tooth flank of a gear; a prediction unit that predicts the noise level of the gear based on the calculated amount of waviness using a machine-learned prediction model that includes the amount of waviness of the tooth surface of the gear as an explanatory variable and has the noise level of the gear as an objective variable.

Citation Information

Patent Citations

  • Tooth surface error evaluation device for gearwheel pair, evaluation program therefor, and manufacturing method for gearwheel pair using the same

    JP2005195360A