JP2020159945A - Abnormality diagnosis method, abnormality diagnosis device and abnormality diagnosis program for rolling bearing - Google Patents

Abnormality diagnosis method, abnormality diagnosis device and abnormality diagnosis program for rolling bearing Download PDF

Info

Publication number
JP2020159945A
JP2020159945A JP2019061225A JP2019061225A JP2020159945A JP 2020159945 A JP2020159945 A JP 2020159945A JP 2019061225 A JP2019061225 A JP 2019061225A JP 2019061225 A JP2019061225 A JP 2019061225A JP 2020159945 A JP2020159945 A JP 2020159945A
Authority
JP
Japan
Prior art keywords
frequency
specific
vibration
rotation
rotation frequency
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
JP2019061225A
Other languages
Japanese (ja)
Other versions
JP7146683B2 (en
Inventor
拓 杉浦
Hiroshi Sugiura
拓 杉浦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Okuma Corp
Original Assignee
Okuma Corp
Okuma Machinery Works Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Okuma Corp, Okuma Machinery Works Ltd filed Critical Okuma Corp
Priority to JP2019061225A priority Critical patent/JP7146683B2/en
Publication of JP2020159945A publication Critical patent/JP2020159945A/en
Application granted granted Critical
Publication of JP7146683B2 publication Critical patent/JP7146683B2/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

To estimate the kind of damage of a bearing even when a bearing specification is unknown.SOLUTION: An abnormality diagnosis method for rolling bearing includes: measuring and recording vibration acceleration when measurement start is commanded together with a diagnosis rotation frequency, and recording the rotation frequency and the vibration acceleration in association with each other (S1 to S6); calculating a rotation synchronous component emphasis waveform in S8 when the total measurement is completed in determination in S7; and assuming a ratio fundamental frequency and a ratio vibration source rotation frequency, determining whether or not there exists a vibration peak at a position of ratio fundamental frequency of rotation synchronous component emphasis waveform×natural number±ratio vibration source rotation frequency, and calculating a ratio (coincidence rate) when there exists the vibration peak in S9. When retrieval has been completed in the predetermined retrieval range in the determination of S10, the kind of bearing abnormality is determined according to the value of the ratio vibration source rotation frequency for the ratio fundamental frequency and the ratio vibration source rotation frequency whose the coincidence rate is equal to or higher than the predetermined threshold in S11, a degree of damage is further diagnosed in S12, and a diagnosis result is displayed in S13.SELECTED DRAWING: Figure 4

Description

本発明は、工作機械等に用いられて主軸等の回転体を支持する転がり軸受の異常を診断する方法及び装置、プログラムに関するものである。 The present invention relates to a method, an apparatus, and a program for diagnosing an abnormality of a rolling bearing used in a machine tool or the like to support a rotating body such as a spindle.

回転体を支持する転がり軸受に内輪の損傷などの異常が生じると振動が発生する。軸受の異常によって発生する力は単純な正弦波状ではないため、高調波の周波数成分の振動が同時に観測される。この際に発生する振動の周波数(特徴周波数)は、回転周波数に比例しており、回転体の回転周波数と軸受諸元から算出することが可能である。
例えば特許文献1では、振動加速度を測定してエンベロープ処理および周波数分析等を行った波形を、所定の関係式に基づいて算出した内輪、外輪、転動体などの特徴周波数とともに表示し、特徴周波数に対応する振動ピークがあるか否かを判断できるようにすることで、転がり軸受の損傷部位(内輪、外輪、転動体など)を特定する手法が示されている。
特許文献2では、振動を測定してエンベロープ処理および周波数分析し、所定の関係式に基づいて算出した内輪、外輪、転動体のそれぞれの特徴周波数の値を抽出して、損傷部位(内輪、外輪、転動体)でそれぞれ異なるしきい値を用いて診断する手法が示されている。
Vibration occurs when an abnormality such as damage to the inner ring occurs in the rolling bearing that supports the rotating body. Since the force generated by the bearing abnormality is not a simple sinusoidal shape, vibrations of the frequency components of the harmonics are observed at the same time. The frequency of vibration (characteristic frequency) generated at this time is proportional to the rotation frequency, and can be calculated from the rotation frequency of the rotating body and the bearing specifications.
For example, in Patent Document 1, a waveform obtained by measuring vibration acceleration and performing envelope processing, frequency analysis, etc. is displayed together with characteristic frequencies of inner rings, outer rings, rolling elements, etc. calculated based on a predetermined relational expression, and used as the characteristic frequencies. A method of identifying a damaged part (inner ring, outer ring, rolling element, etc.) of a rolling bearing is shown by making it possible to determine whether or not there is a corresponding vibration peak.
In Patent Document 2, vibration is measured, envelope processing and frequency analysis are performed, and the value of each characteristic frequency of the inner ring, outer ring, and rolling element calculated based on a predetermined relational expression is extracted, and the damaged portion (inner ring, outer ring). , Rolling bodies), and methods for diagnosing using different threshold values are shown.

特開昭63−297813号公報Japanese Unexamined Patent Publication No. 63-297813 特許第5146008号公報Japanese Patent No. 5146008

しかしながら、損傷部位の特定や損傷部位に応じたしきい値を用いるためには、基本周波数や特徴周波数を算出するための軸受諸元が必要となる。このため、軸受メーカの販売する軸受診断装置においては、その軸受メーカが軸受諸元を把握している自社の軸受については型式を入力することで診断ができるが、他社の軸受については軸受諸元の手入力を要求する実装となっていることが一般的である。この場合、軸受の諸元が入手できない場合には軸受を診断することができないといった課題がある。また、機械の製造者を除くと、その機械で使用されている軸受の型式を把握することさえ困難であるという課題がある。 However, in order to identify the damaged part and use the threshold value according to the damaged part, bearing specifications for calculating the fundamental frequency and the characteristic frequency are required. For this reason, in the bearing diagnostic equipment sold by a bearing manufacturer, it is possible to diagnose the bearings of the company whose bearing manufacturer knows the bearing specifications by inputting the model, but for the bearings of other companies, the bearing specifications. Generally, the implementation requires manual input of. In this case, there is a problem that the bearing cannot be diagnosed if the specifications of the bearing are not available. In addition, there is a problem that it is difficult to even grasp the model of the bearing used in the machine except for the manufacturer of the machine.

そこで、本発明は、上記問題に鑑みなされたものであって、診断対象の機械で使用されている軸受の軸受諸元が把握できない場合であっても、軸受の損傷の種類を推定することができる転がり軸受の異常診断方法及び異常診断装置、異常診断プログラムを提供することを目的とする。 Therefore, the present invention has been made in view of the above problems, and even when the bearing specifications of the bearing used in the machine to be diagnosed cannot be grasped, the type of bearing damage can be estimated. It is an object of the present invention to provide an abnormality diagnosis method, an abnormality diagnosis device, and an abnormality diagnosis program for rolling bearings.

上記目的を達成するために、請求項1に記載の発明は、回転体を支持する転がり軸受の異常を診断する方法であって、
前記回転体の振動を複数回測定する振動測定ステップと、
前記振動を周波数分析して周波数ごとの振動の大きさをそれぞれ求め、前記振動の周波数を回転周波数で除算して得られる無次元量を比周波数として、各回の測定で得られた前記振動の大きさを共通の前記比周波数ごとに算出し、前記比周波数ごとの振動の大きさについて係数が全て同符号の線型和を算出し、算出した値に基づいて回転同期成分強調波形を決定する回転同期成分強調波形算出ステップと、
前記回転同期成分強調波形の振動ピーク位置の規則性に基づいて、前記転がり軸受において力の発生する方向が変化する周波数である振動源回転周波数と前記回転周波数との比である比振動源回転周波数を推定し、前記比振動源回転周波数の値が1の場合には、前記転がり軸受の内輪損傷と判断し、前記比振動源回転周波数の値が前記転がり軸受の転動体の公転周波数を前記回転周波数で除算した値のとり得る範囲である場合には、前記転動体の損傷と判断し、前記比振動源回転周波数の値が0の場合には、前記転がり軸受の外輪損傷と判断する軸受損傷種類判別ステップと、を実行することを特徴とする。
請求項2に記載の発明は、請求項1の構成において、前記比振動源回転周波数の推定は、前記回転同期成分強調波形において、前記転がり軸受の損傷に対応する比基本周波数と前記比振動源回転周波数とを仮定して算出される特定の比周波数に振動ピークが存在する割合である一致率を算出し、前記一致率が所定のしきい値を超過する場合には、仮定した前記比基本周波数と仮定した前記比振動源回転周波数とにより表現される規則性が前記回転同期成分強調波形にあると判断し、仮定した前記比振動源回転周波数を採用することを特徴とする。
請求項3に記載の発明は、請求項1の構成において、前記比振動源回転周波数の推定は、前記回転同期成分強調波形において、ピーク対を構成する2つの振動ピークの間隔をピーク間距離とし、前記ピーク間距離が等しい2組のピーク対について算出したそれぞれのピーク対を構成する2つの前記振動ピークの前記比周波数の平均の比が自然数比の場合に、前記ピーク間距離の2分の1を前記比振動源回転周波数とする、又はある2つの振動ピークの前記比周波数の比が自然数比である場合に、前記比振動源回転周波数を0とすることを特徴とする。
請求項4に記載の発明は、請求項1乃至3の何れかの構成において、前記軸受損傷種類判別ステップは、前記転がり軸受の損傷に対応する比基本周波数と前記比振動源回転周波数とを推定する処理を含むものであり、前記転がり軸受の損傷の種類が判別された前記比基本周波数と前記比振動源回転周波数との組を用いて、判別された損傷の程度をさらに診断することを特徴とする。
上記目的を達成するために、請求項5に記載の発明は、回転体を支持する転がり軸受の異常を診断する方法であって、
前記回転体の振動を複数回測定する振動測定ステップと、
前記振動を周波数分析して周波数ごとの振動の大きさをそれぞれ求め、前記振動の周波数を回転周波数で除算して得られる無次元量を比周波数として、各回の測定で得られた前記振動の大きさを共通の前記比周波数ごとに算出し、前記比周波数ごとの振動の大きさについて係数が全て同符号の線型和を算出し、算出した値に基づいて回転同期成分強調波形を決定する回転同期成分強調波形算出ステップと、
前記回転同期成分強調波形を入力とし、内輪傷の有無、転動体傷の有無、外輪傷の有無の少なくともひとつを出力とする教師データを用いて学習した機械学習モデルを用い、前記回転同期成分強調波形を入力して前記転がり軸受の損傷の種類を判別する軸受損傷種類判別ステップと、を実行することを特徴とする。
請求項6に記載の発明は、請求項1乃至5の何れかの構成において、前記振動測定ステップでは、異なる複数の回転周波数で振動を測定することを特徴とする。
請求項7に記載の発明は、請求項1乃至6の何れかの構成において、前記回転同期成分強調波形算出ステップでは、各回の測定で得られた前記振動の大きさを共通の前記比周波数ごとに算出する際に、ある比周波数の振動の大きさとして、その比周波数の前後に所定の幅を持たせて当該幅内の振動の大きさの最大値を採用することを特徴とする。
In order to achieve the above object, the invention according to claim 1 is a method for diagnosing an abnormality of a rolling bearing that supports a rotating body.
A vibration measurement step for measuring the vibration of the rotating body a plurality of times, and
The magnitude of the vibration obtained in each measurement is obtained by frequency-analyzing the vibration to obtain the magnitude of the vibration for each frequency, and using the dimensionless quantity obtained by dividing the frequency of the vibration by the rotation frequency as the specific frequency. Rotation synchronization is calculated for each of the common specific frequencies, the linear sum of all the coefficients having the same sign is calculated for the magnitude of vibration for each specific frequency, and the rotation synchronization component emphasis waveform is determined based on the calculated value. Component emphasis waveform calculation step and
Based on the regularity of the vibration peak position of the rotation synchronization component emphasis waveform, the specific vibration source rotation frequency which is the ratio of the vibration source rotation frequency which is the frequency at which the direction in which the force is generated in the rolling bearing changes to the rotation frequency. When the value of the specific vibration source rotation frequency is 1, it is determined that the inner ring of the rolling bearing is damaged, and the value of the specific vibration source rotation frequency is the rotation frequency of the rolling element of the rolling bearing. If the value divided by the frequency is within the range that can be taken, it is judged that the rolling element is damaged, and if the value of the specific vibration source rotation frequency is 0, it is judged that the outer ring of the rolling bearing is damaged. It is characterized by executing a type determination step and.
According to the second aspect of the present invention, in the configuration of the first aspect, the estimation of the specific vibration source rotation frequency is the specific fundamental frequency corresponding to the damage of the rolling bearing and the specific vibration source in the rotation synchronization component emphasis waveform. The coincidence rate, which is the ratio at which the vibration peak exists at a specific specific frequency calculated assuming the rotation frequency, is calculated, and when the coincidence rate exceeds a predetermined threshold value, the assumed ratio basic It is characterized in that the regularity expressed by the specific vibration source rotation frequency assumed to be a frequency is determined to be in the rotation synchronization component emphasized waveform, and the assumed specific vibration source rotation frequency is adopted.
According to the third aspect of the present invention, in the configuration of the first aspect, the estimation of the specific vibration source rotation frequency uses the distance between two vibration peaks constituting the peak pair as the inter-peak distance in the rotation synchronization component emphasized waveform. When the average ratio of the specific frequencies of the two vibration peaks constituting each peak pair calculated for two sets of peak pairs having the same peak-to-peak distance is a natural number ratio, it is halved of the peak-to-peak distance. It is characterized in that 1 is set to the specific vibration source rotation frequency, or the specific vibration source rotation frequency is set to 0 when the ratio of the specific frequencies of two vibration peaks is a natural number ratio.
In the invention according to claim 4, in any of the configurations of claims 1 to 3, the bearing damage type determination step estimates a specific fundamental frequency and the specific vibration source rotation frequency corresponding to the damage of the rolling bearing. It is characterized in that the degree of the determined damage is further diagnosed by using the pair of the specific fundamental frequency and the specific vibration source rotation frequency for which the type of damage of the rolling bearing is determined. And.
In order to achieve the above object, the invention according to claim 5 is a method for diagnosing an abnormality of a rolling bearing that supports a rotating body.
A vibration measurement step for measuring the vibration of the rotating body a plurality of times, and
The magnitude of the vibration obtained in each measurement is obtained by frequency-analyzing the vibration to obtain the magnitude of the vibration for each frequency, and using the dimensionless quantity obtained by dividing the frequency of the vibration by the rotation frequency as the specific frequency. Rotation synchronization is calculated for each of the common specific frequencies, the linear sum of all the coefficients having the same sign is calculated for the magnitude of vibration for each specific frequency, and the rotation synchronization component emphasis waveform is determined based on the calculated value. Component emphasis waveform calculation step and
Using a machine learning model learned using teacher data that inputs the rotation-synchronized component emphasis waveform and outputs at least one of the presence / absence of inner ring damage, the presence / absence of rolling body damage, and the presence / absence of outer ring damage, the rotation-synchronization component emphasis It is characterized in that the bearing damage type determination step of inputting a waveform and determining the damage type of the rolling bearing is executed.
The invention according to claim 6 is characterized in that, in any of the configurations of claims 1 to 5, the vibration measurement step measures vibration at a plurality of different rotation frequencies.
According to the invention of claim 7, in any of the configurations of claims 1 to 6, in the rotation synchronization component emphasis waveform calculation step, the magnitude of the vibration obtained in each measurement is set for each of the common specific frequencies. When calculating the above, it is characterized in that a predetermined width is provided before and after the specific frequency as the magnitude of vibration of a certain specific frequency, and the maximum value of the magnitude of vibration within the width is adopted.

上記目的を達成するために、請求項8に記載の発明は、回転体を支持する転がり軸受の異常を診断する装置であって、
前記回転体の振動を複数回測定する振動測定手段と、
前記振動を周波数分析して周波数ごとの振動の大きさをそれぞれ求め、前記振動の周波数を回転周波数で除算して得られる無次元量を比周波数として、各回の測定で得られた前記振動の大きさを共通の前記比周波数ごとに算出し、前記比周波数ごとの振動の大きさについて係数が全て同符号の線型和を算出し、算出した値に基づいて回転同期成分強調波形を決定する回転同期成分強調波形算出手段と、
前記回転同期成分強調波形の振動ピーク位置の規則性に基づいて、前記転がり軸受において力の発生する方向が変化する周波数である振動源回転周波数と前記回転周波数との比である比振動源回転周波数を推定し、前記比振動源回転周波数の値が1の場合には、前記転がり軸受の内輪損傷と判断し、前記比振動源回転周波数の値が前記転がり軸受の転動体の公転周波数を前記回転周波数で除算した値のとり得る範囲である場合には、前記転動体の損傷と判断し、前記比振動源回転周波数の値が0の場合には、前記転がり軸受の外輪損傷と判断する軸受損傷種類判別手段と、を備えることを特徴とする。
請求項9に記載の発明は、請求項8の構成において、前記軸受損傷種類判別手段は、前記比振動源回転周波数の推定に当たり、前記回転同期成分強調波形において、前記転がり軸受の損傷に対応する比基本周波数と前記比振動源回転周波数とを仮定して算出される特定の比周波数に振動ピークが存在する割合である一致率を算出し、前記一致率が所定のしきい値を超過する場合には、仮定した前記比基本周波数と仮定した前記比振動源回転周波数とにより表現される規則性が前記回転同期成分強調波形にあると判断し、仮定した前記比振動源回転周波数を採用することを特徴とする。
請求項10に記載の発明は、請求項8の構成において、前記軸受損傷種類判別手段は、前記比振動源回転周波数の推定に当たり、前記回転同期成分強調波形において、ピーク対を構成する2つの振動ピークの間隔をピーク間距離とし、前記ピーク間距離が等しい2組のピーク対について算出したそれぞれのピーク対を構成する2つの前記振動ピークの前記比周波数の平均の比が自然数比の場合に、前記ピーク間距離の2分の1を前記比振動源回転周波数とする、又はある2つの振動ピークの前記比周波数の比が自然数比である場合に、前記比振動源回転周波数を0とすることを特徴とする。
請求項11に記載の発明は、請求項8乃至10の何れかの構成において、前記軸受損傷種類判別手段は、前記転がり軸受の損傷に対応する比基本周波数と前記比振動源回転周波数とを推定すると共に、前記転がり軸受の損傷の種類が判別された前記比基本周波数と前記比振動源回転周波数との組を用いて、判別された損傷の程度をさらに診断することを特徴とする。
請求項12に記載の発明は、請求項8乃至11の何れかの構成において、前記軸受損傷種類判別手段は、前記転がり軸受の損傷に対応する比基本周波数と、前記比振動源回転周波数と、前記比基本周波数と前記比振動源回転周波数とから算出される特定の比周波数とを用いて前記転がり軸受の損傷の種類ごとの前記振動ピーク位置の存在を確認するものであり、前記比基本周波数と前記比振動源回転周波数とを合わせて表示する表示手段を備えることを特徴とする。
請求項13に記載の発明は、請求項8乃至11の何れかの構成において、前記軸受損傷種類判別手段は、前記転がり軸受の損傷に対応する比基本周波数と、前記比振動源回転周波数と、前記比基本周波数と前記比振動源回転周波数とから算出される特定の比周波数とを用いて前記転がり軸受の損傷の種類ごとの前記振動ピーク位置の存在を確認するものであり、前記特定の比周波数の位置と前記回転同期成分強調波形とを合わせて表示する表示手段を備えることを特徴とする。
上記目的を達成するために、請求項14に記載の発明は、転がり軸受の異常診断プログラムであって、
所定の回転周波数で測定された回転体の振動が前記回転周波数と共に入力されたコンピュータに、請求項1乃至7の何れかに記載の転がり軸受の異常診断方法における回転同期成分強調波形算出ステップと軸受損傷種類判別ステップとを実行させることを特徴とする。
In order to achieve the above object, the invention according to claim 8 is an apparatus for diagnosing an abnormality of a rolling bearing that supports a rotating body.
A vibration measuring means for measuring the vibration of the rotating body a plurality of times,
The magnitude of the vibration obtained in each measurement is obtained by frequency-analyzing the vibration to obtain the magnitude of the vibration for each frequency, and using the dimensionless quantity obtained by dividing the frequency of the vibration by the rotation frequency as the specific frequency. Rotation synchronization is calculated for each of the common specific frequencies, the linear sum of all the coefficients having the same sign is calculated for the magnitude of vibration for each specific frequency, and the rotation synchronization component emphasis waveform is determined based on the calculated value. Component emphasis waveform calculation means and
Based on the regularity of the vibration peak position of the rotation synchronization component emphasis waveform, the specific vibration source rotation frequency which is the ratio of the vibration source rotation frequency which is the frequency at which the direction in which the force is generated in the rolling bearing changes to the rotation frequency. When the value of the specific vibration source rotation frequency is 1, it is determined that the inner ring of the rolling bearing is damaged, and the value of the specific vibration source rotation frequency is the rotation frequency of the rolling element of the rolling bearing. If the value divided by the frequency is within the range that can be taken, it is judged that the rolling element is damaged, and if the value of the specific vibration source rotation frequency is 0, it is judged that the outer ring of the rolling bearing is damaged. It is characterized by comprising a type discrimination means.
According to the invention of claim 9, in the configuration of claim 8, the bearing damage type determining means corresponds to the damage of the rolling bearing in the rotation synchronization component emphasis waveform in estimating the specific vibration source rotation frequency. When the coincidence rate, which is the ratio of the vibration peaks existing at a specific specific frequency calculated by assuming the specific fundamental frequency and the specific vibration source rotation frequency, is calculated, and the coincidence rate exceeds a predetermined threshold value. It is determined that the regularity expressed by the assumed specific fundamental frequency and the assumed specific vibration source rotation frequency lies in the rotation synchronization component emphasized waveform, and the assumed specific vibration source rotation frequency is adopted. It is characterized by.
According to the invention of claim 10, in the configuration of claim 8, the bearing damage type determining means estimates the specific vibration source rotation frequency, and two vibrations forming a peak pair in the rotation synchronization component emphasized waveform. When the interval between peaks is defined as the inter-peak distance, and the average ratio of the specific frequencies of the two vibration peaks constituting each of the two peak pairs having the same inter-peak distance is the natural number ratio. Half of the distance between the peaks is set as the specific vibration source rotation frequency, or when the ratio of the specific frequencies of two vibration peaks is a natural number ratio, the specific vibration source rotation frequency is set to 0. It is characterized by.
According to the invention of claim 11, in any of the configurations of claims 8 to 10, the bearing damage type determining means estimates a specific fundamental frequency and the specific vibration source rotation frequency corresponding to the damage of the rolling bearing. At the same time, the degree of the determined damage is further diagnosed by using the set of the specific fundamental frequency and the specific vibration source rotation frequency for which the type of damage of the rolling bearing is determined.
According to the invention of claim 12, in any of the configurations of claims 8 to 11, the bearing damage type determining means has a specific fundamental frequency corresponding to the damage of the rolling bearing, the specific vibration source rotation frequency, and the like. The existence of the vibration peak position for each type of damage of the rolling bearing is confirmed by using the specific specific frequency calculated from the specific fundamental frequency and the specific vibration source rotation frequency, and the specific fundamental frequency is confirmed. It is characterized by providing a display means for displaying the specific vibration source rotation frequency together with the above.
The invention according to claim 13 has the configuration according to any one of claims 8 to 11, wherein the bearing damage type determining means has a specific fundamental frequency corresponding to the damage of the rolling bearing, the specific vibration source rotation frequency, and the like. The existence of the vibration peak position for each type of damage of the rolling bearing is confirmed by using the specific specific frequency calculated from the specific fundamental frequency and the specific vibration source rotation frequency, and the specific ratio is confirmed. It is characterized by including a display means for displaying the position of the frequency and the rotation-synchronized component emphasized waveform together.
In order to achieve the above object, the invention according to claim 14 is an abnormality diagnosis program for rolling bearings.
The rotation-synchronized component-enhanced waveform calculation step and the bearing in the method for diagnosing an abnormality of a rolling bearing according to any one of claims 1 to 7 in which vibration of a rotating body measured at a predetermined rotation frequency is input together with the rotation frequency. It is characterized in that a damage type determination step is executed.

本発明によれば、測定した振動に対し、回転体の回転周波数に比例した周波数の振動成分を強調する回転同期成分強調波形を決定し、回転同期成分強調波形の振動ピーク位置の規則性に基づいて転がり軸受における損傷の種類を判別するので、軸受諸元を用いることなく振動ピークの規則性から軸受異常の種類を判断することが可能となる。
特に、請求項4及び11の発明によれば、上記効果に加えて、軸受諸元の値を必要とすることなく特徴周波数の値が算出できるため、従来特徴周波数の値を算出するために軸受諸元の値が必要であった診断手法が実行可能となる。よって、軸受異常の種類だけでなく異常の程度も精度よく求めることができる。
特に、請求項5の発明によれば、上記効果に加えて、軸受損傷種類判別ステップでは、回転同期成分強調波形を入力とし、内輪傷の有無、転動体傷の有無、外輪傷の有無の少なくともひとつを出力とする教師データを用いて学習した機械学習モデルを用い、回転同期成分強調波形を入力して転がり軸受の損傷の種類を判別するので、人間が考える処理よりも汎用的に振動ピークの規則性から軸受異常の種類を判断することが可能となる。
特に、請求項6の発明によれば、上記効果に加えて、回転同期成分強調波形算出ステップでは、異なる複数の回転周波数で測定された振動から回転同期成分強調波形を算出するので、さらに外乱の影響が低減されて、振動ピークの規則性を捉え易くなるため、軸受異常の推定精度が向上する。
特に、請求項7の発明によれば、上記効果に加えて、振動ピークの比周波数がばらついても確実に軸受損傷に起因する振動ピークを強調した回転同期成分強調波形を算出することが可能となる。
特に、請求項12の発明によれば、上記効果に加えて、比基本周波数と比振動源回転周波数とを合わせて表示する表示手段を備えることで、数字として表示される周波数の値を用いて他の軸受診断処理を行うことが可能となる。
特に、請求項13の発明によれば、上記効果に加えて、特定の比周波数の位置と回転同期成分強調波形とを合わせて表示する表示手段を備えることで、振動ピークの規則性を正しく抽出できているか否かを視覚的に把握できるため、診断結果の妥当性を容易に検証できる。また、各損傷の種類において着目すべき比周波数がわかりやすくなる。
According to the present invention, a rotation-synchronized component-emphasized waveform that emphasizes a vibration component having a frequency proportional to the rotation frequency of the rotating body is determined with respect to the measured vibration, and is based on the regularity of the vibration peak position of the rotation-synchronized component-enhanced waveform. Since the type of damage in the rolling bearing is determined, it is possible to determine the type of bearing abnormality from the regularity of the vibration peak without using the bearing specifications.
In particular, according to the inventions of claims 4 and 11, in addition to the above effects, the value of the characteristic frequency can be calculated without requiring the value of the bearing specifications. Therefore, the bearing is used to calculate the value of the conventional characteristic frequency. Diagnostic methods that required specification values can now be implemented. Therefore, not only the type of bearing abnormality but also the degree of abnormality can be accurately determined.
In particular, according to the invention of claim 5, in addition to the above effects, in the bearing damage type determination step, the rotation-synchronized component emphasized waveform is input, and at least the presence or absence of inner ring scratches, the presence or absence of rolling body scratches, and the presence or absence of outer ring scratches. Using a machine learning model learned using teacher data that outputs one, the type of damage to the rolling bearing is determined by inputting the rotation synchronization component emphasis waveform, so the vibration peak is more versatile than the processing that humans think. It is possible to determine the type of bearing abnormality from the regularity.
In particular, according to the invention of claim 6, in addition to the above effect, in the rotation synchronization component emphasis waveform calculation step, the rotation synchronization component emphasis waveform is calculated from the vibrations measured at a plurality of different rotation frequencies, so that the disturbance is further disturbed. Since the influence is reduced and the regularity of the vibration peak is easily grasped, the estimation accuracy of the bearing abnormality is improved.
In particular, according to the invention of claim 7, in addition to the above effect, it is possible to reliably calculate a rotation-synchronized component-enhanced waveform that emphasizes the vibration peak caused by bearing damage even if the specific frequency of the vibration peak varies. Become.
In particular, according to the invention of claim 12, in addition to the above effect, by providing a display means for displaying the specific fundamental frequency and the specific vibration source rotation frequency together, the value of the frequency displayed as a numerical value can be used. It is possible to perform other bearing diagnostic processing.
In particular, according to the invention of claim 13, in addition to the above effect, the regularity of the vibration peak is correctly extracted by providing the display means for displaying the position of the specific specific frequency and the rotation synchronization component emphasized waveform together. Since it is possible to visually grasp whether or not it is completed, the validity of the diagnosis result can be easily verified. In addition, the specific frequency to be noted for each type of damage becomes easy to understand.

転がり軸受の異常診断装置の機能ブロック図である。It is a functional block diagram of the abnormality diagnosis device of a rolling bearing. 外輪および内輪が損傷している場合の回転同期成分強調波形である。This is a rotation-synchronized component-enhanced waveform when the outer ring and the inner ring are damaged. 転動体が損傷している場合の回転同期成分強調波形である。It is a rotation-synchronized component-enhanced waveform when the rolling element is damaged. 異常診断方法のフローチャートである。It is a flowchart of an abnormality diagnosis method. 異常診断方法の他の例のフローチャートである。It is a flowchart of another example of an abnormality diagnosis method. 異常診断方法の他の例のフローチャートである。It is a flowchart of another example of an abnormality diagnosis method. 診断回転周波数の表示と測定開始の指令を行う画面である。This is a screen for displaying the diagnostic rotation frequency and instructing the start of measurement. 図4の方法による異常診断結果を表示する画面である。It is a screen which displays the abnormality diagnosis result by the method of FIG. 図5の方法による異常診断結果を表示する画面である。It is a screen which displays the abnormality diagnosis result by the method of FIG. 図6の方法による異常診断結果を表示する画面である。It is a screen which displays the abnormality diagnosis result by the method of FIG.

以下、本発明の実施の形態を図面に基づいて説明する。
図1は転がり軸受の異常診断装置を工作機械の主軸に対して取り付けて使用する場合の構成を示した機能ブロック図で、この図に基づいて具体的に説明する。
工作機械100において、主軸1が転がり軸受である軸受7を介して主軸ハウジング2に対して回転可能に取り付けられており、主軸1には加工を行うための工具3が固定されている。モータ4は主軸1を駆動する。モータ4には速度検出器5が設けられて、測定されたモータ4の回転周波数が制御装置6に入力されるようになっている。制御装置6は、速度検出器5で測定されたモータ4の回転周波数を測定者によって制御装置6に入力された指令回転周波数に保つようにモータ4へ供給する電流の制御を行っている。表示部8には、速度検出器5で測定されたモータ4の現在の回転周波数が表示される。
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a functional block diagram showing a configuration in which an abnormality diagnosis device for rolling bearings is attached to a spindle of a machine tool and used, and will be specifically described with reference to this diagram.
In the machine tool 100, the spindle 1 is rotatably attached to the spindle housing 2 via a bearing 7 which is a rolling bearing, and a tool 3 for machining is fixed to the spindle 1. The motor 4 drives the spindle 1. A speed detector 5 is provided in the motor 4, so that the measured rotation frequency of the motor 4 is input to the control device 6. The control device 6 controls the current supplied to the motor 4 so as to keep the rotation frequency of the motor 4 measured by the speed detector 5 at the command rotation frequency input to the control device 6 by the measurer. The display unit 8 displays the current rotation frequency of the motor 4 measured by the speed detector 5.

コンピュータを含む異常診断装置200において、振動測定手段としての振動センサ9は、工作機械100の軸受7の損傷によって生じる振動を測定可能な位置に磁力により取り付けられている。
表示手段としての表示・操作部13は、測定者が制御装置6へ入力すべき指令回転周波数や診断結果を表示するとともに、表示部8に表示されたモータ4の現在の回転周波数が表示・操作部13に表示された値と一致したか否かを測定者が判断した結果を入力する操作が可能となっている。振動センサ9で測定される振動加速度は、A/D変換部10でデジタル値に変換される。記憶部11は、表示部8に表示されたモータ4の現在の回転周波数が表示・操作部13に表示された値と一致したと測定者が判断したときから既定のデータ長となるときまでの振動加速度のデジタル値を振動測定時の回転周波数とともに記憶する。
In the abnormality diagnosis device 200 including a computer, the vibration sensor 9 as a vibration measuring means is magnetically attached to a position where vibration generated by damage to the bearing 7 of the machine tool 100 can be measured.
The display / operation unit 13 as a display means displays the command rotation frequency and the diagnosis result to be input to the control device 6 by the measurer, and displays / operates the current rotation frequency of the motor 4 displayed on the display unit 8. It is possible to input the result of the measurer's determination as to whether or not the value matches the value displayed in the unit 13. The vibration acceleration measured by the vibration sensor 9 is converted into a digital value by the A / D converter 10. From the time when the measurer determines that the current rotation frequency of the motor 4 displayed on the display unit 8 matches the value displayed on the display / operation unit 13, the storage unit 11 reaches the default data length. The digital value of vibration acceleration is stored together with the rotation frequency at the time of vibration measurement.

演算部12は、記憶部11に記憶された振動測定時の回転周波数と振動加速度より、予め記憶された異常診断プログラムに基づき、乗算、フーリエ変換、絶対値の算出、内挿処理を行って、比周波数(周波数を回転周波数で割って算出される無次元量)ごとの振動の大きさを算出し、複数の振動測定時の回転周波数の振動の大きさについて係数が全て正の線型和を算出することで回転同期成分強調波形を算出する。さらに、回転同期成分強調波形のピーク位置の規則性より、後述する比基本周波数と比振動源回転周波数とを推定し、比振動源回転周波数の値に応じて、損傷が内輪・転動体・外輪の何れに生じているかを判断する。すなわち、演算部12は、回転同期成分強調波形算出手段と軸受損傷種類判別手段として機能する。 The calculation unit 12 performs multiplication, Fourier conversion, absolute value calculation, and insertion processing based on the abnormality diagnosis program stored in advance from the rotation frequency and vibration acceleration stored in the storage unit 11 at the time of vibration measurement. Calculate the magnitude of vibration for each specific frequency (a dimensionless quantity calculated by dividing the frequency by the rotation frequency), and calculate the linear sum with all positive coefficients for the magnitude of vibration of the rotation frequency when measuring multiple vibrations. By doing so, the rotation-synchronized component emphasis waveform is calculated. Furthermore, the specific fundamental frequency and the specific vibration source rotation frequency, which will be described later, are estimated from the regularity of the peak position of the rotation synchronization component emphasis waveform, and the damage is caused to the inner ring, rolling element, and outer ring according to the value of the specific vibration source rotation frequency. Determine which of the above is occurring. That is, the calculation unit 12 functions as a rotation synchronization component emphasis waveform calculation means and a bearing damage type determination means.

軸受7において、内輪傷が存在する場合、転動体が内輪傷を通過する周波数(基本周波数)とその自然数倍の周波数の力が発生する。内輪傷の方向は主軸1の回転とともに回転するため、力の発生する方向は回転周波数で変化する。
軸受7において、転動体傷が存在する場合、転動体傷が内輪または外輪に接する周波数(基本周波数)とその自然数倍の周波数の力が発生する。傷のある転動体の位置は転動体の保持器が自転する(転動体が公転する)周波数で変化するため、力の発生する方向は転動体の保持器が自転する(転動体が公転する)周波数で変化する。保持器の自転(転動体の公転)は、主軸の回転よりも必ず遅くなるので、力の発生する方向が変化する周波数は0Hzより大きく回転周波数より小さい値となる。
軸受7において、外輪傷が存在する場合、転動体が外輪傷を通過する周波数(基本周波数)とその自然数倍の周波数の力が発生する。外輪傷の方向は主軸1の回転に依らず一定であるため、力の発生する方向は変化しない。
ここでは力の発生する方向が変化する周波数を振動源回転周波数と呼称する。力の発生する方向の変化しない場合は、振動源回転周波数が便宜上0Hzであるとみなしてよい。
In the bearing 7, when the inner ring scratch is present, a force is generated at a frequency (fundamental frequency) at which the rolling element passes through the inner ring scratch and a frequency several times that of the natural number. Since the direction of the inner ring scratches rotates with the rotation of the spindle 1, the direction in which the force is generated changes with the rotation frequency.
In the bearing 7, when a rolling element scratch is present, a force is generated at a frequency (fundamental frequency) at which the rolling body scratch is in contact with the inner ring or the outer ring and a frequency several times that of the natural number. Since the position of the damaged rolling element changes at the frequency at which the rolling element cage rotates (the rolling element revolves), the direction in which the force is generated causes the rolling element retainer to rotate (the rolling element revolves). It changes with frequency. Since the rotation of the cage (revolution of the rolling element) is always slower than the rotation of the spindle, the frequency at which the direction in which the force is generated changes is larger than 0 Hz and smaller than the rotation frequency.
In the bearing 7, when the outer ring scratch is present, a force is generated at a frequency (fundamental frequency) at which the rolling element passes through the outer ring scratch and a frequency several times that of the natural number. Since the direction of the outer ring scratch is constant regardless of the rotation of the spindle 1, the direction in which the force is generated does not change.
Here, the frequency at which the direction in which the force is generated changes is referred to as the vibration source rotation frequency. If the direction in which the force is generated does not change, the vibration source rotation frequency may be regarded as 0 Hz for convenience.

ところで、力の発生する方向が振動源回転周波数で変化する場合、固定された座標系でその力を見ると、力の周波数(基本周波数の自然数倍)に対して、振動源回転周波数だけ前後した2つの周波数の力が生じていることになる。線形とみなせるシステムでは、力と同じ周波数の振動加速度が生じるため、固定された座標系で振動加速度を測定し、周波数分析すると、軸受が損傷している場合には、理想的には振動源回転周波数の2倍だけ離れた周波数に振動ピークの対が観測され、その2つの振動ピークの平均の周波数が、基本周波数の自然数倍となるような位置に複数観測される。なお、外輪傷の場合は振動源回転周波数を0Hzとみなすことで、同一の周波数の振動ピークを2回カウントして2つの振動ピークとみなすこととする。この規則性を捉え、振動源回転周波数を推定することで異常の種類を判断することが可能となる。ただし、現実的には、振動を測定し周波数分析するだけでは、振動しにくい周波数の振動ピークはノイズに埋もれてしまい、この規則性を確実に捉えることは難しい。 By the way, when the direction in which the force is generated changes with the vibration source rotation frequency, when the force is viewed in a fixed coordinate system, the vibration source rotation frequency is before and after the force frequency (natural multiple of the fundamental frequency). This means that the forces of the two frequencies are generated. In a system that can be regarded as linear, vibration acceleration occurs at the same frequency as the force, so if the vibration acceleration is measured in a fixed coordinate system and frequency analysis is performed, if the bearing is damaged, the vibration source rotation is ideal. A pair of vibration peaks is observed at a frequency that is twice the frequency apart, and a plurality of vibration peak pairs are observed at positions where the average frequency of the two vibration peaks is a natural number of times the fundamental frequency. In the case of an outer ring scratch, the vibration source rotation frequency is regarded as 0 Hz, and the vibration peaks having the same frequency are counted twice and regarded as two vibration peaks. By grasping this regularity and estimating the vibration source rotation frequency, it is possible to determine the type of abnormality. However, in reality, it is difficult to reliably grasp this regularity because the vibration peaks at frequencies that are difficult to vibrate are buried in noise simply by measuring the vibration and analyzing the frequency.

回転に同期する軸受損傷に起因する振動は、繰り返し測定したり異なる回転周波数で測定したりしたとしても、振動の生じる周波数は常に回転周波数の定数倍である。振動の周波数の回転周波数に対する比の無次元量を比周波数と呼称することにすると、回転に同期する軸受損傷に起因する振動は、回転周波数に依らず同じ比周波数に振動ピークが観測されることになる。このため、複数回測定して比周波数ごとの振動の大きさを算出し、複数回の測定で得られた振動の大きさについて係数が全て正の線型和を算出してホワイトノイズを低減したり、複数の回転周波数で測定して比周波数ごとの振動の大きさを算出し、複数回の測定で得られた振動の大きさについて係数が全て正の線型和を算出してホワイトノイズや回転に同期しない外乱成分を低減したりすることで、回転に同期した振動を強調した波形(回転同期成分強調波形)を算出することができ、軸受損傷に起因する振動ピークの規則性を捉えることが容易となる。 The vibration caused by the bearing damage synchronized with the rotation is always a constant multiple of the rotation frequency, even if it is repeatedly measured or measured at a different rotation frequency. If the non-dimensional amount of the ratio of the vibration frequency to the rotation frequency is called the specific frequency, the vibration peak caused by the bearing damage synchronized with the rotation is observed at the same specific frequency regardless of the rotation frequency. become. For this reason, white noise can be reduced by measuring multiple times to calculate the magnitude of vibration for each specific frequency, and calculating the sum of the waveforms with positive coefficients for the magnitude of vibration obtained from multiple measurements. , Calculate the magnitude of vibration for each specific frequency by measuring at multiple rotation frequencies, and calculate the linear sum with all positive coefficients for the magnitude of vibration obtained in multiple measurements for white noise and rotation. By reducing unsynchronized disturbance components, it is possible to calculate a waveform that emphasizes vibration synchronized with rotation (rotation-synchronized component-emphasized waveform), and it is easy to capture the regularity of vibration peaks caused by bearing damage. It becomes.

なお、係数に異符号の値がある場合には、軸受損傷に起因する振動ピークの振動の大きさを打ち消しあうことになるため、回転に同期した軸受損傷に起因する振動を強調する効果が得られない。全ての係数が1の線型和をとった場合、回転同期成分強調波形は比周波数毎の振動の大きさの合計値となる。全ての係数を測定回数の逆数として線型和をとった場合、回転同期成分強調波形は比周波数毎の振動の大きさの平均値となる。合計値であっても平均値であっても回転同期成分強調波形のピーク位置は同一となるためピーク位置の規則性を議論する上ではいずれであってもよい。また、係数を同一に限らなくとも概ねピーク位置の規則性について同様の議論が成立する。例えば、回転周波数が大きいほど軸受損傷により生じる力の大きさが大きくなる傾向があるため、測定する回転周波数範囲が広い場合には低い回転周波数で測定したデータの影響度合いが小さくなってしまう。このため、回転周波数のべき乗の逆数を係数として線型和をとることで、異なる回転周波数で得られた振動の大きさを平等に扱うことも可能である。 If the coefficients have different sign values, the magnitudes of vibrations of the vibration peaks caused by bearing damage cancel each other out, so that the effect of emphasizing vibrations caused by bearing damage synchronized with rotation can be obtained. I can't. When all the coefficients take a linear sum of 1, the rotation-synchronized component-emphasized waveform becomes the total value of the magnitudes of vibrations for each specific frequency. When the linear sum is taken with all the coefficients as the reciprocals of the number of measurements, the rotation-synchronized component-emphasized waveform becomes the average value of the magnitude of vibration for each specific frequency. Since the peak positions of the rotation-synchronized component-enhanced waveforms are the same regardless of whether they are the total value or the average value, any of them may be used in discussing the regularity of the peak positions. Moreover, even if the coefficients are not limited to the same, the same argument holds about the regularity of the peak position. For example, the larger the rotation frequency, the larger the magnitude of the force generated by the bearing damage tends to be. Therefore, when the measurement rotation frequency range is wide, the influence of the data measured at the low rotation frequency becomes small. Therefore, it is possible to treat the magnitudes of vibrations obtained at different rotation frequencies equally by taking the linear sum with the reciprocal of the power of the rotation frequency as a coefficient.

複数の回転周波数で測定する場合、振動の基本周波数と振動源回転周波数とをそれぞれ回転周波数で除算して得られる無次元量の比基本周波数と比振動源回転周波数とを用いた方が議論しやすい。但し、比基本周波数は、軸受の構造によって様々な値をとるため、例えばある軸受の内輪傷の比基本周波数が別の軸受の外輪傷の比基本周波数と同じ大きさとなることもある。このため、比基本周波数の値に着目しても軸受の異常の種類を判別することは不可能である。しかし、前述したように、振動源回転周波数は、軸受損傷の種類によって回転周波数の何倍となるかが異なり、その大小関係に逆転が起こることはない。よって、比振動源回転周波数に換算すると、内輪損傷の場合は1、転動体損傷の場合は、後述する回転周波数に対する転動体公転の比の取りうる範囲の値(工作機械主軸に用いられる軸受の場合には0.45前後の値が多い)、外輪損傷の場合は0となる。 When measuring at multiple rotation frequencies, it is better to use the non-dimensional amount of the fundamental frequency and the specific vibration source rotation frequency obtained by dividing the fundamental frequency of vibration and the vibration source rotation frequency by the rotation frequency. Cheap. However, since the specific fundamental frequency takes various values depending on the structure of the bearing, for example, the specific fundamental frequency of the inner ring scratch of one bearing may be the same as the specific fundamental frequency of the outer ring scratch of another bearing. Therefore, it is impossible to determine the type of bearing abnormality by focusing on the value of the specific fundamental frequency. However, as described above, the rotation frequency of the vibration source differs depending on the type of bearing damage, and the magnitude relationship does not reverse. Therefore, when converted to the specific vibration source rotation frequency, 1 in the case of inner ring damage, and in the case of rolling element damage, a value within the range in which the ratio of rolling element revolution to the rotational frequency, which will be described later, can be taken (for bearings used for machine tool spindles). In the case, the value is often around 0.45), and in the case of outer ring damage, it becomes 0.

図2に外輪および内輪が損傷している場合の回転同期成分強調波形、図3に転動体が損傷している場合の回転同期成分強調波形を示す。これらの図に基づいて、軸受損傷の種類毎の振動ピーク位置の規則性を説明する。なお、図2において、内輪損傷時の振動ピークに関連する箇所は破線で、外輪損傷時の振動ピークに関連する箇所は実線で、注釈を記した。図3において、転動体損傷時の振動ピークに関連する箇所は実線で注釈を記した。 FIG. 2 shows a rotation-synchronized component-enhanced waveform when the outer ring and the inner ring are damaged, and FIG. 3 shows a rotation-synchronized component-enhanced waveform when the rolling element is damaged. Based on these figures, the regularity of the vibration peak position for each type of bearing damage will be described. In FIG. 2, the part related to the vibration peak when the inner ring is damaged is marked with a broken line, and the part related to the vibration peak when the outer ring is damaged is marked with a solid line. In FIG. 3, the parts related to the vibration peak at the time of rolling element damage are marked with solid lines.

内輪損傷があると、図2に示すように、同一の比振動源回転周波数(=1)の2倍だけ離れた2本の振動ピーク(ピーク対)が複数箇所に存在する。さらに、それぞれのピーク対の平均の比周波数は必ず比基本周波数の自然数倍となるため、比基本周波数が未知であっても、ピーク対の平均の比周波数が自然数比となっていることを確認することで内輪損傷による振動ピークであると確信することができる。
外輪損傷があると、図2に示すように、比周波数の比が自然数比となるような振動ピークが複数箇所に存在する。軸受の比基本周波数は自然数とならないように設計されることが一般的であるため、さらにそれらの振動ピークの比周波数のいくつかが自然数でなければ、比基本周波数が自然数でないと推測できるため、外輪損傷による振動ピークであると確信することができる。なお、外輪損傷の場合、対でない1本の振動ピークが等間隔に存在するだけであるが、便宜上、比振動源回転周波数0だけ離れた2本の振動ピーク(ピーク対)が存在するとみなすことで他の種類の異常と同じ計算処理を用いることができる。
転動体損傷があると、図3に示すように、同一の比振動源回転周波数の2倍だけ離れた2本の振動ピーク(ピーク対)が複数箇所に存在する。さらに、それぞれのピーク対の平均の比周波数は必ず比基本周波数の自然数倍となるため、比基本周波数が未知であっても、ピーク対の平均の比周波数が自然数比となっていることを確認することで転動体損傷による振動ピークであると確信することができる。なお、比振動源回転周波数の値は必ず1より小さい値となるため、内輪損傷との識別が可能である。
また、回転同期成分強調波形の振動ピーク位置の規則性(回転同期成分強調波形の比基本周波数×自然数±比振動源回転周波数の位置に振動ピークが存在する)を捉える方法として、ここではピーク対の平均(外輪の場合はピーク)の比周波数が自然数比となっていることを確認する手法を例に挙げたが他の手法であっても構わない。
When the inner ring is damaged, as shown in FIG. 2, two vibration peaks (peak pairs) separated by twice the same specific vibration source rotation frequency (= 1) are present at a plurality of locations. Furthermore, since the average specific frequency of each peak pair is always a natural number multiple of the specific fundamental frequency, even if the specific fundamental frequency is unknown, the average specific frequency of the peak pairs is the natural number ratio. By checking, it can be confirmed that the vibration peak is due to damage to the inner ring.
When the outer ring is damaged, as shown in FIG. 2, there are vibration peaks at a plurality of locations such that the ratio of the specific frequencies becomes a natural number ratio. Since the ratio fundamental frequency of the bearing is generally designed not to be a natural number, and if some of the specific frequencies of those vibration peaks are not natural numbers, it can be inferred that the specific fundamental frequencies are not natural numbers. It can be convinced that it is a vibration peak due to damage to the outer ring. In the case of outer ring damage, only one unpaired vibration peak exists at equal intervals, but for convenience, it should be considered that there are two vibration peaks (peak pairs) separated by the specific vibration source rotation frequency 0. You can use the same calculation process as other types of anomalies.
When there is rolling element damage, as shown in FIG. 3, two vibration peaks (peak pairs) separated by twice the rotation frequency of the same specific vibration source are present at a plurality of locations. Furthermore, since the average specific frequency of each peak pair is always a natural number multiple of the specific fundamental frequency, even if the specific fundamental frequency is unknown, the average specific frequency of the peak pairs is the natural number ratio. By checking, it can be confirmed that the vibration peak is due to rolling element damage. Since the value of the specific vibration source rotation frequency is always smaller than 1, it can be distinguished from the inner ring damage.
In addition, as a method of capturing the regularity of the vibration peak position of the rotation synchronization component emphasis waveform (the vibration peak exists at the position of the ratio fundamental frequency × natural number ± specific vibration source rotation frequency of the rotation synchronization component emphasis waveform), here, the peak pair The method of confirming that the specific frequency of the average (peak in the case of the outer ring) is the natural number ratio is given as an example, but other methods may be used.

図4は、転がり軸受の異常診断を行う方法のフローチャートを示したものであり、このフローチャートに基づいて具体的に説明する。
まず、予め登録されている診断回転周波数の全条件のうちの1つが表示・操作部13に図7のように表示される(S1)。
次に、診断装置の使用者は、表示された診断回転周波数を制御装置6へ指令する(S2)。
次に、診断装置の使用者は、表示部8を確認して、診断対象の軸受に支持された回転体が表示・操作部13に表示された回転周波数で回転しているかを判断し、一致している場合には表示・操作部13より測定開始を指令する(S3)。
すると、異常診断装置200では、振動加速度を測定して記録し(S4)、必要なデータ長となったら測定完了と判断してS6へ移行する(S5)。S6では、振動測定時の回転周波数と振動加速度とを対応付けて記録する。
S7で、予め登録されている診断回転周波数の全条件の測定が完了しているかを判断し、全条件の測定が完了していればS8へ移行する。完了してない場合にはS1に戻り、測定が完了していない別の診断回転周波数を表示する。ここでのS4〜S7までが振動測定ステップとなる。
FIG. 4 shows a flowchart of a method for diagnosing an abnormality of a rolling bearing, and a specific description will be given based on this flowchart.
First, one of all the conditions of the diagnostic rotation frequency registered in advance is displayed on the display / operation unit 13 as shown in FIG. 7 (S1).
Next, the user of the diagnostic device commands the displayed diagnostic rotation frequency to the control device 6 (S2).
Next, the user of the diagnostic device confirms the display unit 8 and determines whether the rotating body supported by the bearing to be diagnosed is rotating at the rotation frequency displayed on the display / operation unit 13. If so, the display / operation unit 13 commands the start of measurement (S3).
Then, the abnormality diagnosis device 200 measures and records the vibration acceleration (S4), and when the required data length is reached, determines that the measurement is completed and shifts to S6 (S5). In S6, the rotation frequency at the time of vibration measurement and the vibration acceleration are recorded in association with each other.
In S7, it is determined whether the measurement of all the conditions of the diagnostic rotation frequency registered in advance is completed, and if the measurement of all the conditions is completed, the process proceeds to S8. If it is not completed, the process returns to S1 and another diagnostic rotation frequency for which the measurement is not completed is displayed. Here, S4 to S7 are vibration measurement steps.

S8では、まず振動加速度をフーリエ変換し振幅スペクトル密度を算出する。予め決めておいた全条件の測定に対して共通の比周波数分解能となるように振幅スペクトル密度の絶対値について補間処理を行い比周波数毎の振動の大きさを算出する。それぞれの診断回転周波数の振動の大きさについて係数が全て正の線型和を算出することで回転同期成分強調波形を算出する(回転同期成分強調波形算出ステップ)。全N条件のうちi番目の診断回転周波数をfROT(i)、線型和を取る際のi番目の条件の振動の大きさV(i)に対する係数をA(i)、回転同期成分強調波形をWとする。このとき回転同期成分強調波形は以下の数1で得られる。振動の大きさの合計を使う場合は、係数A(i)は以下の数2を用いればよい。振動の大きさの平均を使う場合は、係数A(i)は以下の数3を用いればよい。回転周波数により軸受損傷により生じる力がfROT(i)のM乗に比例すると近似できる場合に全ての回転周波数のデータを平等に扱うためには、係数A(i)を以下の数4とすればよい。回転周波数がfROTbestのとき最も測定精度が良いことがわかっている場合には、回転周波数がfROTbestに近いときの振動の大きさの影響度合いが大きくなるように以下の数5のような係数A(i)を用いてもよい。すなわち、数1〜5の係数A(i)は任意の一部又は全部を選択できる。ここで、振動の大きさV(i)と回転同期成分強調波形Wは、比周波数毎に値を持つベクトル、A(i)はスカラーである。 In S8, first, the vibration acceleration is Fourier transformed to calculate the amplitude spectral density. The magnitude of vibration for each specific frequency is calculated by performing interpolation processing on the absolute value of the amplitude spectral density so that the specific frequency resolution is common to the measurements under all predetermined conditions. The rotation-synchronized component-emphasized waveform is calculated by calculating the linear sum with all positive coefficients for the magnitude of vibration at each diagnostic rotation frequency (rotation-synchronized component-enhanced waveform calculation step). Of all N conditions, the i-th diagnostic rotation frequency is f ROT (i), the coefficient for the vibration magnitude V (i) of the i-th condition when taking the linear sum is A (i), and the rotation synchronization component emphasized waveform. Let be W. At this time, the rotation-synchronized component-enhanced waveform can be obtained by the following equation 1. When using the total magnitude of vibration, the coefficient A (i) may be the following equation 2. When using the average of the magnitudes of vibration, the coefficient A (i) may be the following equation 3. In order to treat all rotation frequency data equally when the force generated by bearing damage due to rotation frequency can be approximated to be proportional to the M power of f ROT (i), the coefficient A (i) should be set to the following equation 4. Just do it. Most if the measurement accuracy is found to be good, coefficients as the following equation 5 such that the magnitude of the influence level of vibration when the rotational frequency is close to f ROTbest increases when the rotational frequency f ROTbest A (i) may be used. That is, any part or all of the coefficients A (i) of Equations 1 to 5 can be selected. Here, the magnitude V (i) of vibration and the rotation-synchronized component emphasis waveform W are vectors having values for each specific frequency, and A (i) is a scalar.

Figure 2020159945
Figure 2020159945
Figure 2020159945
Figure 2020159945
Figure 2020159945
Figure 2020159945
Figure 2020159945
Figure 2020159945
Figure 2020159945
Figure 2020159945

次に、S9では、比基本周波数と比振動源回転周波数とを仮定し、回転同期成分強調波形の比基本周波数×自然数±比振動源回転周波数の位置に振動ピークが存在するかそれぞれ判断する。ある比周波数に振動ピークが存在するか否かの判断は、回転同期成分強調波形と回転同期成分強調波形の移動平均を比較し、回転同期成分強調波形の値の方が大きければ、その比周波数には振動ピークが存在すると判断すればよい。そして、比基本周波数×自然数±比振動源回転周波数の位置に振動ピークが存在した割合(一致率)を算出する。 Next, in S9, the specific fundamental frequency and the specific vibration source rotation frequency are assumed, and it is determined whether or not the vibration peak exists at the position of the specific fundamental frequency × the natural number ± the specific vibration source rotation frequency of the rotation synchronization component emphasized waveform. To determine whether or not there is a vibration peak at a certain specific frequency, compare the moving averages of the rotation-synchronized component-emphasized waveform and the rotation-synchronized component-emphasized waveform, and if the value of the rotation-synchronized component-enhanced waveform is larger, the specific frequency. It may be judged that the vibration peak exists in. Then, the ratio (match rate) at which the vibration peak exists at the position of the specific fundamental frequency × natural number ± specific vibration source rotation frequency is calculated.

次に、S10では、既定された探索範囲の比基本周波数と比振動源回転周波数との組み合わせについて全て探索完了しているか判断し、完了していればS11へ移行する。
S11では、一致率が既定のしきい値以上である比基本周波数と比振動源回転周波数について、比振動源回転周波数の値に応じて軸受異常の種類(内輪損傷、転動体損傷、外輪損傷)を判断する。
さらに、S12では、S11で軸受異常の種類が特定された比基本周波数と比振動源回転周波数の組と診断回転周波数とから、基本周波数と特徴周波数の値を算出し、損傷の程度(修理が必要なレベルか否か)を診断する。例えば先の特許文献2に開示されているように、算出した基本周波数及び特徴周波数を、各周波数ごとに予め設定したしきい値と比較すれば、損傷の程度を診断することができる。ここでのS9〜S12までが軸受損傷種類判別ステップとなる。
そして、S13では、図8に示すような診断結果を表示・操作部13に表示する。ここでは比振動源回転周波数が外輪に対応する0、内輪に対応する1の場合に一致率が高く算出され損傷していると判断されたが、S12の診断では異常(修理が必要)と判定するしきい値を超過しなかったため、損傷の程度では「正常範囲」と表示され、「外輪と内輪が損傷していますが、修理が必要な水準ではありません」との文字も表示される。
Next, in S10, it is determined whether or not all the combinations of the specific fundamental frequency and the specific vibration source rotation frequency in the predetermined search range have been searched, and if so, the process proceeds to S11.
In S11, with respect to the specific fundamental frequency and the specific vibration source rotation frequency whose matching rate is equal to or higher than the predetermined threshold value, the type of bearing abnormality (inner ring damage, rolling element damage, outer ring damage) according to the value of the specific vibration source rotation frequency. To judge.
Further, in S12, the values of the fundamental frequency and the characteristic frequency are calculated from the set of the specific fundamental frequency, the specific vibration source rotation frequency, and the diagnostic rotation frequency for which the type of bearing abnormality is specified in S11, and the degree of damage (repair is performed). Diagnose whether it is the required level). For example, as disclosed in Patent Document 2 above, the degree of damage can be diagnosed by comparing the calculated fundamental frequency and characteristic frequency with a preset threshold value for each frequency. The steps S9 to S12 here are the bearing damage type determination steps.
Then, in S13, the diagnosis result as shown in FIG. 8 is displayed on the display / operation unit 13. Here, when the specific vibration source rotation frequency is 0 corresponding to the outer ring and 1 corresponding to the inner ring, the matching rate is calculated to be high and it is judged to be damaged, but in the diagnosis of S12, it is judged to be abnormal (repair required). Since the threshold value was not exceeded, the degree of damage is displayed as "normal range", and the text "The outer ring and inner ring are damaged, but the level does not require repair" is also displayed.

図5は、転がり軸受の異常診断を行う他の方法のフローチャートを示したものであり、このフローチャートに基づいて具体的に説明する。
S1〜S8までは図4の方法と同じである。まず、予め登録されている診断回転周波数の全条件のうちの1つが表示・操作部13に図7のように表示される(S1)。診断装置の使用者は、表示された診断回転周波数を制御装置6へ指令する(S2)。診断装置の使用者は、表示部8を確認して、診断対象の軸受に支持された回転体が表示・操作部13に表示された回転周波数で回転しているかを判断し、一致している場合には表示・操作部13より測定開始を指令する(S3)。振動加速度を測定して記録し(S4)、必要なデータ長となったら測定完了と判断してS6へ移行する(S5)。振動測定時の回転周波数と振動加速度を対応付けて記録する(S6)。予め登録されている診断回転周波数の全条件の測定が完了しているかを判断し、全条件の測定が完了していればS8へ移行する。完了してない場合にはS1に戻り、測定が完了していない別の診断回転周波数を表示する。
S8では、まず振動加速度をフーリエ変換し振幅スペクトル密度を算出する。予め決めておいた全条件の測定に対して共通の比周波数分解能となるように振幅スペクトル密度の絶対値について補間処理を行い比周波数毎の振動の大きさを算出する。それぞれの診断回転周波数の振動の大きさについて係数が全て正の線型和を算出することで回転同期成分強調波形を算出する(回転同期成分強調波形算出ステップ)。
FIG. 5 shows a flowchart of another method for diagnosing an abnormality of a rolling bearing, and will be specifically described based on this flowchart.
S1 to S8 are the same as the method of FIG. First, one of all the conditions of the diagnostic rotation frequency registered in advance is displayed on the display / operation unit 13 as shown in FIG. 7 (S1). The user of the diagnostic device commands the displayed diagnostic rotation frequency to the control device 6 (S2). The user of the diagnostic device checks the display unit 8 to determine whether the rotating body supported by the bearing to be diagnosed is rotating at the rotation frequency displayed on the display / operation unit 13, and agrees. In that case, the display / operation unit 13 commands the start of measurement (S3). The vibration acceleration is measured and recorded (S4), and when the required data length is reached, it is determined that the measurement is completed and the process proceeds to S6 (S5). The rotation frequency at the time of vibration measurement and the vibration acceleration are recorded in association with each other (S6). It is determined whether the measurement of all the conditions of the diagnostic rotation frequency registered in advance is completed, and if the measurement of all the conditions is completed, the process proceeds to S8. If it is not completed, the process returns to S1 and another diagnostic rotation frequency for which the measurement is not completed is displayed.
In S8, first, the vibration acceleration is Fourier transformed to calculate the amplitude spectral density. The magnitude of vibration for each specific frequency is calculated by performing interpolation processing on the absolute value of the amplitude spectral density so that the specific frequency resolution is common to the measurements under all predetermined conditions. The rotation-synchronized component-emphasized waveform is calculated by calculating the linear sum with all positive coefficients for the magnitude of vibration at each diagnostic rotation frequency (rotation-synchronized component-enhanced waveform calculation step).

そして、S109では、まず振動ピークが存在すると判断された比周波数を4つ選択する。ある比周波数に振動ピークが存在するか否かの判断は、回転同期成分強調波形と回転同期成分強調波形の移動平均とを比較し、回転同期成分強調波形の値の方が大きければ、その比周波数には振動ピークが存在すると判断すればよい。
次に、S110では、選択された4つの振動ピークの比周波数F、F、F、Fと、公約数を持たない異なる自然数N、Nとの組み合わせが以下の数6および数7の関係式を満たすか否かを判断する。なお、F、F、F、Fには同一の比周波数が選ばれても良い(FとF、FとFがそれぞれ同一の比周波数のときに外輪損傷の振動ピークを検出できる)。
Then, in S109, first, four specific frequencies determined to have vibration peaks are selected. To determine whether or not there is a vibration peak at a certain specific frequency, compare the moving average of the rotation-synchronized component-emphasized waveform and the rotation-synchronized component-emphasized waveform, and if the value of the rotation-synchronized component-enhanced waveform is larger, the ratio It may be determined that the frequency has a vibration peak.
Next, in S110, the combination of the specific frequencies F 1 , F 2 , F 3 , F 4 of the four selected vibration peaks and the different natural numbers N 1 and N 2 having no common divisor is the following number 6 and It is determined whether or not the relational expression of Equation 7 is satisfied. Incidentally, the vibration of the F 1, F 2, F 3 , the outer ring damaged when may be chosen the same ratio frequency (F 1 and F 2, F 3 and F 4 are identical ratio frequency Each of the F 4 Peaks can be detected).

Figure 2020159945
Figure 2020159945
Figure 2020159945
Figure 2020159945

S110の判別で、4つの振動ピークの比周波数F、F、F、Fと、公約数を持たない異なる自然数N、Nの組み合わせが数6および数7の関係式を満たす場合、S111へ移行し、関係式を満たさない場合、S109に戻って異なる比周波数を4つ選択してS110の判別を再度行う。
S111では、まず数6および数7の関係式を満たす振動ピークの比周波数の組み合わせについて、以下の数8、数9より比基本周波数R、比振動源回転周波数Rを算出する。
In the discrimination of S110, the specific frequencies F 1 , F 2 , F 3 , F 4 of the four vibration peaks and the combination of different natural numbers N 1 and N 2 having no common divisor satisfy the relational expression of the equations 6 and 7. In this case, the process proceeds to S111, and if the relational expression is not satisfied, the process returns to S109, four different specific frequencies are selected, and S110 is determined again.
In S111, first, the specific fundamental frequency R 0 and the specific vibration source rotation frequency R 1 are calculated from the following equations 8 and 9 for the combination of the specific frequencies of the vibration peaks satisfying the relational expressions of equations 6 and 7.

Figure 2020159945
Figure 2020159945
Figure 2020159945
Figure 2020159945

そして、算出した比振動源回転周波数Rの値に応じて軸受異常の種類(内輪損傷、転動体損傷、外輪損傷)を判断し、S112で、図9に示すような診断結果を表示・操作部13に表示する。ここでは比振動源回転周波数が転動体に対応する0.45と算出されたため、転動体が損傷している可能性ありと表示される。ここでのS109〜S111が軸受損傷種類判別ステップとなる。 Then, the bearing abnormality type according to the calculated value of the ratio vibration source rotation frequency R 1 (inner ring damage, the rolling element damage, the outer ring injury) determines, in S112, the display and operation of the diagnosis result as shown in FIG. 9 Displayed in unit 13. Here, since the specific vibration source rotation frequency is calculated to be 0.45 corresponding to the rolling element, it is displayed that the rolling element may be damaged. Here, S109 to S111 are bearing damage type determination steps.

図6は、機械学習を用いて転がり軸受の異常診断を行う他の方法のフローチャートを示したものであり、このフローチャートに基づいて具体的に説明する。
ここでは軸受異常が既知の軸受について後述する方法で算出される回転同期成分強調波形を入力とし、内輪損傷がある(1)か否か(0)、転動体損傷がある(1)か否か(0)、外輪損傷がある(1)か否か(0)、を出力とする教師データを用いて外部の学習器(図1に図示しない)において、予め機械学習させておき、学習済み数理モデルを図1の演算部12に備えている。数理モデルの形態としては、例えば、多層ニューラルネットワークなどを用いることができる。軸受異常が既知の軸受について評価用データ(回転同期成分強調波形)の値を入力した場合の出力の分布より、内輪、転動体、外輪が異常か否かを判断するしきい値を予め決定しておき、演算部12に備えている。
FIG. 6 shows a flowchart of another method for diagnosing an abnormality of a rolling bearing by using machine learning, and will be specifically described based on this flowchart.
Here, for a bearing whose bearing abnormality is known, the rotation-synchronized component-emphasized waveform calculated by the method described later is input, and whether or not there is inner ring damage (1) or not (0) and whether or not there is rolling element damage (1). Using the teacher data that outputs (0) and whether or not the outer ring is damaged (1) (0), machine learning is performed in advance in an external learner (not shown in FIG. 1), and the learned mathematics. The model is provided in the arithmetic unit 12 of FIG. As the form of the mathematical model, for example, a multi-layer neural network or the like can be used. For bearings with known bearing abnormalities, the threshold value for determining whether the inner ring, rolling element, and outer ring are abnormal is determined in advance from the output distribution when the value of evaluation data (rotation synchronization component emphasis waveform) is input. It is provided in the arithmetic unit 12.

S1〜S8までは図4の方法と同じである。予め登録されている診断回転周波数の全条件のうちの1つが表示・操作部13に図7のように表示される(S1)。診断装置の使用者は、表示された診断回転周波数を制御装置6へ指令する(S2)。診断装置の使用者は、表示部8を確認して、診断対象の軸受に支持された回転体が表示・操作部13に表示された回転周波数で回転しているかを判断し、一致している場合には表示・操作部13より測定開始を指令する(S3)。振動加速度を測定して記録し(S4)、必要なデータ長となったら測定完了と判断してS6へ移行する(S5)。振動測定時の回転周波数と振動加速度を対応付けて記録する(S6)。予め登録されている診断回転周波数の全条件の測定が完了しているかを判断し、全条件の測定が完了していればS8へ移行する。完了してない場合にはS1に戻り、測定が完了していない別の診断回転周波数を表示する。
S8では、まず振動加速度をフーリエ変換し振幅スペクトル密度を算出する。予め決めておいた全条件の測定に対して共通の比周波数分解能となるように振幅スペクトル密度の絶対値について補間処理を行い比周波数毎の振動の大きさを算出する。それぞれの診断回転周波数の振動の大きさについて係数が全て正の線型和を算出することで回転同期成分強調波形を算出する(回転同期成分強調波形算出ステップ)。
S1 to S8 are the same as the method of FIG. One of all the conditions of the diagnostic rotation frequency registered in advance is displayed on the display / operation unit 13 as shown in FIG. 7 (S1). The user of the diagnostic device commands the displayed diagnostic rotation frequency to the control device 6 (S2). The user of the diagnostic device checks the display unit 8 to determine whether the rotating body supported by the bearing to be diagnosed is rotating at the rotation frequency displayed on the display / operation unit 13, and agrees. In that case, the display / operation unit 13 commands the start of measurement (S3). The vibration acceleration is measured and recorded (S4), and when the required data length is reached, it is determined that the measurement is completed and the process proceeds to S6 (S5). The rotation frequency at the time of vibration measurement and the vibration acceleration are recorded in association with each other (S6). It is determined whether the measurement of all the conditions of the diagnostic rotation frequency registered in advance is completed, and if the measurement of all the conditions is completed, the process proceeds to S8. If it is not completed, the process returns to S1 and another diagnostic rotation frequency for which the measurement is not completed is displayed.
In S8, first, the vibration acceleration is Fourier transformed to calculate the amplitude spectral density. The magnitude of vibration for each specific frequency is calculated by performing interpolation processing on the absolute value of the amplitude spectral density so that the specific frequency resolution is common to the measurements under all predetermined conditions. The rotation-synchronized component-emphasized waveform is calculated by calculating the linear sum with all positive coefficients for the magnitude of vibration at each diagnostic rotation frequency (rotation-synchronized component-enhanced waveform calculation step).

そして、S209では、回転同期成分強調波形を学習済み数理モデルに入力し、内輪損傷があるか否か、転動体損傷があるか否か、外輪損傷があるか否か、の出力値をそれぞれ算出する(軸受損傷種類判別ステップ)。
S209で、それぞれの出力値の値が予め設定されたしきい値を超過していれば異常と判断し、S210で、図10に示すような診断結果を表示・操作部13に表示する。図10には、学習済み数理モデルに入力する回転同期成分強調波形がグラフとして、学習済み数理モデルからの出力値、異常か否かを判断するしきい値、異常か否かを判断した判定結果が表示されている。
Then, in S209, the rotation-synchronized component-emphasized waveform is input to the trained mathematical model, and the output values of whether or not there is inner ring damage, whether or not there is rolling element damage, and whether or not there is outer ring damage are calculated. (Bearing damage type determination step).
In S209, if the value of each output value exceeds a preset threshold value, it is determined to be abnormal, and in S210, the diagnosis result as shown in FIG. 10 is displayed on the display / operation unit 13. In FIG. 10, the rotation synchronization component emphasis waveform input to the trained mathematical model is shown as a graph, and the output value from the trained mathematical model, the threshold value for determining whether or not it is abnormal, and the determination result for determining whether or not it is abnormal. Is displayed.

このように、上記各形態の異常診断方法及び異常診断装置200、異常診断プログラムによれば、測定した振動に対し、主軸1の回転周波数に比例した周波数の振動成分を強調する回転同期成分強調波形を算出し、回転同期成分強調波形の振動ピーク位置の規則性に基づいて軸受7における損傷の種類を判別するので、軸受諸元を用いることなく振動ピークの規則性から軸受異常の種類を判断することが可能となる。
さらに、比基本周波数と比振動源回転周波数を推定する処理の場合、基本周波数や特徴周波数の算出が行える。このため、基本周波数や特徴周波数を算出するために軸受諸元を利用する必要があった従来の診断技術(例えば特許文献2)を採用して損傷レベルの診断をさらに行うことも可能となる。ある回転周波数における基本周波数が必要な場合は、比基本周波数の値に回転周波数を乗算することで算出することが可能である。ある回転周波数における、転動体の保持器が自転する(転動体が公転する)周波数が必要な場合には、転動体損傷と判断された比基本周波数と比振動源回転周波数の組み合わせの比振動源回転周波数の値に回転周波数を乗算することで算出することが可能である。この損傷レベルの診断処理は、図5の異常診断方法においても採用できる。
As described above, according to the abnormality diagnosis method of each form, the abnormality diagnosis device 200, and the abnormality diagnosis program, the rotation synchronization component emphasis waveform that emphasizes the vibration component having a frequency proportional to the rotation frequency of the spindle 1 with respect to the measured vibration. Is calculated, and the type of damage in the bearing 7 is determined based on the regularity of the vibration peak position of the rotation-synchronized component emphasized waveform. Therefore, the type of bearing abnormality is determined from the regularity of the vibration peak without using the bearing specifications. It becomes possible.
Further, in the case of the process of estimating the specific fundamental frequency and the specific vibration source rotation frequency, the fundamental frequency and the characteristic frequency can be calculated. Therefore, it is possible to further diagnose the damage level by adopting a conventional diagnostic technique (for example, Patent Document 2) that requires the use of bearing specifications in order to calculate the fundamental frequency and the characteristic frequency. When a fundamental frequency at a certain rotation frequency is required, it can be calculated by multiplying the value of the ratio fundamental frequency by the rotation frequency. When a frequency at which the cage of the rolling element rotates (the rolling element revolves) at a certain rotation frequency is required, the specific vibration source of the combination of the specific fundamental frequency and the specific vibration source determined to be the rolling element damage. It can be calculated by multiplying the value of the rotation frequency by the rotation frequency. This damage level diagnostic process can also be adopted in the abnormality diagnosis method of FIG.

特にここでは、振動測定ステップでは、異なる複数の回転周波数で振動を測定し、回転同期成分強調波形算出ステップでは、振動を周波数分析して共通の比周波数に対応する振動の大きさをそれぞれ算出し、係数が全て正の線型和をとるようにしている(S8)ので、さらに外乱の影響が低減されて、振動ピークの規則性を捉え易くなるため、軸受異常の推定精度が向上する。 In particular, here, in the vibration measurement step, vibration is measured at a plurality of different rotation frequencies, and in the rotation synchronization component emphasis waveform calculation step, the vibration is frequency-analyzed to calculate the magnitude of vibration corresponding to a common specific frequency. Since all the coefficients have a positive linear sum (S8), the influence of disturbance is further reduced, and the regularity of the vibration peak is easily grasped, so that the estimation accuracy of the bearing abnormality is improved.

また、図6に示す異常診断方法では、回転同期成分強調波形を入力とし、内輪傷の有無、転動体傷の有無、外輪傷の有無の少なくともひとつを出力とする教師データを用いて学習した機械学習モデルを用い、回転同期成分強調波形を入力して軸受損傷の種類を判別するので、人間が考える処理よりも汎用的に振動ピークの規則性から軸受異常の種類を判断することが可能となる。
そして、異常診断装置200では、軸受損傷に対応する比基本周波数と比振動源回転周波数とを合わせて表示・操作部13に表示するので、数字として表示される周波数の値を用いて異常診断装置200とは別の異常診断装置による軸受診断処理を行うことが可能となる。
さらにここでは、特定の比周波数の位置と回転同期成分強調波形とを合わせて表示・操作部13に表示するので、振動ピークの規則性を正しく抽出できているか否かを視覚的に把握できるため、診断結果の妥当性を容易に検証できる。また、各損傷の種類において着目すべき比周波数がわかりやすくなる。
Further, in the abnormality diagnosis method shown in FIG. 6, a machine learned using teacher data in which a rotation-synchronized component-emphasized waveform is input and at least one of the presence / absence of an inner ring injury, the presence / absence of a rolling body injury, and the presence / absence of an outer ring injury is output. Since the type of bearing damage is determined by inputting the rotation-synchronized component-enhanced waveform using the learning model, it is possible to determine the type of bearing abnormality from the regularity of the vibration peak, which is more general than the processing considered by humans. ..
Then, in the abnormality diagnosis device 200, since the specific fundamental frequency corresponding to the bearing damage and the specific vibration source rotation frequency are displayed on the display / operation unit 13, the abnormality diagnosis device uses the value of the frequency displayed as a numerical value. It is possible to perform the bearing diagnosis process by an abnormality diagnosis device different from the 200.
Further, here, since the position of the specific specific frequency and the rotation-synchronized component emphasis waveform are displayed together on the display / operation unit 13, it is possible to visually grasp whether or not the regularity of the vibration peak can be extracted correctly. , The validity of the diagnosis result can be easily verified. In addition, the specific frequency to be noted for each type of damage becomes easy to understand.

なお、上記各例では、診断対象の軸受に支持された回転体の回転周波数を直接指令可能な例を示したが、ベルトや歯車を介して工作機械主軸を回転させるモータの支持軸受を診断する場合は、減速比を考慮して、モータの支持軸受の回転周波数が診断周波数となるように指定してやればよい。
また、表示・操作部13に表示された診断回転周波数に診断対象の軸受の回転周波数を合わせることで、振動加速度と回転周波数を対応付けて記憶する方法を示したが、診断対象の軸受の回転周波数を表示・操作部13より診断装置の使用者が入力する方法や、異常診断装置200が速度検出器5の情報を直接取り込めるようにする方法でもよい。
さらに、回転周波数(単位はHzを用いることが多い)と回転速度(工作機械主軸では、単位min−1が用いられることが多い)は、1Hz=60min−1の関係がある同一の量であり、どちらを用いてもよい。
In each of the above examples, the rotation frequency of the rotating body supported by the bearing to be diagnosed is directly commanded. However, the bearing of the motor that rotates the machine tool spindle via a belt or a gear is diagnosed. In this case, the rotation frequency of the support bearing of the motor may be specified as the diagnostic frequency in consideration of the reduction ratio.
Further, a method of storing the vibration acceleration and the rotation frequency in association with each other by matching the rotation frequency of the bearing to be diagnosed with the diagnosis rotation frequency displayed on the display / operation unit 13 has been shown. The frequency may be input by the user of the diagnostic device from the display / operation unit 13, or the abnormality diagnostic device 200 may directly capture the information of the speed detector 5.
Further, the rotation frequency (the unit is often Hz) and the rotation speed (the unit min -1 is often used in the machine tool spindle) are the same quantity having a relationship of 1 Hz = 60 min -1 . , Either may be used.

一方、すべりのない場合の、転がり軸受の転動体の公転周波数を回転周波数で除算した値(転動体異常の場合の比振動源回転周波数R)は、転動体直径d、ピッチ円直径D、接触角αを用いて、以下の数10で表現される。 On the other hand, when there is no slip, (the ratio vibration source in the case of rolling element abnormal rotation frequency R 1) the revolution frequency of the rolling elements of the rolling bearing dividing the value in the rotational frequency, the rolling element diameter d, a pitch circle diameter D, It is expressed by the following several tens using the contact angle α.

Figure 2020159945
Figure 2020159945

数10より、転がり軸受の転動体の公転周波数を回転周波数で除算した値(転動体異常の場合の比振動源回転周波数R)はピッチ円直径に対する転動体直径の比が大きいほど小さくなることがわかる。転がり軸受を構成する上で最低3個の転動体が必要となるので、3個の転動体を密に並べた構造の軸受においてピッチ円直径に対する転動体直径の比が最大となるため、幾何学的に決定される転がり軸受の転動体の公転周波数を回転周波数で除算した値(転動体異常の場合の比振動源回転周波数R)の下限値は、(1−√3÷2)÷2≒0.0669である。数10より、転動体直径を無限小にした場合が、転がり軸受の転動体の公転周波数を回転周波数で除算した値(転動体異常の場合の比振動源回転周波数R)が最大となることがわかる。このため、幾何学的には、転がり軸受の転動体の公転周波数を回転周波数で除算した値(転動体異常の場合の比振動源回転周波数R)の上限値は0.5である。よって、転がり軸受の転動体の公転周波数を回転周波数で除算した値(転動体異常の場合の比振動源回転周波数R)がとり得る範囲は、最も広く見積もっても(1−√3÷2)÷2以上から0.5以下である。この範囲は、内輪異常の場合の比振動源回転周波数である1、外輪異常の場合の比振動源回転周波数である0を含んでいないため、内輪異常・転動体異常・外輪異常のいずれに分類すればよいか判別できないという状況は生じない。
工作機械主軸に用いられる軸受の場合、転動体異常の場合の比振動源回転周波数は0.45前後であるため、仮定する比振動源回転周波数は0、0.4以上0.5以下、1とするなどさらに限定して探索を行ってもよい。
Than the number 10, (the ratio vibration source in the case of rolling element abnormal rotation frequency R 1) the revolution frequency of the rolling elements of the rolling bearing divided by the rotational frequency value be smaller the larger the ratio of the rolling element diameter to the pitch circle diameter I understand. Since a minimum of three rolling elements are required to form a rolling bearing, the ratio of the rolling element diameter to the pitch circle diameter is the maximum in a bearing having a structure in which three rolling elements are closely arranged. The lower limit of the value obtained by dividing the revolving frequency of the rolling element of the rolling bearing (specific vibration source rotation frequency R 1 in the case of an abnormality of the rolling element) by the rotation frequency is (1-√3 / 2) / 2 ≈0.0669. From several tens, when the rolling element diameter is made infinitesimal, the value obtained by dividing the rolling element rolling element of the rolling bearing by the rotation frequency (specific vibration source rotation frequency R 1 in the case of rolling element abnormality) becomes the maximum. I understand. Therefore, the geometric, the upper limit value of the division value in the rotation frequency of the revolution frequency of the rolling elements of the rolling bearing (rolling element in case of abnormal ratios vibration source rotation frequency R 1) is 0.5. Therefore, it may take a range (ratio vibration source rotation frequency R 1 in the case of the rolling element error) the revolution frequency division value in the rotation frequency of the rolling elements of the rolling bearing, even estimates widest (1-√3 ÷ 2 ) ÷ 2 or more to 0.5 or less. Since this range does not include 1 which is the specific vibration source rotation frequency in the case of inner ring abnormality and 0 which is the specific vibration source rotation frequency in the case of outer ring abnormality, it is classified into any of inner ring abnormality, rolling element abnormality, and outer ring abnormality. There is no situation in which it is not possible to determine what to do.
In the case of bearings used for machine tool spindles, the specific vibration source rotation frequency in the case of a rolling element abnormality is around 0.45, so the assumed specific vibration source rotation frequency is 0, 0.4 or more and 0.5 or less, 1 The search may be further limited, such as.

そして、各形態では、測定毎に比周波数分解能が一致しない場合に比周波数分解能を統一する方法として補間処理を行う例を示したが、軸受の接触角の変化や回転周波数の誤差により軸受損傷に起因する振動ピークの周波数がX%程度前後する場合は、比周波数分解能統一後のある比周波数の振動の大きさとして採用する値はその比周波数の±X%の範囲の最大値を採用するとよい。このように比周波数の前後に所定の幅を持たせて当該幅内の最大値を採用することにより、軸受の接触角の変化や回転周波数の誤差によって振動ピークの位置がX%ずれたとしても、必ずある比周波数において軸受損傷に起因する振動ピークの振動の大きさを合算した値が得られることになる。なお、比周波数分解能が荒いため、ある比周波数の±X%の範囲にデータが存在しない場合には、±比周波数分解能分の範囲の最大値を採用する。なお、Xの値の決定は、例えば実際に複数の回転周波数において、軸受諸元が既知の異常軸受に支持された主軸の振動を測定し、軸受諸元から算出される振動ピークの比周波数に最も近い振動ピークの比周波数を求め、複数の回転周波数において求めた比周波数の標準偏差の数倍(例えば3〜6倍)を、複数の回転周波数において求めた比周波数の平均で割って算出するといった方法がある。このように、一旦軸受諸元が既知の異常軸受でばらつきの度合いX%を把握しておけば、このX%を軸受諸元が未知の軸受に対しても用いることができる。 Then, in each form, an example of performing interpolation processing as a method of unifying the specific frequency resolution when the specific frequency resolution does not match for each measurement is shown, but the bearing may be damaged due to a change in the contact angle of the bearing or an error in the rotation frequency. When the frequency of the resulting vibration peak is around X%, the maximum value in the range of ± X% of the specific frequency should be adopted as the value to be adopted as the magnitude of vibration of a certain specific frequency after unifying the specific frequency resolution. .. By providing a predetermined width before and after the specific frequency and adopting the maximum value within the width in this way, even if the position of the vibration peak shifts by X% due to a change in the contact angle of the bearing or an error in the rotation frequency. , The total value of the vibration magnitudes of the vibration peaks caused by bearing damage at a certain specific frequency can always be obtained. Since the specific frequency resolution is rough, if there is no data in the range of ± X% of a certain specific frequency, the maximum value in the range of ± specific frequency resolution is adopted. The value of X is determined by, for example, actually measuring the vibration of the spindle supported by an abnormal bearing whose bearing specifications are known at a plurality of rotation frequencies, and determining the specific frequency of the vibration peak calculated from the bearing specifications. The specific frequency of the closest vibration peak is calculated, and it is calculated by dividing several times the standard deviation of the specific frequencies obtained at multiple rotation frequencies (for example, 3 to 6 times) by the average of the specific frequencies obtained at multiple rotation frequencies. There is such a method. In this way, once the degree of variation X% is known for an abnormal bearing whose bearing specifications are known, this X% can be used for a bearing whose bearing specifications are unknown.

一方、各形態では、係数が全て正の線型和をとることで回転同期成分強調波形を算出し、回転同期成分強調波形と回転同期成分強調波形の移動平均を比較し、回転同期成分強調波形の値の方が大きければ、その比周波数には振動ピークが存在すると判断するような局所的に最大となる振動ピークを抽出する例を示したが、係数が全て負の線型和をとることで回転同期成分強調波形を算出し、回転同期成分強調波形と回転同期成分強調波形の移動平均を比較し、回転同期成分強調波形の値の方が小さければ、その比周波数には振動ピークが存在すると判断する局所的に最小となる振動ピークを抽出する手法でも全く同様の議論が成立する。
また、軸受損傷種類判別ステップで用いる回転同期成分強調波形の比周波数の範囲は、算出可能な全ての比周波数の範囲とする必要はなく、軸受損傷に起因する振動ピークの規則性を捉え易い範囲にすることが望ましい。
さらに、機械学習を用いて転がり軸受の異常診断を行う方法の教師データの出力として、0と1の2値を与える手法を示したが、損傷度合いを定量的に表現する連続的な値を教師データとすることも可能である。
加えて、比周波数、比基本周波数、比振動源回転周波数、これらの値と比較される値、の全てに同一の値を乗算しても、全く同じ議論が成立する。つまり、例えば比振動源回転周波数が1と一致するか否かを判断することは、比振動源回転周波数×30Hzが1×30Hzと一致するか否かを判断することと同義である。
On the other hand, in each form, the rotation-synchronized component-emphasized waveform is calculated by taking a linear sum with all the coefficients positive, and the moving averages of the rotation-synchronized component-enhanced waveform and the rotation-synchronized component-enhanced waveform are compared, and the rotation-synchronized component-enhanced waveform An example was shown in which if the value is larger, the vibration peak that is locally maximized is extracted so that it is judged that there is a vibration peak at the specific frequency, but the rotation is caused by taking the sum of the waveforms in which all the coefficients are negative. Calculate the synchronization component emphasis waveform, compare the moving average of the rotation synchronization component emphasis waveform and the rotation synchronization component emphasis waveform, and if the value of the rotation synchronization component emphasis waveform is smaller, it is judged that there is a vibration peak at the specific frequency. The same argument holds for the method of extracting the locally minimum vibration peak.
Further, the range of the specific frequency of the rotation synchronization component emphasized waveform used in the bearing damage type determination step does not have to be the range of all the calculable specific frequencies, and the regularity of the vibration peak caused by the bearing damage can be easily grasped. It is desirable to.
Furthermore, as the output of the teacher data of the method of diagnosing the abnormality of rolling bearings using machine learning, a method of giving two values of 0 and 1 was shown, but a continuous value that quantitatively expresses the degree of damage is taught. It can also be used as data.
In addition, even if the same value is multiplied by the specific frequency, the specific fundamental frequency, the specific vibration source rotation frequency, and the value to be compared with these values, the exact same argument holds. That is, for example, determining whether or not the specific vibration source rotation frequency matches 1 is synonymous with determining whether or not the specific vibration source rotation frequency × 30 Hz coincides with 1 × 30 Hz.

そして、振動測定ステップは、一定の回転周波数のときに実行する場合に限らない。例えば工作機械の主軸のような回転負荷が小さい回転体の場合、惰性回転をさせた場合の単位時間当たりの回転周波数の変化は非常に小さい。このため、回転体の振動を複数回測定する振動測定ステップは、予め設定した回転周波数に主軸を制御した後惰性回転させ、停止若しくはある回転数に至るまでに繰り返し振動測定を行うようにしてもよい。このとき、振動測定を行ったそれぞれの時間帯では、各時間帯の平均の回転周波数で回転していると見なして比周波数を算出し、同様の処理を行えばよい。
また、比周波数ごとの振動の大きさについて係数が全て同符号の線型和を算出した値に対して、定数を加算・減算・乗算・除算したり、比周波数の変化に対して単調増加又は減少する値を加算・減算したり、比周波数の変化に対して単調増加又は減少する値でさらに全て同符号である値を乗算・除算したりしても差し支えない。すなわち、振動ピーク位置を変化させない処理であれば本発明に包含できる。
The vibration measurement step is not limited to the case where the vibration measurement step is executed at a constant rotation frequency. For example, in the case of a rotating body having a small rotational load such as a spindle of a machine tool, the change in the rotation frequency per unit time when inertial rotation is performed is very small. Therefore, in the vibration measurement step of measuring the vibration of the rotating body a plurality of times, the spindle is controlled to a preset rotation frequency and then coasted, and the vibration is repeatedly measured until it stops or reaches a certain rotation speed. Good. At this time, in each time zone in which the vibration measurement is performed, the specific frequency may be calculated by assuming that the vibration is rotating at the average rotation frequency in each time zone, and the same processing may be performed.
In addition, constants are added / subtracted / multiplied / divided with respect to the calculated linear sum of all coefficients having the same sign for the magnitude of vibration for each specific frequency, and monotonically increased or decreased with respect to changes in the specific frequency. It is also possible to add or subtract the values to be added, or to multiply or divide the values that monotonically increase or decrease with respect to the change in the specific frequency and all have the same sign. That is, any process that does not change the vibration peak position can be included in the present invention.

その他、本実施例では、工作機械と異常診断装置とを別体で示しているが、異常診断装置を制御装置に内蔵しても良い。
また、異常診断装置を複数の工作機械と有線或いは無線で通信可能とし、工作機械側で振動を測定してデータを取得しつつ、異常診断プログラムに基づいて異常診断方法を工作機械毎に実行するようにしてもよい。さらに、振動の測定から異常診断までを連続して行う場合に限らず、振動の測定データを記憶部に保存しておき、所定のタイミングで異常診断プログラムによる異常診断を行うようにしても差し支えない。
In addition, in this embodiment, the machine tool and the abnormality diagnosis device are shown separately, but the abnormality diagnosis device may be built in the control device.
In addition, the abnormality diagnosis device can communicate with a plurality of machine tools by wire or wirelessly, and while the machine tool measures vibration and acquires data, the abnormality diagnosis method is executed for each machine tool based on the abnormality diagnosis program. You may do so. Further, not only when the vibration measurement to the abnormality diagnosis are continuously performed, the vibration measurement data may be saved in the storage unit and the abnormality diagnosis may be performed by the abnormality diagnosis program at a predetermined timing. ..

1・・主軸、2・・主軸ハウジング、3・・工具、4・・モータ、5・・速度検出器、6・・制御装置、7・・軸受、8・・表示部、9・・振動センサ、10・・A/D変換器、11・・記憶部、12・・演算部、13・・表示・操作部、100・・工作機械、200・・異常診断装置。 1 ... Spindle, 2 ... Spindle housing, 3 ... Tools, 4 ... Motors, 5 ... Speed detectors, 6 ... Control devices, 7 ... Bearings, 8 ... Display, 9 ... Vibration sensors 10, ・ ・ A / D converter, 11 ・ ・ storage unit, 12 ・ ・ calculation unit, 13 ・ ・ display ・ operation unit, 100 ・ ・ machine tool, 200 ・ ・ abnormality diagnosis device.

Claims (14)

回転体を支持する転がり軸受の異常を診断する方法であって、
前記回転体の振動を複数回測定する振動測定ステップと、
前記振動を周波数分析して周波数ごとの振動の大きさをそれぞれ求め、前記振動の周波数を回転周波数で除算して得られる無次元量を比周波数として、各回の測定で得られた前記振動の大きさを共通の前記比周波数ごとに算出し、前記比周波数ごとの振動の大きさについて係数が全て同符号の線型和を算出し、算出した値に基づいて回転同期成分強調波形を決定する回転同期成分強調波形算出ステップと、
前記回転同期成分強調波形の振動ピーク位置の規則性に基づいて、前記転がり軸受において力の発生する方向が変化する周波数である振動源回転周波数と前記回転周波数との比である比振動源回転周波数を推定し、前記比振動源回転周波数の値が1の場合には、前記転がり軸受の内輪損傷と判断し、前記比振動源回転周波数の値が前記転がり軸受の転動体の公転周波数を前記回転周波数で除算した値のとり得る範囲である場合には、前記転動体の損傷と判断し、前記比振動源回転周波数の値が0の場合には、前記転がり軸受の外輪損傷と判断する軸受損傷種類判別ステップと、
を実行することを特徴とする転がり軸受の異常診断方法。
It is a method of diagnosing abnormalities in rolling bearings that support rotating bodies.
A vibration measurement step for measuring the vibration of the rotating body a plurality of times, and
The magnitude of the vibration obtained in each measurement is obtained by frequency-analyzing the vibration to obtain the magnitude of the vibration for each frequency, and using the dimensionless quantity obtained by dividing the frequency of the vibration by the rotation frequency as the specific frequency. Rotation synchronization is calculated for each of the common specific frequencies, the linear sum of all the coefficients having the same sign is calculated for the magnitude of vibration for each specific frequency, and the rotation synchronization component emphasis waveform is determined based on the calculated value. Component emphasis waveform calculation step and
Based on the regularity of the vibration peak position of the rotation synchronization component emphasis waveform, the specific vibration source rotation frequency which is the ratio of the vibration source rotation frequency which is the frequency at which the direction in which the force is generated in the rolling bearing changes to the rotation frequency. When the value of the specific vibration source rotation frequency is 1, it is determined that the inner ring of the rolling bearing is damaged, and the value of the specific vibration source rotation frequency is the rotation frequency of the rolling element of the rolling bearing. If the value divided by the frequency is within the range that can be taken, it is judged that the rolling element is damaged, and if the value of the specific vibration source rotation frequency is 0, it is judged that the outer ring of the rolling bearing is damaged. Type determination step and
An abnormality diagnosis method for rolling bearings, which is characterized by performing.
前記比振動源回転周波数の推定は、前記回転同期成分強調波形において、前記転がり軸受の損傷に対応する比基本周波数と前記比振動源回転周波数とを仮定して算出される特定の比周波数に振動ピークが存在する割合である一致率を算出し、前記一致率が所定のしきい値を超過する場合には、仮定した前記比基本周波数と仮定した前記比振動源回転周波数とにより表現される規則性が前記回転同期成分強調波形にあると判断し、仮定した前記比振動源回転周波数を採用することを特徴とする請求項1に記載の転がり軸受の異常診断方法。 The estimation of the specific vibration source rotation frequency vibrates to a specific specific frequency calculated by assuming the specific basic frequency corresponding to the damage of the rolling bearing and the specific vibration source rotation frequency in the rotation synchronization component emphasis waveform. A rule expressed by calculating the coincidence rate, which is the ratio at which peaks exist, and when the coincidence rate exceeds a predetermined threshold value, the assumed specific fundamental frequency and the assumed specific vibration source rotation frequency. The method for diagnosing an abnormality of a rolling bearing according to claim 1, wherein it is determined that the property is in the rotation-synchronized component-emphasized waveform, and the assumed specific vibration source rotation frequency is adopted. 前記比振動源回転周波数の推定は、前記回転同期成分強調波形において、ピーク対を構成する2つの振動ピークの間隔をピーク間距離とし、前記ピーク間距離が等しい2組のピーク対について算出したそれぞれのピーク対を構成する2つの前記振動ピークの前記比周波数の平均の比が自然数比の場合に、前記ピーク間距離の2分の1を前記比振動源回転周波数とする、又はある2つの振動ピークの前記比周波数の比が自然数比である場合に、前記比振動源回転周波数を0とすることを特徴とする請求項1に記載の転がり軸受の異常診断方法。 The estimation of the specific vibration source rotation frequency is calculated for two sets of peak pairs having the same peak-to-peak distance, with the distance between the two vibration peaks constituting the peak pair as the peak-to-peak distance in the rotation-synchronized component-emphasized waveform. When the average ratio of the specific frequencies of the two vibration peaks constituting the peak pair is a natural number ratio, half of the distance between the peaks is set as the specific vibration source rotation frequency, or two vibrations. The method for diagnosing an abnormality of a rolling bearing according to claim 1, wherein the specific vibration source rotation frequency is set to 0 when the ratio of the specific frequencies of the peaks is a natural number ratio. 前記軸受損傷種類判別ステップは、前記転がり軸受の損傷に対応する比基本周波数と前記比振動源回転周波数とを推定する処理を含むものであり、前記転がり軸受の損傷の種類が判別された前記比基本周波数と前記比振動源回転周波数との組を用いて、判別された損傷の程度をさらに診断することを特徴とする請求項1乃至3の何れかに記載の転がり軸受の異常診断方法。 The bearing damage type determination step includes a process of estimating the specific basic frequency corresponding to the damage of the rolling bearing and the specific vibration source rotation frequency, and the ratio at which the type of damage of the rolling bearing is determined. The method for diagnosing an abnormality of a rolling bearing according to any one of claims 1 to 3, wherein the degree of damage determined is further diagnosed by using a set of a basic frequency and the specific vibration source rotation frequency. 回転体を支持する転がり軸受の異常を診断する方法であって、
前記回転体の振動を複数回測定する振動測定ステップと、
前記振動を周波数分析して周波数ごとの振動の大きさをそれぞれ求め、前記振動の周波数を回転周波数で除算して得られる無次元量を比周波数として、各回の測定で得られた前記振動の大きさを共通の前記比周波数ごとに算出し、前記比周波数ごとの振動の大きさについて係数が全て同符号の線型和を算出し、算出した値に基づいて回転同期成分強調波形を決定する回転同期成分強調波形算出ステップと、
前記回転同期成分強調波形を入力とし、内輪傷の有無、転動体傷の有無、外輪傷の有無の少なくともひとつを出力とする教師データを用いて学習した機械学習モデルを用い、前記回転同期成分強調波形を入力して前記転がり軸受の損傷の種類を判別する軸受損傷種類判別ステップと、
を実行することを特徴とする転がり軸受の異常診断方法。
It is a method of diagnosing abnormalities in rolling bearings that support rotating bodies.
A vibration measurement step for measuring the vibration of the rotating body a plurality of times, and
The magnitude of the vibration obtained in each measurement is obtained by frequency-analyzing the vibration to obtain the magnitude of the vibration for each frequency, and using the dimensionless quantity obtained by dividing the frequency of the vibration by the rotation frequency as the specific frequency. Rotation synchronization is calculated for each of the common specific frequencies, the linear sum of all the coefficients having the same sign is calculated for the magnitude of vibration for each specific frequency, and the rotation synchronization component emphasis waveform is determined based on the calculated value. Component emphasis waveform calculation step and
Using a machine learning model learned using teacher data that inputs the rotation-synchronized component emphasis waveform and outputs at least one of the presence / absence of inner ring damage, the presence / absence of rolling body damage, and the presence / absence of outer ring damage, the rotation-synchronization component emphasis Bearing damage type determination step for inputting a waveform to determine the type of damage to the rolling bearing, and
An abnormality diagnosis method for rolling bearings, which is characterized by performing.
前記振動測定ステップでは、異なる複数の回転周波数で振動を測定することを特徴とする請求項1乃至5の何れかに記載の転がり軸受の異常診断方法。 The method for diagnosing an abnormality of a rolling bearing according to any one of claims 1 to 5, wherein in the vibration measurement step, vibration is measured at a plurality of different rotation frequencies. 前記回転同期成分強調波形算出ステップでは、各回の測定で得られた前記振動の大きさを共通の前記比周波数ごとに算出する際に、ある比周波数の振動の大きさとして、その比周波数の前後に所定の幅を持たせて当該幅内の振動の大きさの最大値を採用することを特徴とする請求項1乃至6の何れかに記載の転がり軸受の異常診断方法。 In the rotation synchronization component emphasis waveform calculation step, when the magnitude of the vibration obtained in each measurement is calculated for each of the common specific frequencies, the magnitude of the vibration of a certain specific frequency is set before and after the specific frequency. The method for diagnosing an abnormality of a rolling bearing according to any one of claims 1 to 6, wherein a predetermined width is provided and the maximum value of the magnitude of vibration within the width is adopted. 回転体を支持する転がり軸受の異常を診断する装置であって、
前記回転体の振動を複数回測定する振動測定手段と、
前記振動を周波数分析して周波数ごとの振動の大きさをそれぞれ求め、前記振動の周波数を回転周波数で除算して得られる無次元量を比周波数として、各回の測定で得られた前記振動の大きさを共通の前記比周波数ごとに算出し、前記比周波数ごとの振動の大きさについて係数が全て同符号の線型和を算出し、算出した値に基づいて回転同期成分強調波形を決定する回転同期成分強調波形算出手段と、
前記回転同期成分強調波形の振動ピーク位置の規則性に基づいて、前記転がり軸受において力の発生する方向が変化する周波数である振動源回転周波数と前記回転周波数との比である比振動源回転周波数を推定し、前記比振動源回転周波数の値が1の場合には、前記転がり軸受の内輪損傷と判断し、前記比振動源回転周波数の値が前記転がり軸受の転動体の公転周波数を前記回転周波数で除算した値のとり得る範囲である場合には、前記転動体の損傷と判断し、前記比振動源回転周波数の値が0の場合には、前記転がり軸受の外輪損傷と判断する軸受損傷種類判別手段と、
を備えることを特徴とする転がり軸受の異常診断装置。
A device for diagnosing abnormalities in rolling bearings that support rotating bodies.
A vibration measuring means for measuring the vibration of the rotating body a plurality of times,
The magnitude of the vibration obtained in each measurement is obtained by frequency-analyzing the vibration to obtain the magnitude of the vibration for each frequency, and using the dimensionless quantity obtained by dividing the frequency of the vibration by the rotation frequency as the specific frequency. Rotation synchronization is calculated for each of the common specific frequencies, the linear sum of all the coefficients having the same sign is calculated for the magnitude of vibration for each specific frequency, and the rotation synchronization component emphasis waveform is determined based on the calculated value. Component emphasis waveform calculation means and
Based on the regularity of the vibration peak position of the rotation synchronization component emphasis waveform, the specific vibration source rotation frequency which is the ratio of the vibration source rotation frequency which is the frequency at which the direction in which the force is generated in the rolling bearing changes to the rotation frequency. When the value of the specific vibration source rotation frequency is 1, it is determined that the inner ring of the rolling bearing is damaged, and the value of the specific vibration source rotation frequency is the rotation frequency of the rolling element of the rolling bearing. If the value divided by the frequency is within the range that can be taken, it is judged that the rolling element is damaged, and if the value of the specific vibration source rotation frequency is 0, it is judged that the outer ring of the rolling bearing is damaged. Type discrimination means and
An abnormality diagnostic device for rolling bearings, which comprises.
前記軸受損傷種類判別手段は、前記比振動源回転周波数の推定に当たり、前記回転同期成分強調波形において、前記転がり軸受の損傷に対応する比基本周波数と前記比振動源回転周波数とを仮定して算出される特定の比周波数に振動ピークが存在する割合である一致率を算出し、前記一致率が所定のしきい値を超過する場合には、仮定した前記比基本周波数と仮定した前記比振動源回転周波数とにより表現される規則性が前記回転同期成分強調波形にあると判断し、仮定した前記比振動源回転周波数を採用することを特徴とする請求項8に記載の転がり軸受の異常診断装置。 The bearing damage type determining means calculates the specific vibration source rotation frequency by assuming the specific fundamental frequency corresponding to the damage of the rolling bearing and the specific vibration source rotation frequency in the rotation synchronization component emphasis waveform. The coincidence rate, which is the ratio at which the vibration peak exists at a specific specific frequency, is calculated, and when the coincidence rate exceeds a predetermined threshold value, the specific vibration source assumed to be the assumed specific fundamental frequency. The abnormality diagnostic apparatus for a rolling bearing according to claim 8, wherein the regularity expressed by the rotation frequency is determined to be in the rotation synchronization component emphasized waveform, and the assumed specific vibration source rotation frequency is adopted. .. 前記軸受損傷種類判別手段は、前記比振動源回転周波数の推定に当たり、前記回転同期成分強調波形において、ピーク対を構成する2つの振動ピークの間隔をピーク間距離とし、前記ピーク間距離が等しい2組のピーク対について算出したそれぞれのピーク対を構成する2つの前記振動ピークの前記比周波数の平均の比が自然数比の場合に、前記ピーク間距離の2分の1を前記比振動源回転周波数とする、又はある2つの振動ピークの前記比周波数の比が自然数比である場合に、前記比振動源回転周波数を0とすることを特徴とする請求項8に記載の転がり軸受の異常診断装置。 In estimating the specific vibration source rotation frequency, the bearing damage type determining means sets the distance between two vibration peaks constituting the peak pair as the peak-to-peak distance in the rotation-synchronized component-emphasized waveform, and the peak-to-peak distance is equal 2. When the average ratio of the specific frequencies of the two vibration peaks constituting each peak pair calculated for the pair of peak pairs is a natural number ratio, half of the distance between the peaks is the specific vibration source rotation frequency. The abnormality diagnostic apparatus for rolling bearings according to claim 8, wherein the specific vibration source rotation frequency is set to 0 when the ratio of the specific frequencies of two vibration peaks is a natural number ratio. .. 前記軸受損傷種類判別手段は、前記転がり軸受の損傷に対応する比基本周波数と前記比振動源回転周波数とを推定すると共に、前記転がり軸受の損傷の種類が判別された前記比基本周波数と前記比振動源回転周波数との組を用いて、判別された損傷の程度をさらに診断することを特徴とする請求項8乃至10の何れかに記載の転がり軸受の異常診断装置。 The bearing damage type determining means estimates the specific basic frequency corresponding to the damage of the rolling bearing and the specific vibration source rotation frequency, and the specific basic frequency and the ratio for which the type of damage of the rolling bearing is determined. The abnormality diagnostic apparatus for rolling bearings according to any one of claims 8 to 10, further diagnosing the degree of damage determined by using a pair with a vibration source rotation frequency. 前記軸受損傷種類判別手段は、前記転がり軸受の損傷に対応する比基本周波数と、前記比振動源回転周波数と、前記比基本周波数と前記比振動源回転周波数とから算出される特定の比周波数とを用いて前記転がり軸受の損傷の種類ごとの前記振動ピーク位置の存在を確認するものであり、
前記比基本周波数と前記比振動源回転周波数とを合わせて表示する表示手段を備えることを特徴とする請求項8乃至11の何れかに記載の転がり軸受の異常診断装置。
The bearing damage type determining means includes a specific fundamental frequency corresponding to damage to the rolling bearing, the specific vibration source rotation frequency, and a specific specific frequency calculated from the specific fundamental frequency and the specific vibration source rotation frequency. Is used to confirm the existence of the vibration peak position for each type of damage to the rolling bearing.
The abnormality diagnostic device for a rolling bearing according to any one of claims 8 to 11, further comprising a display means for displaying the specific fundamental frequency and the specific vibration source rotation frequency together.
前記軸受損傷種類判別手段は、前記転がり軸受の損傷に対応する比基本周波数と、前記比振動源回転周波数と、前記比基本周波数と前記比振動源回転周波数とから算出される特定の比周波数とを用いて前記転がり軸受の損傷の種類ごとの前記振動ピーク位置の存在を確認するものであり、
前記特定の比周波数の位置と前記回転同期成分強調波形とを合わせて表示する表示手段を備えることを特徴とする請求項8乃至11の何れかに記載の転がり軸受の異常診断装置。
The bearing damage type determining means includes a specific fundamental frequency corresponding to damage to the rolling bearing, the specific vibration source rotation frequency, and a specific specific frequency calculated from the specific fundamental frequency and the specific vibration source rotation frequency. Is used to confirm the existence of the vibration peak position for each type of damage to the rolling bearing.
The abnormality diagnosis device for a rolling bearing according to any one of claims 8 to 11, further comprising a display means for displaying the position of the specific specific frequency and the rotation-synchronized component-enhanced waveform together.
所定の回転周波数で測定された回転体の振動が前記回転周波数と共に入力されたコンピュータに、請求項1乃至7の何れかに記載の転がり軸受の異常診断方法における回転同期成分強調波形算出ステップと軸受損傷種類判別ステップとを実行させることを特徴とする転がり軸受の異常診断プログラム。 The rotation synchronization component emphasis waveform calculation step and the bearing in the method for diagnosing an abnormality of a rolling bearing according to any one of claims 1 to 7 in which the vibration of a rotating body measured at a predetermined rotation frequency is input together with the rotation frequency. An abnormality diagnosis program for rolling bearings, which comprises performing a damage type determination step.
JP2019061225A 2019-03-27 2019-03-27 Rolling Bearing Abnormality Diagnosis Method, Abnormality Diagnosis Device, Abnormality Diagnosis Program Active JP7146683B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2019061225A JP7146683B2 (en) 2019-03-27 2019-03-27 Rolling Bearing Abnormality Diagnosis Method, Abnormality Diagnosis Device, Abnormality Diagnosis Program

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2019061225A JP7146683B2 (en) 2019-03-27 2019-03-27 Rolling Bearing Abnormality Diagnosis Method, Abnormality Diagnosis Device, Abnormality Diagnosis Program

Publications (2)

Publication Number Publication Date
JP2020159945A true JP2020159945A (en) 2020-10-01
JP7146683B2 JP7146683B2 (en) 2022-10-04

Family

ID=72642984

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2019061225A Active JP7146683B2 (en) 2019-03-27 2019-03-27 Rolling Bearing Abnormality Diagnosis Method, Abnormality Diagnosis Device, Abnormality Diagnosis Program

Country Status (1)

Country Link
JP (1) JP7146683B2 (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112798280A (en) * 2021-02-05 2021-05-14 山东大学 Rolling bearing fault diagnosis method and system
CN113049252A (en) * 2021-03-25 2021-06-29 成都天佑路航轨道交通科技有限公司 Fault detection method for train bearing box
WO2023037882A1 (en) * 2021-09-09 2023-03-16 ローム株式会社 Machine learning device
CN116026598A (en) * 2023-03-30 2023-04-28 山东梁轴科创有限公司 Bearing vibration detecting system
WO2023127248A1 (en) * 2021-12-27 2023-07-06 株式会社神戸製鋼所 Rolling bearing abnormality detection device and rolling bearing abnormality detection method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH09257651A (en) * 1996-03-21 1997-10-03 Matsushita Electric Works Ltd Method for detecting damage of bearing
JP2001021453A (en) * 1999-07-09 2001-01-26 Nsk Ltd Method and device for diagnosing anomaly in bearing
JP2011259624A (en) * 2010-06-09 2011-12-22 Fuji Electric Co Ltd Method and device for removing high frequency electromagnetic vibration component of vibration data of rolling bearing section and method and device for diagnosing rolling bearing of rotary machine
JP2012026481A (en) * 2010-07-21 2012-02-09 Hitachi Constr Mach Co Ltd Method for detecting damage of bearing
JP2017101954A (en) * 2015-11-30 2017-06-08 日本精工株式会社 Mechanical facility evaluation method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH09257651A (en) * 1996-03-21 1997-10-03 Matsushita Electric Works Ltd Method for detecting damage of bearing
JP2001021453A (en) * 1999-07-09 2001-01-26 Nsk Ltd Method and device for diagnosing anomaly in bearing
JP2011259624A (en) * 2010-06-09 2011-12-22 Fuji Electric Co Ltd Method and device for removing high frequency electromagnetic vibration component of vibration data of rolling bearing section and method and device for diagnosing rolling bearing of rotary machine
JP2012026481A (en) * 2010-07-21 2012-02-09 Hitachi Constr Mach Co Ltd Method for detecting damage of bearing
JP2017101954A (en) * 2015-11-30 2017-06-08 日本精工株式会社 Mechanical facility evaluation method

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112798280A (en) * 2021-02-05 2021-05-14 山东大学 Rolling bearing fault diagnosis method and system
CN112798280B (en) * 2021-02-05 2022-01-04 山东大学 Rolling bearing fault diagnosis method and system
CN113049252A (en) * 2021-03-25 2021-06-29 成都天佑路航轨道交通科技有限公司 Fault detection method for train bearing box
WO2023037882A1 (en) * 2021-09-09 2023-03-16 ローム株式会社 Machine learning device
WO2023127248A1 (en) * 2021-12-27 2023-07-06 株式会社神戸製鋼所 Rolling bearing abnormality detection device and rolling bearing abnormality detection method
CN116026598A (en) * 2023-03-30 2023-04-28 山东梁轴科创有限公司 Bearing vibration detecting system

Also Published As

Publication number Publication date
JP7146683B2 (en) 2022-10-04

Similar Documents

Publication Publication Date Title
JP2020159945A (en) Abnormality diagnosis method, abnormality diagnosis device and abnormality diagnosis program for rolling bearing
JP6507297B1 (en) Method and apparatus for diagnosing abnormality of rolling bearing, abnormality diagnosis program
JP5930789B2 (en) Abnormal sound diagnosis device
Su et al. Induction machine condition monitoring using neural network modeling
US11099101B2 (en) Method for estimating bearing fault severity for induction motors
JP4592123B2 (en) Unsteady signal analyzer and medium on which unsteady signal analysis program is recorded
US20150233792A1 (en) Methods and Apparatuses for Defect Diagnosis in a Mechanical System
US6087796A (en) Method and apparatus for determining electric motor speed using vibration and flux
JP2019178973A (en) Abnormality diagnosis method and abnormality diagnosis device of rolling bearing
US20110238329A1 (en) Method and a device for detecting abnormal changes in play in a transmission unit of a movable mechanical unit
JP2008292288A (en) Bearing diagnostic device for reduction gear
CN110346591A (en) Machine rotational speed is determined based on rumble spectrum figure
JP6511573B1 (en) Method and apparatus for diagnosing abnormality of rolling bearing, abnormality diagnosis program
JP2017101954A (en) Mechanical facility evaluation method
US20140236537A1 (en) Method of determining stationary signals for the diagnostics of an electromechanical system
US20200217753A1 (en) Method and apparatus for identifying gear tooth numbers in a gearbox
WO2020255446A1 (en) Operation sound diagnosis system, operation sound diagnosis method, and machine learning device for operation sound diagnosis system
JP7083293B2 (en) Status monitoring method and status monitoring device
JPH11129145A (en) Device and method for diagnosing shape precision of work, and recording medium
JP6742563B1 (en) Driving sound diagnosis system, driving sound diagnosis method, and machine learning device for driving sound diagnosis system
WO2019220542A1 (en) Rolling equipment diagnosis device and diagnosis method
JP7097338B2 (en) Equipment diagnostic equipment, equipment diagnostic methods, and equipment diagnostic programs
EP3929460B1 (en) Anomaly detection system and anomaly detection method
JPH10333743A (en) Method and device for simple abnormality diagnosis of facility accompanied by load variation
Alekseev et al. Data measurement system of compressor units defect diagnosis by vibration value

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20211012

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20220810

TRDD Decision of grant or rejection written
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20220823

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20220921

R150 Certificate of patent or registration of utility model

Ref document number: 7146683

Country of ref document: JP

Free format text: JAPANESE INTERMEDIATE CODE: R150