WO2022061515A1 - 基于人耳听觉特性检测轴承故障的方法和设备 - Google Patents

基于人耳听觉特性检测轴承故障的方法和设备 Download PDF

Info

Publication number
WO2022061515A1
WO2022061515A1 PCT/CN2020/116836 CN2020116836W WO2022061515A1 WO 2022061515 A1 WO2022061515 A1 WO 2022061515A1 CN 2020116836 W CN2020116836 W CN 2020116836W WO 2022061515 A1 WO2022061515 A1 WO 2022061515A1
Authority
WO
WIPO (PCT)
Prior art keywords
psychoacoustic
sound
vibration
bearing
annoyance
Prior art date
Application number
PCT/CN2020/116836
Other languages
English (en)
French (fr)
Inventor
邱志
霍华
王勇
陈士玮
Original Assignee
舍弗勒技术股份两合公司
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 舍弗勒技术股份两合公司 filed Critical 舍弗勒技术股份两合公司
Priority to DE112020007629.6T priority Critical patent/DE112020007629T5/de
Priority to CN202080101967.2A priority patent/CN115917268A/zh
Priority to PCT/CN2020/116836 priority patent/WO2022061515A1/zh
Publication of WO2022061515A1 publication Critical patent/WO2022061515A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
    • G01H1/00—Measuring characteristics of vibrations in solids by using direct conduction to the detector
    • G01H1/003—Measuring characteristics of vibrations in solids by using direct conduction to the detector of rotating machines
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
    • G01H3/00—Measuring characteristics of vibrations by using a detector in a fluid
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts
    • G01M13/04—Bearings
    • G01M13/045—Acoustic or vibration analysis

Definitions

  • the present application relates to the technical field of bearings.
  • the present application specifically relates to methods and apparatus for detecting bearing faults.
  • bearing fault detection is mainly based on simple vibration amplitude measurement and envelope demodulation or resonance demodulation methods.
  • Chinese patent document CN 102840907 B discloses a method for extracting and analyzing vibration signal characteristics of rolling bearings, wherein the method obtains an envelope signal through interpolation resampling technology and performs envelope spectrum analysis on it.
  • the problem encountered in practice is that the basic vibration and sound measurement, such as the measurement of parameters such as amplitude, RMS value, and decibel value, often cannot reflect the true state of bearing operation.
  • the method of demodulating the vibration envelope often requires professional equipment for testing and analysis, and even requires post-processing of the collected data to obtain the result, which is time-consuming and troublesome.
  • the main shaft bearing often uses the human ear to judge the abnormal sound of the bearing, and then decide whether to release it.
  • the accuracy of bearing fault identification is high, but for less experienced inspectors, the accuracy of bearing fault identification is very low.
  • Such subjective judgments are highly volatile and can easily lead to misjudgments.
  • the purpose of the present application is to provide a method and apparatus for detecting bearing failure, by means of the provided method and apparatus, bearing failure can be objectively determined based on human hearing characteristics.
  • the above objects are achieved in one aspect by a method for detecting bearing faults based on human hearing characteristics.
  • the detection method includes:
  • the sound/vibration-psychoacoustic annoyance model can directly adopt a known psychoacoustic annoyance model, such as a general psychoacoustic annoyance model.
  • a known psychoacoustic annoyance model can be optimized for the application of bearing fault detection.
  • the detection method further comprises: selecting a sample bearing and optimizing the sound/vibration-psychoacoustic annoyance model by means of fault characteristics of the sample bearing.
  • a targeted model optimization can be implemented in a simple manner, in particular prior to the inspection of bearing products.
  • the detection methods provided here are more sensitive to bearing faults and thus more effective in detecting bearing faults.
  • the weighting of the different psychoacoustic parameters in the sound/vibration-psychoacoustic annoyance model can advantageously be adjusted by means of the fault characteristics of the sample bearing.
  • step a) the order of step a) relative to step b) and step c) is not limited.
  • a sound/vibration-psychoacoustic annoyance model preferably an optimized sound/vibration-psychoacoustic annoyance model, has been obtained before step b) is performed.
  • step b) the bearing is brought into operation, eg by means of a test platform, in which case acoustic and/or vibration signals are acquired by sensors.
  • the acoustic and/or vibrational signals are time-domain signals.
  • the sound and/or vibration signal is a sound pressure signal and/or a vibration acceleration signal.
  • the psychoacoustic parameter includes at least one of loudness, sharpness, roughness, and waviness.
  • the psychoacoustic parameters involved ie loudness, sharpness, roughness and/or waviness, can be selected according to the specific application.
  • the psychoacoustic parameters include at least loudness, sharpness.
  • psychoacoustic parameters include loudness, sharpness, roughness and waviness.
  • the detected sound and/or vibration signal can be described in a relatively comprehensive manner with psychoacoustic parameters.
  • the detection method further comprises:
  • a psychoacoustic parameter model for at least one of loudness, sharpness, roughness and waviness, preferably loudness and sharpness, particularly preferably all four, respectively, or preferably adopt a known psychoacoustic parameter parametric model.
  • the detected sound and/or vibration signals are respectively input into various psychoacoustic parameter models, thereby obtaining the corresponding characteristic loudness, characteristic sharpness, characteristic roughness and/or characteristic fluctuation degree on the Bark domain.
  • the corresponding total loudness, total sharpness, total roughness and/or total waviness can be input in a subsequent step d) as input variables to a sound/vibration-psychoacoustic annoyance model, preferably an optimized sound/vibration, respectively - in the psychoacoustic annoyance model.
  • the evaluation criteria for bearing failure levels are established by means of a sound/vibration-psychoacoustic annoyance model, in particular an optimized sound/vibration-psychoacoustic annoyance model.
  • the bearing failure class is a division of whether the bearing is operating normally or not. It is also possible that the bearing failure class is a more detailed classification of bearing failures.
  • the evaluation criteria for bearing failure grades thus obtained have the advantage of high objectivity because they are established based on the sound/vibration-psychoacoustic annoyance model.
  • the bearing failure level is judged by comparing the sound/vibration-psychoacoustic annoyance index obtained in step d) with the bearing failure level evaluation standard.
  • the sound/vibration-psychoacoustic annoyance index that is, the VS-PA value (English vibration & sound psychoacoustic annoyance), which is an output variable of the sound/vibration-psychoacoustic annoyance model, is an objective variable based on the auditory characteristics of the human ear.
  • the bearing fault judgment based on the comparison of the VS-PA value and the objective bearing fault grade judgment standard is also objective.
  • the object of the present application is achieved in another aspect by an apparatus for detecting bearing faults based on the auditory characteristics of the human ear.
  • the testing equipment includes:
  • a processor which is used to execute:
  • the inspection device when the bearing is brought into operation, for example by means of a test platform, the inspection device starts working.
  • the detection device here performs most of the steps in the aforementioned method of detecting bearing faults based on human hearing characteristics.
  • a memory unit of the detection device in particular a processor of the detection device, stores a sound/vibration-psychoacoustic annoyance model, in particular an optimized sound/vibration-psychoacoustic annoyance model.
  • the processor may recall a stored sound/vibration-psychoacoustic annoyance model, particularly an optimized sound/vibration-psychoacoustic annoyance model.
  • the apparatus further includes an indicator device, wherein the indicator device is used to output the judged bearing failure level.
  • the indicating device is a display device, so that the sound/vibration-psychoacoustic annoyance index and/or the result of comparison with the bearing failure level evaluation standard, ie, the bearing failure level, can be visually displayed.
  • the indicating device is a sound output device, so that the inspecting personnel can easily obtain the sound/vibration-psychoacoustic annoyance index and/or the result of comparing it with the bearing failure grade evaluation standard through sound, that is, the bearing failure. grade.
  • testing equipment should also include other necessary components, such as power supplies.
  • the inspection device is designed as a hand-held device, so that the bearing can be inspected in a simple and flexible manner and is substantially independent of factors such as the position of the test platform of the bearing.
  • the method and apparatus for detecting bearing faults provided in this paper, it is possible to simulate the characteristics of human hearing to detect the running condition of the bearing, so as to identify the faulty bearing.
  • the method provided here is highly sensitive to bearing faults, and thus can reliably detect bearing faults.
  • the sound/vibration-psychoacoustic annoyance model proposed here is objective and not affected by subjective factors, so it has high reliability and stability. Compared with the traditional spectrum analysis method, the input of bearing size and rotational speed is not required to judge the bearing failure, so the detection equipment can be simplified.
  • FIG. 1 is a block diagram of the basic steps of a method for detecting bearing failure, according to one embodiment.
  • FIG. 2 is a block diagram of steps for calculating psychoacoustic parameters, according to one embodiment.
  • FIG. 3 is a block diagram of steps for optimizing a sound/vibration-psychoacoustic annoyance model, according to one embodiment.
  • FIG. 4 is an illustration of a bearing failure rating criterion according to one embodiment.
  • Figure 1 shows a block diagram of the basic steps of a method for detecting bearing faults according to one embodiment.
  • the detection method according to the present embodiment on the one hand, it is preferable to first a) obtain a sound/vibration-psychoacoustic annoyance model in advance.
  • the sound and/or vibration signals of the bearing are also acquired, in particular by means of sensors b) in the detection device.
  • the sound and/or vibration signal is a sound pressure signal or a vibration acceleration signal in the time domain.
  • Psychoacoustic parameters are then calculated based on the acquired sound and/or vibration signals, in particular by means of a processor c) in the detection device.
  • VS-PA value sound/vibration-psychoacoustic annoyance index
  • the detection method it is possible to simulate the auditory characteristics of the human ear to detect the running condition of the bearing, so as to identify the faulty bearing.
  • the detection method is not affected by subjective factors and has high objectivity, so the detection reliability and stability are high.
  • the input of bearing size and rotational speed is not required to judge the bearing fault, and the detection equipment can be simplified here.
  • Figure 2 shows a block diagram of steps for calculating psychoacoustic parameters, according to one embodiment.
  • the psychoacoustic parameters include four items in total, namely, loudness, sharpness, roughness, and fluctuation.
  • the sound/vibration signal preferably the sound pressure signal or the vibration acceleration signal in the time domain detected by the sensor in the detection device is input into the loudness model, sharpness model, roughness model and waviness model, respectively.
  • the characteristic loudness of the sound/vibration signal in the Bark domain is obtained by means of the loudness model, and then the total loudness can be obtained by integrating the characteristic loudness in the Bark domain.
  • the characteristic sharpness of the sound/vibration signal on the Bark domain is obtained by means of the sharpness model, and then the total sharpness can be obtained by integrating the characteristic sharpness over the Bark domain.
  • the characteristic roughness of the sound/vibration signal on the Bark domain is obtained by means of the roughness model, and then the total roughness can be obtained by integrating the characteristic roughness on the Bark domain.
  • the characteristic fluctuation degree of the sound/vibration signal on the Bark domain can be obtained by means of the fluctuation degree model, and then the total fluctuation degree can be obtained by integrating the characteristic fluctuation degree on the Bark domain.
  • the resulting total loudness, total sharpness, total roughness and/or total waviness can be input in subsequent steps as input variables to a sound/vibration-psychoacoustic annoyance model, preferably an optimized sound/vibration, respectively - in the psychoacoustic annoyance model.
  • Figure 3 shows a block diagram of steps for optimizing a sound/vibration-psychoacoustic annoyance model according to one embodiment.
  • a sample bearing can be selected and an optimization variable can be selected with the aid of the fault characteristics of the sample bearing, and a general or conventional sound/vibration-psychoacoustic annoyance model can be optimized with the aid of a suitable optimization algorithm.
  • the results of the analysis of the fault characteristics of a sample bearing can be used here to adjust the weights of different psychoacoustic parameters in the sound/vibration-psychoacoustic annoyance model.
  • the weights of loudness and sharpness can be correspondingly increased and the weights of roughness and waviness can be correspondingly reduced.
  • the detection method thus provided is more sensitive to bearing faults, and thus can more effectively detect bearing faults.
  • FIG. 3 also shows that it is possible to establish a bearing failure rating criterion by means of an optimized sound/vibration-psychoacoustic annoyance model.
  • the evaluation standard of bearing failure level can be represented by, for example, the relationship between the sound/vibration-psychoacoustic annoyance index (VS-PA value) and the bearing failure level.
  • FIG. 4 is an illustration of a bearing failure rating criterion according to one embodiment.
  • the sound/vibration-psychoacoustic annoyance index (VS-PA value) obtained from a sound/vibration-psychoacoustic annoyance model ranges from 0 to 1 in the range.
  • a possible criterion for evaluating the bearing failure level can be obtained, that is, when the VS-PA value is in the range of 0 to 0.4, the bearing is regarded as normal; when the VS-PA value is in the range of 0 to 0.4 When in the range of 0.4 to 0.8, the bearing is considered to have minor failure; and when the VS-PA value is in the range of 0.8 to 1, the bearing is considered to have serious failure.
  • an apparatus for detecting bearing faults based on human hearing characteristics is provided.
  • the detection device is designed as a hand-held device.
  • the detection device includes a sensor, a processor, an indicating device, and a power supply for supplying power to the sensor, the processor, and the indicating device.
  • the memory unit of the processor of the detection device stores a preferably optimized sound/vibration-psychoacoustic annoyance model.
  • the inspection equipment starts to work.
  • the sound and/or vibration signals of the bearing are detected by means of sensors.
  • psychoacoustic parameters are calculated based on the detected sound and/or vibration signals by means of a processor, and the calculated psychoacoustic parameters are input into a sound/vibration-psychoacoustic annoyance model to obtain a sound/vibration-psychoacoustic annoyance
  • the bearing failure level is judged based on the sound/vibration-psychoacoustic annoyance index.
  • the sound/vibration-psychoacoustic annoyance index and/or the result of comparing it with the bearing fault grade evaluation standard, that is, the bearing fault grade can be output by the indicating device.
  • the indicating device is a display device through which the inspector can read the sound/vibration-psychoacoustic annoyance index and/or the bearing failure level.
  • the indicating device is a sound output device, the inspector can also know the sound/vibration-psychoacoustic annoyance index and/or the bearing failure level, eg by voice.
  • the method and device for detecting bearing failure provided according to the embodiments of this document, it is possible to simulate the characteristics of human hearing to detect the running condition of the bearing, so as to identify the faulty bearing. Since the sound/vibration-psychoacoustic annoyance model is optimized for bearing failures, the methods presented here are highly sensitive to bearing failures and thus can reliably detect bearing failures. The sound/vibration-psychoacoustic annoyance model proposed here is objective and not affected by subjective factors, so it has high reliability and stability. Compared with the traditional spectrum analysis method, the input of bearing size and rotational speed is not required to judge the bearing failure, so the detection equipment can be simplified.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

本申请涉及一种基于人耳听觉特性检测轴承故障的方法和设备。所述方法包括：a）获得声音/振动-心理声学烦恼度模型；b）获取轴承的声音和/或振动信号；c）基于获取的声音和/或振动信号计算心理声学参数；d）将计算得到的心理声学参数输入到获得的声音/振动-心理声学烦恼度模型中得到声音/振动-心理声学烦恼度指数；e）基于得到的声音/振动-心理声学烦恼度指数判断轴承故障等级。所述设备包括传感器以及处理器。

Description

基于人耳听觉特性检测轴承故障的方法和设备 技术领域
本申请涉及轴承技术领域。本申请具体涉及用于检测轴承故障的方法以及设备。
背景技术
机床行业所用的主轴轴承在零件加工精度方面具有关键作用。因此，出厂前的轴承检测对于机床的运行维护尤为重要。主轴轴承在出厂前通常进行基本的振动测试检验，从而检查轴承是否异常。
目前，轴承故障检测主要是基于简单的振动幅值的测量以及包络解调或共振解调的方法来进行的。例如在中国专利文献CN 102840907 B中公开了一种对滚动轴承的振动信号特征提取和分析方法，其中，该方法通过插值重采样技术获得包络信号并且对其作包络谱分析。然而在实际中遇到的问题在于，基础的振动声音测量、例如对振幅、RMS值、分贝值等参数的测量往往不能反映轴承运转的真实状态。通过振动包络解调的方法往往又需要专业的设备进行测试和分析，甚至需要对采集到的数据进行后处理才能得到结果，比较费时和麻烦。
另外，主轴轴承在出厂前也经常借助人耳来对轴承异音进行判别，进而决定是否放行。对于经验丰富的检测人员，轴承故障判别的准确率较高，然而对于经验较少的检测人员，轴承故障判别的准确率很低。这种主观的判断存在较大的波动性，容易导致误判。
发明内容
因此，本申请的目的在于提供一种用于检测轴承故障的方法和设备，借助提供的方法和设备能够基于人耳听觉特性客观地判别轴承故障。
上述目的在一方面通过一种基于人耳听觉特性检测轴承故障的方法实 现。该检测方法包括：
a)获得声音/振动-心理声学烦恼度模型；
b)获取轴承的声音和/或振动信号；
c)基于获取的声音和/或振动信号计算心理声学参数；
d)将计算得到的心理声学参数输入到获得的声音/振动-心理声学烦恼度模型中，得到声音/振动-心理声学烦恼度指数；
e)基于得到的声音/振动-心理声学烦恼度指数判断轴承故障等级。
在步骤a)中，声音/振动-心理声学烦恼度模型可以直接采用已知的心理声学烦恼度模型，例如普适的心理声学烦恼度模型。
备选地且优选地，尤其在步骤a)中，可以针对轴承故障检测的应用对已知的心理声学烦恼度模型进行优化。
在此，特别优选地，检测方法还包括：选取样本轴承并且借助样本轴承的故障特征优化声音/振动-心理声学烦恼度模型。由此可以尤其在轴承产品检测前以简单的方式地实现有针对性的模型优化。在此提供的检测方法对轴承故障更加敏感，进而能够更有效地检测轴承故障。
在此，有利地，可以借助样本轴承的故障特征调整不同心理声学参数在声音/振动-心理声学烦恼度模型中的权重。
在本文的范围中，不限定步骤a)相对步骤b)以及步骤c)的先后顺序。然而，有利的是，在执行步骤b)前已经获得声音/振动-心理声学烦恼度模型、优选经优化的声音/振动-心理声学烦恼度模型。
在步骤b)中，例如借助测试平台使得轴承处于运行状态，在这种情况下由传感器获取声音和/或振动信号。
在一种优选的实施方式中，声音和/或振动信号是时域信号。
在一种优选的实施方式中，声音和/或振动信号是声压信号和/或振动加速度信号。
在一种优选的实施方式中，心理声学参数包括响度、尖锐度、粗糙度和波动度中的至少一项。在这种情况下，能够根据具体的应用选定涉及的心理声学参数，即响度、尖锐度、粗糙度和/或波动度。
有利地，心理声学参数至少包括响度、尖锐度。
特别有利地，心理声学参数包括响度、尖锐度、粗糙度和波动度。在此，能够以心理声学参数比较全面地描述检测到的声音和/或振动信号。
在一种有利的实施方式中，尤其在步骤c)中，检测方法还包括：
c1)获得针对每项的心理声学参数的心理声学参数模型；
c2)将检测到的声音和/或振动信号分别输入到各项心理声学参数模型中，并且分别计算在Bark域上的各项特征心理声学参数；
c3)在Bark域上分别对各项特征心理声学参数进行积分，由此得到各项总心理声学参数，其中，各项总心理声学参数能够用于在步骤d)中输入到声音/振动-心理声学烦恼度模型中。
在此，特别地，首先需要针对响度、尖锐度、粗糙度和波动度中的至少一项、优选响度和尖锐度、特别优选全部四项分别建立心理声学参数模型或优选采用已知的心理声学参数模型。然后，将检测到的声音和/或振动信号分别输入到各项心理声学参数模型中，由此得到在Bark域上的相应的特征响度、特征尖锐度、特征粗糙度和/或特征波动度。通过在Bark域上分别对特征响度、特征尖锐度、特征粗糙度和/或特征波动度进行积分，可以得到相应的总响度、总尖锐度、总粗糙度和/或总波动度。相应的总响度、总尖锐度、总粗糙度和/或总波动度可以在随后的步骤d)中作为输入变量被分别输入到声音/振动-心理声学烦恼度模型、优选经优化的声音/振动-心理声学烦恼度模型中。
在一种优选的实施方式中，借助声音/振动-心理声学烦恼度模型、尤其经优化的声音/振动-心理声学烦恼度模型建立轴承故障等级评判标准。例如，轴承故障等级是对轴承运行状态正常与否的划分。也可行的是，轴承故障等级对轴承故障进行更为细化的分级。由此得到的轴承故障等级评判标准因为是基于声音/振动-心理声学烦恼度模型建立的而具有客观性高的优点。
在此，优选地，尤其在步骤e)中，通过在步骤d)中得到的声音/振动-心理声学烦恼度指数与轴承故障等级评判标准的比对来判断轴承故障等级。在此，作为声音/振动-心理声学烦恼度模型的输出变量的声音/振动-心理声学烦恼度指数、即VS-PA值(英文vibration & sound psychoacoustic  annoyance)是基于人耳听觉特性的客观的变量。在这种情况下，借助VS-PA值与客观的轴承故障等级评判标准的比对进行的轴承故障判断也具有客观性。
本申请的目的在另一方面通过一种基于人耳听觉特性检测轴承故障的设备实现。该检测设备包括：
传感器，其用于检测轴承的声音和/或振动信号；以及
处理器，其用于执行：
基于所检测到的声音和/或振动信号计算心理声学参数，
将计算得到的心理声学参数输入到声音/振动-心理声学烦恼度模型中得到声音/振动-心理声学烦恼度指数，并且
基于声音/振动-心理声学烦恼度指数判断轴承故障等级。
在这种情况下，当例如借助测试平台使得轴承处于运行状态时，检测设备开始工作。检测设备在此执行在前述的基于人耳听觉特性检测轴承故障的方法中的大部分步骤。
在此优选地，检测设备、尤其检测设备的处理器的存储单元存储有声音/振动-心理声学烦恼度模型、尤其经优化的声音/振动-心理声学烦恼度模型。在执行特定步骤时，处理器可以调用所存储的声音/振动-心理声学烦恼度模型、尤其经优化的声音/振动-心理声学烦恼度模型。
在一种优选的实施方式中，该设备还包括指示装置，其中，指示装置用于输出所判断的轴承故障等级。
特别优选地，指示装置是显示装置，从而能够直观地显示声音/振动-心理声学烦恼度指数和/或其与轴承故障等级评判标准进行比对后的结果、即轴承故障等级。
备选地，指示装置是声音输出器件，从而能够简便地通过声音使得检测人员获知声音/振动-心理声学烦恼度指数和/或其与轴承故障等级评判标准进行比对后的结果、即轴承故障等级。
可以想到的是，检测设备还应包括其他必要的零部件，例如电源等。
在一种有利的实施方式中，检测设备构造为手持式设备，从而可以简便灵活地对轴承进行检测，并且基本不受例如轴承的测试平台的位置等因 素的影响。
借助根据本文提供的用于检测轴承故障的方法和设备，能够模拟人耳听觉特性来对轴承运行情况进行检测，从而识别出故障轴承。另外由于针对轴承故障对声音/振动-心理声学烦恼度模型进行优化，在此提供的方法对轴承故障敏感度高，由此可以可靠地检测轴承故障。在此提出的声音/振动-心理声学烦恼度模型具有客观性，不受主观因素影响，因此可靠性及稳定性较高。与传统的频谱分析方法相比，不需要轴承尺寸及转速的输入来判断轴承故障，因而可以简化地构造检测设备。
附图说明
下面将参考附图来描述本申请示例性实施例的特征、优点和技术效果。附图示出：
图1是根据一种实施例的用于检测轴承故障的方法的基本步骤的框图。
图2是根据一种实施例的用于计算心理声学参数的步骤的框图。
图3是根据一种实施例的用于优化声音/振动-心理声学烦恼度模型的步骤的框图。
图4是根据一种实施例的轴承故障等级评判标准的图示。
具体实施方式
图1示出了根据一种实施例的用于检测轴承故障的方法的基本步骤的框图。
如图1所示，在根据本实施例的检测方法中，在一方面，首先优选预先地a)获得声音/振动-心理声学烦恼度模型。
在另一方面，还尤其借助检测设备中的传感器b)获取轴承的声音和/或振动信号。在本实施例中，声音和/或振动信号是在时域上的声压信号或振动加速度信号。然后，尤其借助检测设备中的处理器c)基于获取的声音和/或振动信号计算心理声学参数。
然后，尤其借助检测设备中的处理器，d)将计算得到的心理声学参 数输入到获得的声音/振动-心理声学烦恼度模型中，得到声音/振动-心理声学烦恼度指数(VS-PA值)，进而e)基于得到的声音/振动-心理声学烦恼度指数(VS-PA值)判断轴承故障等级。
借助根据本实施例的检测方法，能够模拟人耳听觉特性来对轴承运行情况进行检测，从而识别出故障轴承。该检测方法不受主观因素影响，客观性高，因此检测的可靠性及稳定性较高。另外，与传统的频谱分析方法相比，不需要轴承尺寸及转速的输入来判断轴承故障，在此可以简化地构造检测设备。
图2示出了根据一种实施例的用于计算心理声学参数的步骤的框图。在本实施例中，心理声学参数共包括四项，即响度、尖锐度、粗糙度和波动度。在此，将由检测设备中的传感器检测的在时域上的声音/振动信号、优选声压信号或振动加速度信号分别输入到响度模型、尖锐度模型、粗糙度模型和波动度模型中。
在此，借助响度模型得到声音/振动信号在Bark域上的特征响度，然后通过在Bark域上对特征响度进行积分可以得到总响度。类似地，借助尖锐度模型得到声音/振动信号在Bark域上的特征尖锐度，然后通过在Bark域上对特征尖锐度进行积分可以得到总尖锐度。类似地，借助粗糙度模型得到声音/振动信号在Bark域上的特征粗糙度，然后通过在Bark域上对特征粗糙度进行积分可以得到总粗糙度。类似地，借助波动度模型得到声音/振动信号在Bark域上的特征波动度，然后通过在Bark域上对特征波动度进行积分可以得到总波动度。由此得到的总响度、总尖锐度、总粗糙度和/或总波动度可以在随后的步骤中作为输入变量被分别输入到声音/振动-心理声学烦恼度模型、优选经优化的声音/振动-心理声学烦恼度模型中。
图3示出了根据一种实施例的用于优化声音/振动-心理声学烦恼度模型的步骤的框图。在本实施例中，例如可以选取样本轴承并且借助样本轴承的故障特征选取优化变量，并且借助合适的优化算法对普适的或者说常规的声音/振动-心理声学烦恼度模型进行优化。例如，在此可以利用对样本轴承的故障特征的分析结果调整不同心理声学参数在声音/振动-心理声学烦恼度模型中的权重。在此，具体地，例如可以相应增大响度和尖锐度 的权重并且相应减小粗糙度和波动度的权重。由此提供的检测方法对轴承故障更加敏感，进而能够更有效地检测轴承故障。
此外，图3还示出，能够借助经优化的声音/振动-心理声学烦恼度模型建立轴承故障等级评判标准。轴承故障等级评判标准例如可以通过声音/振动-心理声学烦恼度指数(VS-PA值)与轴承故障等级的关系表示。
图4是根据一种实施例的轴承故障等级评判标准的图示。
在本实施例中，由声音/振动-心理声学烦恼度模型、优选经优化的声音/振动-心理声学烦恼度模型得到的声音/振动-心理声学烦恼度指数(VS-PA值)在0至1的范围中。在此，根据声音/振动-心理声学烦恼度模型可以获得一种可能的轴承故障等级评判标准，即当VS-PA值在0至0.4的范围中时，轴承视为正常；当VS-PA值在0.4至0.8的范围中时，轴承视为具有轻微故障；并且当VS-PA值在0.8至1的范围中时，轴承视为具有严重故障。
在根据本文的另外的实施例中，提供了一种基于人耳听觉特性检测轴承故障的设备。检测设备在本实施例中构造为手持式设备。
检测设备包括传感器、处理器、指示装置以及向传感器、处理器、指示装置供电的电源。在本实施例中，检测设备的处理器的存储单元存储有优选经优化的声音/振动-心理声学烦恼度模型。
当借助测试平台使得轴承处于运行状态时，检测设备开始工作。首先，借助传感器检测轴承的声音和/或振动信号。随后，借助处理器基于所检测到的声音和/或振动信号计算心理声学参数，并且将计算得到的心理声学参数输入到声音/振动-心理声学烦恼度模型中得到声音/振动-心理声学烦恼度指数，最后基于声音/振动-心理声学烦恼度指数判断轴承故障等级。
在本实施例中，声音/振动-心理声学烦恼度指数和/或其与轴承故障等级评判标准进行比对后的结果、即轴承故障等级可以利用指示装置输出。优选地，指示装置是显示装置，检查人员可以通过显示装置读取声音/振动-心理声学烦恼度指数和/或轴承故障等级。替代地或附加地，指示装置是声音输出器件，检查人员还可以通过例如语音获知声音/振动-心理声学烦恼度指数和/或轴承故障等级。
借助根据本文的实施例提供的用于检测轴承故障的方法和设备，能够模拟人耳听觉特性来对轴承运行情况进行检测，从而识别出故障轴承。由于针对轴承故障对声音/振动-心理声学烦恼度模型进行优化，在此提供的方法对轴承故障敏感度高，由此可以可靠地检测轴承故障。在此提出的声音/振动-心理声学烦恼度模型具有客观性，不受主观因素影响，因此可靠性及稳定性较高。与传统的频谱分析方法相比，不需要轴承尺寸及转速的输入来判断轴承故障，因而可以简化地构造检测设备。
虽然在上述说明中示例性地描述了可能的实施例，但是应该理解到，仍然通过所有已知的和此外技术人员容易想到的技术特征和实施方式的组合存在大量实施例的变化。此外还应该理解到，示例性的实施方式仅仅作为一个例子，这种实施例绝不以任何形式限制本发明的保护范围、应用和构造。通过前述说明更多地是向技术人员提供一种用于转化至少一个示例性实施方式的技术指导，其中，只要不脱离权利要求书的保护范围，便可以进行各种改变，尤其是关于所述部件的功能和结构方面的改变。

Claims (15)

  1. 一种基于人耳听觉特性检测轴承故障的方法，所述方法包括：
    a)获得声音/振动-心理声学烦恼度模型；
    b)获取轴承的声音和/或振动信号；
    c)基于获取的声音和/或振动信号计算心理声学参数；
    d)将计算得到的心理声学参数输入到获得的声音/振动-心理声学烦恼度模型中，得到声音/振动-心理声学烦恼度指数；
    e)基于得到的声音/振动-心理声学烦恼度指数判断轴承故障等级。
  2. 根据权利要求1所述的方法，其特征在于，
    选取样本轴承并且借助所述样本轴承的故障特征优化所述声音/振动-心理声学烦恼度模型。
  3. 根据权利要求2所述的方法，其特征在于，所述方法包括：
    借助所述样本轴承的故障特征调整不同心理声学参数在所述声音/振动-心理声学烦恼度模型中的权重。
  4. 根据前述权利要求中任一项所述的方法，其特征在于，
    所述声音和/或振动信号是时域信号。
  5. 根据前述权利要求中任一项所述的方法，其特征在于，
    所述声音和/或振动信号是声压信号和/或振动加速度信号。
  6. 根据前述权利要求中任一项所述的方法，其特征在于，所述心理声学参数包括响度、尖锐度、粗糙度和波动度中的至少一项。
  7. 根据权利要求6所述的方法，其特征在于，所述心理声学参数至少包括响度、尖锐度。
  8. 根据权利要求7所述的方法，其特征在于，所述心理声学参数包括响度、尖锐度、粗糙度和波动度。
  9. 根据权利要求6至8中任一项所述的方法，其特征在于，所述方法还包括：
    c1)获得针对每项的心理声学参数的心理声学参数模型；
    c2)将检测到的声音和/或振动信号分别输入到各项心理声学参数模型 中，并且分别计算在Bark域上的各项特征心理声学参数；
    c3)在Bark域上分别对各项特征心理声学参数进行积分，由此得到各项总心理声学参数，其中，所述各项总心理声学参数用于在步骤d)中输入到所述声音/振动-心理声学烦恼度模型中。
  10. 根据前述权利要求中任一项所述的方法，其特征在于，
    借助所述声音/振动-心理声学烦恼度模型建立轴承故障等级评判标准。
  11. 根据权利要求10所述的方法，其特征在于，所述方法还包括：
    通过所述声音/振动-心理声学烦恼度指数与所述轴承故障等级评判标准的比对来判断轴承故障等级。
  12. 一种基于人耳听觉特性检测轴承故障的设备，所述设备包括：
    传感器，其用于检测轴承的声音和/或振动信号；以及
    处理器，其用于执行：
    基于所检测到的声音和/或振动信号计算心理声学参数，
    将计算得到的心理声学参数输入到声音/振动-心理声学烦恼度模型中得到声音/振动-心理声学烦恼度指数，并且
    基于所述声音/振动-心理声学烦恼度指数判断轴承故障等级。
  13. 根据权利要求11所述的设备，其特征在于，所述设备还包括指示装置，其中，所述指示装置用于输出所判断的轴承故障等级。
  14. 根据权利要求13所述的设备，其特征在于，所述指示装置是显示装置。
  15. 根据权利要求12至14中任一项所述的设备，其特征在于，所述设备构造为手持式设备。
PCT/CN2020/116836 2020-09-22 2020-09-22 基于人耳听觉特性检测轴承故障的方法和设备 WO2022061515A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
DE112020007629.6T DE112020007629T5 (de) 2020-09-22 2020-09-22 Verfahren und Vorrichtung zur Erfassung von Lagerfehlern auf der Grundlage der Höreigenschaften des menschlichen Ohres
CN202080101967.2A CN115917268A (zh) 2020-09-22 2020-09-22 基于人耳听觉特性检测轴承故障的方法和设备
PCT/CN2020/116836 WO2022061515A1 (zh) 2020-09-22 2020-09-22 基于人耳听觉特性检测轴承故障的方法和设备

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/116836 WO2022061515A1 (zh) 2020-09-22 2020-09-22 基于人耳听觉特性检测轴承故障的方法和设备

Publications (1)

Publication Number Publication Date
WO2022061515A1 true WO2022061515A1 (zh) 2022-03-31

Family

ID=80844702

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/116836 WO2022061515A1 (zh) 2020-09-22 2020-09-22 基于人耳听觉特性检测轴承故障的方法和设备

Country Status (3)

Country Link
CN (1) CN115917268A (zh)
DE (1) DE112020007629T5 (zh)
WO (1) WO2022061515A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114878938A (zh) * 2022-05-06 2022-08-09 常州蓝山机电科技有限公司 适用于导电滑环故障检测的数据处理方法及装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102252749A (zh) * 2010-05-17 2011-11-23 松下电器产业株式会社 品质检查方法和品质检查装置
JP2013221877A (ja) * 2012-04-18 2013-10-28 Panasonic Corp 異常検査方法および異常検査装置
CN103558029A (zh) * 2013-10-22 2014-02-05 重庆建设摩托车股份有限公司 一种发动机异响故障在线诊断系统和诊断方法
CN109668626A (zh) * 2018-12-25 2019-04-23 东莞材料基因高等理工研究院 一种基于人机交互界面的噪声声品质评价方法
CN110688712A (zh) * 2019-10-11 2020-01-14 湖南文理学院 汽车风振噪声声品质客观烦恼度评价指标及其计算方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102840907B (zh) 2012-09-18 2014-05-14 河南省电力公司电力科学研究院 早期故障状态下滚动轴承振动信号特征提取和分析方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102252749A (zh) * 2010-05-17 2011-11-23 松下电器产业株式会社 品质检查方法和品质检查装置
JP2013221877A (ja) * 2012-04-18 2013-10-28 Panasonic Corp 異常検査方法および異常検査装置
CN103558029A (zh) * 2013-10-22 2014-02-05 重庆建设摩托车股份有限公司 一种发动机异响故障在线诊断系统和诊断方法
CN109668626A (zh) * 2018-12-25 2019-04-23 东莞材料基因高等理工研究院 一种基于人机交互界面的噪声声品质评价方法
CN110688712A (zh) * 2019-10-11 2020-01-14 湖南文理学院 汽车风振噪声声品质客观烦恼度评价指标及其计算方法

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
KANE P V, ANDHARE AB: "Application of psychoacoustics for gear fault diagnosis using artificial neural network", APPLICATION OF PSYCHOACOUSTICS FOR GEAR FAULT DIAGNOSIS USING ARTIFICIAL NEURAL NETWORK, vol. 35, no. 3, 1 September 2016 (2016-09-01), pages 207 - 220, XP055918118, DOI: 10.1177/0263092316660915 *
KANE P.V.; ANDHARE A.B.: "Critical evaluation and comparison of psychoacoustics, acoustics and vibration features for gear fault correlation and classification", MEASUREMENT., INSTITUTE OF MEASUREMENT AND CONTROL. LONDON, GB, vol. 154, 13 January 2020 (2020-01-13), GB , XP086066733, ISSN: 0263-2241, DOI: 10.1016/j.measurement.2020.107495 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114878938A (zh) * 2022-05-06 2022-08-09 常州蓝山机电科技有限公司 适用于导电滑环故障检测的数据处理方法及装置

Also Published As

Publication number Publication date
CN115917268A (zh) 2023-04-04
DE112020007629T5 (de) 2023-07-13

Similar Documents

Publication Publication Date Title
Mechefske et al. Fault detection and diagnosis in low speed rolling element bearings Part I: The use of parametric spectra
CA2184649C (en) System and method for machining process characterization using mechanical signature analysis
CN106323452B (zh) 一种设备异音的检测方法及检测装置
US6801864B2 (en) System and method for analyzing vibration signals
KR101178962B1 (ko) 극저속 회전 기계의 이상 진단 방법 및 장치
JP2008292288A (ja) 減速機の軸受診断装置
WO2022061515A1 (zh) 基于人耳听觉特性检测轴承故障的方法和设备
JP3966500B2 (ja) 異音による不良検査装置
JP3743001B2 (ja) 異常判定方法および装置
CA2882930C (en) Rotating stall detection through ratiometric measure of the sub-synchronous band spectrum
JP3020349B2 (ja) 異常検出方法及び装置
CN111122085B (zh) 一种基于功率分布特性的结构装配质量的评价方法
JP2005283227A (ja) 異音検査方法およびその装置
Jiang et al. Rolling bearing quality evaluation based on a morphological filter and a Kolmogorov complexity measure
US11624687B2 (en) Apparatus and method for detecting microcrack using orthogonality analysis of mode shape vector and principal plane in resonance point
JP2006300524A (ja) 異音検査方法
JP3132388B2 (ja) 回転機械の異音検査方法およびその装置
CN108387308A (zh) 变速箱啸叫下线检测方法及其检测系统
JP3473008B2 (ja) 品質判定方法および装置
JP2877901B2 (ja) 騒音判別方法
JPH06300826A (ja) 電動機の検査装置
JP3554445B2 (ja) 異音判定装置
CN112268688B (zh) 一种错误数据识别方法及装置
TWI680370B (zh) 時間序列訊號之分析方法
JP7396327B2 (ja) 鋼管の加工性評価方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20954369

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20954369

Country of ref document: EP

Kind code of ref document: A1