WO2020082217A1 - 机器人故障诊断方法、系统及存储装置 - Google Patents

机器人故障诊断方法、系统及存储装置 Download PDF

Info

Publication number
WO2020082217A1
WO2020082217A1 PCT/CN2018/111261 CN2018111261W WO2020082217A1 WO 2020082217 A1 WO2020082217 A1 WO 2020082217A1 CN 2018111261 W CN2018111261 W CN 2018111261W WO 2020082217 A1 WO2020082217 A1 WO 2020082217A1
Authority
WO
WIPO (PCT)
Prior art keywords
robot
audio information
actual
fault
characteristic frequency
Prior art date
Application number
PCT/CN2018/111261
Other languages
English (en)
French (fr)
Inventor
王鹏
Original Assignee
深圳配天智能技术研究院有限公司
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 深圳配天智能技术研究院有限公司 filed Critical 深圳配天智能技术研究院有限公司
Priority to CN201880087304.2A priority Critical patent/CN111684213A/zh
Priority to PCT/CN2018/111261 priority patent/WO2020082217A1/zh
Publication of WO2020082217A1 publication Critical patent/WO2020082217A1/zh

Links

Classifications

    • FMECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24HEATING; RANGES; VENTILATING
    • F24FAIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00Control or safety arrangements
    • F24F11/89Arrangement or mounting of control or safety devices

Definitions

  • the invention relates to the technical field of robots, and in particular to a robot fault diagnosis method, system and storage device.
  • the movement of the robot usually requires multiple parts to move together to complete. Compared with the general mechanical structure, the robot has many parts and complex movement. In order to ensure the normal operation of the robot, all parts of the robot need to be in a normal state, or the degree of wear and aging is at an acceptable level.
  • the inventor of the present invention found in the practice of the prior art that the method of fault diagnosis of the robot usually requires the removal of the robot, and then each component is checked one by one, for example, by visual inspection, temperature measurement, individual test . On the one hand, this inspection method has complicated procedures and requires a lot of time; on the other hand, since the robot must be removed, it affects the normal working time of the robot.
  • the invention provides a robot fault diagnosis method, system and storage device, which can be used to solve the above problems.
  • a technical solution adopted by the present invention is to provide a robot fault diagnosis method, which includes: obtaining reference audio information when the robot has no fault; making the robot run and recording the actual audio when the robot is running Information; and compare the actual audio information with the reference audio information, and judge the robot's failure according to the comparison result.
  • a robot fault diagnosis system which includes a recorder, a memory, and a processor coupled to each other, wherein the memory stores program instructions, which can be The processor loads and executes a robot fault diagnosis method, which includes: obtaining reference audio information when the robot is free of trouble; running the robot and recording the actual audio information when the robot is operating; and combining the actual audio information with the reference audio information Make a comparison, and judge the robot's failure based on the comparison result.
  • another technical solution adopted by the present invention is to provide a device with a storage function, in which program instructions are stored, and the program instructions can be loaded and execute the aforementioned robot fault diagnosis method.
  • the beneficial effect of the present invention is: by directly recording the actual audio information when the robot is in operation, and comparing it with the reference audio information when the robot is not in trouble, it is possible to easily determine the failure of the robot without removing the robot. Therefore, the present invention can help shorten test maintenance time and improve repair efficiency.
  • FIG. 1 is a schematic flowchart of an embodiment of the robot fault diagnosis method of the present invention
  • FIG. 2 is a schematic flowchart of another embodiment of the robot fault diagnosis method of the present invention.
  • FIG. 3 is a schematic structural view of an embodiment of the robot fault diagnosis system of the present invention.
  • FIG. 4 is a schematic flowchart of another embodiment of the robot fault diagnosis method of the present invention.
  • FIG. 1 is a schematic flowchart of an embodiment of a robot fault diagnosis method of the present invention. The method includes the following steps:
  • Step S101 Obtain reference audio information under the condition that the robot has no fault
  • the reference audio information may be obtained by testing before the robot leaves the factory, and transmitted to the user terminal in an appropriate electronic format, or may be obtained by the user or the manufacturer when the robot is normally used.
  • the robot When the robot is shipped from the factory, it is fully debugged and all parts are almost not worn, or it meets the factory test standards so that it can operate normally, or the various movement indicators of the robot during normal operation (such as displacement, speed, acceleration, etc.) and When other indicators (such as temperature, noise level, etc.) meet the preset standards, it can be considered that the robot is not faulty.
  • the robot can be a household service robot or an industrial application robot, such as a sweeping robot, an articulated robot, a welding robot, etc.
  • the reference audio information may be directly obtained audio information in the time domain, or processed audio information in the time domain, or processed audio information in the frequency domain.
  • Step S102 Run the robot and record the actual audio information when the robot is running.
  • step S102 the robot is first operated. This may include uniform or variable speed operation of a single component of the robot, common or variable speed operation of multiple components, or uniform or variable speed operation of all components.
  • the actual audio information generated when the robot is running can be recorded through a smart terminal, computer, microphone, or other terminal device with audio recording.
  • the robot or some parts of the robot are running at a constant speed, the constant motion of the robot can be recorded.
  • the methods for recording audio information may include but are not limited to the following two types: one is to directly place the device for recording audio information in a suitable location close to the robot so that it can collect audio signals that may be generated by the robot during operation; The other is to add a sensor device to the robot device, and the audio signal is directly obtained by the sensor device. Similar to the reference audio information, the actual audio information may include audio information in the time domain obtained by direct measurement, may be audio information in the processed time domain, or may be audio information in the processed frequency domain.
  • Step S103 Compare the actual audio information with the reference audio information, and determine the fault condition of the robot according to the comparison result.
  • the actual audio information in the time domain may be compared with the reference audio information in the time domain acquired in step S101. For example, by comparing the amplitude, Information such as density or audio signal energy determines whether the actual audio information is the same as or similar to the reference audio information.
  • the acquired actual audio information is audio information in the frequency domain
  • the actual audio information in the frequency domain may be compared with the reference audio information in the frequency domain acquired in step S101. For example, by comparing characteristic frequencies, Information such as the amplitude corresponding to the characteristic frequency, the sideband frequency of the characteristic frequency, or low-frequency noise determines whether the actual audio information is the same as or similar to the reference audio information.
  • a comparison threshold can be set for each relevant information or parameter through empirical values or historical data.
  • the difference between the relevant information or parameters of the actual audio information and the reference audio information is less than the comparison threshold, the actual audio information and the The reference audio information is the same or similar, otherwise it is considered that they are different or not similar.
  • the actual audio information and the reference audio information are not the same or not similar, it is determined that the robot may be malfunctioning.
  • this Inventions can help shorten test maintenance time and improve repair efficiency.
  • the reference audio information acquired in step S101 may be one or more first reference audio information in which one or more components move independently when the robot has no faults. That is to say, when the robot is fault-free, make one or more of these parts move separately, and record the audio information when these parts move independently. For example, you can make these parts move at the rated speed Movement, or variable speed movement from 0 to rated movement speed. Accordingly, in order to facilitate subsequent comparison, in step S102, one or more components of the robot may be separately moved, and one or more first actual audio information when the one or more components are individually moved may be recorded.
  • the actual movement of one or more parts of the robot should be basically the same as the movement when recording the first reference audio of each part of the robot, that is to say, if the first reference audio is recorded, the part A is operated at the rated movement speed At a constant speed, then in step S102, the component A is also moved at a constant speed at the rated motion speed. If the component B performs a variable speed movement from 0 to the rated motion speed when recording the first reference audio, then the component is also made at step S102 B from 0 to rated speed for variable speed movement. In this way, in step 103, the first reference audio information can be compared with the first actual audio information, so as to determine the fault condition of the corresponding component.
  • the reference audio information acquired in step S101 may be the second reference audio information when multiple components move together under the condition that the robot is free of faults. That is to say, when the robot is free of faults, multiple parts of these parts are moved together, and the audio information when these parts move together is recorded.
  • multiple components that have a cooperative relationship with each other or move together to achieve a specific motion of the robot may be selected, such as adjacent mechanical arms, adjacent components driven by gears or gear sets, and so on. Accordingly, in order to facilitate subsequent comparison, in step S102, multiple components of the robot may be moved together, and the second actual audio information of the robot when the multiple components move together is recorded.
  • the actual common movement of multiple parts of the robot should be basically the same as the movement when recording the first reference audio of the common movement of multiple parts of the robot, that is, if parts A and B are recorded when the first reference audio is recorded Both are moving at a constant speed at the rated movement speed, then in step S102, parts A and B are also moved at a constant speed at the rated movement speed. If the rotation of parts A and B is the same (or opposite) when recording the second reference audio, then Correspondingly, in step S102, parts A and B are turned in the same direction (or vice versa). In this way, it is possible to compare the second reference audio information with the second actual audio information in step 103, so as to determine the fault condition of each of the corresponding multiple components and / or the cooperation relationship and connection structure between them Failure conditions, etc.
  • FIG. 2 is a schematic flowchart of another embodiment of the robot fault diagnosis method of the present invention.
  • the method includes the following steps:
  • Step S201 Obtain the reference audio frequency spectrum when the robot has no faults.
  • the reference audio frequency spectrum is reference audio information in the frequency domain, which may include information such as the characteristic frequency and the amplitude corresponding to each characteristic frequency.
  • Step S202 Run the robot, and record the audio signal in the time domain when the robot is running.
  • Step S203 Remove the environmental noise in the audio signal, and normalize the audio signal after removing the environmental noise to update the audio signal.
  • the recorded audio signal may include only the sound emitted when the robot is running, or may include other sounds of the surrounding environment.
  • the environmental noise may be removed first.
  • the main microphone records the audio signals of the robot's operation (including the environmental noise of the robot's environment).
  • the auxiliary microphone only records the environmental noise. To superimpose, that is, the environmental noise in the audio signal recorded by the main microphone can be removed.
  • the audio signal can also be normalized (that is, the amplitude at the point of maximum amplitude of the recorded audio signal is adjusted to 1, and at other points Amplitude is proportionally scaled) to obtain the updated audio signal.
  • other effective processing can also be performed, such as low-pass or high-pass filtering.
  • Step S204 Fourier transform the audio signal to obtain the actual audio frequency spectrum in the frequency domain when the robot is running.
  • sampling frequency for example, 256Hz, 512Hz, or 1024Hz, etc.
  • FFT fast Fourier transform
  • Step S205 Compare the reference audio spectrum with the actual audio spectrum to determine whether the actual audio spectrum and the reference audio spectrum are the same or similar.
  • the comparison method may be: comparing the actual characteristic frequency in the actual audio spectrum and the amplitude corresponding to each actual characteristic frequency with the reference characteristic frequency in the reference audio spectrum and the amplitude corresponding to each reference characteristic frequency, and comparing the actual audio spectrum Whether the actual characteristic frequency in A and the amplitude corresponding to each actual characteristic frequency are the same or similar to the reference characteristic frequency in the reference audio spectrum and the amplitude corresponding to each reference characteristic frequency. Considering that there are errors in the actual situation, the above judgment can be made by setting preset range values for relevant information or parameters through experience values or historical data.
  • a preset range value can be set for the characteristic frequency or the amplitude corresponding to the characteristic frequency, for example, the first preset range can be set to 0.5 Hz, 1 Hz, or 2 Hz, etc.
  • the first preset range can be set to 0.5 Hz, 1 Hz, or 2 Hz, etc.
  • the second preset range may be set to the characteristic frequency corresponding to the reference audio spectrum 5%, 10%, or 15% of the amplitude, etc.
  • the difference between the amplitude at a characteristic frequency of the actual audio spectrum and the amplitude at the corresponding characteristic frequency of the reference audio spectrum is less than the second preset range, they are considered to be The same or similar.
  • Step S206 Determine the fault condition of the robot according to the comparison result.
  • step S204 If the comparison result obtained in step S204 indicates that the actual audio frequency spectrum is not the same as or not close to the reference audio frequency spectrum, then it is determined that the robot has a fault, and then step S207 is continued. If the comparison result shows that the actual audio frequency spectrum is the same as or similar to the reference audio frequency spectrum, then it is judged that the robot has no malfunction.
  • S207 According to the actual characteristic frequency and the reference characteristic frequency, it is determined that the fault of the robot exists in the parts, the connection structure, or the installation method.
  • the characteristic frequency reflects the sound of a specific frequency corresponding to the parts and connection structure, for example, the characteristic frequency included in the sound of a motor running at 600r / min may be 10Hz and its harmonic frequency (for example, 20Hz, 30Hz, etc.)
  • the characteristic frequency contained in the sound of the gear transmission connection structure may be the frequency of gear meshing and its harmonic frequency.
  • the characteristic frequency can also reflect whether there is a problem in the installation method between multiple components. For example, if the components of the two common rotation shafts are poorly aligned, the sound emitted when they move together may appear to correspond to their speed The characteristic frequency of 2 times the harmonics of the harmonics.
  • the characteristic frequency of the higher harmonics corresponding to their rotation speed or the characteristic frequency of low-frequency noise may appear in the sound produced by their joint motion. Therefore, in the case where it is determined in step S206 that the robot has a fault, it can be determined whether the fault of the robot exists in the parts, the connection structure, or the installation method according to the actual characteristic frequency and the reference characteristic frequency. If the reference audio frequency spectrum and the actual audio frequency spectrum are both the frequency spectrum when a single component of the robot moves, then one or more characteristic frequencies in the frequency spectrum correspond to the component or the installation method of the component.
  • the different characteristic frequencies in the frequency spectrum correspond to different parts, connection structures, or different parts installation methods. Finding the characteristic frequency in the actual audio spectrum that is different from the reference audio spectrum (added characteristic frequency or characteristic frequency with changed amplitude) can determine whether the robot's fault exists in parts, connection structure or installation method.
  • the removal robot can be detected one by one, and the fault condition of the robot can be determined and the fault point can be determined under the fault condition, thereby providing effective help for the robot's fault handling.
  • a solution to the failure of the component, the connection structure, or the installation method may also be provided.
  • a list of fault handling methods based on historical experience or fault diagnosis theory can be pre-stored in the database of the fault diagnosis system, where corresponding repairs are provided for various types of faults or faults of various parts, connection structures, and installation methods. Suggest. When the fault point of the robot is determined in step S207, the fault or the fault point is retrieved in the fault handling method list accordingly, and the maintenance suggestions recorded in the list are provided to the user.
  • the solution when it is determined that the rotation axis 1 of the robot is defective, the solution may be to replace the rotation axis, or when it is determined that the connection 2 of the robot is defective, the solution may be to check whether the screws at the connection are tightened, etc. .
  • the foregoing recording of the actual audio information during the operation of the robot may use an audio recorder with a dynamic range of 0-40KHZ or 0-50KHZ. In this way, sounds that cannot be sensed by human hearing can be fully recorded, thereby avoiding the omission of robot failure points.
  • FIG. 3 is a schematic structural diagram of a robot fault diagnosis system provided by the present invention.
  • the robot fault diagnosis system includes a recorder 301, a processor 302, and a memory 303 coupled to each other, and these components are interconnected through a communication bus 304.
  • the recorder 301 is used to record actual audio information.
  • an audio recording device with a dynamic range greater than 0 to 40 kHz can be used.
  • the processor 302 may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the system to perform desired functions, such as in other embodiments.
  • display components or voice components can also be added to better inform the user of the fault.
  • the memory 303 stores program instructions, and the processor 302 can run the program instructions to implement the robot fault diagnosis method of the present invention described above.
  • the memory 303 may also store various application programs and various data, such as various data used and / or generated by the application programs. It can be understood that, in other embodiments, the memory 303 may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory.
  • the volatile memory may include, for example, random access memory (RAM) and / or cache memory.
  • the non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, and the like.
  • the system has a simple structure and relies on audio information to diagnose faults, which can bring convenience to fault diagnosis and improve the efficiency of robot or maintenance operations.
  • FIG. 4 a schematic flowchart of another embodiment of the robot fault diagnosis method of the present invention.
  • the method includes the following steps:
  • the user collects the abnormal sound audio and environmental noise of the actual operation of the robot through a portable microphone device such as a notebook, mobile phone, etc., uses the set filter to remove the influence of environmental noise, and further removes the influence of the difference in the measured distance through normalization, and then The fast Fourier transform (FFT) of the audio can establish the spectrum information database under the running state of the robot.
  • a portable microphone device such as a notebook, mobile phone, etc.
  • the spectrogram of each axis running at different speeds by comparing the characteristic frequency of the actual audio information collected with the characteristic frequency interval of each component under normal conditions, if the frequency domain of the actual audio information collected
  • the characteristic frequency is the same as or close to the characteristic frequency during normal operation of the component, there is no problem with the component corresponding to the actual characteristic frequency; otherwise, the component has a fault.
  • the faulty component can be found according to the frequency of the problem, and a solution can be found based on the fault.
  • the functions described in the above embodiments are implemented in software and sold or used as independent products, they can be stored in a device with a storage function, that is, the present application also provides a storage device that stores a program.
  • the program data in the storage device can be executed to implement the fault diagnosis method of the robot in the above embodiment, and the storage device includes but is not limited to a USB disk, an optical disk, a server, or a hard disk.

Landscapes

  • Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Mechanical Engineering (AREA)
  • General Engineering & Computer Science (AREA)
  • Manipulator (AREA)

Abstract

一种机器人故障诊断方法、系统及存储装置,涉及机器人领域,该方法包括:获取机器人无故障情况下的参考音频信息(S101);使机器人运转,并记录机器人运转时的实际音频信息(S102);将实际音频信息与参考音频信息进行对比,根据对比结果判断机器人的故障情况(S103)。通过直接记录机器人运转时的实际音频信息,并将其与机器人无故障情况下的音频信息进行对比,可以在不拆除机器人的情况下方便地判断机器人的故障情况,帮助缩短测试维护时间,提高维修效率。

Description

机器人故障诊断方法、系统及存储装置
【技术领域】
本发明涉及机器人技术领域,特别是涉及机器人故障诊断方法、系统及存储装置。
【背景技术】
机器人的运动通常需要多个零部件共同运动来配合完成,与一般的机械结构相比,机器人的零件多、运动复杂。为了保证机器人的正常工作,需要机器人的各零部件都处在正常状态,或者磨损老化的程度处于可接收的水平。
本发明的发明人在对现有技术的实践过程中发现,对机器人进行故障诊断的方法通常需要将机器人拆除,而后对各零部件逐一检查,例如通过目视检查、温度测量、单独测试等方式。这样的检查方式一方面工序复杂,需要耗费较多的时间;另一方面,由于必须将机器人拆除,因此对机器人的正常工作时间造成影响。
综上所述,有必要提供一种机器人故障诊断方法以解决上述问题。
【发明内容】
本发明提供一种机器人故障诊断方法、系统及存储装置,可用于解决上述问题。
为了解决上述技术问题,本发明采用的一个技术方案是:提供一种机器人故障诊断方法,该方法包括:获取机器人无故障情况下的参考音频信息;使机器人运转,并记录机器人运转时的实际音频信息;以及将实际音频信息与参考音频信息进行对比,根据对比结果判断机器人的故障情况。
为解决上述技术问题,本发明采用的另一个技术方案是:提供一种机器人故障诊断系统,其包括互相耦合的记录器、存储器、和处理器,其中存储器存储有程序指令,该程序指令可被处理器加载并执行一种机器人故障诊断方法,该方法包括:获取机器人无故障情况下的参考音频信息;使机器人运转,并记录机器人运转时的实际音频信息;以及将实际音频信息与参考音频信息进行对比,根据对比结果判断机器人的故障情况。
为解决上述技术问题,本发明采用的另一个技术方案是:提供一种具有存储功能的装置,其中存储有程序指令,该程序指令可被加载并执行前述机器人故障诊断方法。
本发明的有益效果是:通过直接记录机器人运转时的实际音频信息,并将其与机器人无故障情况下的参考音频信息进行对比,可以在不拆除机器人的情况下方便地判断机器人的故障情况,因此,本发明可以帮助缩短测试维护时间,提高维修效率。
【附图说明】
图1是本发明机器人故障诊断方法的一实施例的流程示意图;
图2是本发明机器人故障诊断方法的又一实施例的流程示意图;
图3是本发明机器人故障诊断系统一实施例的结构示意图;
图4是本发明机器人故障诊断方法的另一实施例的流程示意图。
【具体实施方式】
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅是本发明的一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
如图1所示,图1是本发明机器人故障诊断方法的一实施例的流程示意图。该方法包括以下步骤:
步骤S101:获取机器人无故障情况下的参考音频信息;
在步骤S101中,参考音频信息可以通过在机器人出厂前测试得到,以适当的电子格式传送到用户端,或者,也可以由用户或厂家在机器人正常使用的情况下测试得到。当机器人出厂时经过充分调试并且所有零部件都几乎没有磨损,或者达到出厂的测试标准从而可以正常运转,又或者机器人在正常运转时的各项运动指标(例如,位移、速度、加速度等)和其他指标(例如温度、噪声水平等)符合预设的标准时,可以认为机器人是没有故障的。其中机器人可以是家庭服务类机器人或者工业运用类机器人,例如扫地机器人、关节型机器人,焊接机器人等。参考音频信息可以是直接获得的时域上的音频信息,也可以是经过处理的时域上的音频信息,也可以是经过处理后频域上的音频信息。
步骤S102:使机器人运转,并记录机器人运转时的实际音频信息。
在步骤S102中,首先使机器人运转,这可以包括使机器人的单个零部件的匀速运转或变速运转,使多个零部件共同匀速运转或变速运转,或者使所有零部件匀速运转或变速运转,在机器人运转时,可以通过智能终端、电脑、麦克风或者其它具有音频记录的终端设备来记录机器人运转时产生的实际音频信息,当机器人或者机器人的部分零部件匀速运转时,可以记录其匀速运动情况下的实际音频信息,而当机器人或机器人的部分零部件变速运转时,可以记录其在整个速度范围内的音频信息、其中部分(一个或多个)速度范围内的音频信息,或者,也可以只记录特定运转速度(例如,额定转速)下的音频信息。记录音频信息的方式可包括但不限于以下两种:一种是直接将用于记录音频信息的设备放置在靠近机器人的合适位置,使其可以采集到机器人在运行过程中可能产生的音频信号;另一种是在机器人设备上增加传感器装置,由传感器装置直接获取音频信号。与参考音频信息类似,实际音频信息可包含直接测量获得的时域上的音频信息,可以是经过处理的时域上的音频信息,也可以是经过处理后频域上的音频信息。
步骤S103:将实际音频信息与参考音频信息进行对比,根据对比结果判断机器人的故障情况。
当获取的实际音频信息是时域上的音频信息时,可以将该时域上的实际音频信息与步骤S101中获取的时域上的参考音频信息进行对比,例如,可以通过比对幅值、密集程度或者音频信号能量等信息来判断实际音频信息与参考音频信息是否相同或者相近。当获取的实际音频信息是频域上的音频信息时,可以将该频域上的实际音频信息与步骤S101中获取的频域上的参考音频信息进行对比,例如,可以通过比对特征频率、特征频率对应的幅值、特征频率的边带频率或者低频噪声等信息来判断实际音频信息与参考音频信息是否相同或者相近。此外,可以通过经验值或者历史数据为各相关信息或参数设定一比较阈值,当实际音频信息和参考音频信息各相关信息或参数之间的差小于该比较阈值时,可认为实际音频信息和参考音频信息是相同或相近的,反之则认为它们不相同或不相近。当根据比较结果认为实际音频信息和参考音频信息是相同或相近时,可相应确定机器人不存在故障,反之,当实际音频信息和参考音频信息不相同或不相近时,则确定机器人可能存在故障。
在本实施例中,通过直接记录机器人运转的实际音频信息,并将其与机器人无故障情况下的音频信息进行对比,可以在不拆除机器人的情况下方便地判断机器人的故障情况,因此,本发明可以帮助缩短测试维护时间,提高维修效率。
在一些实施例中,步骤S101中获取的参考音频信息可以是在机器人无故障情况下,一个或多个零部件单独运动的一个或多个第一参考音频信息。也就是说,在机器人无故障的情况下,使这些零部件中的一个或多个分别单独运动,并记录这些零部件单独运动时的音频信息,例如,可以使这些零部件分别以额定运动速度运动,或者从0至额定运动速度的范围内变速运动。相应地,为了方便后续对比,在步骤S102中,可以使机器人的一个或多个零部件分别单独运动,并记录一个或多个零部件单独运动时的一个或多个第一实际音频信息。机器人的一个或多个零部件实际的运动应该与记录机器人各零部件单独运动的第一参考音频时的运动基本相同,也就是说,若记录第一参考音频时零部件A以额定运动速度作匀速运动,那么在步骤S102中也使零部件A以额定运动速度作匀速运动,若记录第一参考音频时零部件B从0至额定运动速度作变速运动,那么在步骤S102中也使零部件B从0至额定运动速度作变速运动。这样,就可以在步骤103中将第一参考音频信息与第一实际音频信息进行对比,从而判断对应的零部件的故障情况。
在另一些实施例中,步骤S101中获取的参考音频信息可以是在机器人无故障情况下,多个零部件共同运动时的第二参考音频信息。也就是说,在机器人无故障的情况下,使这些零部件中的多个零部件共同运动,并记录这些零部件共同运动时的音频信息。在本实施例中,可以选择相互之间有配合关系或者共同运动以实现机器人的特定运动的多个零部件,例如相邻的机械臂、由齿轮或齿轮组传动的相邻部件等。相应地,为了方便后续对比,在步骤S102中,可以使机器人的多个零部件共同运动,并记录机器人在多个零部件共同运动时的第二实际音频信息。类似地,机器人的多个零部件实际的共同运动应该与记录机器人多个零部件共同运动的第一参考音频时的运动基本相同,也就是说,若记录第一参考音频时零部件A和B均以额定运动速度作匀速运动,那么在步骤S102中也使零部件A和B以额定运动速度作匀速运动,若记录第二参考音频时零部件A和B的转向相同(或相反),那么相应地在步骤S102中也使零部件A和B的转向相同(或相反)。这样,就可以在步骤103中将第二参考音频信息与第二实际音频信息进行对比,从而判断对应的多个零部件中的每一个的故障情况和/或它们之间的配合关系、连接结构的故障情况等。
请参阅图2,图2是本发明机器人故障诊断方法的另一实施例的流程示意图。该方法包括以下步骤:
步骤S201:获取机器人无故障情况下的参考音频频谱。
其中,该参考音频频谱为频域上的参考音频信息,其中可以包括特征频率和各特征频率对应的幅值等信息。
步骤S202:使机器人运转,并记录机器人运转时在时域上的音频信号。
步骤S203:将音频信号中的环境噪声去除,并对去除环境噪声后的音频信号进行归一化处理,以更新该音频信号。
记录的音频信号可以仅包括机器人运转时发出的声音,或者也可以包括周围环境的其他声音。当其中包括周围的环境噪声时,为了提高故障诊断的精度,可以首先将环境噪声去除。例如,可以是设置主副两个麦克风,主麦克风记录机器人运转的音频信号(其中也包括机器人所处环境的环境噪声),副麦克风仅记录环境噪声,通过将主、副麦克风记录的音频信号反向叠加,即可以将主麦克风记录的音频信号中的环境噪声去除。为了消除声音采集装置与机器人之间的距离对测量结果的影响,还可以对音频信号做归一化处理(即将记录的音频信号中幅值最大点处的幅值调整为1,其它点处的幅值等比例进行缩放),从而得到更新后的音频信号。当然为了使音频信息的数据更加有利于观察,还可进行其它有效的处理,例如低通或者高通滤波等。
步骤S204:将该音频信号进行傅里叶变换,得到机器人运转时在频域上的实际音频频谱。
使用适当的采样频率(例如,256Hz、 512Hz或者1024Hz等)对音频信号进行采样,从而实现音频信号的模数转换。进而对采样得到的信号进行傅里叶变换(例如快速傅里叶变换FFT),得到机器人运转时在频域上的实际音频频谱。实际音频频谱中包含了机器人运转时发出的声音的实际特征频率和各实际特征频率对应的幅值。
步骤S205:将参考音频频谱和实际音频频谱对比,判断实际音频频谱和参考音频频谱是否相同或相近。
其中,对比方法可以是:将实际音频频谱中的实际特征频率和各实际特征频率对应的幅值与参考音频频谱中的参考特征频率和各参考特征频率对应的幅值进行对比,比较实际音频频谱中的实际特征频率和各实际特征频率对应的幅值与参考音频频谱中的参考特征频率和各参考特征频率对应的幅值是否相同或者相近。考虑到实际情况下存在误差,可以通过经验值或者历史数据为相关信息或参数设定预设范围值来进行上述判断。在本实施例中可以为特征频率或特征频率对应的幅值设定预设范围值,例如,可以将第一预设范围设定为0.5Hz、1Hz或2Hz等,当实际音频频谱和参考音频频谱的特征频率的差小于第一预设范围值时,可认为实际音频频谱和参考音频频谱是相同或相近的,或者,可以将第二预设范围设定为参考音频频谱对应特征频率处的幅值的5%、10%、或15%等,当实际音频频谱的一特征频率处的幅值与参考音频频谱对应特征频率处的幅值的差小于第二预设范围时,认为它们是相同或相近的。
应当注意,在另一些实施例中,也可以判断实际音频频谱与参考音频频谱相比是否出现其它新的特征频率,如是,则说明它们二者不相同。
步骤S206:根据对比结果判断所述机器人的故障情况。
如果步骤S204中得到的对比结果表明实际音频频谱与参考音频频谱不相同或不相近,那么判断该机器人存在故障,此时继续执行步骤S207。如果对比结果表明实际音频频谱与参考音频频谱相同或相近,那么判断该机器人不存在故障。
S207:根据实际特征频率和参考特征频率,确定机器人的故障存在于零部件、连接结构或者安装方式上。
特征频率反映的是零部件和连接结构等对应的特定频率的声音,例如,以600r/min运转的电机的声音中包含的特征频率可能为10Hz及其谐波频率(例如,20Hz、30Hz等),以齿轮传动的连接结构的声音中包含的特征频率可能为齿轮啮合的频率及其谐波频率。此外,特征频率还可以反映多个部件之间的安装方式中是否存在问题,例如,若两共旋转轴的部件安装时对中度较差,它们共同运动时发出的声音中可能出现对应其转速的2倍谐波的特征频率,若两连接在一起的部件安装存在松动,那么它们共同运动发出的声音中可能出现对应其转速的高次谐波的特征频率或者类似低频噪声的特征频率。因此,在步骤S206判断机器人存在故障的情况下,可以根据实际特征频率和参考特征频率来确定机器人的故障存在于零部件、连接结构还是安装方式上。若参考音频频谱和实际音频频谱均为机器人单个零部件运动时的频谱,则该频谱中的一个或多个特征频率就对应该零部件或者该零部件的安装方式。若参考音频频谱和实际音频频谱均为机器人多个零部件共同运动时的频谱,则该频谱中的不同特征频率就对应不同零部件、连接结构或者不同零部件的安装方式。找出实际音频频谱中与参考音频频谱不同的特征频率(新增特征频率或者幅值改变的特征频率),就可以确定机器人的故障存在于零部件、连接结构还是安装方式上。
以上实施例中,通过将机器人的实际音频频谱与参考音频频谱进行对比,并根据对比结果判断机器人的故障情况,进而可以利用实际特征频率和参考特征频率确定机器人的故障点。因此,本实施例可以在避免了拆除机器人逐一检测,可以判断故障情况下判断机器人的故障情况,确定故障点,从而为机器人的故障处理提供有效的帮助。
在一些实施例中,还可根据实际音频信息和参考音频信息的对比结果,提供所述零部件、所述连接结构或者所述安装方式的故障的解决方案。具体地,可以在故障诊断系统的数据库中预存根据历史经验或者故障诊断理论建立的故障处置方法清单,其中对各类型的故障或者各零部件、连接结构、安装方式的故障都提供了相应地维修建议。当在步骤S207中确定了机器人的故障点后,相应地在故障处置方法清单中检索该故障或者该故障点,并将清单中记录的维修建议提供给用户。例如:当确定是机器人的转动轴1存在故障时,提供方案可以是更换该转动轴,或者当确定是机器人的连接处2存在故障时,提供方案可以是检查该连接处的螺丝是否紧固等。
可选地,前述记录所述机器人运转时的实际音频信息可以使用动态范围为0-40KHZ或0-50KHZ等的音频记录仪。这样,可以充分记录人的听觉无法感知的声音,从而避免遗漏机器人的故障点。
请参阅图3所示,图3是本发明提供的机器人故障诊断系统的结构示意图。该机器人故障诊断系统包括包括互相耦合的记录器301、处理器302和存储器303,这些组件通过通信总线304互连。
记录器301用于记录实际的音频信息,例如,可以使用动态范围大于0~40kHz的音频记录设备。
处理器302可以是中央处理单元(CPU)或者具有数据处理能力和/或指令执行能力的其它形式的处理单元,并且可以控制所述系统中的其它组件以执行期望的功能,例如在其它实施例中,还可以添加显示组件或者语音组件等,将故障情况更好地向用户端提示。
存储器303保存有程序指令,处理器302可以运行所述程序指令,以实现上文所述的本发明的机器人故障诊断方法。存储器303中还可以存储各种应用程序和各种数据,例如所述应用程序使用和/或产生的各种数据等。可以理解,在其它一些实施例中,存储器303可以包括各种形式的计算机可读存储介质,例如易失性存储器和/或非易失性存储器。所述易失性存储器例如可以包括随机存取存储器(RAM)和/或高速缓冲存储器(cache)等。所述非易失性存储器例如可以包括只读存储器(ROM)、硬盘、闪存等。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
本系统结构简单,依靠音频信息诊断故障,可以为故障诊断带来方便,提高机器人或者维修作业效率。
如图4所示,本发明机器人故障诊断方法的另一实施例的流程示意图。该方法包括以下步骤:
用户通过笔记本、手机等便携麦克风设备采集机器人实际运转的异响音频和环境噪声,使用设定的滤波器去除环境噪声影响,进一步通过归一化处理去除测量距离差异影响,然后对采集的异响音频进行快速傅里叶变换(FFT),就可建立机器人运转状态下的频谱信息库。比如在本实施例中各轴在不同速率下运转时的频谱图,通过将采集到的实际音频信息的特征频率与正常情况下各部件的特征频率区间对比,如果采集的实际音频信息的频域特征频率在与该部件正常运转时的特征频率相同或相近,则该实际特征频率对应的零部件没有问题,否则,该零部件存在故障。进而可以根据问题频率点找出该故障零部件,并依据该故障寻找处理方案。
上述实施例所述功能如果以软件形式实现并作为独立的产品销售或使用时,可存储在一个具有存储功能的装置中,即,本申请还提供一种存储有程序的存储装置。存储装置中程序数据能够被执行以实现上述实施例中的机器人的故障诊断方法,该存储装置包括但不限于U盘、光盘、服务器或者硬盘等。
以上所述仅为本发明的实施方式,并非因此限制本发明的专利范围,凡是利用本发明说明书及附图内容所作的等效结构或等效流程变换,或直接或间接运用在其他相关的技术领域,均同理包括在本发明的专利保护范围内。

Claims (19)

  1. 一种机器人的故障诊断方法,其特征在于,包括:
    获取机器人无故障情况下的参考音频信息;
    使所述机器人运转,并记录所述机器人运转时的实际音频信息;以及
    将所述实际音频信息与所述参考音频信息进行对比,根据对比结果判断所述机器人的故障情况。
  2. 根据权利要求1所述的故障诊断方法,其特征在于:
    所述获取机器人无故障情况下的参考音频信息的步骤包括:分别获取所述机器人无故障情况下,一个或多个零部件单独运动的一个或多个第一参考音频信息;
    所述使所述机器人运转,并记录所述机器人运转时的实际音频信息的步骤包括:使所述机器人的所述一个或多个零部件分别单独运动,并记录所述一个或多个零部件单独运动时的一个或多个第一实际音频信息;
    所述将所述实际音频信息与所述参考音频信息进行对比的步骤包括:将所述机器人的所述一个或多个第一实际音频信息与所述一个或多个第一参考音频信息进行对比。
  3. 根据权利要求1所述的故障诊断方法,其特征在于:
    所述获取机器人无故障情况下的参考音频信息的步骤包括:获取所述机器人无故障情况下,在多个零部件共同运动时的第二参考音频信息;
    所述使所述机器人运转,并记录所述机器人运转时的实际音频信息的步骤包括:使所述机器人的所述多个零部件共同运动,并记录所述机器人在所述多个零部件共同运动时的第二实际音频信息;
    所述将所述实际音频信息与所述参考音频信息进行对比的步骤包括:将所述机器人的所述第二实际音频信息与所述第二参考音频信息进行对比。
  4. 根据权利要求1所述的故障诊断方法,其特征在于:
    所述参考音频信息包括所述机器人无故障情况下的参考音频频谱;
    所述记录所述机器人运转时的实际音频信息的步骤包括:
    记录所述机器人运转时在时域上的音频信号;
    将所述音频信号进行傅里叶变换,得到所述机器人运转时在频域上的实际音频频谱;
    所述将所述实际音频信息与所述参考音频信息进行对比的步骤包括:
    将所述实际音频频谱中的实际特征频率和各所述实际特征频率对应的幅值与所述参考音频频谱中的参考特征频率和各参考特征频率对应的幅值进行对比。
  5. 根据权利要求4所述的故障诊断方法,其特征在于,所述根据对比结果判断所述机器人的故障情况的步骤包括:
    当所述实际特征频率与所述参考特征频率不相同,或者所述实际特征频率与所述参考特征频率的差值大于第一预设范围,则判断所述机器人存在故障;或者
    当所述实际特征频率对应的幅值与所述参考特征频率对应的幅值的差值大于第二预设范围,则判断所述机器人存在故障。
  6. 根据权利要求5所述的故障诊断方法,其特征在于,还包括:
    当判断所述机器人存在故障时,根据所述实际特征频率和所述参考特征频率,确定所述机器人的故障存在于零部件、连接结构或者安装方式上。
  7. 根据权利要求6所述的故障诊断方法,其特征在于,在所述确定所述机器人的故障存在于所述零部件、所述连接结构或者所述安装方式上的步骤之后,还包括:
    根据所述实际音频信息和所述参考音频信息的对比结果,提供所述零部件、所述连接结构或者所述安装方式的故障的解决方案。
  8. 根据权利要求4所述的故障诊断方法,其特征在于,在所述将所述音频信号进行傅里叶变换的步骤之前,还包括:
    将所述音频信号中的环境噪声去除;以及
    对去除环境噪声后的所述音频信号进行归一化处理,以更新所述音频信号。
  9. 根据权利要求1所述的故障诊断方法,其特征在于,所述记录所述机器人运转时的实际音频信息的步骤包括:
    使用动态范围大于0~40kHz的音频记录仪记录所述机器人运转时的所述实际音频信息。
  10. 一种机器人的故障诊断系统,其特征在于,包括互相耦合的记录器、存储器、处理器,所述记录器用于记录机器人运转时的实际音频信息,所述存储器存储有程序指令,所述程序指令可被处理器加载并执行一种机器人故障诊断方法,该方法包括:获取机器人无故障情况下的参考音频信息;使机器人运转,并记录机器人运转时的实际音频信息;将所述实际音频信息与所述参考音频信息进行对比,根据对比结果判断所述机器人的故障情况。
  11. 根据权利要求10所述的故障诊断系统,所述方法还包括:
    所述获取机器人无故障情况下的参考音频信息的步骤包括:分别获取所述机器人无故障情况下,一个或多个零部件单独运动的一个或多个第一参考音频信息;
    所述使所述机器人运转,并记录所述机器人运转时的实际音频信息的步骤包括:使所述机器人的所述一个或多个零部件分别单独运动,并记录所述一个或多个零部件单独运动时的一个或多个第一实际音频信息;
    所述将所述实际音频信息与所述参考音频信息进行对比的步骤包括:将所述机器人的所述一个或多个第一实际音频信息与所述一个或多个第一参考音频信息进行对比。
  12. 根据权利要求10所述的故障诊断系统,所述方法还包括:
    所述获取机器人无故障情况下的参考音频信息的步骤包括:获取所述机器人无故障情况下,在多个零部件共同运动时的第二参考音频信息;
    所述使所述机器人运转,并记录所述机器人运转时的实际音频信息的步骤包括:使所述机器人的所述多个零部件共同运动,并记录所述机器人在所述多个零部件共同运动时的第二实际音频信息;
    所述将所述实际音频信息与所述参考音频信息进行对比的步骤包括:将所述机器人的所述第二实际音频信息与所述第二参考音频信息进行对比。
  13. 根据权利要求10所述的故障诊断系统,所述方法还包括:
    所述参考音频信息包括所述机器人无故障情况下的参考音频频谱;
    所述记录所述机器人运转时的实际音频信息的步骤包括:
    记录所述机器人运转时在时域上的音频信号;
    将所述音频信号进行傅里叶变换,得到所述机器人运转时在频域上的实际音频频谱;
    所述将所述实际音频信息与所述参考音频信息进行对比的步骤包括:
    将所述实际音频频谱中的实际特征频率和各所述实际特征频率对应的幅值与所述参考音频频谱中的参考特征频率和各参考特征频率对应的幅值进行对比。
  14. 根据权利要求13所述的故障诊断系统,所述方法还包括:
    当所述实际特征频率与所述参考特征频率不相同,或者所述实际特征频率与所述参考特征频率的差值大于第一预设范围,则判断所述机器人存在故障;或者
    当所述实际特征频率对应的幅值与所述参考特征频率对应的幅值的差值大于第二预设范围,则判断所述机器人存在故障。
  15. 根据权利要求14所述的故障诊断系统,所述方法还包括:
    当判断所述机器人存在故障时,根据所述实际特征频率和所述参考特征频率,确定所述机器人的故障存在于零部件、连接结构或者安装方式上。
  16. 根据权利要求15所述的故障诊断系统,所述方法在所述确定所述机器人的故障存在于所述零部件、所述连接结构或者所述安装方式上的步骤之后,还包括:
    根据所述实际音频信息和所述参考音频信息的决策结果,提供所述零部件、所述连接结构或者所述安装方式的故障的解决方案。
  17. 根据权利要求13所述的故障诊断系统,所述方法在所述将所述音频信号进行傅里叶变换的步骤之前,还包括:
    所述处理器将所述音频信号中的环境噪声去除;以及
    对去除环境噪声后的所述音频信号进行归一化处理,以更新所述音频信号。
  18. 根据权利要求10所述的故障诊断系统,所述方法还包括:所述记录器使用动态范围大于0~40kHz的音频记录仪记录所述机器人运转时的所述实际音频信息。
  19. 一种存储装置,其上存储有程序指令,其特征在于,所述程序指令被处理器执行时实现权利要求1-9任一项所述方法的步骤。
PCT/CN2018/111261 2018-10-22 2018-10-22 机器人故障诊断方法、系统及存储装置 WO2020082217A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201880087304.2A CN111684213A (zh) 2018-10-22 2018-10-22 机器人故障诊断方法、系统及存储装置
PCT/CN2018/111261 WO2020082217A1 (zh) 2018-10-22 2018-10-22 机器人故障诊断方法、系统及存储装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/111261 WO2020082217A1 (zh) 2018-10-22 2018-10-22 机器人故障诊断方法、系统及存储装置

Publications (1)

Publication Number Publication Date
WO2020082217A1 true WO2020082217A1 (zh) 2020-04-30

Family

ID=70330810

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/111261 WO2020082217A1 (zh) 2018-10-22 2018-10-22 机器人故障诊断方法、系统及存储装置

Country Status (2)

Country Link
CN (1) CN111684213A (zh)
WO (1) WO2020082217A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114571459B (zh) * 2022-03-22 2023-07-21 浙江钱江机器人有限公司 一种工业机器人故障诊断方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20090010430A (ko) * 2007-07-23 2009-01-30 주식회사 싸이언 기계 장치의 고장 진단 장치
CN103744021A (zh) * 2013-12-23 2014-04-23 煤炭科学研究总院 一种电机故障监控装置及方法
CN108469109A (zh) * 2018-03-01 2018-08-31 广东美的制冷设备有限公司 设备异常的检测方法、装置、系统、空调器及存储介质
CN108638128A (zh) * 2018-05-24 2018-10-12 哈工大机器人(合肥)国际创新研究院 一种工业机器人的实时异常监测方法及其系统

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10026415B2 (en) * 2014-03-17 2018-07-17 Koninklijke Philips N.V. Noise suppression
CN105095918B (zh) * 2015-09-07 2018-06-26 上海交通大学 一种多机器人系统故障诊断方法
CN105675306B (zh) * 2016-01-12 2018-08-10 天津理工大学 一种汽车自检测方法及检测系统
CN105882687A (zh) * 2016-05-13 2016-08-24 南京雅信科技集团有限公司 道岔转辙机的故障类别分析方法
CN106934421B (zh) * 2017-03-16 2020-11-06 山东大学 基于2dpca及svm的电力变压器故障检测方法
CN107963239B (zh) * 2017-11-16 2020-10-23 北京宇航系统工程研究所 一种基于音频的运载火箭故障检测装置及检测方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20090010430A (ko) * 2007-07-23 2009-01-30 주식회사 싸이언 기계 장치의 고장 진단 장치
CN103744021A (zh) * 2013-12-23 2014-04-23 煤炭科学研究总院 一种电机故障监控装置及方法
CN108469109A (zh) * 2018-03-01 2018-08-31 广东美的制冷设备有限公司 设备异常的检测方法、装置、系统、空调器及存储介质
CN108638128A (zh) * 2018-05-24 2018-10-12 哈工大机器人(合肥)国际创新研究院 一种工业机器人的实时异常监测方法及其系统

Also Published As

Publication number Publication date
CN111684213A (zh) 2020-09-18

Similar Documents

Publication Publication Date Title
TWI474023B (zh) 馬達故障診斷方法及其診斷裝置
JP6388102B1 (ja) 情報端末器及び機械部品診断システム
JP5460160B2 (ja) 設備機器の診断装置
JP6835265B2 (ja) 計測診断装置、及び計測診断方法
JP6728808B2 (ja) 計測診断装置、及び計測診断方法
JP6281273B2 (ja) 音響装置の検査装置及び音響装置の検査方法並びに音響装置の検査プログラム
WO2011036815A1 (ja) モニタリング装置
WO2022100084A1 (zh) 一种工业设备异常振动检测装置
US20190170704A1 (en) Rotating machine abnormality detection device and method, and rotating machine
JP7083293B2 (ja) 状態監視方法および状態監視装置
WO2020082217A1 (zh) 机器人故障诊断方法、系统及存储装置
US20210018906A1 (en) Feature Value Extraction Apparatus, Predicted-Failure-Evidence Diagnosis Apparatus, Design Assistance Apparatus, and Predicted-Failure-Evidence Diagnosis Operation Method
JP2010203866A (ja) 機器の異常検出装置
CN115310490A (zh) 基于多域特征与敏感特征选择的旋转设备故障分析方法
KR101752298B1 (ko) 회전익 진동 기반 건전성 감시 장치 및 이를 이용하는 감시 방법
WO2021060500A1 (ja) 信号処理装置、信号処理方法およびプログラム
JP2004020484A (ja) 異常監視装置および異常監視プログラム
WO2011125233A1 (ja) 機器の異常検出装置
WO2017212645A1 (ja) 軸受診断装置および軸受診断方法、並びに回転機器およびその保守方法
JP6808405B2 (ja) 回転機器の監視装置
JP7040920B2 (ja) 軸受の状態監視装置及び異常診断方法
CN201843767U (zh) 透平式压缩机故障音频分析诊断装置
Hsia et al. Real-Time Monitor and Control System for Abnormal Motor Vibrations
JP4209793B2 (ja) 音響信号に基づく異常診断方法及び該方法を実行するために用いるプログラム
KR20200128902A (ko) 세탁기 소음불량 검출시스템 및 이를 이용한 소음불량 검출 방법

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: 18937819

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 06/09/2021)

122 Ep: pct application non-entry in european phase

Ref document number: 18937819

Country of ref document: EP

Kind code of ref document: A1