WO2019024450A1 - 一种设备故障检测方法及装置 - Google Patents
一种设备故障检测方法及装置 Download PDFInfo
- Publication number
- WO2019024450A1 WO2019024450A1 PCT/CN2018/072544 CN2018072544W WO2019024450A1 WO 2019024450 A1 WO2019024450 A1 WO 2019024450A1 CN 2018072544 W CN2018072544 W CN 2018072544W WO 2019024450 A1 WO2019024450 A1 WO 2019024450A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- detection model
- detection
- target device
- sound signal
- sampling data
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M99/00—Subject matter not provided for in other groups of this subclass
- G01M99/005—Testing of complete machines, e.g. washing-machines or mobile phones
-
- 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
- G01H1/12—Measuring characteristics of vibrations in solids by using direct conduction to the detector of longitudinal or not specified vibrations
- G01H1/14—Frequency
-
- 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
- G01H1/12—Measuring characteristics of vibrations in solids by using direct conduction to the detector of longitudinal or not specified vibrations
- G01H1/16—Amplitude
Definitions
- the present invention relates to the field of detection technologies, and in particular, to a device fault detection method and apparatus.
- the vibrating sound signal analyzer is equipped with a dedicated sensor probe device and a dedicated human-machine interaction device.
- the user inputs setting parameters through the human-computer interaction operating device, samples the vibration sound signal through the sensor probe device, and analyzes the sampling data to obtain a fault detection result.
- the prior art method requires the user to skillfully use the vibration sound signal analyzer, which requires high degree of specialization, high personnel detection and maintenance cost, less types of detectable vibration sound signals, and user sampling data. Most of the analysis relies on its own experience. If the experience is insufficient, it will not be able to accurately locate the fault and affect the rapid recovery of the production equipment.
- the present invention provides the following technical solutions:
- a device fault detection method is applied to a cloud server, including:
- a preliminary fault diagnosis result for the target device is obtained based on the first detection model and the first sampled data.
- the method further includes:
- the method further includes:
- the detection result is: the second vibration sound signal is not in the detection model library
- the method further includes:
- the updated detection model library is sent to the client, so that the user performs failure analysis directly on the client based on the detection model library.
- the method further includes:
- a device fault detecting device is applied to a cloud server, including:
- a data obtaining module configured to obtain first sampling data of a first vibro-acoustic signal for the target device by a signal detector deployed in the target device;
- a model acquisition module configured to retrieve a first detection model corresponding to the type of the first vibration sound signal in a pre-established detection model library
- a diagnosis result obtaining module configured to obtain a preliminary fault diagnosis result for the target device based on the first detection model and the first sampling data.
- a diagnostic result output module is further included, configured to:
- the target device After obtaining the preliminary fault diagnosis result for the target device based on the first detection model and the first sampling data, outputting the preliminary fault diagnosis result, so that the user according to the preliminary fault diagnosis result
- the target device After obtaining the preliminary fault diagnosis result for the target device based on the first detection model and the first sampling data, outputting the preliminary fault diagnosis result, so that the user according to the preliminary fault diagnosis result
- the target device After obtaining the preliminary fault diagnosis result for the target device based on the first detection model and the first sampling data, outputting the preliminary fault diagnosis result, so that the user according to the preliminary fault diagnosis result
- the target device performs corresponding processing.
- the method further includes:
- a sampling data and a detection result receiving module configured to receive a second sampling data of the second vibration sound signal of the target device and a detection result corresponding to the second sampling data, where the detection result is: in the detection model The result obtained by the user analyzing the second sampled data by the client when there is no detection model corresponding to the type of the second vibration sound signal in the library;
- a model establishing module configured to establish, according to the second sampling data and the detection result, a second detection model corresponding to a type of the second vibration sound signal
- a detection model library update module configured to update the detection model library based on the second detection model.
- the method further includes a detection model library sending module, configured to:
- the updated detection model library is sent to the client, so that the user performs failure analysis directly on the client based on the detection model library.
- the method further includes:
- a feedback information receiving module configured to receive feedback information of the user for the preliminary fault diagnosis result
- a detection model adjustment module configured to adjust the first detection model according to the feedback information.
- the cloud server obtains first sampling data of the first vibration sound signal for the target device by using a signal detector deployed in the target device, and extracts and searches in the pre-established detection model library.
- a first detection model corresponding to the type of the first vibration sound signal based on the first detection model and the first sampling data, obtains a preliminary failure diagnosis result for the target device.
- the sampling data is analyzed, and the preliminary fault diagnosis result of the target device is obtained, the diagnosis accuracy is improved, the personnel detection and maintenance cost is reduced, and the target device is quickly Recovery increases the variety of detectable vibration sound signals.
- FIG. 1 is a structural block diagram of a device fault detection system according to an embodiment of the present invention.
- FIG. 2 is a flowchart of implementing a device fault detection method according to an embodiment of the present invention
- FIG. 3 is a structural block diagram of a signal detector according to an embodiment of the present invention.
- FIG. 4 is a schematic structural diagram of a device fault detecting apparatus according to an embodiment of the present invention.
- the core of the present invention is to provide a device fault detection method, which can be applied to a cloud server, and the cloud server can be connected to a signal detector deployed in the target device, as shown in FIG.
- the sampling data is analyzed, and the preliminary fault diagnosis result of the target device is obtained, the diagnosis accuracy is improved, the personnel detection and maintenance cost is reduced, and the target device is quickly Recovery increases the variety of detectable vibration sound signals.
- FIG. 2 is a flowchart of an implementation of a device fault detection method according to an embodiment of the present invention. The method may include the following steps:
- S201 Obtain first sampling data of a first vibration sound signal for the target device by using a signal detector deployed in the target device.
- the target device can be any device to be fault detected.
- a signal detector can be deployed in the target device.
- the signal detector can detect the first vibration signal of the target device to obtain the first sample data.
- the network detector of the signal detector can be configured through the client, so that the signal detector can communicate with the cloud server, as shown in FIG. 1 .
- the cloud server communicates with the signal detector, and the first sampled data can be obtained directly from the signal detector, or the cloud server communicates with the client, and the first sampled data is obtained from the signal detector through the client.
- the signal detector may include a vibration sound signal sensing detecting unit, a master chip unit, a communication unit, and a power conversion unit.
- the vibration sound signal sensing detection unit generally refers to a three-axis acceleration sensor for sampling frequency, amplitude and the like of the vibration sound signal, and sends the sampling data to the main control chip unit;
- the main control chip unit may be a microcontroller MCU, micro a processor MPU, or other form of programmable logic device for processing sampled data, such as filtering, averaging, fast Fourier transform FFT, etc., and transmitting the processed sampled data to the communication unit;
- the communication unit is configured to
- the processed sample data is encapsulated into a data format that can be transmitted through a network, such as a json format, and sent to the cloud server through a wireless network communication method or a wired network communication method.
- the specific communication methods may be wireless Wi-Fi, narrowband Internet of Things NB-IoT, ultra long-range low-power data transmission
- the main control chip unit and the communication unit may be two independent units, or may be an integrated unit.
- the communication mode between the main control chip unit and the vibration sound signal sensing detection unit or between the main control chip unit and the communication unit may be serial communication, or other types of communication methods, such as high and low levels, ADC signals, PWM signals, etc. Wait.
- the client can be a mobile application APP, a computer PC client, or a tablet ipad application, etc., and is responsible for implementing the human-computer interaction function of the detection.
- the current mobile phone or PC is quite popular, and the operation of the client will become a skill for everyone.
- the user interacts with the signal detector and the cloud server through the client man-machine operation interface, and the operation is simple, which saves the production cost and the maintenance cost of the analysis instrument, and reduces the requirement for the user specialization degree.
- the software update speed in the client of the present invention is faster, which will greatly improve the speed of future product iteration update.
- step S202 After the cloud server obtains the first sampling data by using the signal detector deployed in the target device, the operation of step S202 may be continued.
- S202 Acquire a first detection model corresponding to the type of the first vibration sound signal in a pre-established detection model library.
- the detection model library may be pre-established in the cloud server, and the detection model corresponding to the type of the plurality of vibration sound signals is stored in the detection model library.
- the cloud server can obtain sampling data corresponding to various types of vibration sound signals in advance, and detection results of each sampling data. For each type of vibration sound signal, the sampling data corresponding to the vibration sound signal of the type and the detection result corresponding to each sampling data are analyzed, and a detection model corresponding to the type can be established.
- the user can select or set the type of the first vibrating sound signal on the client, and the cloud server obtains the type of the first vibrating sound signal through the client.
- the cloud server obtains the first sampled data
- the first sampled data is identified and analyzed by using the prior data, and the type of the corresponding first vibration sound signal is obtained.
- the cloud server obtains the first vibration sound signal type corresponding to the first sample data, and retrieves a first detection model corresponding to the type from the detection model library.
- S203 Obtain a preliminary fault diagnosis result for the target device based on the first detection model and the first sampling data.
- the cloud server obtains first sampling data of the first vibrating sound signal for the target device by using a signal detector deployed in the target device, and retrieves the pre-established detection model library.
- a first detection model corresponding to the type of the first vibration sound signal based on the first detection model and the first sampling data, obtains a preliminary fault diagnosis result for the target device.
- the sampling data is analyzed, and the preliminary fault diagnosis result of the target device is obtained, the diagnosis accuracy is improved, the personnel detection and maintenance cost is reduced, and the target device is quickly Recovery increases the variety of detectable vibration sound signals.
- the method may further include the following steps:
- the preliminary fault diagnosis result is output, so that the user can process the target device according to the preliminary fault diagnosis result.
- the initial fault diagnosis result may be output on the client or the cloud server, so that the user processes the target device by viewing the preliminary fault diagnosis result. For example, when the initial fault diagnosis result is that the belt is loose, the user can shut down the production equipment, stop it, adjust the belt to a proper tightness, and then start the production equipment to continue running. Therefore, equipment failures can be detected and checked in time to avoid affecting the production efficiency of the equipment.
- the method may further include the following steps:
- Step 1 receiving the detection result corresponding to the second sampling data and the second sampling data of the second vibration sound signal of the target device, and the detection result is: when there is no detection model corresponding to the type of the second vibration sound signal in the detection model library The result obtained by the user analyzing the second sampled data through the client;
- Step 2 establishing a second detection model corresponding to the type of the second vibration sound signal according to the second sampling data and the detection result;
- Step 3 Update the detection model library based on the second detection model.
- the second sample data may be sent to the client first, and the user checks the detection model library of the cloud server through the client. Whether there is a detection model corresponding to the type of the second vibration sound signal. If yes, the second sampled data is sent to the cloud server, and the cloud server analyzes it to obtain a corresponding detection result. If it does not exist, the second sample data can be analyzed by the user to obtain a corresponding detection result.
- the second sample data may be directly sent to the cloud server.
- the cloud server checks whether there is a detection model corresponding to the type of the second vibration sound signal in the detection model library. If it exists, the second sampled data is directly analyzed to obtain a corresponding detection result. If it does not exist, the second sampling data may be sent to the client, and the second sampling data is analyzed by the user to obtain a corresponding detection result.
- the client may send the detection result corresponding to the second sampling data and the second sampling data to the cloud server.
- the cloud server establishes a second detection model corresponding to the type of the second vibration sound signal according to the second sampled data and the detection result, adds the second detection model to the detection model library, and updates the detection model library to complete a set of self-learning Modeling process.
- the cloud server receives the sampled data of the vibration sound signal of the type, the corresponding detection model can be retrieved from the detection model library to obtain the preliminary fault detection result, so that the empirical data can be passed on and used.
- the self-learning modeling process improves the detection adaptability of the new vibration sound signal: when encountering the type of vibration sound signal that the cloud server cannot support, the present invention can obtain the learning by analyzing the vibration sound signal of the type by the user. Data, that is, what kind of equipment, what kind of components, what kind of vibration sound signal caused the fault phenomenon. Thereby the corresponding detection results are obtained. The learning data and the detection result are sent to the cloud server, and the model is established, and the whole process of self-learning and memory is completed.
- the method may further include the following steps:
- the updated detection model library is sent to the client, so that the user performs fault analysis based on the detection model library directly on the client.
- the updated detection model library can be sent to the client. Specifically, only the currently updated detection model may be sent to the client, or all the detection models in the detection model library may be sent to the client.
- the detection model library of the client is consistent with the detection model library of the cloud server. In the case that the client and the cloud server have no network connection, the user can directly perform fault analysis based on the detection model library on the client.
- the method may further include the following steps:
- Step 1 Receive feedback information of the user for the preliminary fault diagnosis result
- Step 2 Adjust the first detection model according to the feedback information.
- the user After the user obtains the preliminary fault diagnosis result for the target device, the user can verify the obtained preliminary fault diagnosis result, and send the verified feedback information to the cloud server.
- the cloud server can adjust the first detection model in the detection model library according to the feedback information.
- the corresponding first detection model in the detection model library is: when the intensity of the vibration sound signal is 40 dB to 60 dB, the corresponding failure detection result is that the belt is loose; when the vibration sound When the intensity of the signal is 60dB to 80dB, the corresponding fault detection result is that the motor shaft is bent; when the strength of the vibration sound signal is higher than 80dB, the corresponding fault detection result is the link break of the chain transmission mechanism.
- the cloud server obtains the vibration sound signal strength of the first sampled data to be 58 dB
- the first fault detection result is obtained by calling the corresponding first detection model in the model library, and the preliminary fault detection result is that the belt is loose.
- the feedback information for the preliminary fault diagnosis result can be sent to the cloud server, and the cloud server can adjust the interval of the vibration sound signal strength in the first detection model. In order to make the detection model more accurate and improve the accuracy of the diagnosis results.
- the embodiment of the present invention further provides a device fault detecting apparatus, which is applied to a cloud server, and a device fault detecting apparatus described below and a device fault detecting method described above may correspond to each other.
- a device fault detecting apparatus which is applied to a cloud server
- a device fault detecting apparatus described below and a device fault detecting method described above may correspond to each other.
- the device includes the following modules:
- a data obtaining module 401 configured to obtain first sampling data of a first vibration sound signal for the target device by using a signal detector deployed in the target device;
- the model retrieval module 402 is configured to retrieve a first detection model corresponding to the type of the first vibration sound signal in a pre-established detection model library;
- the diagnosis result obtaining module 403 is configured to obtain a preliminary fault diagnosis result for the target device based on the first detection model and the first sampling data.
- the cloud server obtains first sampling data of the first vibrating sound signal for the target device by using a signal detector deployed in the target device, and extracts and searches in a pre-established detection model library.
- a first detection model corresponding to the type of the first vibration sound signal based on the first detection model and the first sampling data, obtains a preliminary fault diagnosis result for the target device.
- the sampling data is analyzed, and the preliminary fault diagnosis result of the target device is obtained, the diagnosis accuracy is improved, the personnel detection and maintenance cost is reduced, and the target device is quickly Recovery increases the variety of detectable vibration sound signals.
- a diagnostic result output module is further included, configured to:
- the preliminary fault diagnosis result After obtaining the preliminary fault diagnosis result for the target device based on the first detection model and the first sampling data, the preliminary fault diagnosis result is output, so that the user performs corresponding processing on the target device according to the preliminary fault diagnosis result.
- the method further includes:
- the sampling data and the detection result receiving module are configured to receive the detection result corresponding to the second sampling data and the second sampling data of the second vibration sound signal of the target device, and the detection result is: there is no second vibration sound signal in the detection model library When the type corresponds to the detection model, the result obtained by the user analyzing the second sampled data through the client;
- a model establishing module configured to establish, according to the second sampled data and the detection result, a second detection model corresponding to the type of the second vibration sound signal
- the detection model library update module is configured to update the detection model library based on the second detection model.
- the method further includes a detection model library sending module, configured to:
- the updated detection model library is sent to the client, so that the user performs fault analysis based on the detection model library directly on the client.
- the method further includes:
- a feedback information receiving module configured to receive feedback information of the user for the preliminary fault diagnosis result
- the detection model adjustment module is configured to adjust the first detection model according to the feedback information.
- the steps of a method or algorithm described in connection with the embodiments disclosed herein can be implemented directly in hardware, a software module executed by a processor, or a combination of both.
- the software module can be placed in random access memory (RAM), memory, read only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or technical field. Any other form of storage medium known.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Telephonic Communication Services (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
一种设备故障检测方法及装置，应用于云端服务器，该方法包括：通过部署在目标设备中的信号检测器获得针对目标设备的第一振动声音信号的第一采样数据（S201）；在预先建立的检测模型库中调取与第一振动声音信号的类型对应的第一检测模型（S202）；基于第一检测模型和第一采样数据，获得针对目标设备的初步故障诊断结果（S203）。该方法提高了诊断的准确性，减少了人员检测和维护成本，使得目标设备得以快速恢复，增加了可检测的振动声音信号种类。
Description
本申请要求于2017年8月1日提交中国专利局、申请号为201710647420.2、发明名称为“一种设备故障检测方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
本发明涉及检测技术领域，特别是涉及一种设备故障检测方法及装置。
在生产过程中，生产设备难免会发生故障，使得生产设备无法正常运行。而生产设备是否能够正常运行直接影响着生产效率，所以及时对设备故障进行检测和排查是非常重要的。
目前，多是由用户通过独立设计的振动声音信号分析仪对生产设备进行检测。振动声音信号分析仪中配备有专用的传感器探头装置和专门的人机交互操作装置。用户通过人机交互操作装置输入设置参数，通过传感器探头装置对振动声音信号进行采样，并对采样数据进行分析，得到故障检测结果。
现有技术的这种方法需要用户熟练使用振动声音信号分析仪，对其专业化程度要求较高，人员检测和维护成本较高，可检测的振动声音信号种类少，且，用户对采样数据的分析多是依赖于自身经验，如果经验不足将无法准确定位故障，影响生产设备的快速恢复。
发明内容
本发明的目的是提供一种设备故障检测方法及装置，以提高故障诊断的准确性，减少人员检测和维护成本，使得目标设备得以快速恢复。
为解决上述技术问题，本发明提供如下技术方案：
一种设备故障检测方法，应用于云端服务器，包括：
通过部署在目标设备中的信号检测器获得针对所述目标设备的第一振动声音信号的第一采样数据；
在预先建立的检测模型库中调取与所述第一振动声音信号的类型对应的第一检测模型；
基于所述第一检测模型和所述第一采样数据，获得针对所述目标设备的初步故障诊断结果。
在本发明的一种具体实施方式中，在所述基于所述第一检测模型和所述第一采样数据，获得针对所述目标设备的初步故障诊断结果之后，还包括：
输出所述初步故障诊断结果，以使用户根据所述初步故障诊断结果对所述目标设备进行相应处理。
在本发明的一种具体实施方式中，还包括：
接收针对所述目标设备的第二振动声音信号的第二采样数据和所述第二采样数据对应的检测结果，所述检测结果为：在所述检测模型库中没有所述第二振动声音信号的类型对应的检测模型时，由用户通过客户端对所述第二采样数据进行分析得到的结果；
根据所述第二采样数据和所述检测结果，建立与所述第二振动声音信号的类型对应的第二检测模型；
基于所述第二检测模型，更新所述检测模型库。
在本发明的一种具体实施方式中，还包括：
将更新后的所述检测模型库发送给所述客户端，以使所述用户直接在所述客户端上基于所述检测模型库进行故障分析。
在本发明的一种具体实施方式中，还包括：
接收用户针对所述初步故障诊断结果的反馈信息；
根据所述反馈信息，调整所述第一检测模型。
一种设备故障检测装置，应用于云端服务器，包括：
数据获得模块，用于通过部署在目标设备中的信号检测器获得针对所述目标设备的第一振动声音信号的第一采样数据；
模型调取模块，用于在预先建立的检测模型库中调取与所述第一振动声音信号的类型对应的第一检测模型；
诊断结果获得模块，用于基于所述第一检测模型和所述第一采样数据， 获得针对所述目标设备的初步故障诊断结果。
在本发明的一种具体实施方式中，还包括诊断结果输出模块，用于：
在所述基于所述第一检测模型和所述第一采样数据，获得针对所述目标设备的初步故障诊断结果之后，输出所述初步故障诊断结果，以使用户根据所述初步故障诊断结果对所述目标设备进行相应处理。
在本发明的一种具体实施方式中，还包括：
采样数据和检测结果接收模块，用于接收针对所述目标设备的第二振动声音信号的第二采样数据和所述第二采样数据对应的检测结果，所述检测结果为：在所述检测模型库中没有所述第二振动声音信号的类型对应的检测模型时，由用户通过客户端对所述第二采样数据进行分析得到的结果；
模型建立模块，用于根据所述第二采样数据和所述检测结果，建立与所述第二振动声音信号的类型对应的第二检测模型；
检测模型库更新模块，用于基于所述第二检测模型，更新所述检测模型库。
在本发明的一种具体实施方式中，还包括检测模型库发送模块，用于：
将更新后的所述检测模型库发送给所述客户端，以使所述用户直接在所述客户端上基于所述检测模型库进行故障分析。
在本发明的一种具体实施方式中，还包括：
反馈信息接收模块，用于接收用户针对所述初步故障诊断结果的反馈信息；
检测模型调整模块，用于根据所述反馈信息，调整所述第一检测模型。
应用本发明实施例所提供的技术方案，云端服务器通过部署在目标设备中的信号检测器获得针对目标设备的第一振动声音信号的第一采样数据，在预先建立的检测模型库中调取与第一振动声音信号的类型对应的第一检测模型，基于第一检测模型和第一采样数据，获得针对目标设备的初步故障诊断结果。通过调取的预先建立的检测模型库中的检测模型，对采样数据进行分析，得到目标设备的初步故障诊断结果，提高了诊断的准确性，减少了人员检测和维护成本，使得目标设备得以快速恢复，增加了可检测的振动声音信号种类。
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本发明实施例中一种设备故障检测系统的结构框图；
图2为本发明实施例中一种设备故障检测方法的实施流程图；
图3为本发明实施例中一种信号检测器的结构框图；
图4为本发明实施例中一种设备故障检测装置的结构示意图。
为了使本技术领域的人员更好地理解本发明方案，下面结合附图和具体实施方式对本发明作进一步的详细说明。显然，所描述的实施例仅仅是本发明一部分实施例，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。
本发明的核心是提供一种设备故障检测方法，该方法可以应用于云端服务器，云端服务器可以与部署在目标设备中的信号检测器相连，如图1所示。通过信号检测器获得针对目标设备的第一振动声音信号的第一采样数据，在预先建立的检测模型库中调取与第一振动声音信号的类型对应的第一检测模型，基于第一检测模型和第一采样数据，获得针对目标设备的初步故障诊断结果。通过调取的预先建立的检测模型库中的检测模型，对采样数据进行分析，得到目标设备的初步故障诊断结果，提高了诊断的准确性，减少了人员检测和维护成本，使得目标设备得以快速恢复，增加了可检测的振动声音信号种类。
参见图2，为本发明实施例中一种设备故障检测方法的实施流程图， 该方法可以包括以下步骤：
S201：通过部署在目标设备中的信号检测器获得针对目标设备的第一振动声音信号的第一采样数据。
目标设备可以是待进行故障检测的任意一台设备。在目标设备中可以部署信号检测器。信号检测器可以对目标设备的第一振动信号进行检测，获得第一采样数据。在实际应用中，可以通过客户端对信号检测器进行网络配置，使信号检测器可以与云端服务器通信连接，如图1所示。云端服务器与信号检测器通信，可以直接从信号检测器中获得第一采样数据，或者，云端服务器与客户端通信，通过客户端从信号检测器中获得第一采样数据。
如图3所示，信号检测器可以包括振动声音信号传感检测单元、主控芯片单元、通信单元和电源转换单元。振动声音信号传感检测单元通常指三轴加速度传感器，用于采样振动声音信号的频率、振幅等信息，并将采样数据发送给主控芯片单元；主控芯片单元可以是微控制器MCU、微处理器MPU、或其它形式的可编程逻辑器件，用于对采样数据进行处理，如滤波、求平均值、快速傅立叶变换FFT等，并将处理后的采样数据传送给通信单元；通信单元用于将处理后的采样数据封包成可通过网络传输的数据格式，如json格式等，并通过无线网络通信方式或有线网络通信方式发送给云端服务器。具体的通信方式可以是无线Wi-Fi、窄带物联网NB-IoT、超长距低功耗数据传输Lora、第四代移动通信技术4G网络信号或其它可数据传输至云端的通信方式。
其中，主控芯片单元与通信单元可以是独立的两个单元，也可以是一体的单元。主控芯片单元与振动声音信号传感检测单元间或主控芯片单元与通信单元间的通信方式可以是串口通信，也可以是其它种类的通信方式，如：高低电平、ADC信号、PWM信号等等。
客户端可以是手机应用端APP、电脑PC客户端、或平板电脑ipad应用端等，负责实施检测的人机交互功能。当下手机或PC已相当普及，对客户端的操作将成为人人具备的技能。在本发明实施例中，用户通过客户端人机操作界面与信号检测器和云端服务器交互，操作简单，节约了分析 仪器的生产成本与维护成本，降低了对用户专业化程度要求。相较于现有技术中对分析仪硬件的依赖，本发明客户端中的软件更新速度更快，将极大地提高未来产品迭代更新的速度。
云端服务器通过部署在目标设备中的信号检测器获得第一采样数据后，可以继续执行步骤S202的操作。
S202：在预先建立的检测模型库中调取与第一振动声音信号的类型对应的第一检测模型。
在本发明实施例中，云端服务器中可以预先建立检测模型库，检测模型库中存储有多种振动声音信号的类型对应的检测模型。
在实际应用中，云端服务器可以预先获得各种类型的振动声音信号对应的采样数据，及各采样数据的检测结果。针对每种类型的振动声音信号，对该类型的振动声音信号对应的采样数据及各采样数据对应的检测结果进行分析，可以建立该类型对应的检测模型。
用户可以在客户端上选择或设置第一振动声音信号的类型，云端服务器通过客户端获得第一振动声音信号的类型。或者，云端服务器获得第一采样数据后，利用先验数据对第一采样数据进行识别分析，得出对应的第一振动声音信号的类型。
云端服务器获得第一采样数据对应的第一振动声音信号类型，从检测模型库中调取与该类型对应的第一检测模型。
S203：基于第一检测模型和第一采样数据，获得针对目标设备的初步故障诊断结果。
在预先建立的检测模型库中调取与第一振动声音信号的类型对应的第一检测模型，将第一采样数据输入到对应的第一检测模型中，得到第一采样数据对应的检测结果，从而获得针对目标设备的初步故障诊断结果。
应用本发明实施例所提供的技术方案，云端服务器通过部署在目标设备中的信号检测器获得针对该目标设备的第一振动声音信号的第一采样数据，在预先建立的检测模型库中调取与第一振动声音信号的类型对应的第一检测模型，基于该第一检测模型和第一采样数据，获得针对目标设备的初步故障诊断结果。通过调取的预先建立的检测模型库中的检测模型，对 采样数据进行分析，得到目标设备的初步故障诊断结果，提高了诊断的准确性，减少了人员检测和维护成本，使得目标设备得以快速恢复，增加了可检测的振动声音信号种类。
在本发明的一种具体实施方式中，在步骤S103之后，该方法还可以包括以下步骤：
输出初步故障诊断结果，以使用户根据初步故障诊断结果对目标设备进行相应处理。
当云端服务器获得针对目标设备的初步故障诊断结果之后，可以将该初步故障诊断结果在客户端或云端服务器上输出，以使用户通过查看初步故障诊断结果对目标设备进行相应处理。如当输出初步故障诊断结果为皮带松动时，用户可以关闭生产设备，使其停止运行，将皮带调节到合适的松紧程度后，再启动生产设备继续运行。从而可以对设备故障进行及时的检测和排查，避免影响设备的生产效率。
在本发明的一种具体实施方式中，该方法还可以包括以下步骤：
步骤一：接收针对目标设备的第二振动声音信号的第二采样数据和第二采样数据对应的检测结果，检测结果为：在检测模型库中没有第二振动声音信号的类型对应的检测模型时，由用户通过客户端对第二采样数据进行分析得到的结果；
步骤二：根据第二采样数据和检测结果，建立与第二振动声音信号的类型对应的第二检测模型；
步骤三：基于第二检测模型，更新检测模型库。
在实际应用中，目标设备中部署的信号检测器采集到第二振动声音信号的第二采样数据后，可以先将第二采样数据发送给客户端，用户通过客户端查看云端服务器的检测模型库中是否存在与第二振动声音信号的类型对应的检测模型。如果存在，则将第二采样数据发送给云端服务器，云端服务器对其进行分析，得到相应的检测结果。如果不存在，则可以由用户对第二采样数据进行分析得到相应的检测结果。
或者，信号检测器采集到第二振动声音信号的第二采样数据后，可以直接将第二采样数据发送给云端服务器。云端服务器查看检测模型库中是 否存在与第二振动声音信号的类型对应的检测模型。如果存在，则直接对第二采样数据进行分析，得到相应的检测结果。如果不存在，则可以将第二采样数据发送给客户端，由用户对第二采样数据进行分析得到相应的检测结果。
客户端可以将第二采样数据和第二采样数据对应的检测结果发送到云端服务器。云端服务器根据第二采样数据和检测结果，建立与第二振动声音信号的类型对应的第二检测模型，将第二检测模型加入到检测模型库中，更新检测模型库，从而完成一套自学习建模流程。待下次云端服务器接收到该类型的振动声音信号的采样数据时，可从检测模型库中调取对应的检测模型得出初步故障检测结果，使经验数据得以传承和沿用。
该自学习建模流程，提高了对新型振动声音信号的检测适应性：当遇到云端服务器无法支持的振动声音信号种类时，本发明可以通过用户对该类型的振动声音信号进行分析，获得学习数据，即：何种设备何种部件何种振动声音信号引发了何种故障现象。从而得到相应的检测结果。将该学习数据和检测结果发送到云端服务器，实现模型建立，完成自学习与记忆的全过程。
在本发明的一种具体实施方式中，该方法还可以包括以下步骤：
将更新后的检测模型库发送给客户端，以使用户直接在客户端上基于检测模型库进行故障分析。
云端服务器对检测模型库进行更新后，可以将更新后的检测模型库发送给客户端。具体的，可以是只把当前更新的检测模型发送给客户端，也可以是把检测模型库中所有的检测模型发给客户端。使得客户端的检测模型库与云端服务器的检测模型库保持一致，在客户端与云端服务器无网络连接的情况下，用户可以直接在客户端上基于检测模型库进行故障分析。
在本发明的一种具体实施方式中，该方法还可以包括以下步骤：
步骤一：接收用户针对初步故障诊断结果的反馈信息；
步骤二：根据反馈信息，调整第一检测模型。
当用户获得针对目标设备的初步故障诊断结果后，用户可以对获得的初步故障诊断结果进行核实，将核实后的反馈信息发送到云端服务器。云 端服务器可以根据反馈信息对检测模型库中的第一检测模型进行调整。
如振动声音信号的类型为振动声音信号强度时，检测模型库中的对应的第一检测模型为：当振动声音信号的强度为40dB到60dB时，对应的故障检测结果为皮带松动；当振动声音信号的强度为60dB到80dB时，对应的故障检测结果为电机转轴弯曲；当振动声音信号的强度高于80dB时，对应的故障检测结果为链传动机构的链节断裂。在当前设备故障检测过程中，云端服务器获得第一采样数据的振动声音信号强度为58dB时，通过调用检测模型库中对应的第一检测模型，得出初步故障检测结果为皮带松动。用户进行核实后，发现是电机转轴弯曲，那么可以将针对该初步故障诊断结果的反馈信息发送到云端服务器，云端服务器可以对第一检测模型中振动声音信号强度的区间进行调整。以使得检测模型更加准确，提高诊断结果的准确性。
相对于上面的方法实施例，本发明实施例还提供了一种设备故障检测装置，应用于云端服务器，下文描述的一种设备故障检测装置与上文描述的一种设备故障检测方法可相互对应参照。
参见图4，该装置包括以下模块：
数据获得模块401，用于通过部署在目标设备中的信号检测器获得针对目标设备的第一振动声音信号的第一采样数据；
模型调取模块402，用于在预先建立的检测模型库中调取与第一振动声音信号的类型对应的第一检测模型；
诊断结果获得模块403，用于基于第一检测模型和第一采样数据，获得针对目标设备的初步故障诊断结果。
应用本发明实施例所提供的装置，云端服务器通过部署在目标设备中的信号检测器获得针对该目标设备的第一振动声音信号的第一采样数据，在预先建立的检测模型库中调取与第一振动声音信号的类型对应的第一检测模型，基于该第一检测模型和第一采样数据，获得针对目标设备的初步故障诊断结果。通过调取的预先建立的检测模型库中的检测模型，对采样数据进行分析，得到目标设备的初步故障诊断结果，提高了诊断的准确性，减少了人员检测和维护成本，使得目标设备得以快速恢复，增加了可检测 的振动声音信号种类。
在本发明的一种具体实施方式中，还包括诊断结果输出模块，用于：
在基于第一检测模型和第一采样数据，获得针对目标设备的初步故障诊断结果之后，输出初步故障诊断结果，以使用户根据初步故障诊断结果对目标设备进行相应处理。
在本发明的一种具体实施方式中，还包括：
采样数据和检测结果接收模块，用于接收针对目标设备的第二振动声音信号的第二采样数据和第二采样数据对应的检测结果，检测结果为：在检测模型库中没有第二振动声音信号的类型对应的检测模型时，由用户通过客户端对第二采样数据进行分析得到的结果；
模型建立模块，用于根据第二采样数据和检测结果，建立与第二振动声音信号的类型对应的第二检测模型；
检测模型库更新模块，用于基于第二检测模型，更新检测模型库。
在本发明的一种具体实施方式中，还包括检测模型库发送模块，用于：
将更新后的检测模型库发送给客户端，以使用户直接在客户端上基于检测模型库进行故障分析。
在本发明的一种具体实施方式中，还包括：
反馈信息接收模块，用于接收用户针对初步故障诊断结果的反馈信息；
检测模型调整模块，用于根据反馈信息，调整第一检测模型。
本说明书中各个实施例采用递进的方式描述，每个实施例重点说明的都是与其它实施例的不同之处，各个实施例之间相同或相似部分互相参见即可。对于实施例公开的装置而言，由于其与实施例公开的方法相对应，所以描述的比较简单，相关之处参见方法部分说明即可。
专业人员还可以进一步意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、计算机软件或者二者的结合来实现，为了清楚地说明硬件和软件的可互换性，在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现 不应认为超出本发明的范围。
结合本文中所公开的实施例描述的方法或算法的步骤可以直接用硬件、处理器执行的软件模块，或者二者的结合来实施。软件模块可以置于随机存储器(RAM)、内存、只读存储器(ROM)、电可编程ROM、电可擦除可编程ROM、寄存器、硬盘、可移动磁盘、CD-ROM、或技术领域内所公知的任意其它形式的存储介质中。
本文中应用了具体个例对本发明的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本发明的技术方案及其核心思想。应当指出，对于本技术领域的普通技术人员来说，在不脱离本发明原理的前提下，还可以对本发明进行若干改进和修饰，这些改进和修饰也落入本发明权利要求的保护范围内。
Claims (10)
- 一种设备故障检测方法，其特征在于，应用于云端服务器，包括：通过部署在目标设备中的信号检测器获得针对所述目标设备的第一振动声音信号的第一采样数据；在预先建立的检测模型库中调取与所述第一振动声音信号的类型对应的第一检测模型；基于所述第一检测模型和所述第一采样数据，获得针对所述目标设备的初步故障诊断结果。
- 根据权利要求1所述的方法，其特征在于，在所述基于所述第一检测模型和所述第一采样数据，获得针对所述目标设备的初步故障诊断结果之后，还包括：输出所述初步故障诊断结果，以使用户根据所述初步故障诊断结果对所述目标设备进行相应处理。
- 根据权利要求1所述的方法，其特征在于，还包括：接收针对所述目标设备的第二振动声音信号的第二采样数据和所述第二采样数据对应的检测结果，所述检测结果为：在所述检测模型库中没有所述第二振动声音信号的类型对应的检测模型时，由用户通过客户端对所述第二采样数据进行分析得到的结果；根据所述第二采样数据和所述检测结果，建立与所述第二振动声音信号的类型对应的第二检测模型；基于所述第二检测模型，更新所述检测模型库。
- 根据权利要求3所述的方法，其特征在于，还包括：将更新后的所述检测模型库发送给所述客户端，以使所述用户直接在所述客户端上基于所述检测模型库进行故障分析。
- 根据权利要求1至4任一项所述的方法，其特征在于，还包括：接收用户针对所述初步故障诊断结果的反馈信息；根据所述反馈信息，调整所述第一检测模型。
- 一种设备故障检测装置，其特征在于，应用于云端服务器，包括：数据获得模块，用于通过部署在目标设备中的信号检测器获得针对所 述目标设备的第一振动声音信号的第一采样数据；模型调取模块，用于在预先建立的检测模型库中调取与所述第一振动声音信号的类型对应的第一检测模型；诊断结果获得模块，用于基于所述第一检测模型和所述第一采样数据，获得针对所述目标设备的初步故障诊断结果。
- 根据权利要求6所述的装置，其特征在于，还包括诊断结果输出模块，用于：在所述基于所述第一检测模型和所述第一采样数据，获得针对所述目标设备的初步故障诊断结果之后，输出所述初步故障诊断结果，以使用户根据所述初步故障诊断结果对所述目标设备进行相应处理。
- 根据权利要求6所述的装置，其特征在于，还包括：采样数据和检测结果接收模块，用于接收针对所述目标设备的第二振动声音信号的第二采样数据和所述第二采样数据对应的检测结果，所述检测结果为：在所述检测模型库中没有所述第二振动声音信号的类型对应的检测模型时，由用户通过客户端对所述第二采样数据进行分析得到的结果；模型建立模块，用于根据所述第二采样数据和所述检测结果，建立与所述第二振动声音信号的类型对应的第二检测模型；检测模型库更新模块，用于基于所述第二检测模型，更新所述检测模型库。
- 根据权利要求8所述的装置，其特征在于，还包括检测模型库发送模块，用于：将更新后的所述检测模型库发送给所述客户端，以使所述用户直接在所述客户端上基于所述检测模型库进行故障分析。
- 根据权利要求6至9任一项所述的装置，其特征在于，还包括：反馈信息接收模块，用于接收用户针对所述初步故障诊断结果的反馈信息；检测模型调整模块，用于根据所述反馈信息，调整所述第一检测模型。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710647420.2 | 2017-08-01 | ||
| CN201710647420.2A CN107421763B (zh) | 2017-08-01 | 2017-08-01 | 一种设备故障检测方法及装置 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2019024450A1 true WO2019024450A1 (zh) | 2019-02-07 |
Family
ID=60436483
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2018/072544 WO2019024450A1 (zh) | 2017-08-01 | 2018-01-15 | 一种设备故障检测方法及装置 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN107421763B (zh) |
| WO (1) | WO2019024450A1 (zh) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107421763B (zh) * | 2017-08-01 | 2019-11-15 | 上海庆科信息技术有限公司 | 一种设备故障检测方法及装置 |
| CN109104705A (zh) * | 2018-07-13 | 2018-12-28 | 广州市森锐科技股份有限公司 | 一种基于nb-iot的智能终端设备及其系统 |
| CN110749372A (zh) * | 2018-07-18 | 2020-02-04 | 上海数深智能科技有限公司 | 一种电机振动移动智能诊断系统及其使用方法 |
| CN110674184B (zh) * | 2019-09-06 | 2023-10-17 | 创新先进技术有限公司 | 一种构造异动检测模型库的方法、系统及设备 |
| CN111336100A (zh) * | 2020-04-10 | 2020-06-26 | 中国恩菲工程技术有限公司 | 水泵故障诊断系统 |
| CN116541800A (zh) * | 2023-07-06 | 2023-08-04 | 利维智能（深圳）有限公司 | 基于振动和声音数据的融合诊断方法、系统、设备及介质 |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101126929A (zh) * | 2007-09-05 | 2008-02-20 | 东北大学 | 连采机远程实时故障预测及诊断方法与装置 |
| CN101780909A (zh) * | 2009-09-16 | 2010-07-21 | 新疆大学 | 无机房电梯检测诊断系统 |
| CN102621971A (zh) * | 2012-04-17 | 2012-08-01 | 上海探能实业有限公司 | 一种保证风电机组正常运行的共享维护系统及其实现方法 |
| CN102778358A (zh) * | 2012-06-04 | 2012-11-14 | 上海东锐风电技术有限公司 | 故障预测模型建立方法及系统、风机监测预警系统及方法 |
| US20130090831A1 (en) * | 2010-06-14 | 2013-04-11 | Inha-Industry Partnership Institute | System for fault detection and diagnosis of aircraft engine and method thereof |
| CN103558048A (zh) * | 2013-11-13 | 2014-02-05 | 重庆大学 | Android平台上的虚拟式机械故障诊断仪和方法 |
| CN104568438A (zh) * | 2014-10-07 | 2015-04-29 | 芜湖扬宇机电技术开发有限公司 | 发动机轴承故障检测系统及方法 |
| CN106323664A (zh) * | 2016-09-22 | 2017-01-11 | 珠海格力电器股份有限公司 | 空调机组的振动测试及诊断方法、装置及空调机组 |
| CN107421763A (zh) * | 2017-08-01 | 2017-12-01 | 上海庆科信息技术有限公司 | 一种设备故障检测方法及装置 |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008298527A (ja) * | 2007-05-30 | 2008-12-11 | Toshiba Corp | 回転機械の振動診断方法及びその振動診断装置 |
| CN103914617B (zh) * | 2014-03-25 | 2017-02-01 | 北京交通大学 | 地铁车辆转向架轴承的故障诊断方法 |
| CN106527339B (zh) * | 2016-12-14 | 2019-04-23 | 东北大学 | 一种基于工业云的高可靠的选矿设备故障诊断系统及方法 |
| CN106961249B (zh) * | 2017-03-17 | 2019-02-19 | 广西大学 | 一种光伏阵列故障诊断和预警方法 |
-
2017
- 2017-08-01 CN CN201710647420.2A patent/CN107421763B/zh active Active
-
2018
- 2018-01-15 WO PCT/CN2018/072544 patent/WO2019024450A1/zh active Application Filing
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101126929A (zh) * | 2007-09-05 | 2008-02-20 | 东北大学 | 连采机远程实时故障预测及诊断方法与装置 |
| CN101780909A (zh) * | 2009-09-16 | 2010-07-21 | 新疆大学 | 无机房电梯检测诊断系统 |
| US20130090831A1 (en) * | 2010-06-14 | 2013-04-11 | Inha-Industry Partnership Institute | System for fault detection and diagnosis of aircraft engine and method thereof |
| CN102621971A (zh) * | 2012-04-17 | 2012-08-01 | 上海探能实业有限公司 | 一种保证风电机组正常运行的共享维护系统及其实现方法 |
| CN102778358A (zh) * | 2012-06-04 | 2012-11-14 | 上海东锐风电技术有限公司 | 故障预测模型建立方法及系统、风机监测预警系统及方法 |
| CN103558048A (zh) * | 2013-11-13 | 2014-02-05 | 重庆大学 | Android平台上的虚拟式机械故障诊断仪和方法 |
| CN104568438A (zh) * | 2014-10-07 | 2015-04-29 | 芜湖扬宇机电技术开发有限公司 | 发动机轴承故障检测系统及方法 |
| CN106323664A (zh) * | 2016-09-22 | 2017-01-11 | 珠海格力电器股份有限公司 | 空调机组的振动测试及诊断方法、装置及空调机组 |
| CN107421763A (zh) * | 2017-08-01 | 2017-12-01 | 上海庆科信息技术有限公司 | 一种设备故障检测方法及装置 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107421763A (zh) | 2017-12-01 |
| CN107421763B (zh) | 2019-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2019024450A1 (zh) | 一种设备故障检测方法及装置 | |
| US11615794B2 (en) | Voice recognition system, server, display apparatus and control methods thereof | |
| US10649060B2 (en) | Sound source localization confidence estimation using machine learning | |
| US20190371322A1 (en) | Method, apparatus and system for testing intelligent voice device | |
| US8964995B2 (en) | Acoustic diagnosis and correction system | |
| JP2020511804A (ja) | 音声サービスを提供するための方法、装置およびサーバ | |
| Chen et al. | Multiple fault condition recognition of gearbox with sequential hypothesis test | |
| US11615344B2 (en) | Condition monitoring device, method, and storage medium | |
| JP2020095258A (ja) | 問題騒音の発音源を識別するための騒音データの人工知能装置および前処理方法 | |
| WO2023093919A2 (zh) | 一种洗衣机异音检测方法、装置、电子设备及存储介质 | |
| WO2021213135A1 (zh) | 音频处理方法、装置、电子设备和存储介质 | |
| Yin et al. | Integration of deep learning and soft robotics for a biomimetic approach to nonlinear sensing | |
| CN115277261B (zh) | 基于工控网络病毒的异常机器智能识别方法、装置、设备 | |
| CN105809948A (zh) | 红外控制设备自动化测试方法及系统 | |
| JPWO2015011791A1 (ja) | 異常検知評価システム | |
| US11934581B2 (en) | Terminal vibration evaluation method and apparatus in game scenario, medium, and device | |
| CN111798865A (zh) | 终端马达的检测方法、装置、终端及存储介质 | |
| CN112985867B (zh) | 舵机测试方法、装置、设备及存储介质 | |
| CN113704044B (zh) | 一种接口验证系统、方法及芯片 | |
| WO2022100075A1 (zh) | 性能检测方法、装置、电子设备和计算机可读介质 | |
| KR20220056782A (ko) | 기계 모니터링을 위한 시스템과 방법 | |
| KR102202126B1 (ko) | 전류센서 및 유동감지센서 패턴정보를 이용한 장비 이상유무 판단 방법 및 장치 | |
| CN109029696B (zh) | 共振检测方法、装置及存储介质 | |
| CN114666258B (zh) | 接口测试方法、装置、电子设备和介质 | |
| KR102275571B1 (ko) | 스마트 단말 기반 소리 및 진동을 이용한 기기 진단 서비스 제공 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: 18841150 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 18841150 Country of ref document: EP Kind code of ref document: A1 |