Machine diagnostic apparatus and machine diagnostic method

• US 10,295,995 B2
• Filed: 01/20/2015
• Issued: 05/21/2019
• Est. Priority Date: 01/20/2015
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Hiruta, Tomoaki, Maki, Kohji, Kato, Tetsuji, Atarshi, Yoshitaka
• Primary Examiner(s): Khuu, Hien D
• Application Number:
  US15/544,298
• Publication Number:
  US 20180011480A1
• Time in Patent Office: 1,582 Days
• Field of Search: 702183
• CPC Class Codes: G01R 31/343 in operation
• CPC Class Codes: G05B 23/0208 characterized by the configuration of the monitoring system
• CPC Class Codes: G05B 23/0227 Qualitative history assessment, whereby the type of data acted upon, e.g. waveforms, images or patterns, is not relevant, e.g. rule based assessment; if-then decisions

Abstract

An operation mode specifying unit specifies an operation mode of a machine by comparing time-series data of an amplitude and a frequency of measurement data obtained from a sensor with definition data of the operation mode of the machine created in advance by an operation mode data creation unit. In addition, an abnormality diagnosis unit performs processing of cluster analysis for the measurement data obtained from the sensor or the like, and diagnoses abnormality of the machine according to diagnosis procedure information that is set in advance depending on the set operation mode and an abnormality mode of the machine.

Claims

1. A machine abnormality diagnostic apparatus for an existing machine when an operation mode is unknown comprising:

a storage device;

an input/output device;

a display device; and

a processor communicatively coupled to the storage device, the input/output device and the display device, wherein the processor;

acquires via the input/output device, time-series measurement data measured by a sensor from the sensor attached to the existing machine having one or more operation modes;

stores the time-series measurement data in the storage device;

fits time-series measurement data into operation mode categories, wherein each operation mode category is defined by at least one extraction condition;

determines whether the at least one extraction condition is input via the input/output device;

selects one piece of time-series measurement data from the time-series measurement data for defining the one or more operation modes of the existing machine;

acquires from the one piece of time-series measurement data amplitude, frequency and change of amount of time;

determines if amplitude, frequency and change of amount of time for every piece of the time-series measurement data is acquired;

creates operation mode data using the acquired amplitude, frequency and change of amount of time for every piece of the time-series measurement data;

saves the operation mode data in the storage device;

acquires the time-series measurement data used for creating the operation mode data;

converts the time-series measurement data into time-series measurement data of amplitude and frequency;

compares the time-series measurement data of amplitude and frequency with the saved operation mode data;

specifies the operation mode of the existing machine when the comparison is a match and save the operation mode of the existing machine in the storage device;

wherein the storage device stores abnormality mode data which includes a plurality of abnormality modes, and each abnormality mode has at least one corresponding operation mode and a diagnosis procedure;

wherein the processor compares the operation mode of the existing machine to a matching operation mode found in the abnormality mode data;

selects the diagnosis procedure that corresponds to the matching operation mode; and

runs the selected diagnosis procedure on the time-series measurement data saved on the storage device to determine an abnormality of the time-series measurement data.

2. The machine diagnostic apparatus according to claim 1,

wherein the processor via the display device displays the time-series data of the amplitude and the frequency of the time-series measurement data acquired in the past when creating the operation mode data, reads the extraction conditions for extracting each operation mode category that a user inputs based on the time-series data of the displayed amplitude and the frequency, and creates the operation mode data using information of a time period designated by the read extraction conditions and the time-series data of the amplitude and the frequency of the measurement data included in the time period.

3. The machine diagnostic apparatus according to claim 1,

wherein when the abnormality of the machine is detected, the processor via the display device displays diagnosis result data configured to include the abnormality mode out of the plurality of abnormality modes and the operation mode of the existing machine that is specified at a time point at which the abnormality is detected.

4. The machine diagnostic apparatus according to claim 3,

wherein the processor via the display displays the acquired time-series measurement data or a frequency analysis result thereof in a time range including a time point at which the abnormality is detected in a case where an input for selecting one piece of diagnosis result data among one or more pieces of diagnosis result data displayed in the display device is received.

5. A machine abnormality diagnostic method for an existing machine when an operation mode is unknown,

wherein a computer that is connected to a sensor attached to the existing machine executes:

a sensor data acquisition process of;

acquiring time-series measurement data measured by the sensor from the sensor, and

storing the time-series measurement data in a storage device;

an operation mode data creation process;

fitting time-series measurement data into operation mode categories, wherein each operation mode category is defined by at least one extraction condition,

determining whether the at least one extraction condition is input via an input/output device,

selecting one piece of time-series measurement data from the time-series measurement data for defining the one or more operation modes of the existing machine,

acquiring from the one piece of time-series measurement data amplitude, frequency and change of amount of time,

determining if amplitude, frequency and change of amount of time for every piece of the time-series measurement data is acquired,

creating operation mode data using the acquired amplitude, frequency and change of amount of time for every piece of the time-series measurement data, and

saving the operation mode data in the storage device;

an operation mode specifying process of;

acquiring the time-series measurement data used for creating the operation mode data,

converting the time-series measurement data into time-series measurement data of amplitude and frequency,

comparing the time-series measurement data of amplitude and frequency with the saved operation mode data, and

specifying the operation mode of the existing machine when the comparison is a match and save the operation mode of the existing machine in the storage device; and

an abnormality diagnosis process of;

storing abnormality mode data on the storage device, which includes a plurality of abnormality modes, and each abnormality mode has at least one corresponding operation mode and a diagnosis procedure,

comparing the operation mode of the existing machine to a matching operation mode found in the abnormality mode data;

selecting the diagnosis procedure that corresponds to the matching operation mode, and

running the selected diagnosis procedure on the time-series measurement data saved on the storage device to determine an abnormality of the time-series measurement data.

6. The machine diagnostic method according to claim 5,

wherein in the operation mode data creation process, the computer displays the time-series data of the amplitude and the frequency of the time-series measurement data acquired in the past via the sensor data acquisition process to a display device when creating the operation mode data, reads the extraction conditions for extracting each operation mode that a user inputs based on the time-series data of the displayed amplitude and the frequency, and creates the operation mode data using information of a time period designated by the read extraction conditions and the time-series data of the amplitude and the frequency of the measurement data included in the time period.

7. The machine diagnostic method according to claim 6,

wherein in the abnormality diagnosis process, when abnormality of the existing machine is detected, the computer displays diagnosis result data configured to include the abnormality mode out of the plurality of abnormality modes and the operation mode of the existing machine that is specified at a time point at which the abnormality is detected to the display device.

8. The machine diagnostic method according to claim 7,

wherein in the abnormality diagnosis process, the computer further displays the time-series measurement data acquired by the sensor data acquisition process or a frequency analysis result thereof in a time range including a time point at which the abnormality is detected in a case where an input for selecting one piece of diagnosis result data among one or more pieces of diagnosis result data displayed in the display device is received.