Abnormality diagnosis method and device therefor
In industrial machine abnormality diagnosis, if the machine is diagnosed to have abnormality, then sensor data from the machine needs to be sent to a management center for causal analysis. However, since machines operated at a remote site cannot always communicate with a management center, it has been found that, in some cases, sensor data that has failed to be sent from a machine remains in the memory of the machine, resulting in lack of available memory capacity. In view of this, the present invention determines beforehand whether the diagnosed machine will run out of available memory capacity before the completion of sending the amount of sensor data required for causal analysis for the machine, and instructs a maintenance person to recover memory. This determination as to whether the machine will run out of available memory capacity before the completion of sending the amount of sensor data required for the causal analysis for the machine, is made as follows: (1) first, the machine predicts the run-out date on which the machine will run out of memory capacity for storing sensor data generated in the machine, and sends a notification of the predicted run-out date to the management center for the machine; and (2) next, from the amount of sensor data required for the causal analysis and the reception rate of sensor data, the management center calculates the number of days required to retrieve the necessary data for the causal analysis and determines whether the management center can retrieve the data by the predicted run-out date.
1. An abnormality diagnosis device arranged with a machine, which is a construction machine or an industrial machine, to diagnose abnormalities in the machine, the abnormality diagnosis device comprising:
a first processor programmed to perform an abnormality diagnosis based on sensor data measured by sensors attached the machine using data mining, and produce abnormality diagnosis data;
a memory which, if an abnormality is found by the first processor during the abnormality diagnosis, receives and stores the sensor data corresponding to the abnormality;
a transmitter configured to transmit the abnormality diagnosis data and the sensor data to a data center through a communication channel arranged between the abnormality diagnosis device and the data center; and
a second processor programmed to predict an available capacity run-out date on which the memory will run out of remaining available capacity when a communication speed between the machine and the data center has dropped and not all unsent sensor data can be sent to the data center.
2. The abnormality diagnosis device according to claim 1 , further comprising
a third processor programmed to predict a data retrieval completion date on which retrieval of an amount of sensor data required for causal analysis is completed, based on the amount of data successfully sent to the data center.
3. The abnormality diagnosis device according to claim 2 , wherein
prediction results of the available capacity run-out date of the memory and the data retrieval completion date are compared with each other, and a determination is made whether or not the memory will run out of the available capacity before the retrieval of the sensor data required for the causal analysis is completed.
4. An abnormality diagnosis method comprising:
receiving, in an abnormality diagnosis device, sensor data from sensors attached to a machine, which is a construction machine or an industrial machine;
carrying out an abnormality diagnosis based on the sensor data using data mining to produce abnormality diagnosis data;
if an abnormality is found during the abnormality diagnosis, receiving and storing the sensor data corresponding to the abnormality in a memory;
sending the abnormality diagnosis data and the sensor data to a data center through a communication channel arranged between the abnormality diagnosis device and the data center; and
predicting a run-out date on which a memory of the machine runs out of remaining available capacity when a communication speed between the machine and the data center has dropped and not all unsent sensor data can be sent to the data center.
5. The abnormality diagnosis method according to claim 4 , further comprising
predicting a data retrieval completion date on which retrieval of an amount of sensor data required for causal analysis is completed, based an amount of data successfully sent to the data center.
6. The abnormality diagnosis method according to claim 5 , wherein
prediction results of the available capacity run-out date of the memory and the data retrieval completion date are compared with each other, and a determination is made whether or not the memory will run out of the available capacity before the retrieval of the sensor data required for the causal analysis is completed.