JP2017182414A - Failure prediction device, failure prediction method, and failure prediction program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately predict failure occurring in an object device.SOLUTION: A failure prediction device comprises acquisition means 11a for acquiring history of occurrence frequency of abnormal phenomena in each unit of the device and causing alarm data storage means to store the history as alarm data; state transition data storage means 12b for storing a condition of transition of a prediction level according to the occurrence frequency of abnormal phenomena; failure prediction data storage means 12c for accumulating and storing a prediction level for each unit; failure prediction means 11b for obtaining a new prediction level for each unit, as a prediction result, according to the state transition data with respect to the current prediction level included in the failure prediction data and the occurrence frequency of abnormal phenomena included in the alarm data; and prediction data update means 11c for updating the failure prediction data using the prediction result.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a failure prediction device, a failure prediction method, and a failure prediction program for predicting a failure occurring in a target device.

  In recent years, device failure prediction has been performed in devices of various scales including plants and the like. By predicting a failure occurring in the apparatus, it is possible to prevent the apparatus from being stopped due to the occurrence of the failure.

  As a device failure prediction method, there is a method of collecting records of failures that have occurred in the past and formulating the huge number of failure records. In this method, the prediction accuracy is poor when the recorded data collection period is short or when there is a small amount of data to be recorded due to the fact that no failure has occurred. Also, the prediction accuracy is poor when there is an error in the data. Furthermore, at the beginning of the operation of the apparatus or when a device or the like is added in the apparatus, there is no failure record and the failure cannot be predicted.

  Further, if the occurrence of a failure is predicted using the failure record and the device is repaired or the like, it can be said that the possibility of failure of the corresponding item is extremely low. However, since the failure record does not change, in the newly obtained prediction, it is predicted that there is still a high possibility of a failure even though the possibility of the failure is low, and the reliability of the prediction result is low.

  As a method of predicting plant failure, there is a technology that collects sensor values measured in the plant, detects failure signs, and notifies them when the values exceed the limit values, rather than using failure records. Yes (see, for example, Patent Document 1). The failure prediction method described in Patent Literature 1 can automatically predict a failure using a sensor value and notify that it is a predetermined time before the occurrence of an abnormality. However, the failure prediction result obtained only from the current sensor value does not reflect the past progress, and is not sufficiently accurate.

  In addition, when the failure prediction result is reported in the form of a percentage indicating the possibility of failure occurrence or the probability of occurrence, how to interpret the prediction result by the operator referring to the result depends on experience and knowledge Different. Therefore, the response according to the prediction result depends on the experience and knowledge of the operator, the response of each operator is not unified, and even if the same situation occurs, different responses may be taken.

JP 2009-75692 A

  As described above, conventionally, it has been difficult to perform highly accurate failure prediction and to enable unified response from the failure prediction.

  In view of the above problems, an object of the present invention is to provide a failure prediction device, a failure prediction method, and a failure prediction program capable of performing failure prediction with high accuracy and making a unified response according to the result.

  In order to achieve the above object, the first invention is a failure prediction apparatus for predicting the occurrence of a failure in a device composed of a plurality of devices, wherein at least sensor measurement values are input from each device and input The history of the number of occurrences of abnormal phenomena in each device for each predetermined period is acquired from an external device that collects and manages as abnormal phenomena when the data meets a predetermined abnormality determination condition, and is stored in the alarm data storage means as alarm data The acquisition means to be stored, and the prediction level that expresses the degree of possibility of failure in each device and the necessity of operator's work to prevent the occurrence of failure due to abnormal phenomenon occurring in the device, step by step, State transition data storage means for storing transition conditions according to the number of occurrences of abnormal phenomena in each device as state transition data, and a prediction level predicted for each device Failure prediction data storage means that accumulates and stores as failure prediction data, and the current prediction level included in the failure prediction data and the number of occurrences of abnormal phenomena included in the alarm data, according to the state transition data, for each device A failure prediction means for obtaining a new prediction level as a prediction result of occurrence of a failure, and a prediction data update means for updating the failure prediction data with the prediction result obtained by the failure prediction means. .

  In addition, the second invention corresponds to the device when the record data indicating that the operation for preventing the failure of the device is executed according to the prediction result is input from the input terminal used by the operator to the device. A reset means for storing the identifier of the prediction level to be stored in the reset data storage means as reset data, and the failure prediction means, when obtaining a new prediction level, of the current prediction level included in the failure prediction data Among them, the predicted level of the identifier included in the reset data is replaced with an initial value with a low possibility of occurrence of failure and the necessity of work to prevent the occurrence of failure, and the number of occurrences of abnormal phenomena included in the alarm data The occurrence number of the abnormal phenomenon corresponding to the identifier included in the reset data is replaced with zero.

  According to a third aspect of the present invention, when the failure prediction means predicts a failure of a new device that is not included in the failure prediction data, the possibility of occurrence of failure and the necessity of work for preventing failure are low. The value is used as the current prediction level of the device.

  According to a fourth aspect of the present invention, the failure prediction data is data that associates an identifier of a device, a period, and a prediction level of the device obtained from a history of abnormal phenomena that have occurred in the period. The updating means updates the failure prediction data by adding a new prediction level and a new occurrence period specifying the prediction level.

  According to a fifth aspect of the present invention, a prediction level of a device by the failure prediction means, a prediction level for a predetermined period obtained for the device included in the failure prediction data, and occurrence of an abnormal phenomenon used for specifying the prediction level The apparatus further comprises a result output means for outputting data including the history of the number of times to the output terminal used by the operator as a prediction result.

  In addition, according to a sixth aspect of the present invention, when the reset signal is input from the input terminal used by the operator to operate the reset of the prediction level for any of the devices, the reset means includes the failure prediction data. The predicted level of the device specified by the reset signal is updated to an initial value with a low possibility of failure occurrence and the necessity of work to prevent the failure occurrence, and the past abnormal phenomenon of the device included in the alarm data is also updated. The number of occurrences is updated to zero.

  Further, the seventh invention is characterized in that when the acquisition means acquires simulation data of the number of occurrences of an alarm used for simulation from an input terminal used by an operator, the alarm data is updated with the simulation data. To do.

  The eighth invention is a failure prediction method for predicting the occurrence of a failure in an apparatus composed of a plurality of devices, wherein at least sensor measurement values are input from each device, and the input data is a predetermined abnormality. Acquiring a history of the number of occurrences of abnormal phenomena in each device for each predetermined period from an external device that is collected and managed as an abnormal phenomenon when the judgment condition is met, and storing the history in the alarm data storage means as alarm data; The degree of possibility that a failure will occur in each device included in the failure prediction data stored in the prediction data storage means and the degree of necessity of the operator's work to prevent the occurrence of failure due to an abnormal phenomenon occurring in the device Each unit stored in the state transition data storage means for the current prediction level expressed in stages and the number of occurrences of abnormal phenomena included in the alarm data In accordance with the state transition data which is a condition for transition of the prediction level according to the number of occurrences of the abnormal phenomenon in the step, a new prediction level is obtained for each device as a prediction result of the failure occurrence, and the failure And a step of updating the prediction data.

  The invention of claim 9 includes a plurality of devices from an external device that receives at least sensor measurement values from each device and collects and manages as an abnormal phenomenon when the input data meets a predetermined abnormality determination condition. A failure prediction program for predicting the occurrence of a failure in a device comprising: an acquisition means for acquiring a history of the number of occurrences of abnormal phenomena in each device for each predetermined period, and storing the history as alarm data in an alarm data storage means; The degree of possibility that a failure will occur in each device included in the failure prediction data stored in the failure prediction data storage means and the necessity of the operator's work to prevent the occurrence of a failure due to an abnormal phenomenon occurring in the device Stored in the state transition data storage means for the current prediction level representing the degree in steps and the number of occurrences of abnormal phenomena included in the alarm data A failure prediction means for obtaining a new prediction level as a failure occurrence prediction result for each device in accordance with state transition data which is a condition for transition of a prediction level according to the number of occurrences of abnormal phenomena in each device, and the failure prediction means The information processing apparatus is made to realize prediction data update means for updating the failure prediction data based on the prediction result obtained in step (b).

  According to the present invention, it is possible to predict failure with high accuracy and to make a unified response according to the result.

It is a block diagram explaining the structure of the failure prediction apparatus which concerns on 1st Embodiment. It is a data block diagram of the alarm data which the failure prediction apparatus of FIG. 1 utilizes. It is a data block diagram of the failure prediction data which the failure prediction apparatus of FIG. 1 produces | generates. It is a data block diagram of the prediction result which the failure prediction apparatus of FIG. 1 produces | generates. It is a flowchart explaining the failure prediction process in the failure prediction apparatus of FIG. It is a flowchart explaining the change process in the failure prediction apparatus of FIG. It is a flowchart explaining the simulation process in the failure prediction apparatus of FIG.

  Below, the failure prediction apparatus which concerns on embodiment of this invention is demonstrated. The failure prediction apparatus according to the present invention predicts the occurrence of a failure in a target device composed of a plurality of devices, and provides the prediction result to an operator such as an inspection maintenance staff. Based on the prediction result, the operator can prevent a failure in the apparatus by repairing the equipment constituting the apparatus. Moreover, when the failure prediction apparatus 10 inputs the simulation value of the alarm of each apparatus 20a-20z of the apparatus 20, it can obtain the simulation result at the time of occurrence of each alarm.

  Description will be made assuming that the failure prediction apparatus 10 predicts a failure of the apparatus 20 including the plurality of devices 20a to 20z with reference to FIG. The device 20 that is predicted to fail is connected to an alarm data collection management device 30 that collects the occurrence history of abnormal phenomena in the devices 20a to 20z. Further, the failure prediction device 10 predicts a failure of the target device 20 based on alarm data that is an occurrence history of an abnormal phenomenon received from the alarm data collection management device 30. Furthermore, the failure prediction apparatus 10 is connected to an output terminal 40 used by an operator, and transmits a failure prediction result to the output terminal 40. The failure prediction apparatus 10 is connected to an input terminal 50 used by an operator, and receives various data and requests from the input terminal 50.

  Here, the “abnormal phenomenon” refers to a phenomenon in which the devices 20a to 20z constituting the device 20 are not operating normally even if it cannot be immediately determined that the target device 20 has failed. When this abnormal phenomenon occurs a plurality of times or when it occurs for a long time, the apparatus 20 may break down, so that the occurrence of the abnormal phenomenon can be considered as a sign of failure. For example, when the operation of the devices 20a to 20z is not completed within a certain time, it can be considered as an abnormal phenomenon. Further, for example, when the measured value of the sensor exceeds the limit value, it can be considered as an abnormal phenomenon. Furthermore, it can be considered an abnormal phenomenon when the total of the sensor measurement values for the two devices exceeds the predetermined range even though the sensor measurement values for the two devices are within the predetermined range.

<Alarm data collection management device>
The alarm data collection management device 30 is connected to the device 20 that is the target of failure prediction, detects an abnormal phenomenon that occurs in each device 20a to 20z that constitutes the device 20, and determines the type and time of occurrence of the abnormal phenomenon that has occurred. The history is stored in the storage device and managed. In addition, the alarm data collection management device 30 transmits the number of occurrences of each abnormal phenomenon occurring in a predetermined period to the failure prediction device 10 as alarm data.

  The alarm data collection management device 30 receives sensor measurement values and the like from the device 20 and collects and manages them as an abnormal phenomenon when a predetermined abnormality condition is met. For example, the alarm data collection management device 30 transmits alarm data to the failure prediction device 10 in response to a request signal received from the failure prediction device 10. Alternatively, the alarm data collection management device 30 transmits alarm data to the failure prediction device 10 at regular timing.

<Output terminal / Input terminal>
The output terminal 40 is, for example, a display display or a printer that is used by an operator to confirm the failure prediction obtained by the failure prediction apparatus 10.

  The input terminal 50 is, for example, a keyboard, a mouse, a touch panel, and operation buttons that are used by an operator to input data and operation signals.

  The output terminal 40 and the input terminal 50 may be an integrated device. In addition, since a plurality of operators may hold different output terminals and input terminals, a plurality of output terminals 40 and input terminals 50 may exist.

<Failure prediction device>
As shown in FIG. 1, the failure prediction apparatus 10 includes an acquisition unit 11 a that acquires a history of the number of occurrences of abnormal phenomena in each device 20 a to 20 z for each predetermined period, and a failure has occurred in each device for each device 20 a to 20 z. A failure prediction means 11b that obtains a prediction level as a prediction result of occurrence of failure, stepwise indicating the degree of possibility of failure and the level of necessity of operator's work to prevent occurrence of failure due to abnormal phenomenon occurring in the device; From the prediction results obtained, the prediction data update means 11c for updating the failure prediction data 12c, the result output means 11d for outputting the prediction results obtained by the failure prediction means 11b to the output terminal 40, and the input terminal 50, When the record data indicating that the work for preventing the occurrence of the failure is executed according to the prediction result is input, the identifier of the corresponding prediction level is used as the reset data 12d. And a reset means 11e for storing the 憶 device 12. In this embodiment, it is assumed that the work mode is defined by expressing the prediction level with alphabets or numbers. This definition is intended to unify the operations of maintenance personnel.

  The failure prediction device 10 is an information processing device including a CPU 11 and a storage device 12. In this failure prediction device 10, the failure prediction program P stored in the storage device 12 is read and executed, so that the CPU 11 obtains the acquisition means 11a, failure prediction means 11b, prediction data update means 11c, and result output means 11d. Then, processing as the reset unit 11e and the change unit 11f is executed.

  In addition to the failure prediction program P, the storage device 12 stores alarm data 12a, state transition data 12b, failure prediction data 12c, and reset data 12d. Here, “reset data 12d” refers to, for example, an identifier for identifying the prediction level column of failure prediction data 12c (hereinafter referred to as “prediction level identifier”).

  Note that the failure prediction apparatus 10 does not need to be configured by one information processing apparatus, and may be configured by a plurality of information processing apparatuses. For example, each of the data 12a to 12c may be stored in an external server device that constitutes an information processing device different from the CPU 11, and each of the data 12a to 12c is stored in a plurality of different server devices. Also good.

<Alarm data>
For example, as shown in FIG. 2A, the alarm data 12a includes “period” and “number of times each abnormal phenomenon (first alarm to n-th alarm) has occurred in the devices 20a to 20z of the apparatus 20 during the period. Is associated with the data. In the example shown in FIG. 2A, one day is set as the period. Therefore, for example, on “December 15, 2015”, an abnormal phenomenon defined as “first alarm” occurs “once” and an abnormal phenomenon defined as “second alarm” occurs on “six times”. It can be seen that the “third alarm” and the “nth alarm” have not occurred. Also, on “December 14, 2015”, an abnormal phenomenon defined as “second alarm” occurs “three times” and an abnormal phenomenon defined as “nth alarm” occurs “twice”. It can be seen that the “first alarm” and the “third alarm” have not occurred.

<State transition data>
The state transition data 12b includes a “device identifier” that identifies a target device, a “device identifier” that identifies a device included in the device, a “state change” that defines a transition of the prediction level of the device, a prediction “Logic” that specifies the conditions for obtaining the level transition, the number of occurrences of each abnormal phenomenon (the first alarm to the nth alarm) that are used in the logic, and the logic The “monitoring period” and the “number of occurrences within the monitoring period” used in the above are associated with each other.

  In FIG. 1, only one device 20 is connected to the alarm data collection management device 30, and the state transition data 12b indicates the transition of the prediction level of this one device 20 (identifier “device A”). It is data to prescribe. A plurality of devices 20 may be connected to the alarm data collection management device 30, and a plurality of alarm data collection management devices 30 may be connected to the failure prediction device 10. When there are a plurality of devices 20 as described above, the state transition data 12b defines the transition of the prediction level of each device 20.

<< Failure prediction data >>
For example, as shown in FIG. 3A, the failure prediction data 12c is data in which a “period” is associated with a “prediction level” obtained from a history of abnormal phenomena (alarms) that occurred during the period. In the example shown in FIG. 3A, a unit of one day is set as the period.

  For example, a prediction level of “December 15, 2015” is obtained from an alarm generated before “December 14, 2015” and a prediction level of “December 14, 2015”. It can be seen that the “first prediction level” is “0”, the “second prediction level” is “2”, the “third prediction level” is “0”, and the “m-th prediction level” is “1”. Temporarily, the first prediction level is a possibility of failure relating to the device a, the second prediction level is a possibility of failure relating to the device b, the third prediction level is a possibility of failure relating to the device c, and the mth prediction level is a failure relating to the device z. If there is a possibility, the devices a and c are in a state where there is no possibility of failure.

<< Acquisition means >>
The acquisition unit 11a acquires alarm data from the alarm data collection management device 30 that collects and manages the history of abnormal phenomena. At this time, the acquisition unit 11a transmits a request signal for requesting transmission of new alarm data to the alarm data collection management device 30, and receives alarm data transmitted in response thereto.

  For example, the acquisition unit 11a acquires alarm data at regular timing (for example, every day, every 3 hours, etc.). Alternatively, the acquisition unit 11a may acquire the alarm data at a timing when the data is updated in the alarm data collection management device 30 due to the occurrence of the abnormal phenomenon notification.

  The acquisition unit 11 a adds new alarm data acquired from the alarm data collection management device 30 and updates the alarm data 12 a stored in the storage device 12. At this time, the acquisition unit 11a compares the acquired new alarm data with the alarm data 12a already stored in the storage device 12, adds the history that occurred after the previous update of the alarm data 12a, and the alarm data 12a Update.

  For example, when the alarm data 12a including the alarm history “November 1, 2015 to December 15, 2015” as shown in FIG. Assume that alarm data that is a history of alarms that occurred between “November 1, 2015 and December 16, 2015” is acquired. Since the alarm data 12a already includes the history up to “December 15, 2015”, the acquisition unit 11a includes only the history of the alarm “December 16, 2015” as shown in FIG. The alarm data 12a is updated by adding.

  Note that, when transmitting the request signal, the acquisition unit 11a may specify a period that is already included in the alarm data 12a, and may transmit a request signal that requests transmission of alarm data related to a history of alarms that have occurred thereafter. In this case, the acquisition unit 11a updates the alarm data 12a by adding the received alarm data.

《Failure prediction means》
The failure prediction means 11b reads out the alarm data 12a from the storage device 12, and also reads out the state transition data 12b and the failure prediction data 12c, and uses the alarm data 12a, the failure prediction data 12c, and the state transition data 12b, and each of the devices 20a to 20b. A prediction level representing the degree of possibility of failure occurring at 20z and the degree of necessity of operator's work to prevent the occurrence of failure due to the abnormal phenomenon occurring in each device 20a-20z is obtained.

  For example, the failure prediction unit 11b obtains a new prediction level at regular timing. Further, for example, the failure prediction unit 11b obtains a new prediction level at a timing when the acquisition unit 11a updates the alarm data 12a of the storage device 12.

  Specifically, the failure prediction unit 11b calculates the prediction level of the device when the number of occurrences of each alarm stored in the alarm data 12a satisfies the logic condition associated with each device in the state transition data 12b. Transition according to

<< Predictive data update means >>
When the failure prediction unit 11b obtains a new prediction level, the prediction data update unit 11c adds a new prediction level and updates the failure prediction data 12c stored in the storage device 12.

  For example, as shown in FIG. 3A, when failure prediction data 12c including prediction results from “November 1, 2015 to December 15, 2015” is stored in the storage device 12, failure prediction means Assume that 11b obtains a new prediction level using alarm data including a history of alarms generated on “December 16, 2015”. In this case, as shown in FIG. 3B, the prediction data updating unit 11c adds the new prediction level obtained from the history of the alarm that occurred on “December 16, 2015” to obtain the failure prediction data 12c. Update.

  Here, if an identifier of any prediction level is stored as the reset data 12d, the failure prediction unit 11b sets the level of the prediction level identifier to “0” when obtaining a new prediction level. In addition, the number of occurrences of abnormal phenomena corresponding to the prediction level is replaced with “0”. That is, when the failure prediction means 11b predicts a failure of a new device that is not included in the failure prediction data 12c, the failure prediction means 11b sets an initial value with a low possibility of occurrence of failure and a low necessity of work for preventing occurrence of the failure. Use as a prediction level for.

<< Result output means >>
When a new prediction level is obtained by the failure prediction unit 11b, the result output unit 11d transmits a prediction result including the new prediction level to the output terminal 40. At this time, the result output unit 11d extracts the prediction level of the past predetermined period (for example, the past three weeks) from the failure prediction data 12c stored in the storage device 12, and adds it to the newly obtained prediction level, A prediction result including the extracted past prediction level can be output.

  For example, as shown in FIG. 4, the result output unit 11 d outputs, to the output terminal 40 as the result data, the “number of occurrences of alarm” generated during a predetermined period and the “prediction level” obtained during the period, as shown in FIG. can do. The example shown in FIG. 4 shows the “number of occurrences” of the first to n-th alarms that occurred from “November 26, 2015 to December 16, 2015”, and the “predicted level” (first To m-th prediction level). The operator can grasp the history of occurrence of alarms in a predetermined period and the fluctuation of the obtained prediction level. Therefore, an operator who refers to the result data output to the output terminal 40 can be used for grasping the status of the apparatus. Note that the result data preferably includes a history and a prediction level for a period (for example, three weeks) sufficient for the operator to grasp the situation. Further, the operator can know the degree of the possibility of the occurrence of the failure and the necessity of the operator's work for preventing the occurrence of the failure due to the abnormal phenomenon simply by referring to only the latest prediction level. Therefore, a unified response can be made for any operator regardless of the experience and knowledge of the operator.

<Reset means>
When the reset means 11e receives from the input terminal 50 the corresponding record data corresponding to the device for which an alarm has occurred in the failure prediction means 11b, the reset level 11e sets the associated prediction level in the failure prediction data 12c to “0” in the subsequent calculation. And the prediction level identifier is stored as reset data 12d so that the number of occurrences of the associated alarm is reset to “0” and the calculation is performed. The reset unit 11e resets the number of occurrences of the alarm and the prediction level for obtaining the prediction level to “0”, but the number of occurrences of the alarm displayed on the output terminal and the prediction level are reset to “0”. There is no. Since the number of occurrences of alarms and the predicted level already displayed indicate actual results, they are not changed.

  The case where the failure prediction data 12c in FIG. 3B and the alarm data 12a in FIG. 2B are stored in the storage device will be described. In the failure prediction data 12c shown in FIG. 3B, the “second prediction level” is “3”, and the result output unit 11d sends the result data including the value of the “second prediction level” to the output terminal 40. Send. For example, the operator executes a response related to the “second prediction level” to the device 20 in accordance with the result data, so that a failure can be dealt with in advance. Further, after the response, the operator operates the input terminal 50 to input the corresponding record data. When receiving the corresponding record data from the input terminal 50, the reset unit 11e stores the prediction level identifier “second prediction level” as the reset data 12d. Based on the reset data 12d, in the subsequent failure prediction processing, the value of the “second prediction level” is reset to “0” for calculation.

  Further, it is assumed that the “second prediction level” is determined according to the “second alarm”. At this time, the number of occurrences of “second alarm” in the alarm data 12a as shown in FIG. 2B is reset to “0”. Each prediction level may be attributed to a plurality of alarms. In this case, the number of occurrences of the plurality of alarms is reset to “0”. In this way, the reset means 11e resets all the prediction levels and alarm histories related to the alarms that have been dealt with, so that the prediction level to be obtained later using these can be influenced by the effects of alarms that have occurred in the past in the dealt equipment. I will not receive it. The reset unit 11e resets the number of occurrences of the alarm and the prediction level for obtaining the prediction level to “0”, but the number of occurrences of the alarm displayed on the output terminal and the prediction level are reset to “0”. There is no. Since the number of occurrences of alarms and the predicted level already displayed indicate actual results, they are not changed.

  Further, when the reset unit 11e receives from the input terminal 50 a reset signal requesting that the value of the prediction level is reset to “0” and calculated, the reset unit 11e stores the prediction level in the storage device 12 as the reset data 12d. Store the identifier. The reset signal that requests a change in the value of the prediction level includes an identifier of the prediction level to be reset.

  Moreover, the reset means 11e can change the frequency | count of occurrence of the alarm data 12a, when the reset signal which requests | requires the change of the frequency | count of occurrence of the alarm data 12a from the input terminal 50 is received. The reset signal for requesting the change in the number of occurrences includes an identifier of the alarm to be reset. Note that the reset signal may request to change both the prediction level value and the number of occurrences of the alarm, and thus may include both the prediction level identifier and the alarm identifier.

《Changing means》
When the request signal for requesting the change of the parameter of the state transition data 12b is input from the input terminal 50, the changing unit 11f changes the state transition data 12b of the storage device 12 according to the change request. For example, when a request signal for changing a part of the logic value of the state transition data 12b is input, each value is changed, and in the subsequent processing, failure prediction processing is executed under the changed condition. The request signal includes items and values of parameters to be changed.

<Failure prediction processing>
With reference to the flowchart shown in FIG. 5, the failure prediction process performed in the failure prediction apparatus 10 is demonstrated. First, the acquisition unit 11a acquires alarm data from the alarm data collection management device 30 (S1). The acquisition unit 11a stores the acquired alarm data in the storage device 12 as alarm data 12a (S2).

  Next, the failure prediction means 11b executes failure prediction using the alarm data 12a, the state transition data 12b, and the failure prediction data 12c in the storage device 12 (S3). At this time, since no failure prediction has been made for the target device 20 in the past, the failure prediction means 11b executes failure prediction assuming that each prediction level of the failure prediction data 12c is the initial value “0”. To do. That is, when the failure prediction means 11b predicts a failure of a new device that is not included in the failure prediction data 12c, the failure prediction means 11b sets an initial value with a low possibility of occurrence of failure and a low necessity of work for preventing occurrence of the failure. Use as a prediction level for.

  The prediction data update unit 11c stores the prediction result obtained in step S3 in the storage device 12 as failure prediction data 12c (S4). The result output unit 11d outputs the result data including the prediction result obtained in step S3 to the output terminal 40 (S5).

  When the response record data handled by the operator according to the prediction result output to the output terminal 40 in step S5 is received (YES in S6), the reset unit 11e sets the identifier of the corresponding prediction level as the reset data 12d. The data is stored in the storage device 12 (S7).

  When the corresponding record data is not received for the prediction result output in step S5 (NO in S6), or after the reset data 12d is stored in step S7, the acquisition unit 11a again returns from the alarm data collection management device 30. Alarm data is acquired (S8). Further, the acquisition unit 11a adds the acquired alarm data to the storage device 12 and updates the alarm data 12a (S9).

  Subsequently, the failure prediction unit 11b executes failure prediction using the alarm data 12a, the state transition data 12b, and the failure prediction data 12c in the storage device 12 (S10). Here, each prediction level of the failure prediction data 12c used by the failure prediction apparatus 10 is a value obtained in the past.

  The prediction data update unit 11c updates the failure prediction data 12c of the storage device 12 with the prediction result obtained in step S10 (S11). The result output unit 11d outputs the prediction result obtained in step S10 to the output terminal 40 (S12).

  The failure prediction apparatus 10 repeats the processes of steps S6 to S12 until the failure prediction is finished (S13).

  In this way, by using the alarm data 12a that is the history of abnormality and the current prediction level and obtaining a new prediction level according to the state transition data 12b, a highly reliable prediction result can be obtained.

  Note that the alarm data collection management device 30 accumulates alarms from the start of operation of the failure prediction target device 20, for example. Further, the failure prediction device 10 does not need to make a failure prediction at the same time as the start of operation. The failure prediction device 10 is installed after a certain period after the operation is started, and uses the alarm data accumulated in the alarm data collection management device 30 before the installation. Failure prediction can be started. The alarm data collection management device 30 operates independently of failure prediction. Therefore, even if there is a period during which the failure prediction device 10 is stopped due to maintenance or the like, the failure prediction device 10 inputs alarm data during the stop period from the alarm data collection management device 30 when the operation is resumed. Thus, the latest failure prediction can be executed.

<Change processing>
With reference to the flowchart shown in FIG. 6, the parameter changing process executed in the failure prediction apparatus 10 will be described. First, the changing unit 11f receives a request signal for requesting the change of the parameter of the state transition data 12b from the input terminal 50 (S21). The changing unit 11f can consider that a request signal has been received by putting the state transition data file in the folder.

  The changing unit 11f reads the state transition data 12b from the storage device 12, and selects a record to be changed in the request signal (S22). Further, the changing unit 11f updates the parameter value requested to be changed by the request signal in the record selected in step S22, and changes the state transition data 12b (S23).

  Further, the changing unit 11f repeats the processes of steps S22 to S23 until the change is completed for all the records that are requested to be changed by the request signal (S24).

  For example, it is assumed that the state transition data 12b is stored in the storage device 12, and the parameter value is requested to be changed in the request signal. In this case, the changing unit 11f changes the state transition data 12b.

  As described above, by enabling the parameter value of the state transition data 12b to be changed, when the target device for failure prediction is changed by adding or replacing the device in the target device 20, the parameter is changed according to the change of the device. Can be changed. Therefore, it is possible to flexibly cope with the change of the device, and it is possible to obtain a highly reliable prediction result.

<simulation>
As shown in FIG. 1, the failure prediction apparatus 10 can also execute a simulation of a transition of a prediction level at the time of occurrence of each alarm in the apparatus 20 by inputting simulation data from the input terminal 50.

  A simulation process executed in the failure prediction apparatus 10 will be described with reference to the flowchart shown in FIG. First, the acquisition unit 11a acquires simulation data input via the input terminal 50 (S31). Further, the acquisition unit 11a stores the acquired simulation data in the storage device 12 as alarm data 12a (S32). This simulation data includes a simulation value used in place of the alarm history.

  Next, the failure prediction unit 11b executes failure prediction using the alarm data 12a, the state transition data 12b, and the failure prediction data 12c in the storage device 12 (S33). Here, since the prediction level is obtained using the alarm data 12a composed of values for simulation, the obtained prediction result is the simulation result. At this time, when the prediction level used for the simulation is input via the input terminal 50 together with the input of the simulation data, the failure prediction unit 11b uses the value. When the prediction level for simulation is not input, the initial prediction level can be used, or the prediction level obtained in the past failure prediction can be used.

  The prediction data update unit 11c updates the failure prediction data 12c of the storage device 12 with the prediction result obtained in step S33 (S34). Thereby, the prediction level of the simulation result is stored in the storage device 12 as the failure prediction data 12c.

  The result output unit 11d outputs the prediction result obtained in step S33 to the output terminal 40 (S35). Thereby, the prediction level of the simulation result is output to the output terminal 40.

  When continuing the simulation process, the failure prediction device 10 repeats the processes of steps S31 to S35 with new simulation data (NO in S36). If the simulation process is to be terminated, the process is terminated (YES in S36). Thereafter, a change process for changing the parameter value of the state transition data 12b may be executed according to the result of the simulation. Ordinary failure prediction processing may be executed. However, simulation and normal failure prediction processing are managed as separate data.

  As mentioned above, although this invention was demonstrated in detail using each embodiment, this invention is not limited to embodiment described in this specification. The scope of the present invention is determined by the description of the claims and the scope equivalent to the description of the claims.

10 ... Failure prediction device 11 ... CPU
11a ... Acquisition means 11b ... Failure prediction means 11c ... Prediction data update means 11d ... Result output means 11e ... Reset means 11f ... Change means 12 ... Storage device 12a ... Alarm data 12b ... State transition data 12c ... Failure prediction data P ... Failure prediction Program 20 ... Device 20a to 20z ... Device 30 ... Alarm data collection management device 40 ... Output terminal 50 ... Input terminal

Claims (9)

A failure prediction device that predicts the occurrence of a failure in a device composed of a plurality of devices,
Input at least sensor measurement values from each device, and collect and manage as an abnormal phenomenon when the input data meets a predetermined abnormality determination condition. An acquisition means for acquiring a history of the number of occurrences and storing it in an alarm data storage means as alarm data;
An abnormal phenomenon in each device at a predicted level that indicates the degree of possibility of failure in each device and the necessity of operator's work to prevent the occurrence of failure due to the abnormal phenomenon that occurred in the device. State transition data storage means for storing transition conditions according to the number of occurrences as state transition data;
Failure prediction data storage means for accumulating prediction levels predicted for each device and storing them as failure prediction data;
Failure prediction for obtaining a new prediction level as a prediction result of failure occurrence for each device according to the state transition data with respect to the current prediction level included in the failure prediction data and the number of occurrences of abnormal phenomena included in the alarm data Means,
Prediction data update means for updating the failure prediction data with the prediction result obtained by the failure prediction means;
A failure prediction apparatus comprising: When record data indicating that an operation for preventing the occurrence of a failure of the apparatus has been performed on the device from the input terminal used by the operator is performed, the identifier of the prediction level corresponding to the device is reset. A reset means for storing the data in the reset data storage means as data;
When the failure prediction means obtains a new prediction level, out of the current prediction levels included in the failure prediction data, the prediction level of the identifier included in the reset data is determined based on the possibility of failure occurrence and the failure occurrence. Replacing the initial value with a low necessity for the work to prevent, and out of the number of occurrences of the abnormal phenomenon included in the alarm data, replacing the occurrence number of the abnormal phenomenon corresponding to the identifier included in the reset data with zero. The failure prediction apparatus according to claim 1, wherein the apparatus is a failure prediction apparatus. When the failure prediction means predicts a failure of a new device that is not included in the failure prediction data, the failure prediction means sets the initial value of the device to the initial value with a low possibility of failure occurrence and a low necessity of work to prevent the failure occurrence. It uses as a level. The failure prediction apparatus of Claim 1 or 2 characterized by the above-mentioned. The failure prediction data is data that associates an identifier of a device, a period, and a predicted level of the device obtained from a history of abnormal phenomena occurring in the period,
The said prediction data update means adds a new prediction level and the generation | occurrence | production period of the abnormal phenomenon utilized for specification of the said prediction level, and updates the said failure prediction data of Claim 1 thru | or 3 characterized by the above-mentioned. The failure prediction apparatus according to any one of the preceding claims. Data including the prediction level of the device by the failure prediction means, the prediction level of the predetermined period obtained for the device included in the failure prediction data, and the history of the number of occurrences of abnormal phenomena used for specifying the prediction level, The failure prediction apparatus according to claim 1, further comprising: a result output unit that outputs a prediction result to an output terminal used by an operator. When the reset signal for operating the reset of the prediction level for any device is input from the input terminal used by the operator, the reset means includes the device specified by the reset signal included in the failure prediction data. The prediction level is updated to an initial value with a low possibility of occurrence of failure and the necessity of work for preventing the occurrence of failure, and the number of occurrences of past abnormal phenomena of the device included in the alarm data is updated to zero. The failure prediction apparatus according to claim 2. The acquisition unit updates the alarm data with the simulation data when acquiring the simulation data of the number of occurrences of the alarm used for the simulation from the input terminal used by the operator. The failure prediction apparatus according to claim 1. A failure prediction method for predicting a failure occurrence of a device composed of a plurality of devices,
Occurrence of an abnormal phenomenon in each device every predetermined period from an external device that inputs at least sensor measurement values from each device and collects and manages as an abnormal phenomenon when the input data meets a predetermined abnormality determination condition Acquiring a history of the number of times and storing it in the alarm data storage means as alarm data;
Degree of possibility of failure in each device included in failure prediction data stored in failure prediction data storage means and degree of necessity of operator's work to prevent failure due to abnormal phenomenon occurring in the device The transition of the prediction level according to the number of occurrences of abnormal phenomena in each device stored in the state transition data storage means with respect to the current prediction level that represents stepwise and the number of occurrences of abnormal phenomena included in the alarm data In accordance with the state transition data that is the condition of
Updating the failure prediction data with the obtained prediction results;
A failure prediction method comprising: A failure prediction program for predicting the occurrence of a failure in a device composed of a plurality of devices,
Occurrence of an abnormal phenomenon in each device every predetermined period from an external device that inputs at least sensor measurement values from each device and collects and manages as an abnormal phenomenon when the input data meets a predetermined abnormality determination condition An acquisition means for acquiring a history of the number of times and storing it in the alarm data storage means as alarm data;
Degree of possibility of failure in each device included in failure prediction data stored in failure prediction data storage means and degree of necessity of operator's work to prevent failure due to abnormal phenomenon occurring in the device The transition of the prediction level according to the number of occurrences of abnormal phenomena in each device stored in the state transition data storage means with respect to the current prediction level that represents stepwise and the number of occurrences of abnormal phenomena included in the alarm data In accordance with the state transition data that is the condition of
Prediction data update means for updating the failure prediction data with the prediction result obtained by the failure prediction means;
A failure prediction program for causing an information processing apparatus to function.

Images