TWI710872B - Data processing device, data processing method and storage medium - Google Patents

Abstract

The data processing device as one aspect of the present invention includes an index calculation unit, an exclusion condition calculation unit, and a refining unit. The index calculation unit calculates index data including an index indicating the state of the monitoring target based on the measurement data related to the monitoring target. The exclusion condition calculation unit calculates exclusion conditions for removing the index related to interference from the index data based on the measurement data. The refining part removes the index related to the interference from the index data according to the exclusion condition, thereby refining the index data.

Description

Data processing device, data processing method and storage medium

The embodiment of the present invention relates to a data processing device, a data processing method and a storage medium.

In order to operate the system safely and stably, it is indispensable to check the integrity of the system and implement rectification as necessary. However, the higher the frequency of inspection, the higher the maintenance cost. Therefore, the next inspection plan is created based on the elapsed time since the last maintenance, the operating time of the system, and the like. This type of maintenance method of making inspection plans based on time is called Time Based Maintenance (TBM).

In recent years, due to the development of wireless technology, the low price of sensors, etc., real-time monitoring of the system status through remote monitoring has become the mainstream. If the status of the system can be grasped, the inspection period can be postponed and unnecessary inspection items can be omitted. This type of maintenance method of making an inspection plan based on the current status of the system is called Condition Based Maintenance (CBM). With the transition from TBM to CBM, we expect to reduce maintenance costs.

However, it is easy to mix various disturbances in the data measured during the operation of the system in order to grasp the current status of the system. Therefore, in order to grasp the current status of the system with high accuracy, it is necessary to remove unnecessary interference from the measurement data.

[Prior Art Literature] [Patent Literature]

[Patent Document 1] Japanese Patent No. 5306902

An embodiment of the present invention determines the interference contained in the measurement data for the monitoring target.

The data processing device as one aspect of the present invention includes an index calculation unit, an exclusion condition calculation unit, and a refining unit. The index calculation unit calculates index data including an index indicating the state of the monitoring target based on the measurement data related to the monitoring target. The exclusion condition calculation unit calculates exclusion conditions for removing the index related to interference from the index data based on the measurement data. The refining part removes the index related to the interference from the index data according to the exclusion condition, thereby refining the index data.

1: Measurement data acquisition department

2: Memory Department

3: Data Processing Department

4: output section

5: Input section

6: Computer device

7: Communication network

8: External device

31: Index calculation department

32: Calculation of exclusion conditions

33: Except outside (refining department)

34: Departure Judgment Department

35: Total Department

37: State Judgment Department

61: processor

62: main memory

63: auxiliary memory device

64: network interface

65: component interface

66: Bus

S101~S107, S201~S204, S301~S304: steps

Fig. 1 is a block diagram showing an example of a data processing apparatus of the first embodiment.

Figure 2 is a diagram illustrating index data.

Fig. 3 is a diagram showing an example of exclusion conditions.

Fig. 4 is a diagram showing the processing result of the removal and departure determination unit.

Fig. 5 is a diagram showing an example of output by an output unit.

FIG. 6 is a diagram showing an example of a schematic flowchart of the overall processing of the data processing apparatus of the first embodiment.

Fig. 7 is a block diagram showing an example of a schematic configuration of a data processing apparatus of the second embodiment.

Fig. 8 is a diagram for explaining correction of exclusion conditions.

FIG. 9 is a diagram showing an example of a flowchart of the update processing of exclusion conditions.

Fig. 10 is a block diagram showing an example of a schematic configuration of a data processing device of the third embodiment.

FIG. 11 is a diagram showing an example of the result of the total number of samples.

Fig. 12 is a diagram showing another example of the result of the total number of samples.

Fig. 13 is a block diagram showing an example of a schematic configuration of a data processing device of the fourth embodiment.

FIG. 14 is a diagram explaining the input for the evaluation of the state determination.

FIG. 15 is a diagram showing an example of a history of a correct solution for state determination.

FIG. 16 is a diagram showing an example of a flow chart for state determination.

Fig. 17 is a block diagram showing an example of a hardware configuration in an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(First Embodiment)

Fig. 1 is a block diagram showing an example of a data processing apparatus of the first embodiment. The data processing device of the first embodiment includes a measurement data acquisition unit 1, a storage unit 2, a data processing unit 3, and an output unit 4. The data processing unit 3 includes: an index calculation unit 31, The exclusion condition calculation unit 32, the exclusion (refining unit) 33, and the departure determination unit 34.

The data processing device of this embodiment calculates index data including an index indicating the state of the monitoring target based on the measurement data related to the monitoring target. Based on this index, detection of abnormality of the monitoring target, determination of whether inspection of the monitoring target is necessary, and the like are performed.

The calculated index can be considered to indicate a state that is difficult to measure with a sensor or the like. For example, it may be an index (performance index) indicating the state of the performance of the monitoring target.

The monitoring target is not particularly limited, and may be a machine or a system including a plurality of machines. In addition, it may also be living things such as humans and animals.

The measurement data refers to time-series data including the measurement results of sensors and the like. The measurement result includes the time of measurement and the measurement value. The sensor only needs to use a well-known sensor.

The measurement item of the measurement data may be an item that can be measured by a known sensor or the like. The measurement site may be the entire monitoring target or a specific site of the monitoring target. For example, as measurement items, temperature, humidity, flow rate, current, voltage, pressure, location, etc. can be considered. In addition, when the monitoring object is a moving body such as a vehicle, it is assumed that the speed, acceleration, etc. of the moving body will also be the measurement items.

It is assumed that the measurement related to the measurement data can be performed arbitrarily. Therefore, it is assumed that the measurement data may also include data measured when the monitoring object is operating and data measured when the monitoring object has stopped.

Furthermore, the measurement results of the monitoring object itself can also be included in the measurement data in. In addition, the settings entered into the monitoring object can also be included in the measurement data. For example, it may also include data indicating the time periods during which the power supply of the monitoring target is turned on (ON) and turned off (OFF). In addition, for example, when the monitoring target has a power saving mode that suppresses power consumption, data indicating the time period during which the monitoring target is in the power saving mode may also be included in the measurement data. In addition, for example, when the monitoring object is an air conditioner, data indicating the state of being cooling, heating, or humidifying may also be included in the measurement data. In addition, when the monitoring object is a vehicle, data indicating states such as driving, accelerating, decelerating, or temporarily stopping can also be included in the measurement data.

In addition, the state of the monitoring target determined by the monitoring target itself or other external devices based on the measurement data may also be included in the measurement data. For example, when the current flowing in the motor built into the monitoring object is measured, and the monitoring object determines that the motor is abnormal based on the measured current value, the monitoring object can also add a value indicating the abnormality of the motor or the monitoring object To the measurement data.

The index contained in the index data is represented by a combination of at least the index value and the time corresponding to the index value. The index value is calculated from one or more measurement values related to one or more measurement items. For example, the index value may be calculated based on the measured values of a plurality of currents in a predetermined period. For example, five current values can also be used to calculate an index value. Alternatively, the index value may be calculated based on one measured value of current and one measured value of engine temperature in the same time zone. Furthermore, as long as the absolute value of the difference between the measurement times is within a predetermined value, it can be regarded as the same time zone.

The time corresponding to the index value can be set to be the same as the measurement time corresponding to the measured value used for the calculation of the index value. When multiple measurement values are different according to the measurement time When calculating the index value, it only needs to be calculated based on the statistical value of a plurality of measurement times. For example, the time corresponding to the index value may be an average value, an intermediate value, or the like of a plurality of measurement time.

Figure 2 is a diagram illustrating index data. The graph represented by the circle represents the index. The vertical axis represents the index value, and the horizontal axis represents the time corresponding to the index value. The broken lines in FIG. 2 indicate the upper limit and the lower limit of the measurement value that is expected to be measured when the monitoring target is normal. That is, when there is an index value between the upper limit and the lower limit, it is expected that the monitoring target is normal. The range of the index value expected to be normal for the monitored object is described as the allowable range. In Fig. 2, hollow circles are used to indicate indexes within the allowable range, and black circles are used to indicate indexes outside the allowable range. Record indicators outside the allowable range as deviation cases.

Departure case indicates that the monitored object departs from the normal state. That is, there is a high possibility that the monitored object is abnormal. However, not all instances of deviations indicate abnormalities in the monitored object.

As described above, the measurement data includes the measurement value when the monitoring object is operating. When the monitoring object is running, compared with the case where the monitoring object has stopped, the possibility of including interference in the measurement data is higher, and the accuracy of the index tends to be lower. For example, when the monitoring object is a vehicle and the current flowing in a specific part is measured, when the vehicle is running, the possibility that interference is included in the measurement data and an abnormal value is included in the measurement value becomes high.

Therefore, there are also cases where interference is mixed into the measurement data, which causes the calculated index to show an abnormal value, which may deviate from the case. Since interference is included in the departure case as mentioned, it is necessary to determine whether the departure case is caused by interference or Caused by an abnormality of the monitored object.

Therefore, the data processing device removes indicators related to interference from the calculated indicator data, thereby refining the indicator data. Moreover, by using refined index data, the status of the monitored object can be grasped with high accuracy.

The internal structure of the data processing device will be described. The measurement data acquisition unit 1 acquires measurement data. The measurement data obtaining unit 1 may be directly obtained from a sensor or the like, or may be obtained indirectly via an external device. In addition, the measurement data acquisition unit 1 may also process the measurement data to calculate the measurement data processed by the data processing unit 3. For example, after excluding unnecessary measurement items, a plurality of measurement data may be combined to calculate one measurement data.

The storage unit 2 stores data used in each process of the data processing unit 3. This information is stored in the memory 2 in advance. In addition, data input into the data processing unit 3, data calculated in each process of the data processing unit 3, etc. may also be memorized, and the memorized data is not particularly limited. Furthermore, the memory unit 2 can also be divided according to the data stored.

The data processing unit 3 processes the measurement data and calculates index data. The details will be explained together with the internal structure.

The output unit 4 outputs data related to the data processing unit 3. For example, it outputs the following exclusion conditions, refined index data, and judgment results based on refined index data. In addition, the data used in the processing of each part and the processing result of each part can also be output.

Furthermore, the data output by the output unit 4 is not particularly limited, and can also be output Data stored in the memory part 2. In addition, the output method of the output unit 4 is not particularly limited. Images, voices, etc. can be output to the display, etc., and the processing results can also be stored in an external storage through electronic files.

The internal structure of the data processing unit 3 will be described. The index calculation unit 31 calculates index data based on the measurement data. The index calculation unit 31 may use a predetermined calculation formula to calculate an index value based on one or more measured values. Alternatively, the measured value itself may be used as the index value. Known ones can be used for the calculation formula.

The exclusion condition calculation unit 32 calculates conditions for removing index data related to interference from the index data. This condition is described as an exclusion condition. Based on the exclusion conditions, each index in the index data is divided into an exclusion target index and an out-of-judgment index index.

The exclusion condition is to determine whether the first measurement data used for the calculation of the index or the second measurement data measured in the same time period as the first measurement data are data measured under a situation where interference is likely to be mixed. As a situation where interference is likely to be mixed, it is sufficient to determine in advance a predetermined operating state corresponding to the monitoring target.

For example, when the index value is calculated from the measured value of the current, it is possible to remove the index at the time of high-speed movement of the monitored object that is likely to be disturbed by using an exclusion condition for the speed measured in the same time period as the measured value of the current. The measurement items used to calculate the index and the measurement items used to calculate the exclusion conditions may also be determined in advance.

The predetermined operating state may be a state in which the monitoring object is only operating, or an operating state in which the output value or the like is above a predetermined value. For example, when the monitoring object is a generator, etc., it can also be set to a state that transmits power above a predetermined value. state. Alternatively, it may be a state where the power consumption of the monitoring target is a predetermined value or more. Alternatively, when the monitoring target is a vehicle, it may be in a state of traveling at a speed higher than a predetermined value. Alternatively, when the internal temperature of the monitoring target is greater than or equal to a predetermined value, it can also be regarded that the monitoring target is operating under a high load. In this way, it can be determined whether the first measurement data or the second measurement data are those measured under a predetermined operating state, and it can be determined that the index calculated by the first measurement data is an index related to interference.

Fig. 3 is a diagram showing an example of exclusion conditions. The exclusion condition in Fig. 3 is a condition for using a decision tree. For example, suppose that the measurement data includes the measurement values of three sensors (a, b, and c). The index is produced based on at least any of the three sensors (a, b, and c), and the exclusion conditions are also produced based on at least any of the three sensors (a, b, and c) .

In the exclusion conditions of FIG. 3, classification is first performed based on the measured value of the sensor a. When the measured value of the sensor a is 3 or more, the index is judged to be out of the judgment target. When the measurement value of the sensor a is less than 3, further classification is performed based on the measurement value of the sensor b. When the measurement value of the sensor b is less than 5.2, the index is determined to be out of the determination target, and when the measurement value of the sensor b is 5.2 or more, the index is determined to be the exclusion target.

In the above exclusion conditions, for example, in the case of calculating the index using all the measured values of the sensor a, the sensor b, and the sensor c, the measurement value of the sensor a is set to 2 and the sensor b is The index calculated when the measured value of is 6 is set to be excluded. In addition, for example, in the case of calculating the index based only on the measurement value of the sensor c, when the measurement value of the 13-point sensor a is 2 and the measurement value of the sensor b is 6, by The index calculated by the measurement value of the sensor c at 13 points is also excluded.

Furthermore, the measurement items of each sensor may be the same or different. For example, both the sensor a and the sensor b may measure the current at the same part. Alternatively, the sensor b may measure the current at a location different from the sensor a. Alternatively, the sensor a may measure the current and the sensor b may measure the voltage.

The exclusion condition can be calculated by using machine learning based on the determination result of the departure determination unit 34. Known methods can be used for machine learning. As a machine learning method, for example, there is a method of using a model that is classified into a group including indicators that depart from the judgment target and a group including indicators that are excluded.

In addition, there is also a method of quantifying the degree of deviation of the index and performing sparse regression to calculate the exclusion condition. The degree of separation can be calculated based on the difference from the reference value or the nearest limit value. In addition, it can also be calculated from statistical values such as the average value and median value of the index.

Except outside 33 (refining department) removes the index data related to interference from the index data according to the exclusion conditions, thereby refining the index data.

The deviation determination unit 34 determines whether each index of the refined index data is out of a predetermined allowable range. Furthermore, the allowable range may specify only either the upper limit value or the lower limit value. That is, if the index is less than the upper limit value or more than the lower limit value, it can also be judged as allowable. In addition, it can be set within a predetermined value before and after the reference value. For example, when the reference value of the current is 10 A, 0.5 A before and after the reference value may be set as the allowable range. In this case, as long as the measured value of the current is included in the range of 9.5A to 10.5A, it is judged to be acceptable.

The allowable range can also vary according to the state of the monitored object. For example, it can be considered that the allowable range of the index related to the measurement data when the monitoring target is moving is different from the allowable range of the index related to the measurement data when the monitoring target has stopped. The status of the monitored object can be judged based on the measurement data.

FIG. 4 is a diagram showing the processing result of the removal of the exterior 33 and the departure determination unit 34. FIG. 4 is also an example of output from the output unit 4. Among the indicators shown in FIG. 2, the indicators excepted by the outer part 33 are represented by triangles and four-corner graphics. In FIG. 4, it is assumed that the exclusion 33 is refined using exclusion conditions including a plurality of conditions as shown in FIG. 3. The conditions constituting the exclusion conditions are described as sub-conditions. The index indicated by the triangle is the index excluded based on the first sub-condition, and the index indicated by the four corners is the index excluded based on the second sub-condition. As shown in Figure 4, even indicators within the allowable range can be excluded.

In Fig. 4, the index leaving the judgment target is indicated by a circle. Among the indexes indicated by circles, indexes outside the allowable range are indicated by black circles. In this way, the departure determination unit 34 determines whether or not each index of the departure determination target is out of the allowable range.

Furthermore, the output unit 4 can also display whether each index is within the allowable range or out of the allowable range, and whether it is an excluded index by changing the shape, color, etc. of the graph as shown in FIG. 4. In addition, the release rate can also be output. The rate of separation is calculated by dividing the number of indicators that are out of the judgment target as the denominator and the number of indicators that are separated from the numerator.

In addition, the output unit 4 may also output the degree of separation of the index of the separation determination target within a predetermined period such as one day and one week. Figure 5 shows the output of the output unit 4 A diagram of an example. Fig. 5 is a box-and-whisker plot showing the distribution of index data for each day. The output unit 4 can also switch the display form in this way.

Next, the flow of processing using each constituent element will be described. FIG. 6 is a diagram showing an example of a schematic flowchart of the overall processing of the data processing apparatus of the first embodiment.

The measurement data acquisition unit 1 acquires measurement data (S101). The index calculation unit 31 calculates index data based on the measurement data (S102). Furthermore, the index value can be calculated every time a measured value is acquired, or a predetermined number of measured values can be acquired and then aggregated to calculate the index value.

The exclusion condition calculation unit 32 calculates the exclusion conditions based on the history of past departure determinations and the measurement data related to the past departure determinations (S103). Furthermore, when there is no history of past departure determinations, a prescribed condition (initial condition) is used as the exclusion condition.

The exclusion 33 excludes the index to be excluded from the index data from the index calculation unit 31 based on the exclusion condition calculated by the exclusion condition calculation unit 32 (S104). In addition, the departure determination unit 34 determines the departure of each index of the refined index data obtained by excluding the exclusion target by the exclusion 33 (S105). In addition, the index departure determination unit 34 updates the departure determination history (S106). In this way, the exclusion conditions are updated in the next processing. Then, the output unit 4 displays the processing result and the like (S107), thereby ending this flow.

In addition, this flowchart is just an example, and as long as the desired processing result can be obtained, the processing sequence and the like are not limited. For example, S103 can also be processed in S101 and S102 Before the treatment. In addition, the processing results of each process are sequentially stored in the storage unit 2, and each component element can also refer to the storage unit 2 to obtain the processing results.

As described above, according to this embodiment, unnecessary interference can be removed from the measurement data. Therefore, according to the present embodiment, even based on measurement data obtained during the operation of the monitoring target and containing many disturbances, the current status of the monitoring target can be grasped with high accuracy. Therefore, as a result, it is possible to take measures such as prolonging the inspection frequency of the monitoring object and omitting unnecessary inspections, and the maintenance cost can be suppressed.

Furthermore, the data processing device may also include multiple devices for transmitting and receiving data through communication or electrical signals. For example, it may be divided into a first device that has the outside 33 and the like to produce refined index data, and a second device that has the separation determination unit 34 and performs separation determination.

(Second Embodiment)

Fig. 7 is a block diagram showing an example of a schematic configuration of a data processing apparatus of the second embodiment. The second embodiment is different from the first embodiment in that it further includes an input unit 5 that accepts input from the user. Description of the same points as in the first embodiment is omitted.

If the output unit 4 outputs the exclusion condition calculated by the exclusion condition calculation unit 32, a user such as a monitoring person, a manager of the monitoring target, etc. can determine whether the output exclusion condition is appropriate. Moreover, there may be situations where it is desired to ease or strengthen the exclusion conditions.

Therefore, in this embodiment, the input unit 5 accepts input related to exclusion conditions. For example, input the parameters required for the creation of exclusion conditions and the calculated division Correction of external conditions, etc. The exclusion condition creation unit creates and corrects exclusion conditions based on the input value accepted by the input unit 5.

Fig. 8 is a diagram for explaining correction of exclusion conditions. The Graphical User Interface (GUI) shown in FIG. 8 is output by the output unit 4 in order to receive corrections from the user.

On the right side of Figure 8, the tree structure is used to show the exclusion conditions calculated by the decision tree. The calculation conditions of the exclusion conditions are shown on the left side of Fig. 8. As calculation conditions, the period, application method, and application conditions of the learning materials used for calculation are displayed. The so-called application method refers to the applied machine learning method. The setting content of the application conditions differs according to the selected application method. In the case of decision trees, as a classification problem, machine learning methods are applied, so the group for classification must be selected. In the example shown in FIG. 8, classification is made between the allowable range and the outside of the allowable range. It can be divided into two groups based on the distance from the reference value.

The GUI of FIG. 8 can change the display content and functions as an input interface for correcting the exclusion conditions. That is, the change of the GUI is input to the input unit 5, and the exclusion condition calculation unit 32 recreates the exclusion condition based on the change.

For example, consider adjusting the threshold of the exclusion condition. In the example of FIG. 8, after changing the threshold value in the tree structure shown on the right, press the "model correction" button displayed on the lower side of the screen to create the exclusion condition with the threshold value corrected again.

Furthermore, it is ideal that the parameters of the decision tree can also be set and changed. As parameters, imagine the tree construction algorithm, the range of pruning, and so on.

In addition, in the case where a model with the desired accuracy cannot be created even if the parameters of the exclusion conditions are adjusted, it is possible to consider changing or rebuilding the model. After determining the necessary items calculated by the exclusion condition on the left side of Fig. 8, press the "model rebuild" button displayed on the lower side of the screen to create a new exclusion condition again.

Furthermore, when there is no problem with the exclusion conditions, you can also press the "model determination" button displayed at the bottom of the screen. After the exclusion condition calculation unit 32 calculates the exclusion conditions, the output unit 4 outputs the GUI of FIG. 8, and the processing of the exclusion 33 and the departure determination unit 34 may not be performed until the "model determination" button is pressed. In this way, processing can also be performed after the user's confirmation, thereby preventing unsatisfactory processing results from being output.

Next, the process of updating the exclusion conditions will be described. FIG. 9 is a diagram showing an example of a flowchart of the update processing of exclusion conditions. It is assumed that this flow is performed between S103 and S104 of the overall processing shown in FIG. 6. In addition, this flow may be performed after the processing of S107, and the processing of S104 may be performed again after the end of this flow.

The output section 4 uses the GUI as shown in FIG. 8 to output exclusion conditions (S201). When the output exclusion condition has not been approved (NO in S202), the input unit 5 obtains the exclusion condition corrected by the GUI (S203). The corrected exclusion condition is delivered to the exclusion condition calculation unit 32, and the exclusion condition calculation unit 32 updates the exclusion condition (S204). The updated exclusion condition is output again by the output unit 4 (S201). In this way, the processing of S201 to S204 is repeated until the exclusion condition is approved. Furthermore, when the exclusion condition is approved (YES in S202), this flow ends.

As above, according to this embodiment, based on the input to the input unit 5, The exclusion condition calculation unit 32 creates or modifies exclusion conditions. With this, it is possible to create exclusion conditions that reflect the user's thoughts, experience, etc., and to ensure the appropriateness of interference removal methods.

(Third Embodiment)

Fig. 10 is a block diagram showing an example of a schematic configuration of a data processing device of the third embodiment. The third embodiment is different from the above embodiment in that the data processing unit 3 further includes the summing unit 35. In FIG. 10, the totalizing part 35 is added to the second embodiment, but the totaling part 35 may be added to other embodiments. The description of the same points as in the above embodiment is omitted.

The totalizing unit 35 totals the number of indexes included in the refined index data. The number of this index is described as the number of samples. That is, it can be said that the totalizing unit 35 totals the number of valid samples of the index data. If the number of samples of index data is above the prescribed threshold, the index data can be considered trustworthy. For example, in the case of the processing result shown in FIG. 4, the number of samples becomes 11 out of the number of indicators that are out of the determination target.

The threshold relative to the number of samples is determined in advance. In addition, the number of samples may be totaled within each predetermined period and in each state of the monitoring target.

FIG. 11 is a diagram showing an example of the result of the total number of samples. It can be considered that the output unit 4 outputs a table as shown in FIG. 11. In Fig. 11, the number of samples was totaled in all three operation modes and five periods. The operation mode indicates the type of the state of the monitoring target. The period can be arbitrarily determined. The length of each period may not be fixed.

For example, when the monitoring object is a vehicle, as the operation mode of the vehicle, it can be considered as having three types: accelerating, decelerating, and temporarily stopping. Status. Furthermore, if the 0 o'clock to 5 o'clock are divided in units of 1 hour, and the number of samples is totaled in each operation mode and each period, a total result as shown in Fig. 11 is created.

The numbers in parentheses in Fig. 11 indicate the threshold value relative to the number of samples. The threshold value can also be different in each period and in each action mode. In the example of FIG. 11, period 4 and period 5 are set to be longer than period 1-period 3. Therefore, the number of thresholds for periods 4 and 5 is set to be larger than the number of thresholds for periods 1 to 3. In this way, the threshold of the number of samples can also be output.

In the example of FIG. 11, the number of samples in operation mode 3 in period 4 is less than the corresponding threshold value. Therefore, it can be considered that the reliability of index data related to period 4 and operation mode 3 is low.

Fig. 12 is a diagram showing another example of the result of the total number of samples. In FIG. 11, the total result is shown in the form of a table, but it may also be shown as a bar graph as in FIG. It may also be a display format in which the number of samples can be designated via the input unit 5 shown in the second embodiment.

The totalizing process by the totalizing unit 35 may be performed after the exclusion 33 excludes the exclusion target index based on exclusion conditions. In other words, in the flow shown in FIG. 6, it only needs to be performed after the processing of S104. The diagram of the flow of this embodiment is omitted.

As described above, according to this embodiment, the number of valid samples of index data is totaled. With this, the number of samples of refined index data can be confirmed, and the reliability of the output index data is guaranteed.

(Fourth Embodiment)

Fig. 13 is a block diagram showing an example of a schematic configuration of a data processing device of the fourth embodiment. The fourth embodiment is different from the above embodiment in that it further includes a state determination condition calculation unit 36 and a state determination unit 37. In FIG. 13, these components are added to the third embodiment, but they may be added to other embodiments. The description of the same points as in the above embodiment is omitted.

In this embodiment, the data processing device directly performs state determination and determines the state of the monitored object. It is conceivable to use the judgment result of the state of the monitoring object for various purposes. For example, a part determined to be normal based on this determination may be omitted from the next inspection. In addition, when it is determined to be abnormal based on this determination, the output unit 4 may output an alarm by an image, sound, or the like. In this case, it can be said that the data processing device is both a state determination device and an abnormality detection device.

In addition, the measurement data is also obtained during the operation of the monitored object. Therefore, if the measurement data is obtained in real time every time the measurement is performed and the state of the monitored object is immediately determined, an abnormality can be detected immediately.

The state determination condition calculation unit 36 calculates a state determination condition as a condition for performing state determination. The state judgment condition is calculated using a learning model for calculating the state judgment condition. This model is described as a learning model for calculating state judgment conditions. The state determination condition calculation learning model is updated by learning the evaluation (correct solution) of the state determination input from the user or the like. Thereby, the accuracy of the state determination condition is improved. As in the case where the exclusion condition calculation unit 32 calculates the exclusion condition, a known method can be used for the learning method. The previous evaluation of the state decision is via the input Part 5.

FIG. 14 is a diagram explaining the input for the evaluation of the state determination. The GUI shown in FIG. 14 is an interface for accepting the evaluation of the status judgment output by the output unit 4. In each period and each operation mode, the release rate and the number of samples are displayed. In the example of FIG. 14, there are two buttons of "inspection can be reduced in labor" and "inspection can not be reduced in labor", and the button is used to answer whether or not the inspection of the part ticked in the check box can be omitted.

The gray areas indicate that the detachment rate or the number of samples is not recognized as normal. In the example of FIG. 14, the separation rate is high in the operation mode 1 of the period 1, and the number of samples is small in the operation mode 3 of the period 4. Therefore, it can be considered that the inspection cannot be omitted at the two locations.

FIG. 15 is a diagram showing an example of a history of a correct solution for state determination. Furthermore, items other than the items shown in FIG. 15 may also be included in the positive solution. It is assumed that the history of such a correct solution is stored in the memory 2 first. Furthermore, the history of the correct solution is used as the learning material, and machine learning is performed as a classification problem for classifying the refined index data to calculate the state determination condition. Known methods can be used for machine learning. In addition, learning can be performed in each operation mode, and the state determination conditions can be calculated in each operation mode.

In this way, the state determination condition calculation unit 36 repeatedly learns based on the state of the monitoring target determined by the state determination unit and data related to the legitimacy of the state, thereby updating the learning model for state determination condition calculation. Calculate based on the updated state judgment condition Use the learning model to calculate the state judgment condition to determine the state The accuracy of the conditions is improved. In addition, when the state determination condition is provided by the user via the input unit 5, the state determination condition calculation unit 36 may be omitted.

The state determination unit 37 determines the state of the monitoring target based on the state determination condition and the determination result of the departure determination unit 34. Furthermore, the state of the monitored object can be classified into two types: normal and abnormal, and can also be classified into three or more types based on deviation from the reference value.

In addition, the state determination unit 37 may also determine whether or not inspection of the monitoring target is necessary based on the determined state of the monitoring target. For example, when the state is determined to be normal, the state determination unit 37 may determine that the predetermined inspection item can be omitted. In addition, for example, when the deviation rate is 5% or more and less than 10%, the state determination unit 37 determines the state of the monitored object as caution, and may also determine that the inspection for the first inspection item can be omitted, but the inspection for the first inspection item cannot be omitted. 2 Inspection of inspection items.

In addition, when the number of samples totaled by the totalizing unit 35 is lower than the threshold value, the state determining unit 37 may not perform the state determination or may add a warning such as low reliability to the determination result. For example, when the number of samples does not satisfy the condition as in the example of FIG. 11, the state cannot be accurately determined, and it may be determined that the inspection cannot be omitted. In addition, the state can also be judged in each operation mode. When there are inspection items corresponding to each operation mode, in the example of FIG. 11, the inspection items related to operation mode 1 and operation mode 2 may be omitted, but the inspection items related to operation mode 3 cannot be omitted.

The state determination result of the state determination unit 37 is output by the output unit 4. It is assumed that the status of the monitoring object, the possibility of omission of inspection, the reason for omission of inspection, the inspection items that can be omitted, etc. are output in the output information. For example, due to the separation rate The message that it is higher than the threshold value and the number of samples in the operation mode 3 in the period 4 is insufficient to omit the inspection can also be displayed on the monitor connected to the data processing device through the output unit 4.

Next, the flow of state determination will be described. FIG. 16 is a diagram showing an example of a flow chart for state determination. It is assumed that this flow is performed after S105 of the overall processing shown in FIG. 6.

The state determination unit 37 determines the state based on the state determination condition based on the index data (S301). The judgment result may be output together with the processing result of the above embodiment, or may be output individually. Then, the output unit 4 outputs a GUI for correct solution input (S302). The user operates the GUI for correct solution input, whereby the input unit 5 accepts the correct solution for state determination (S303). Then, the state determination condition calculation unit 36 updates the state determination condition based on the index data and the correct solution history (S304), and this flow ends. Thereby, in the next processing, the updated state determination conditions can be used, and the accuracy of state determination is improved.

As described above, the data processing device of this embodiment also determines the state of the monitoring target based on index data. In this way, it is possible to automate the determination of the state of the monitoring target and omission of inspection. In addition, the state determination conditions can also be automatically created by learning.

In addition, each processing in the embodiment described above can be realized by a dedicated circuit, or can be realized by software (program). When software (program) is used, the above-described embodiment can be used, for example, by using a general-purpose computer device as the basic hardware, and the central processing device (CPU: Central Processing Unit) and other processors execute programs.

Fig. 17 is a block diagram showing an example of a hardware configuration in an embodiment of the present invention. The data processing device includes a processor 61, a main memory device 62, an auxiliary memory device 63, a component interface 65, and a network interface 64, and is implemented as a computer device 6 connected to these via a bus 66. In addition, the data processing device may further include an input device and an output device.

The data processing device in this embodiment can be realized by installing the programs executed by each device in the computer device 6 in advance, or by storing the programs in a compact disc-read only memory (Compact Disc-Read Only Memory, CD-ROM) and other storage media, or distributed via the network, and suitable for installation in the computer device 6 to achieve.

Furthermore, in FIG. 17, the computer device includes one of each component, but it may include a plurality of the same component. In addition, one computer device is shown in FIG. 17, but the software can also be installed in multiple computer devices. The multiple computer devices execute different parts of the processing of the software, thereby generating processing results. That is, the data processing device may also be configured as a system.

The processor 61 is an electronic circuit including a control device and an arithmetic device of a computer. The processor 61 performs arithmetic processing based on data or programs input from each device or the like of the internal structure of the computer device 6, and outputs the result of the calculation or a control signal to each device or the like. Specifically, the processor 61 executes an operating system (OS) or an application program of the computer device 6 and controls each device constituting the computer device 6.

The processor 61 is not particularly limited as long as it can perform the aforementioned processing. The processor 61 may also be, for example, a general purpose processor, a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a controller, a microcontroller, a state machine, etc. In addition, the processor 61 can also be assembled into a specific-purpose integrated circuit, a field programmable gate array (Field Programmable Gate Array, FPGA), and a programmable logic circuit (Programmable Logic Device (PLD)). )in. In addition, the processor 61 may also include multiple processing devices. For example, it may be a combination of a DSP and a microprocessor, or it may be more than one microprocessor working in cooperation with the DSP core.

The main memory device 62 is a memory device that stores commands and various data executed by the processor 61, and the information stored in the main memory device 62 is directly read by the processor 61. The auxiliary memory device 63 is a memory device other than the main memory device 62. Furthermore, the memory device refers to any electronic component that can store electronic information. As the main memory device 62, it mainly uses random access memory (Random Access Memory, RAM), dynamic random access memory (Dynamic Random Access Memory, DRAM), and static random access memory (Static Random Access Memory, SRAM). Such as volatile memory used for temporary storage of information, but in the embodiment of the present invention, the main memory device 62 is not limited to these volatile memories. The memory devices used as the main memory device 62 and the auxiliary memory device 63 may be volatile memory or non-volatile memory. Non-volatile memory includes Programmable Read Only Memory (PROM), Erasable Programmable Read Only Memory (Erasable Programmable Read Only Memory, EPROM), Electronically Erasable Programmable Read Only Memory (EEPROM), Non-Volatile Random Access Memory (NVRAM), flash memory, magnetic Random access memory (Magnetic Random Access Memory, MRAM), etc. In addition, as the auxiliary memory device 63, a magnetic or optical data storage device may also be used. As a data storage device, it is also possible to use disks such as hard disks, optical discs such as Digital Versatile Disc (DVD), flash memory such as Universal Serial Bus (USB), and tapes, etc. .

Furthermore, if the processor 61 directly or indirectly reads or writes information, or performs both of the main memory device 62 or the auxiliary memory device 63, it can be said that the memory device and the processor are in electrical communication. Furthermore, the main memory device 62 can also be incorporated in the processor. In this case, it can also be said that the main memory device 62 is in electrical communication with the processor.

The network interface 64 is an interface for connecting to a communication network by wireless or wired connection. As long as the network interface 64 is suitable for the existing communication specifications, it is sufficient. The network interface 64 may also be used to send the output result and the like to the external device 8 which is communicatively connected via the communication network 7.

The component interface 65 is an interface such as USB that is connected to an external device 8 that records output results and the like. The external device 8 can be an external storage medium or a storage such as a database. The external storage medium can also be a hard disc drive (Hard Disc Drive, HDD), a recordable disc (Compact Disc-Recordable, CD-R), a rewritable disc (Compact Disc-Rewritable, CD-RW), digital Random storage of multi-function disc Access memory (Digital Versatile Disc-Random Access Memory, DVD-RAM), recordable digital versatile disc (Digital Versatile Disc-Recordable, DVD-R), storage area network (Storage area network, SAN), etc. The recording medium. Alternatively, the external device 8 may also be an output device. For example, it may be a display device for displaying images, or a device for outputting voice or the like. For example, there are Liquid Crystal Display (LCD), Cathode Ray Tube (CRT), Plasma Display Panel (PDP), speakers, etc., but they are not limited to these devices.

In addition, a part or all of the computer device 6, that is, a part or all of the data processing device, may also include a dedicated electronic circuit (ie, hardware) such as a semiconductor integrated circuit in which the processor 61 and the like are mounted. Dedicated hardware can also be constructed in combination with memory devices such as RAM and ROM.

Furthermore, one computer device is shown in FIG. 17, but the software can also be installed in multiple computer devices. The multiple computer devices execute different parts of the processing of the software, thereby generating processing results.

One embodiment of the present invention has been described above, but these embodiments are presented as examples and are not intended to limit the scope of the invention. These new embodiments can be implemented in various other forms, and various omissions, substitutions, and changes can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope or spirit of the invention, and are included in the invention described in the scope of the patent application and its equivalent scope.

1‧‧‧Measurement Data Acquisition Department

2‧‧‧Memory Department

3‧‧‧Data Processing Department

4‧‧‧Output

31‧‧‧Indicator calculation department

32‧‧‧Exclusion condition calculation section

33‧‧‧Except outside (refining department)

34‧‧‧Departure Judgment Department

Claims (9)

A data processing device includes: an index calculation unit that calculates index data including an index indicating the state of the monitoring object based on measurement data related to the monitoring object; and an exclusion condition calculation unit that calculates the value for The exclusion conditions of the index related to interference are removed from the index data; and the refining department removes the index related to the interference from the index data according to the exclusion condition, thereby refining the index data , Wherein the exclusion condition includes a condition for whether the value of the second measurement data measured in the same time period as the first measurement data used for the calculation of the index is within the allowable range, and by removing the second measurement The index data is refined by the index in the time period when the value of the data is not within the allowable range. The data processing device described in the first item of the scope of patent application further includes a departure determination unit that determines whether the index included in the index data refined by the refining unit is out of the allowable range. The data processing device described in the second claim of the patent application further includes a state determination unit that determines the state of the monitoring object from the determination result of the departure determination unit based on a state determination condition. The data processing device according to the third item of the scope of patent application, wherein the state determination unit determines whether or not inspection of the monitoring target is necessary based on the determined state of the monitoring target. The data processing device described in the third of the scope of patent application further includes a state determination condition calculation unit based on the state of the monitoring target determined by the state determination unit, and The determined data related to the validity of the state of the monitoring target updates the state determination condition calculation learning model, and based on the state determination condition calculation learning model, the state determination condition is calculated. For example, the data processing device described in item 1 of the scope of patent application further includes: an output unit that outputs at least the exclusion condition; and an input unit that accepts input of data related to the exclusion condition; and the exclusion condition is calculated The part re-creates the exclusion condition based on the data related to the exclusion condition accepted by the input part. The data processing device described in the first item of the scope of patent application further includes a totalizing unit that totals the number of the indicators contained in the indicator data refined by the refining unit. A data processing method, comprising: calculating index data including an index indicating the state of the monitored object based on measurement data related to the monitored object; and calculating based on the measurement data to remove from the indicator data The step of exclusion conditions of indicators related to interference; and the step of removing the indicators related to the interference from the indicator data according to the exclusion conditions to refine the indicator data, wherein The exclusion condition includes a condition for whether the value of the second measurement data measured in the same time period as the first measurement data used for the calculation of the index is within the allowable range, and by removing the condition of the second measurement data The index data is refined by the index in the time period whose value is not within the allowable range. A storage medium that stores a program for a computer to execute the following steps: a step of calculating index data including an index indicating the state of the monitored object based on measurement data related to the monitored object; The step of removing the exclusion condition of the index related to interference from the index data; and according to the exclusion condition, removing the index related to the interference from the index data, thereby refining the index data , Wherein the exclusion condition includes a condition for whether the value of the second measurement data measured in the same time period as the first measurement data used for the calculation of the index is within the allowable range, and by removing the first measurement data 2 Measure the index in the time period when the value of the data is not within the allowable range to refine the index data.

Images