Fluctuation Detection Device and Fluctuation Detection Method

Abstract

The present invention relates to a fluctuation detection device and a fluctuation detection method. The fluctuation detection device (1) includes: a deviation feature quantity calculation unit (31) that performs frequency analysis on the deviation between the value of the operation instruction information (SP) and the value of the behavior information, and calculates the deviation feature quantity, where the operation instruction information (SP) is input to an operator (10) that drives an operation end of a feedback control system, and the behavior information is any one of the output (MV) from the operator (10) to the operation end or the control quantity of a controlled object; a similarity feature quantity calculation unit (32) that performs frequency analysis on the values of the operation instruction information (SP) and the behavior information respectively, and calculates the similarity feature quantity between the operation instruction information (SP) and the behavior information; and a fluctuation determination unit (4) that determines whether a fluctuation phenomenon occurs and whether maintenance of the operator (10) is required based on the deviation feature quantity and the similarity feature quantity.

Description

Technical Field

[0001] The present invention relates to a hunting detection device and a hunting detection method for detecting hunting. Background Art

[0002] As a method for detecting hunting in feedback control, for example, focusing on the characteristics of hunting such as Figure 10 as shown, where the controlled variable PV such as temperature or flow rate periodically changes, a method using frequency analysis has been proposed (see Patent Document 1). In the detection method disclosed in Patent Document 1, frequency analysis is performed on the controlled variable PV or the operation signal MV, and when a specific frequency exceeds a preset value, it is determined that hunting has occurred.

[0003] As disclosed in Patent Document 1, hunting can be detected by focusing on PV or MV. However, even if hunting can be detected, it is still too early to conclude that an actuator that drives a damper or a valve needs maintenance.

[0004] For example, as in the example Figure 11 shown, when a vibratory action instruction information SP is given to the actuator and the controlled variable PV as indicated is obtained, there is a high possibility that an abnormality has occurred in the controller or the like of the feedback control system, rather than in the actuator that receives the action instruction information SP from the controller. Therefore, the necessity of maintaining the actuator is reduced.

[0005] On the other hand, as in the example Figure 12 shown, when a non-vibratory action instruction information SP is given to the actuator and the controlled variable PV shows vibration in a form different from the instruction, there is a possibility that an abnormality has occurred in the actuator. Therefore, it can be considered that maintenance of the actuator needs to be performed.

[0006] As described above, in the prior art disclosed in Patent Document 1, the action instruction information SP for the actuator is not considered, and thus there are problems such as being unable to determine whether the actuator needs to be maintained.

[0007] Prior Art Documents

[0008] Patent Documents

[0009] Patent Document 1: Japanese Patent Laid-Open No. 07-093002 Summary of the Invention

[0010] Problems to be Solved by the Invention

[0011] The present invention has been made to solve the above problems, and an object thereof is to provide a hunting detection device and a hunting detection method that can determine whether an actuator needs to be maintained while detecting hunting in the detection of hunting phenomena related to control.

[0012] Technical means for solving the problem

[0013] The fluctuation detection device of the present invention includes: a first feature quantity calculation unit configured to perform frequency analysis on the deviation between the value of the operation instruction information and the value of the behavior information and calculate the deviation feature quantity, where the operation instruction information is input to an operator that drives an operation terminal of a feedback control system, and the behavior information is either the output from the operator to the operation terminal or the control quantity of the controlled object; a second feature quantity calculation unit configured to perform frequency analysis on the values of the operation instruction information and the behavior information respectively and calculate the similarity feature quantity between the operation instruction information and the behavior information; and a determination unit configured to determine whether a fluctuation phenomenon occurs and whether it is necessary to maintain the operator based on the deviation feature quantity and the similarity feature quantity.

[0014] Moreover, a structural example of the fluctuation detection device of the present invention further includes: a data extraction unit configured to extract the operation instruction information and the behavior information that meet the specified extraction conditions from the operation instruction information and the behavior information and provide them to the first feature quantity calculation unit and the second feature quantity calculation unit.

[0015] Moreover, in a structural example of the fluctuation detection device of the present invention, the first feature quantity calculation unit uses the maximum intensity among the intensities of each frequency component in the first frequency spectrum obtained by the frequency analysis of the deviation as the deviation feature quantity, and the second feature quantity calculation unit overlaps the second frequency spectrum obtained by the frequency analysis of the operation instruction information with the third frequency spectrum obtained by the frequency analysis of the behavior information, and calculates the similarity feature quantity based on the measure information of the part surrounded by the second frequency spectrum and the third frequency spectrum.

[0016] Moreover, in a structural example of the fluctuation detection device of the present invention, the determination unit determines that a fluctuation phenomenon occurs and it is necessary to maintain the operator when the deviation feature quantity is within a specified first threshold range and the similarity feature quantity is within a specified second threshold range.

[0017] Moreover, the fluctuation detection method of the present invention includes: a first step of performing frequency analysis on the deviation between the value of the operation instruction information and the value of the behavior information and calculating the deviation feature quantity, where the operation instruction information is input to an operator that drives an operation terminal of a feedback control system, and the behavior information is either the output from the operator to the operation terminal or the control quantity of the controlled object; a second step of performing frequency analysis on the values of the operation instruction information and the behavior information respectively and calculating the similarity feature quantity between the operation instruction information and the behavior information; and a third step of determining whether a fluctuation phenomenon occurs and whether it is necessary to maintain the operator based on the deviation feature quantity and the similarity feature quantity.

[0018] Effect of the Invention

[0019] According to the present invention, by providing the first feature quantity calculation unit, the second feature quantity calculation unit, and the determination unit, it is possible to determine whether maintenance of the actuator is required while detecting fluctuations. Therefore, in the present invention, it is possible to select an actuator with a high necessity for maintenance, and thus it is possible to reduce the maintenance cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a block diagram showing the configuration of a fluctuation detection device according to an embodiment of the present invention.

[0021] Figure 2 It is a flowchart for explaining the operation of the fluctuation detection device according to an embodiment of the present invention.

[0022] Figure 3 (A) of Figure 3 (B) of is a diagram showing an example of the deviation between the operation instruction information and the behavior information.

[0023] Figure 4 (A) of Figure 4 (B) of is a diagram showing another example of the deviation between the operation instruction information and the behavior information.

[0024] Figure 5 (A) of Figure 5 (B) of is a diagram showing an example of the operation instruction information and the behavior information in the case where there is an abnormality in the actuator, and an example of the spectrum obtained by frequency analysis of the deviation between the operation instruction information and the behavior information.

[0025] Figure 6 (A) of Figure 6 (B) of is a diagram showing an example of the operation instruction information and the behavior information in the case where there is an abnormality in the feedback control system, and an example of the spectrum obtained by frequency analysis of the deviation between the operation instruction information and the behavior information.

[0026] Figure 7 (A) of Figure 7 (B) of is a diagram showing an example of the spectrum obtained by frequency analysis of the operation instruction information in the case where there is an abnormality in the actuator, and an example of the spectrum obtained by frequency analysis of the behavior information in the case where there is an abnormality in the actuator.

[0027] Figure 8 (A) of Figure 8 (B) of is a diagram showing an example of the spectrum obtained by frequency analysis of the operation instruction information in the case where there is an abnormality in the feedback control system, and an example of the spectrum obtained by frequency analysis of the behavior information in the case where there is an abnormality in the feedback control system.

[0028] Figure 9 The block diagram showing the structure example of a computer of a fluctuation detection device for implementing an embodiment of the present invention.

[0029] Figure 10 The figure showing an example of a fluctuation.

[0030] Figure 11 The figure showing an example of operation instruction information and a control amount.

[0031] Figure 12 The figure showing another example of operation instruction information and a control amount.

[0032] Description of symbols

[0033] 1: Fluctuation detection device

[0034] 2: Operator operation data storage unit

[0035] 3: Feature quantity calculation unit

[0036] 4: Fluctuation determination unit

[0037] 5: Determination result output unit

[0038] 10: Operator

[0039] 30: Data extraction unit

[0040] 31: Deviation feature quantity calculation unit

[0041] 32: Similarity feature quantity calculation unit

[0042] 200: CPU

[0043] 201: Storage device

[0044] 202: Interface device (I / F)

[0045] BV: Behavior information

[0046] E: Deviation

[0047] MV: Operation signal, output

[0048] PV: Control amount

[0049] S1 to S8: Steps

[0050] SP: Operation instruction information Detailed implementation manners

[0051] [Principle of the invention]

[0052] The inventor focused on the deviation between the motion instruction information and the behavior information of the manipulator in order to consider the motion instruction information given to the manipulator. When vibratory motion instruction information is given to the manipulator, if the manipulator operates without a following delay with respect to the motion instruction information, then no vibratory deviation occurs between the motion instruction information and the behavior information. On the other hand, when non-vibratory motion instruction information is given to the manipulator and vibratory behavior information is exhibited in a form different from the instruction, a vibratory deviation occurs. Therefore, the inventor thought that by performing a frequency analysis on the deviation between the motion instruction information and the behavior information, if the maximum spectrum of the analysis result is above a threshold value, it is determined that a fluctuation phenomenon has occurred.

[0053] However, when vibratory motion instruction information is given, but the manipulator has a following delay, a vibratory deviation occurs. Therefore, if a frequency analysis of the deviation is performed, the following delay of the manipulator is also determined to be a fluctuation. Thus, the inventor focused on the waveform similarity between the motion instruction information and the behavior information in order to distinguish between fluctuations and the following delay of the manipulator.

[0054] When the manipulator has a following delay, although there is a phase deviation between the motion instruction information and the behavior information, the amplitude and period become similar waveforms. Therefore, if frequency analysis is performed on the motion instruction information and the behavior information respectively, similar spectra are calculated. On the other hand, when a fluctuation phenomenon occurs, the motion instruction information and the behavior information become waveforms with completely different amplitudes and periods, and thus, even if frequency analysis is performed on the motion instruction information and the behavior information respectively, no spectral similarity is seen.

[0055] In this way, the inventor thought that the similarity between the spectrum of the motion instruction information and the spectrum of the behavior information can be used to determine fluctuations and following delays. Therefore, in the present invention, the maximum spectrum of the deviation between the motion instruction information and the behavior information, and the similarity between the motion instruction information and the behavior information are used to determine whether maintenance of the manipulator is required while detecting fluctuations.

[0056] [Embodiment]

[0057] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Figure 1 It is a block diagram showing the structure of a fluctuation detection device according to an embodiment of the present invention. The fluctuation detection device 1 includes a manipulator operation data storage unit 2, a feature quantity calculation unit 3, a fluctuation determination unit 4, and a determination result output unit 5.

[0058] The feature quantity calculation unit 3 includes a data extraction unit 30, a deviation feature quantity calculation unit 31 (first feature quantity calculation unit), and a similarity feature quantity calculation unit 32 (second feature quantity calculation unit).

[0059] Figure 2The flowchart for explaining the operation of the fluctuation detection device 1. In the operator operation data storage unit 2 of the fluctuation detection device 1, the time-series data of the operation instruction information SP input from an upper controller (not shown) or the like to the operator 10 and the time-series data of the output MV output from the operator 10 to an operation end such as a damper or a valve according to the operation instruction information SP are stored.

[0060] In feedback control, the controller performs a control operation (for example, proportional integral derivative (PID) operation) so that the target value of the control and the control amount of the controlled object are consistent, and outputs the operation instruction information SP. The operation instruction information SP represents the value of the operation amount specified from the controller.

[0061] In this embodiment, the output MV (for example, a current signal, a voltage signal, etc.) of the operator 10 is set as the behavior information BV, but the control amount PV such as temperature or flow rate may also be set as the behavior information BV. In the case where the control amount PV is set as the behavior information BV, the time-series data of the control amount PV measured by the measurement end is stored in the operator operation data storage unit 2.

[0062] The data extraction unit 30 of the fluctuation detection device 1 extracts the time-series data of the operation instruction information SP and the time-series data of the behavior information BV from the operator operation data storage unit 2 ( Figure 2 Step S1). At this time, the data extraction unit 30 may not acquire all the information, but regards the time-series data of the behavior information BV that meets the specified extraction conditions as having the possibility of fluctuation and extracts it, and extracts the time-series data of the operation instruction information SP in the same time range as the time-series data of the behavior information BV.

[0063] The extraction conditions are preset based on past insights. As extraction conditions, for example, there are the following conditions: the case where the period of the behavior information BV is within a specified range, and the case where only one-sided deviation of the positive deviation (BV>SP) and the negative deviation (BV<SP) between the value of the operation instruction information SP and the value of the behavior information BV exceeds a specified threshold, and at least one of these two cases holds.

[0064] The deviation characteristic quantity calculation unit 31 of the fluctuation detection device 1 calculates the deviation E = BV - SP between the value of the operation instruction information SP and the value of the behavior information BV at each moment according to the time-series data of the operation instruction information SP and the time-series data of the behavior information BV extracted by the data extraction unit 30 ( Figure 2 Step S2). For example, if it is Figure 3 the operation instruction information SP and the behavior information BV as in (A) of, then Figure 3 the deviation E as in (B) of is obtained. Moreover, if it is Figure 4action instruction information SP and behavior information BV such as (A) thereof, then obtain Figure 4 a deviation E such as (B) thereof.

[0065] Next, the deviation feature amount calculation unit 31 performs a frequency analysis on the time series data of the calculated deviation E ( Figure 2 step S3), and calculates a deviation feature amount x based on the result of the frequency analysis ( Figure 2 step S4). As a representative method of frequency analysis, there is a fast Fourier transform (FFT). As is well known, when frequency analysis is performed on data, a frequency spectrum representing the intensity distribution of frequency components is obtained. The deviation feature amount calculation unit 31 uses the maximum intensity among the intensities of the respective frequency components in the frequency spectrum as the deviation feature amount x. In this way, index value conversion of the periodically generated deviation E can be achieved.

[0066] Figure 5 (A) thereof is a diagram showing an example of action instruction information SP and behavior information BV in the case where the operator 10 is abnormal, Figure 5 (B) thereof is a diagram showing an example of a frequency spectrum obtained by frequency analysis of the deviation E between the action instruction information SP and the behavior information BV of ( Figure 5 (A) thereof. Figure 6 (A) thereof is a diagram showing an example of action instruction information SP and behavior information BV in the case where an abnormality occurs in a controller or the like of the feedback control system, Figure 6 (B) thereof is a diagram showing an example of a frequency spectrum obtained by frequency analysis of the deviation E between the action instruction information SP and the behavior information BV of ( Figure 6 (A) thereof.

[0067] On the other hand, the similarity feature amount calculation unit 32 of the fluctuation detection device 1 performs frequency analysis on the time series data of the action instruction information SP extracted by the data extraction unit 30 and the time series data of the behavior information BV respectively ( Figure 2 step S5), and calculates a similarity feature amount y between the action instruction information SP and the behavior information BV based on the result of the frequency analysis ( Figure 2 step S6).

[0068] Figure 7 (A) thereof is a diagram showing an example of a frequency spectrum obtained by frequency analysis of the action instruction information SP of ( Figure 5 (A) thereof, Figure 7 (B) thereof is a diagram showing an example of a frequency spectrum obtained by frequency analysis of the behavior information BV of ( Figure 5 (A) thereof. Figure 8 (A) thereof is a diagram showing an example of a frequency spectrum obtained by frequency analysis of the action instruction information SP of ( Figure 6 (A) thereof, Figure 8(B) is a diagram showing an example of a spectrum obtained by frequency analysis of behavior information BV indicating the behavior of (A) of Figure 6 FIG.

[0069] The similarity feature quantity calculation unit 32 overlaps the spectrum of the action instruction information SP and the spectrum of the behavior information BV so that the scales of the frequencies on the horizontal axis are the same and the scales of the intensities on the vertical axis are the same, and calculates the area d (measurement information) of the portion surrounded by the spectrum of the action instruction information SP and the spectrum of the behavior information BV. Next, the similarity feature quantity calculation unit 32 calculates the similarity feature quantity y as follows according to the area d.

[0070] y = 1 / (1 + d) ··· (1)

[0071] In this way, the index value of the similarity between the action instruction information SP and the behavior information BV can be realized. The closer the value of the similarity feature quantity y in this embodiment is to 1, the more similar the action instruction information SP and the behavior information BV are. One purpose of this calculation method of the similarity feature quantity y is to easily understand the similarity. The calculation method in this embodiment is an example, and other calculation methods may also be used.

[0072] In addition, as can be seen from the operations of the deviation feature quantity calculation unit 31 and the similarity feature quantity calculation unit 32, it is necessary to set the value ranges (ranges) of the action instruction information SP and the behavior information BV to be the same. Therefore, the action instruction information SP and the behavior information BV must be normalized in advance so that the minimum values that can be taken are the same and the maximum values that can be taken are the same. When the value ranges of the action instruction information SP and the behavior information BV are the same, normalization is not required. When the value ranges of the action instruction information SP and the behavior information BV are different, for example, it is only necessary to set the normalization to be performed by the data extraction unit 30 of the feature quantity calculation unit 3.

[0073] Next, the fluctuation determination unit 4 of the fluctuation detection device 1 determines whether a fluctuation phenomenon has occurred and whether the operator 10 needs maintenance based on the deviation feature quantity x and the similarity feature quantity y ( Figure 2 Step S7). Table 1 shows the operation of the fluctuation determination unit 4.

[0074] [Table 1]

[0075]

[0076] When the similarity feature quantity y is less than 0.05 (a in Table 1), the action instruction information SP and the behavior information BV are constantly or locally very different. In the case of a in Table 1, the fluctuation determination unit 4 determines that no fluctuation phenomenon has occurred and the operator 10 does not need to be maintained.

[0077] When the deviation characteristic quantity x is less than 0.1 and the similarity characteristic quantity y is 0.05 or more and less than 0.3 (case b in Table 1), or when the deviation characteristic quantity x is less than 0.1 and the similarity characteristic quantity y is 0.3 or more (case e in Table 1), either the operation instruction information SP or the behavior information BV changes locally. In the cases of b and e in Table 1, the fluctuation determination unit 4 determines that no fluctuation phenomenon has occurred and there is no need to maintain the operator 10.

[0078] When the deviation characteristic quantity x is 0.1 or more and less than 0.5 and the similarity characteristic quantity y is 0.05 or more and less than 0.3 (case c in Table 1), when the deviation characteristic quantity x is 0.1 or more and less than 0.5 and the similarity characteristic quantity y is 0.3 or more (case f in Table 1), or when the deviation characteristic quantity x is 0.5 or more and the similarity characteristic quantity y is 0.3 or more (case g in Table 1), large deviations occur periodically. In the cases of c, f, and g in Table 1, the fluctuation determination unit 4 determines that there is a possibility of a fluctuation phenomenon occurring, but there is no need to maintain the operator 10. Figure 6 of (A), Figure 6 of (B), Figure 8 of (A), Figure 8 The examples shown in (B) correspond to the case of g in Table 1.

[0079] When the deviation characteristic quantity x is 0.5 or more and the similarity characteristic quantity y is 0.05 or more and less than 0.3 (case d in Table 1), the fluctuation determination unit 4 determines that a fluctuation phenomenon has occurred and the operator 10 needs to be maintained. Figure 5 of (A), Figure 5 of (B), Figure 7 of (A), Figure 7 The examples shown in (B) correspond to the case of d in Table 1.

[0080] When the deviation characteristic quantity x is large and the similarity between the operation instruction information SP and the behavior information BV indicated by the similarity characteristic quantity y is low, the fluctuation determination unit 4 determines that a fluctuation phenomenon has occurred and the operator 10 needs to be maintained. More specifically, when the deviation characteristic quantity x is within the first threshold range (0.5 ≤ x in this embodiment) and the similarity characteristic quantity y is within the second threshold range (0.05 ≤ y < 0.3 in this embodiment), the fluctuation determination unit 4 determines that a fluctuation phenomenon has occurred and the operator 10 needs to be maintained. The first threshold range and the second threshold range are examples and are not limited to this embodiment.

[0081] The determination result output unit 5 of the fluctuation detection device 1 outputs the determination result of the fluctuation determination unit 4 ( Figure 2 step S8). As a method for outputting the determination result, for example, there are methods such as displaying and informing the content of the determination result or sending the information informing the determination result to the outside.

[0082] Thus, in this embodiment, it is possible to determine whether the manipulator 10 needs to be maintained while detecting fluctuations. Therefore, in this embodiment, the manipulator 10 with a high necessity for maintenance can be selected, and thus the reduction of maintenance costs can be achieved.

[0083] The fluctuation detection device 1 described in this embodiment can be implemented by a computer including a Central Processing Unit (CPU), a storage device, and an interface, and a program for controlling these hardware resources. An example of the structure of the computer is illustrated in Figure 9 .

[0084] The computer includes a CPU 200, a storage device 201, and an interface device (I / F) 202. To the I / F 202, the manipulator 10 and the like are connected. In such a computer, the program for implementing the fluctuation detection method of the present invention is stored in the storage device 201. The CPU 200 executes the processing described in this embodiment according to the program stored in the storage device 201.

[0085] Industrial applicability

[0086] The present invention can be applied to the technology for detecting fluctuations.

Claims

1. A fluctuation detection device, characterized in that, Comprising: A first feature quantity calculation unit configured to perform frequency analysis on the deviation between the value of the action instruction information and the value of the behavior information and calculate a deviation feature quantity, where the action instruction information is input to an actuator that drives an operation end of a feedback control system, and the behavior information is either the output from the actuator to the operation end or the control quantity of a controlled object; A second feature quantity calculation unit configured to perform frequency analysis on the values of the action instruction information and the behavior information respectively and calculate a similarity feature quantity between the action instruction information and the behavior information; And A determination unit configured to determine whether a fluctuation phenomenon occurs and whether the actuator needs to be maintained based on the deviation feature quantity and the similarity feature quantity.

2. The fluctuation detection device according to claim 1, wherein Further comprising: A data extraction unit configured to extract the action instruction information and the behavior information that meet the specified extraction conditions from the action instruction information and the behavior information and provide them to the first feature quantity calculation unit and the second feature quantity calculation unit.

3. The fluctuation detection device according to claim 1 or 2, wherein The first feature quantity calculation unit uses the maximum intensity among the intensities of each frequency component in the first frequency spectrum obtained by the frequency analysis of the deviation as the deviation feature quantity, The second feature quantity calculation unit overlaps the second frequency spectrum obtained by the frequency analysis of the action instruction information with the third frequency spectrum obtained by the frequency analysis of the behavior information, and calculates the similarity feature quantity based on the measure information of the part surrounded by the second frequency spectrum and the third frequency spectrum.

4. The fluctuation detection device according to claim 1 or 2, wherein The determination unit determines that a fluctuation phenomenon occurs and the actuator needs to be maintained when the deviation feature quantity is within a specified first threshold range and the similarity feature quantity is within a specified second threshold range.

5. The fluctuation detection device according to claim 3, wherein The determination unit determines that a fluctuation phenomenon occurs and the actuator needs to be maintained when the deviation feature quantity is within a specified first threshold range and the similarity feature quantity is within a specified second threshold range.

6. A wave detection method, characterized in that, Comprising: A first step of performing frequency analysis on the deviation between the value of the action instruction information and the value of the behavior information and calculating a deviation feature quantity, where the action instruction information is input to an actuator that drives an operation end of a feedback control system, and the behavior information is either the output from the actuator to the operation end or the control quantity of a controlled object; A second step of performing frequency analysis on the values of the action instruction information and the behavior information respectively and calculating a similarity feature quantity between the action instruction information and the behavior information; And A third step of determining whether a fluctuation phenomenon occurs and whether the actuator needs to be maintained based on the deviation feature quantity and the similarity feature quantity.

7. The fluctuation detection method according to claim 6, characterized in that Further comprising: A fourth step of extracting the action instruction information and the behavior information that meet the specified extraction conditions from the action instruction information and the behavior information as the data used in the first step and the second step.

8. The fluctuation detection method according to claim 6 or 7, wherein The first step includes the following steps: taking the maximum intensity among the intensities of the frequency components in the first frequency spectrum obtained by the frequency analysis of the deviation as the deviation characteristic quantity. The second step includes the following steps: superimposing the second frequency spectrum obtained by the frequency analysis of the action indication information and the third frequency spectrum obtained by the frequency analysis of the behavior information, and calculating the similarity characteristic quantity based on the measure information of the part surrounded by the second frequency spectrum and the third frequency spectrum.

9. The fluctuation detection method according to claim 6 or 7, wherein The third step includes the following steps: when the deviation characteristic quantity is within a specified first threshold range and the similarity characteristic quantity is within a specified second threshold range, it is determined that a fluctuation phenomenon occurs and the operator needs to be maintained.

10. The fluctuation detection method according to claim 8, wherein The third step includes the following steps: when the deviation characteristic quantity is within a specified first threshold range and the similarity characteristic quantity is within a specified second threshold range, it is determined that a fluctuation phenomenon occurs and the operator needs to be maintained.

Images