Monitoring devices, monitoring methods and recording media

Abstract

The monitoring device includes: an operation record acquisition unit that acquires operation records for multiple failure cases; a threshold setting unit that sets a threshold; a loss rate acquisition unit that acquires a loss rate representing the degree of wear and tear on the device corresponding to operating time; a failure rate cumulative frequency acquisition unit that acquires the shift of the failure rate cumulative frequency for the multiple failure cases relative to the loss rate; an evaluation index calculation unit that calculates an evaluation index as a value obtained by integrating the failure rate cumulative frequency with the loss rate; and a notification unit that notifies the user that maintenance is required when the loss rate, derived from the relationship between the load rate or load change rate of the monitored device and the threshold, reaches a given management value determined based on the shift of the failure rate cumulative frequency for the failure cases, wherein the threshold setting unit sets the threshold such that the evaluation index is minimized.

Description

Technical Field

[0001] This disclosure relates to monitoring devices, monitoring methods, and procedures. This application claims priority based on Japanese Patent Application No. 2020-130789, filed on July 31, 2020, the contents of which are incorporated herein by reference. Background Technology

[0002] In asset management services that guarantee the value of equipment assets by ensuring the integrity of monitoring data monitoring devices and complete sets of equipment, and by carrying out appropriate maintenance and improving usage methods, it is important to predict when the probability of product wear and tear failures will increase and to carry out appropriate maintenance.

[0003] With the service life of the device as the horizontal axis and the failure rate curve (the so-called bus tag curve) as the vertical axis, TBM (Time-Based Maintenance) has been implemented in the past to predict the probability of wear failure based on the service life.

[0004] Traditional TBMs set maintenance periods without reflecting usage conditions, which sometimes resulted in excessive maintenance costs for equipment with low uptime. Conversely, if extremely high loads were applied, unplanned failures might occur before the maintenance period.

[0005] Therefore, the shift towards Condition Based Maintenance (CBM) – which involves maintaining devices backward from detecting abnormal behavior – has become a trend.

[0006] In Patent Document 1, the following method is used: when a large residual suddenly appears between the multiple regression analysis and principal component analysis model of the observed KPI values ​​when they are normal and the actual observed values, or when the residuals show an increasing tendency, it is evaluated as abnormal.

[0007] Prior art literature

[0008] Patent documents

[0009] Patent Document 1: JP Patent No. 6474564 Summary of the Invention

[0010] -The problem the invention aims to solve-

[0011] However, there are many cases where the period between the point at which the abnormal behavior of the device becomes apparent and the point at which the device stops is relatively short. Therefore, even with CBM, there are sometimes situations that require unplanned shutdowns and emergency maintenance responses.

[0012] This disclosure was made in consideration of the above-mentioned circumstances, and its purpose is to provide a monitoring device, monitoring method and procedure that can suppress excessive maintenance, unplanned shutdowns, etc., and can properly perform maintenance and upkeep.

[0013] -Methods used to solve problems

[0014] According to one aspect of this disclosure, a monitoring device includes: an operation record acquisition unit that acquires operation records for multiple fault cases, representing the shift of load rate or load change rate relative to the operation time up to the fault; a threshold setting unit that sets a threshold for comparing the load rate or load change rate in the operation records of the multiple fault cases; a loss rate acquisition unit that acquires a loss rate, which is a value derived from the relationship between the load rate or load change rate and the threshold and represents the degree of wear and tear on the device corresponding to the operation time; a fault rate cumulative frequency acquisition unit that acquires the shift of the cumulative frequency of the fault rate for the multiple fault cases relative to the loss rate; an evaluation index calculation unit that calculates an evaluation index as a value obtained by integrating the cumulative frequency of the fault rate with the loss rate; and a notification unit that notifies that maintenance is required when the loss rate derived from the relationship between the load rate or load change rate of the monitored device and the threshold reaches a given management value determined based on the shift of the cumulative frequency of the fault rate for the fault cases, wherein the threshold setting unit sets the threshold such that the evaluation index is minimized.

[0015] According to one aspect of this disclosure, the monitoring method includes the following steps: obtaining operation records for multiple failure cases, representing the shift of load rate or load change rate relative to the operating time up to the time of failure; setting a threshold for comparing the load rate or load change rate in the operation records of the multiple failure cases; obtaining a loss rate, which is a value derived from the relationship between the load rate or load change rate and the threshold and represents the degree of wear and tear on the device corresponding to the operating time; obtaining the shift of the cumulative frequency of failure rates of the multiple failure cases relative to the loss rate; calculating an evaluation index as a value obtained by integrating the cumulative frequency of failure rates with the loss rate; and issuing a notification indicating that maintenance is required when the loss rate derived from the relationship between the load rate or load change rate of the device to be monitored and the threshold reaches a given management value determined based on the shift of the cumulative frequency of failure rates of the failure cases. In the step of setting the threshold, the threshold is set such that the evaluation index is minimized.

[0016] According to one aspect of this disclosure, a program causes a computer to perform the following steps: obtaining operation records for multiple failure cases, representing the shift of load rate or load change rate relative to the operating time up to the failure point; setting a threshold for comparing the load rate or load change rate in the operation records of the multiple failure cases; obtaining a loss rate, which is a value derived from the relationship between the load rate or load change rate and the threshold and represents the degree of wear and tear on the device corresponding to the operating time; obtaining the shift of the cumulative frequency of failure rates of the multiple failure cases relative to the loss rate; calculating an evaluation index as a value obtained by integrating the cumulative frequency of failure rates with the loss rate; and issuing a notification indicating that maintenance is required if the loss rate derived from the relationship between the load rate or load change rate of the device to be monitored and the threshold reaches a given management value determined based on the shift of the cumulative frequency of failure rates of the failure cases, wherein in the step of setting the threshold, the threshold is set such that the evaluation index is minimized.

[0017] -Invention Effects-

[0018] According to the monitoring device, monitoring method and recording medium disclosed herein, excessive maintenance, unplanned shutdowns, etc., can be suppressed, and proper maintenance and upkeep can be performed. Attached Figure Description

[0019] Figure 1 This is a diagram showing the structure of the monitoring device according to the first embodiment.

[0020] Figure 2 This is a diagram showing the structure of the monitoring device according to the first embodiment.

[0021] Figure 3 This is a diagram illustrating the processing flow of the monitoring device according to the first embodiment.

[0022] Figure 4 This diagram illustrates the processing of the monitoring device according to the first embodiment.

[0023] Figure 5 This diagram illustrates the processing of the monitoring device according to the first embodiment.

[0024] Figure 6 This diagram illustrates the processing of the monitoring device according to the second embodiment.

[0025] Figure 7 This diagram illustrates the processing of the monitoring device according to the third embodiment. Detailed Implementation

[0026] <First Implementation>

[0027] The following is for reference. Figures 1-5 The monitoring device involved in the first embodiment will be described in detail.

[0028] (Functional structure of the specification information management device)

[0029] Figure 1 , Figure 2 This is a diagram showing the structure of the monitoring system according to the first embodiment.

[0030] Figure 1 The monitoring device 1 shown monitors the operating status of the monitoring device 2 and issues a notification (alarm) indicating the need for maintenance at appropriate intervals. In this embodiment, the monitoring device 2 is a large refrigerator, submersible pump, etc., but it is not limited to this in other embodiments. The monitoring device 1 can be applied to all devices, machines, and equipment.

[0031] The monitoring device 1 receives the operating load rate or load change rate from the device 2 at all times, and sends an alarm when the loss rate (described later) calculated based on the load rate or load change rate reaches a given management value.

[0032] like Figure 1 As shown, the monitoring device 1 includes a CPU 10, a memory 11, a communication interface 12, and a recording medium 13.

[0033] CPU10 performs various functions by following pre-prepared programs. Details regarding the functions of CPU10 will be described later.

[0034] Memory 11 is the so-called main storage device, which provides the storage area required for the operation of CPU 10.

[0035] Communication interface 12 is an interface for establishing a connection with a wide area communication network (such as the Internet).

[0036] Recording medium 13 is a non-volatile, high-capacity storage area such as HDD (Hard Disk Drive) or SSD (Solid State Drive). Recording medium 13 contains operational records DB related to multiple (100 to 500 cases) failure incidents of device 2.

[0037] (Functional structure of CPU)

[0038] Figure 2 This diagram illustrates the function of the CPU in the specification information management device according to the first embodiment.

[0039] like Figure 2As shown, the CPU 10 performs the functions of an operation record acquisition unit 100, a threshold setting unit 101, a loss rate acquisition unit 102, a failure rate cumulative frequency acquisition unit 103, an evaluation index calculation unit 104, a notification unit 105, and a loss rate determination unit 106 by operating according to a pre-prepared program.

[0040] The operation record acquisition unit 100 acquires the operation record DB of failure cases (past operation, failure of device 2). The operation record DB shows the shift of load rate or load change rate relative to the operating time of each device 2 until the failure.

[0041] The threshold setting unit 101 sets a threshold that compares the load rate or load change rate with the operation records of multiple fault cases.

[0042] The loss rate acquisition unit 102 acquires a loss rate that represents the degree of wear and tear on the device 2 corresponding to the operating time. As will be described later, the loss rate is a value derived from the relationship between the load rate or load change rate of the device 2 and a threshold set by the threshold setting unit 101.

[0043] The failure rate cumulative frequency acquisition unit 103 acquires the change of the failure rate cumulative frequency relative to the loss rate of multiple failure cases.

[0044] The evaluation index calculation unit 104 calculates the value obtained by integrating the cumulative failure rate frequency with the loss rate, which is the evaluation index. Furthermore, the threshold setting unit 101 sets a threshold to minimize the evaluation index. This function of the threshold setting unit 101 will be described later.

[0045] The notification unit 105 issues a notification (alarm) indicating that maintenance is required when the loss rate, derived from the relationship between the load rate or load change rate of the monitored device 2 and a threshold, reaches a given management value. Here, the given management value is determined based on the cumulative frequency of failure events.

[0046] The loss rate determination unit 106 determines one of the various "loss rate candidates" derived from the relationship between the load rate or load change rate and the threshold as the loss rate used in the monitoring of the device 2.

[0047] (Processing flow of the monitoring device)

[0048] Figure 3 This is a diagram illustrating the processing flow of the monitoring device according to the first embodiment.

[0049] Figure 4 , Figure 5 This diagram illustrates the processing of the monitoring device according to the first embodiment.

[0050] The following is for reference. Figures 3-5 This section details the processing flow performed by monitoring device 1.

[0051] Figure 3 The processing flow of the monitoring device 1 shown is a processing flow executed in the stage before the monitoring device 2. It is a processing flow used to set management values ​​for the device 2, which is set as the monitoring object, to properly trigger an alarm.

[0052] First, the operation record acquisition unit 100 of the monitoring device 1 acquires operation records DB regarding multiple past failure cases (step S01). In this embodiment, the operation records DB are recorded on the recording medium 13, but this is not a limitation in other embodiments. In other embodiments, the operation records DB are recorded on an external device, and the operation record acquisition unit 100 may access the external device to acquire the operation records DB.

[0053] refer to Figure 4 This example illustrates how to run a database (DB).

[0054] like Figure 4 As shown, the operation record DB records the time example of the load rate or load change rate from the start of operation to the point of failure in the fault case. Here, the load rate is a value that represents the level of load during the operation of device 2. In the case of device 2 being a large refrigeration unit, the load rate can be the power consumption (W) and speed (rpm) of the compressor. In addition, the load rate can be the magnitude of vibration obtained by the vibration sensor, the temperature observation value of the temperature sensor, etc.

[0055] In addition, the load change rate is a value that represents the degree of change in the load rate, and can be calculated, for example, by the time derivative of the observed load rate.

[0056] return Figure 3 Next, the threshold setting unit 101 of the monitoring device 1 sets a threshold TH for comparison with the load rate or load change rate of each operation record DB (step S02). In addition, the threshold TH set here is a temporary value as an initial value, which will be adjusted to the normal optimal value in subsequent steps.

[0057] Next, the loss rate acquisition unit 102 of the monitoring device 1 acquires a "candidate loss rate" (step S03) as a value derived from the relationship between the load rate or load change rate and the threshold TH.

[0058] Here, for reference Figure 4 Detailed explanation of the processing steps S02 and S03.

[0059] The so-called loss rate is a value representing the degree of wear and tear on device 2 corresponding to operating time. In this embodiment, as... Figure 4As shown, the loss rate is selected from the "candidates for loss rate" of 24 modes that include combinations of the following (1) to (3).

[0060] (1) Load rate or load change rate (2 modes)

[0061] (2) Load rate or load change rate from the first to the fourth power (4 modes)

[0062] (3) The following conditions (a) to (c) (3 patterns)

[0063] (a) ...the total amount of time during which the load rate or load change rate is above the threshold TH.

[0064] (b) ...the number of times the load rate or load change rate intersects with the threshold TH.

[0065] (c)...the time integral of the amount by which the load rate or load change rate exceeds the threshold TH.

[0066] return Figure 3 Next, the failure rate accumulation frequency acquisition unit 103 of the monitoring device 1 acquires the failure rate accumulation frequency for the "candidates of loss rate" of the above-mentioned 24 modes respectively (step S04).

[0067] Here, for reference Figure 5 Let's explain the process of step S04 in detail.

[0068] Figure 5 The chart shown uses the loss rate at the time of failure in each failure case as the horizontal axis, and the ratio of the cumulative number of devices that failed up to that loss rate to the total number of devices (cumulative failure rate frequency) as the vertical axis. For example, in Figure 5 In the example shown, the cumulative frequency of failure rate in the loss rate xi is characterized by the following equation (1).

[0069] [Mathematical Expression 1]

[0070]

[0071] In this embodiment, Figure 5 The charts shown are created with 24 versions of each of the aforementioned "candidate loss rates". Furthermore, the units for each of the 24 "candidate loss rates" are different, and normalization within the range of 0 to 1 is performed for comparison purposes (see reference). Figure 5 (The horizontal axis of the chart).

[0072] In addition, Figure 5In the example shown, the lower limit of the cumulative failure rate is set to a value greater than 0% (e.g., 10%), and the upper limit of the cumulative failure rate is set to a value less than 100% (e.g., 90%). In this way, by excluding failure cases that are prone to include outliers (the range where the cumulative failure rate is close to 0 or 100), the accuracy of the determination using the cumulative failure rate can be improved.

[0073] return Figure 3 Next, the evaluation index calculation unit 104 of the monitoring device 1 calculates the evaluation index based on the 24 cumulative frequencies of failure rates obtained in step S04 (step S05).

[0074] Here, for reference Figure 5 Let me explain the process of step S05 in detail.

[0075] The evaluation index is the value obtained by integrating the cumulative frequency of the failure rate (Equation (1)) obtained in step S04 with the loss rate in the range of xL (0) to xH (1), as represented by Equation (2).

[0076] [Mathematical Expression 2]

[0077]

[0078] That is, evaluation indicators become Figure 5 The area of ​​the range from xL to xH in the graph showing the cumulative frequency of failure rates.

[0079] return Figure 3 Next, the threshold setting unit 101 of the monitoring device 1 adjusts the threshold TH so that the evaluation index obtained in step S05 ( Figure 5 The area of ​​the cumulative frequency of failures becomes the minimum (step S06).

[0080] Specifically, the threshold setting unit 101 increases or decreases the threshold TH (initial value) set in step S02. As a result, the loss rate xi changes due to the increase or decrease of the threshold TH (see reference). Figure 4 Therefore, the evaluation index changes in response to the change in the loss rate xi. The threshold setting unit 101 searches for the threshold TH that minimizes the evaluation index while referring to the calculated result of the evaluation index.

[0081] The threshold setting unit 101 determines the minimum threshold TH for each of the 24 cumulative failure rate frequencies. As a result, 24 thresholds TH that minimize the area of ​​each of the 24 cumulative failure rate frequencies are determined.

[0082] Next, the loss rate determination unit 106 of the monitoring device 1 determines one of the "candidates for loss rate" that is the smallest among the 24 "minimum values ​​of evaluation indicators" calculated from the cumulative frequencies of 24 failure rates, and gives the threshold TH of the "minimum value of evaluation indicator". The loss rate determination unit 106 determines the combination of the "candidate for loss rate" and the threshold TH as the loss rate and threshold TH to be used in the monitoring of the device 2 that is set as the monitoring object in the future (step S07).

[0083] Finally, a management value for the loss rate (the threshold for issuing an alarm) is determined. Monitoring device 1 calculates the loss rate using the loss rate determined in step S07 and the threshold TH, based on the operational history data received continuously from the monitored device 2 (step S08). The management value for the loss rate can be determined by the operator based on a graph of the cumulative failure rate obtained using the threshold TH from step S07.

[0084] (Function, Effect)

[0085] refer to Figure 5 This embodiment will explain the function and effect of the monitoring device 1. The threshold setting unit 101 in this embodiment is adjusted to minimize the evaluation index (see reference). Figure 3 Step S06). Therefore, Figure 5 The integral value (area) of the cumulative frequency of failures from xL to xH in the graph is minimized. By minimizing this integral value (area), it is possible to select a value relative to the loss rate ( Figure 5 The horizontal axis) and the cumulative frequency of failures ( Figure 5 The vertical axis (TH) should be set as steeply as possible to represent the threshold value TH for the characteristic that rises sharply. Therefore, by selecting the loss rate as the "management value" before the cumulative failure rate frequency rises sharply, an alarm can be appropriately issued at the stage before signs of a failure begin.

[0086] As described above, the monitoring device 1 according to the first embodiment can perform maintenance and upkeep based on the wear rate according to actual usage (failure cases), thus suppressing the occurrence of excessive maintenance, unplanned shutdowns, etc., and can perform maintenance and upkeep appropriately.

[0087] Furthermore, the monitoring device 1 according to the first embodiment calculates evaluation indicators corresponding to various types (24 modes) of "candidates for loss rate", and sets the corresponding evaluation indicators for each type of "candidate for loss rate" to a minimum threshold TH. Then, the monitoring device 1 (loss rate determination unit 106) determines the candidate for loss rate with the minimum "minimum value of evaluation indicator" under the set threshold TH as the loss rate used for monitoring in the device 2.

[0088] Therefore, even if it is unknown what parameters should be monitored to correctly determine the end of the product's lifespan, the optimal parameters (loss rate) for determining the product's lifespan can be selected from past failure cases.

[0089] <Second Implementation Method>

[0090] Next, refer to Figure 6 The monitoring device involved in the second embodiment will be described in detail.

[0091] (Processing of the monitoring device according to the second embodiment)

[0092] Figure 6 This diagram illustrates the processing of the monitoring device according to the second embodiment.

[0093] The evaluation index calculation unit 104 in this embodiment calculates the macro average of the evaluation index of each classification group of the device 2 classified according to one of the multiple classification methods (described later).

[0094] Furthermore, the threshold setting unit 101 in this embodiment sets a threshold TH so that the macro average becomes the minimum.

[0095] The following is for reference. Figure 6 The processing of the monitoring device 1 involved in this embodiment will be explained in detail.

[0096] The device 2 involved in this embodiment is a large-scale refrigeration machine, and for example, consider the following three classification methods: (α), (β), and (γ).

[0097] (α) Classified only by model number

[0098] (β) Classified only by specifications

[0099] (γ) Classification by combination of model and specification

[0100] The evaluation index calculation unit 104 in this embodiment calculates evaluation indexes for each of the multiple fault cases according to the classification methods (α), (β), and (γ). Specifically, it is as follows.

[0101] The processing of the evaluation index calculation unit 104 will be explained using the classification method (α) as an example.

[0102] The evaluation index calculation unit 104 extracts the operation record DB of the failure cases belonging to the category group "Model A" from all failure cases, sets a given threshold TH (initial value) for the operation record DB, and calculates the evaluation index for each "candidate of loss rate".

[0103] Next, the evaluation index calculation unit 104 extracts the operation record DB of the failure cases belonging to the category group "Model B" from all failure cases, sets a given threshold TH (initial value) for the operation record DB, and calculates the evaluation index for each "candidate of loss rate".

[0104] Then, the classification method determination unit 107 calculates the macro average of the evaluation indicators for "Model A" and "Model B" respectively.

[0105] The threshold setting unit 101 increases or decreases the threshold TH common to "Model A" and "Model B" to search for the macro average to become the minimum threshold TH.

[0106] exist Figure 6 The example shown illustrates that, when using the classification method (α), one of the "candidates for loss rate" at a certain threshold TH yielded a minimum macro-average value of 0.45.

[0107] The monitoring device 1 also adjusts the threshold TH commonly used in each classification group for the classification methods (β) and (γ), and searches for the minimum "candidate of loss rate" and threshold TH by the macro average.

[0108] exist Figure 6 The example shown illustrates that, when using the classification method (β), one of the "candidates for loss rate" at a certain threshold TH yields a minimum macro-average value of 0.47. Furthermore, in Figure 6 The example shown illustrates that, when using classification method (c), one of the "candidates for loss rate" at a certain threshold TH yielded a minimum macro-average value of 0.46.

[0109] Monitoring device 1 uses the minimum "loss rate candidate" and threshold TH as macro averages to monitor device 2.

[0110] As described above, the evaluation index calculation unit 104 calculates the macro average of the evaluation index for each classification group of the device after classification according to the multiple classification methods (α), (β), and (γ). Furthermore, the threshold setting unit 101 sets a threshold TH to minimize the macro average.

[0111] (Function, Effect)

[0112] As described above, according to the monitoring device 1 of the second embodiment, a more suitable "candidate for loss rate" and threshold TH are determined based on the macro-average of the evaluation index calculated according to each classification method.

[0113] <Third Implementation Method>

[0114] Next, refer to Figure 7The monitoring device involved in the third embodiment will be described in detail.

[0115] (Processing of the monitoring device according to the third embodiment)

[0116] Figure 7 This diagram illustrates the processing of the monitoring device according to the third embodiment.

[0117] In addition to the functions of the first embodiment, the loss rate acquisition unit 102 of this embodiment further searches for candidate combinations of evaluation indicators that result in the lowest loss rate. Hereinafter, the processing of the loss rate acquisition unit 102 according to the third embodiment will be described in detail.

[0118] Figure 7 The x1, x2, x3, ... shown are each of the various types (24 modes) of "candidates for loss rate". The loss rate acquisition unit 102 according to the third embodiment uses genetic programming to search for the combination of "candidates for loss rate" that can obtain the minimum evaluation index.

[0119] Specifically, such as Figure 7 As shown, firstly, a portion of the elements of each of the various "candidate loss rates" combinations (a1), (b1), ... belonging to the nth generation are extracted and incorporated into a portion of other combinations. Figure 7 In the example shown, the loss rate acquisition unit 102 incorporates a portion of the elements of combination (a1) into the nth generation combination (b1) to create the (n+1)th generation combination (b2). Similarly, the loss rate acquisition unit 102 incorporates a portion of the elements of combination (b1) into the nth generation combination (a1) to create the (n+1)th generation combination (a2).

[0120] Similar to the first embodiment, while increasing or decreasing the threshold TH, the resulting combinations of multiple types belonging to the (n+1)th generation (a2), (b2), ... are searched and used as loss rates. Figure 5 The horizontal axis of the chart represents the minimum value of the evaluation metric under the given conditions. Then, only the combinations (a2), (b2), ... that further reduce the minimum value of the evaluation metric are selected to create the next generation (n+2th generation) of combinations.

[0121] In this way, instead of setting any one of the "candidates for loss rate" in the 24 modes as "loss rate" alone, "loss rate" is defined as a combination of multiple variables (candidates for loss rate), the minimization of the evaluation index can be further pursued. As a result, the "loss rate" and its management value that can be used for monitoring in device 2 are more appropriate.

[0122] <Postscript>

[0123] The monitoring device, monitoring method, and procedure described in each embodiment are as follows.

[0124] (1) The monitoring device 1 according to the first method includes: an operation record acquisition unit 100, which acquires operation records for multiple fault cases, representing the progression of load rate or load change rate relative to the operation time up to the fault; a threshold setting unit 101, which sets a threshold for comparing the load rate or load change rate in the operation records of the multiple fault cases; and a loss rate acquisition unit 102, which acquires a loss rate that is a value derived from the relationship between the load rate or load change rate and the threshold and represents the degree of wear and tear of the device corresponding to the operation time; therefore The failure rate cumulative frequency acquisition unit 103 acquires the shift in the cumulative frequency of failure rates for the plurality of failure cases relative to the loss rate; the evaluation index calculation unit 104 calculates an evaluation index as a value obtained by integrating the cumulative frequency of failure rates with the loss rate; the notification unit 105 notifies the user that maintenance is required when the loss rate, derived from the relationship between the load rate or load change rate of the monitored device and the threshold, reaches a given management value determined based on the shift in the cumulative frequency of failure rates for the failure cases. Then, the threshold setting unit 101 sets the threshold such that the evaluation index is minimized.

[0125] (2) The monitoring device 1 according to the second method includes: a loss rate determination unit 106, which determines one of a variety of loss rate candidates derived from the relationship between load rate or load change rate and the threshold as the loss rate used in monitoring. Then, an evaluation index calculation unit 104 calculates the evaluation index corresponding to each of the various loss rate candidates, a threshold setting unit 101 sets a threshold for each of the various loss rate candidates such that the corresponding evaluation index becomes the minimum, and the loss rate determination unit 106 determines the loss rate candidate with the minimum value of the evaluation index under the set threshold as the loss rate used in monitoring.

[0126] (3) According to the monitoring device 1 involved in the third method, the loss rate acquisition unit 102 further searches for candidate combinations of evaluation indicators that result in the minimum loss rate.

[0127] (4) According to the monitoring device 1 involved in the fourth method, the loss rate is the total time when the load rate or load change rate is higher than the threshold, the number of times the load rate or load change rate intersects with the threshold, or the time integral value of the amount by which the load rate or load change rate is higher than the threshold.

[0128] (5) According to the monitoring device 1 involved in the fifth method, the evaluation index calculation unit 104 calculates the macro average of the evaluation index of each classification group of the device classified according to one of the multiple classification methods according to each of the multiple classification methods, and the threshold setting unit 101 sets the threshold so that the macro average becomes the minimum.

[0129] (6) The monitoring method involved in the sixth method includes the following steps: obtaining operation records for multiple failure cases, representing the shift of load rate or load change rate relative to the operating time up to the failure point; setting a threshold for comparing the load rate or load change rate in the operation records of the multiple failure cases; obtaining a loss rate, which is a value derived from the relationship between the load rate or load change rate and the threshold and represents the degree of wear and tear of the device corresponding to the operating time; obtaining the shift of the cumulative frequency of failure rate of the multiple failure cases relative to the loss rate; calculating an evaluation index as a value obtained by integrating the cumulative frequency of failure rate with the loss rate; and notifying that maintenance is required when the loss rate derived from the relationship between the load rate or load change rate of the device to be monitored and the threshold reaches a given management value determined based on the shift of the cumulative frequency of failure rate of the failure cases, wherein in the step of setting the threshold, the threshold is set such that the evaluation index is minimized.

[0130] (7) The procedure involved in the seventh method causes the computer to perform the following steps: obtaining operation records for multiple failure cases, representing the shift of load rate or load change rate relative to the operating time up to the failure point; setting a threshold for comparing the load rate or load change rate in the operation records of the multiple failure cases; obtaining a loss rate, which is a value derived from the relationship between the load rate or load change rate and the threshold and represents the degree of wear and tear of the device corresponding to the operating time; obtaining the shift of the cumulative frequency of failure rate of the multiple failure cases relative to the loss rate; calculating an evaluation index as a value obtained by integrating the cumulative frequency of failure rate with the loss rate; and issuing a notification that maintenance is required when the loss rate derived from the relationship between the load rate or load change rate of the device to be monitored and the threshold reaches a given management value determined based on the shift of the cumulative frequency of failure rate of the failure cases. In the step of setting the threshold, the threshold is set such that the evaluation index is minimized.

[0131] Industrial availability

[0132] According to the various methods of the present invention, excessive maintenance, unplanned shutdowns, etc., can be suppressed, and proper maintenance and upkeep can be performed.

[0133] -Symbol Explanation-

[0134] 1. Monitoring device

[0135] 10 CPUs

[0136] 100 Operation Record Acquisition Department

[0137] 101 Threshold Setting Section

[0138] 102 Loss Rate Acquisition Department

[0139] 103 Failure Rate Cumulative Frequency Acquisition Department

[0140] 104 Evaluation Indicators Calculation Department

[0141] 105 Notification Department

[0142] 11 Memory

[0143] 12 communication interfaces

[0144] 13 Recording Media

[0145] 2. Apparatus.

Claims

1. A monitoring device, characterized by Possessing: an operation record acquisition section that acquires, for a plurality of failure cases, operation records indicating the progress of a load rate or a load change rate with respect to the operation time until a failure is reached; a threshold value setting section that sets a threshold value with which the load rate or the load change rate in the operation records of the plurality of failure cases is compared; a deterioration rate acquisition section that acquires a deterioration rate that is a value derived from the relationship between the load rate or the load change rate and the threshold value and indicates the progress of the deterioration of the device corresponding to the operation time; a failure rate cumulative frequency acquisition section that acquires, with respect to the deterioration rate, the progress of the failure rate cumulative frequency of the plurality of failure cases; an evaluation index calculation section that calculates an evaluation index that is a value obtained by integrating the failure rate cumulative frequency with the deterioration rate; and a notification section that, in a case where the deterioration rate derived from the relationship between the load rate or the load change rate of a device set as a monitoring target and the threshold value reaches a given management value decided based on the progress of the failure rate cumulative frequency of the failure cases, performs notification of the meaning that maintenance is required, the threshold value setting section sets the threshold value so that the evaluation index becomes the minimum.

2. The monitoring device according to claim 1, characterized in that the monitoring device possesses: a deterioration rate decision section that decides, as the deterioration rate used in the monitoring, one of a plurality of kinds of deterioration rates derived from the relationship between the load rate or the load change rate and the threshold value, the evaluation index calculation section calculates the evaluation index corresponding to each of the plurality of kinds of deterioration rates, the threshold value setting section sets, for each of the plurality of kinds of deterioration rates, the threshold value so that the corresponding evaluation index becomes the minimum, the deterioration rate decision section decides, as the deterioration rate used in the monitoring, the candidate of the deterioration rate whose minimum value of the evaluation index under the set threshold value is the minimum.

3. The monitoring device according to claim 2, characterized in that the deterioration rate acquisition section further searches for a combination of the candidates of the deterioration rate whose evaluation index becomes the minimum.

4. The monitoring device according to any one of claims 1 to 3, characterized in that the deterioration rate is the total of the time when the load rate or the load change rate is higher than the threshold value, the number of times when the load rate or the load change rate crosses the threshold value, or the time integral value of the amount when the load rate or the load change rate is higher than the threshold value.

5. The monitoring device according to any one of claims 1 to 3, characterized in that the evaluation index calculation section calculates, for each of a plurality of classification methods, the macro average of the evaluation index of each classification group of the device classified by one of the plurality of classification methods, the threshold value setting section sets the threshold value so that the macro average becomes the minimum. having the steps of:

6. A monitoring method characterized by, acquiring, for a plurality of failure cases, operation records indicating the progress of a load rate or a load change rate with respect to the operation time until a failure is reached; setting a threshold value with which the load rate or the load change rate in the operation records of the plurality of failure cases is compared; ​ a deterioration rate that is a value derived from a relationship between the load rate or the load change rate and the threshold value and indicates a degree of progress of deterioration of the device corresponding to the operation time; obtaining a change in a failure rate cumulative frequency of the plurality of failure cases with respect to the deterioration rate; calculating an evaluation index that is a value obtained by integrating the failure rate cumulative frequency with the deterioration rate; and when the deterioration rate derived from the relationship between the load rate or the load change rate of the device set as a monitoring object and the threshold value reaches a given management value decided based on the change in the failure rate cumulative frequency of the failure cases, performing notification of the meaning of requiring maintenance, in the step of setting the threshold value, the threshold value is set so that the evaluation index becomes minimum.

7. A computer-readable recording medium recording a program that causes a computer to execute the steps of: obtaining an operation record indicating a change in a load rate or a load change rate with respect to an operation time until a failure is reached for a plurality of failure cases; setting a threshold value that is compared with the load rate or the load change rate in the operation record of the plurality of failure cases; obtaining a deterioration rate that is a value derived from a relationship between the load rate or the load change rate and the threshold value and indicates a degree of progress of deterioration of the device corresponding to the operation time; obtaining a change in a failure rate cumulative frequency of the plurality of failure cases with respect to the deterioration rate; calculating an evaluation index that is a value obtained by integrating the failure rate cumulative frequency with the deterioration rate; and when the deterioration rate derived from the relationship between the load rate or the load change rate of the device set as a monitoring object and the threshold value reaches a given management value decided based on the change in the failure rate cumulative frequency of the failure cases, performing notification of the meaning of requiring maintenance, in the step of setting the threshold value, the threshold value is set so that the evaluation index becomes minimum.

Images