JP2019028834A - Abnormal value diagnostic device, abnormal value diagnostic method, and program - Google Patents

Abstract

To provide an abnormal value diagnostic apparatus capable of easily and efficiently selecting a system to be inspected without setting individual thresholds for facilities comprising a plurality of systems.SOLUTION: A data extraction unit acquires information on a normal period of a system and extracts data of the normal period of the system as normal period data from a database and also extracts data of the system in a diagnosis period as a diagnosis-purpose period as diagnostic period data from the database. Based on each of the normal period data and the diagnosis period data, an index calculation unit calculates performance indexes of the normal period and the diagnosis period for each system. A deterioration degree calculation unit calculates a degree of deterioration representing a degree of deterioration of a performance index in a diagnosis period for the normal period for each system. Based on the calculated deterioration degree and a predetermined threshold value, an inspection determination unit determines a system to be inspected among all the systems.SELECTED DRAWING: Figure 1

Description

  Embodiments described herein relate generally to an abnormal value diagnosis apparatus, an abnormal value diagnosis method, and a program.

  Maintenance and inspections are regularly performed on air conditioners and heat source units installed in buildings and factories. For example, in the case of a large-scale building or the like, it may take about two years until the inspection work for all the air conditioners is completed. In other words, in this case, once inspected equipment is not inspected for about two years thereafter. The reason why the inspection cycle of the air conditioner and the heat source machine is long in this way is that manual inspection is often performed and the inspection takes a lot of time and money. For this reason, various techniques for automatically diagnosing abnormalities in air conditioners and heat source devices have been proposed for the purpose of reducing inspection costs.

  In the conventional technology, for example, a management index indicating the possibility of unsoundness of the flow control system of the controller in the air conditioner or the heat source device is provided. Then, the opening degree instruction value from the controller and the actual opening degree of the valve are acquired from the control system at a predetermined sampling period. As described above, the management index is calculated from the predetermined period in which the numerical data is acquired, and when the management index exceeds a predetermined threshold, an alarm is issued to display that an abnormality has occurred. Is done.

  However, for example, in a large-scale building or factory, air conditioners or the like may be installed in units of several tens, and their operations differ depending on the load status. Therefore, it is difficult to set a uniform threshold value for a large number of air conditioners and the like. Also, setting the threshold value for each air conditioner itself requires a lot of experience and knowledge and is not easy.

Japanese Patent Laid-Open No. 2006-162025 Japanese Patent Laying-Open No. 2015-109039

  The problem to be solved by the present invention is simple and efficient without setting individual threshold values for facilities consisting of a plurality of systems (for example, but not limited to air conditioners and heat source devices). In particular, an abnormal value diagnosis apparatus, an abnormal value diagnosis method, and a program capable of selecting a system to be inspected are provided.

  The abnormal value diagnosis apparatus of the embodiment includes a data acquisition unit, a database, a system information setting unit, a data extraction unit, an index calculation unit, a deterioration degree calculation unit, a threshold value determination unit, and an inspection determination unit. . The data acquisition unit acquires data from sensors for each system. The database stores the data acquired by the data acquisition unit as a series for each system and for each sensor. A system | strain information setting part sets the information of the normal period which is the period when the said system | strain was a normal state for every said system | strain as system | strain information. The data extraction unit obtains information on the normal period of the system by referring to the system information, extracts data of the normal period of the system from the database as normal period data, and is a period to be diagnosed Data of the system in the diagnosis period is extracted from the database as diagnosis period data. The index calculation unit includes a performance index representing control performance in the normal period based on the extracted normal period data, and a performance index representing control performance in the diagnosis period based on the extracted diagnosis period data. It calculates for every said system | strain. The deterioration degree calculation unit, based on the calculated performance index in the normal period and the calculated performance index in the diagnosis period, a degree of deterioration representing the degree of deterioration of the performance index in the diagnosis period relative to the normal period, It calculates for every said system | strain. The threshold value determination unit determines a threshold value for inspection determination. The inspection determination unit determines a system to be inspected among all the systems based on the deterioration degree calculated by the deterioration degree calculation unit and the threshold value determined by the threshold value determination unit.

1 is a block diagram showing a schematic functional configuration of an abnormal value diagnosis apparatus according to a first embodiment. The flowchart which shows the general | schematic process sequence of the abnormal value diagnostic apparatus of 1st Embodiment. Schematic which shows the structural example of the series information set in the abnormal value diagnostic apparatus of 1st Embodiment. Schematic which shows the structural example of the series information set in the abnormal value diagnostic apparatus of the 3rd modification of 1st Embodiment. The block diagram which shows schematic function structure of the abnormal value diagnostic apparatus of 2nd Embodiment. The block diagram which shows schematic function structure of the abnormal value diagnostic apparatus of 3rd Embodiment. The block diagram which shows schematic function structure of the abnormal value diagnostic apparatus of 4th Embodiment. Schematic which shows an example of the screen on which a display part displays a diagnostic result in the abnormal value diagnostic apparatus of 4th Embodiment. Schematic which shows an example of the screen which a display part displays an inspection plan in the abnormal value diagnostic apparatus of 4th Embodiment.

  Hereinafter, an abnormal value diagnosis apparatus, an abnormal value diagnosis method, and a program according to embodiments will be described with reference to the drawings.

(First embodiment)
FIG. 1 is a block diagram showing a schematic functional configuration of the abnormal value diagnosis apparatus according to the first embodiment. As illustrated, the abnormal value diagnosis apparatus 1 includes a current value acquisition unit 101, a database unit 121, a system information setting unit 141, a diagnostic data extraction unit 161, a preprocessing unit 181, an index calculation unit 201, A deterioration degree calculation unit 221, a threshold value input unit 241, a threshold value determination unit 261, and an inspection determination unit 281 are provided.
Each of these functional units is realized using an electronic circuit, for example. Each functional unit may include a storage unit such as a semiconductor memory or a magnetic hard disk device as necessary. Each function may be realized by a computer and software.
The current value acquisition unit 101 is also referred to as a “data acquisition unit”.
The database unit 121 is also simply referred to as a “database”.
The diagnostic data extraction unit 161 is also simply referred to as a “data extraction unit”.

  The current value acquisition unit 101 acquires various current values related to a device to be controlled. Specifically, the current value acquisition unit 101, for example, sets the valve opening of the air conditioner, the indoor temperature setting value set in the air conditioner, and the current values of the indoor temperature and indoor humidity controlled by the air conditioner, respectively. Obtain from the corresponding sensor. Note that the current value acquisition unit 101 acquires data of these current values in real time. That is, the current value acquisition unit 101 acquires data from sensors for each system.

  The database unit 121 stores the room temperature, the room humidity, the room temperature set value, and the current value of the valve opening acquired by the current value acquisition unit 101 as data. Note that when the current value acquisition unit acquires current value data, for example, at a cycle of 1 minute, the database unit 121 sequentially accumulates and stores data at a cycle of 1 minute in the database. That is, the database unit 121 stores all the data acquired by the current value acquisition unit 101. The database unit 121 stores the data acquired by the current value acquisition unit 101 as a series (time series) for each system and for each sensor.

The system information setting unit 141 sets information on a system that is a diagnosis target for the presence or absence of abnormality. Note that the abnormal value diagnosis apparatus 1 may simultaneously diagnose a plurality of systems, and in this case, the system information setting unit 141 sets a plurality of system information. The system information setting unit 141 sets, for example, information for identifying devices (such as air conditioners) belonging to the system, sensor values, and date information indicating the range of the normal period as information of each system. The normal period is a period in which no abnormality has occurred with respect to the device. The normal period may be different for each system. That is, the system information setting unit 141 sets, as system information, information on a normal period that is at least a period during which the system is in a normal state for each system.
A specific example of the system information will be described later with reference to another drawing.

The diagnostic data extraction unit 161 extracts normal data and diagnostic data from the database based on the system information set by the system information setting unit 141. Here, the normal data represents a sensor value in a normal period of each system set by the system information setting unit 141. The diagnostic data represents a sensor value in a period to be diagnosed (usually, a period immediately before the diagnosis is executed). The diagnosis data does not necessarily have to be a sensor value in the period immediately before the diagnosis is performed. For example, the system information setting unit 141 may set the diagnosis period.
That is, the diagnostic data extraction unit 161 acquires information on the normal period of the system by referring to the set system information. And the data of the normal period of the said system | strain are extracted from the database part 121 as normal period data. Further, the diagnostic data extraction unit 161 extracts data of the system in the diagnostic period that is a period to be diagnosed from the database unit 121 as diagnostic period data.

  The preprocessing unit 181 is a process for removing missing values and abnormal values in order to reduce the possibility of calculation errors or erroneous diagnosis in each of the normal data and diagnostic data of each system extracted by the diagnostic data extraction unit 161. (Pre-processing performed before index calculation) is performed.

The index calculation unit 201 calculates a performance index representing the control performance of each system using the preprocessed normal data and diagnostic data.
That is, the index calculation unit 201 represents a performance index representing the control performance in the normal period based on the extracted normal period data and a performance index representing the control performance in the diagnosis period based on the extracted diagnosis period data. Calculate for each.

The deterioration level calculation unit 221 calculates the deterioration level of each system based on the performance index calculated by the index calculation unit 201. Specifically, the deterioration degree calculation unit 221 calculates, for each system, the degree of deterioration of the system using the performance index obtained from normal data and the performance index obtained from diagnosis data.
That is, the deterioration degree calculation unit 221 calculates a deterioration degree indicating the degree of deterioration of the performance index in the diagnosis period relative to the normal period based on the calculated performance index in the normal period and the calculated performance index in the diagnosis period. Calculate for each. In other words, the deterioration degree calculation unit 221 calculates the performance index in the diagnosis period normalized by the performance index in the normal period as the deterioration degree.

  The threshold value input unit 241 inputs a threshold value for determining whether or not there is an abnormality. This threshold is a threshold related to the degree of deterioration of the performance index.

  The threshold value determination unit 261 determines a threshold value for determining whether there is an abnormality. That is, the threshold determination unit 261 determines a threshold serving as a reference for inspection determination. As described above, here, this threshold value is a threshold value related to the degree of deterioration of the performance index.

The inspection determination unit 281 rearranges all the systems in descending order of possibility of abnormality. At this time, the inspection determination unit 281 is based on the premise that the higher the degree of deterioration of the performance index, the higher the possibility of abnormality. That is, the inspection determination unit 281 rearranges all the systems in descending order of the performance index deterioration degree. Further, the inspection determination unit 281 determines a system to be inspected based on the threshold value determined by the threshold value determination unit 261. Specifically, the inspection determination unit 281 determines that the system is a system to be inspected when the degree of deterioration of the performance index is higher than a threshold value.
That is, the inspection determination unit 281 determines a system to be an inspection target among all the systems based on the deterioration degree calculated by the deterioration degree calculation part 221 and the threshold value determined by the threshold value determination part 261. is there.

  Next, specific processing examples of the preprocessing unit 181, the index calculation unit 201, the deterioration degree calculation unit 221, and the threshold value determination unit 262 will be described in detail.

  As described above, the preprocessing unit 181 removes missing values and abnormal values from time-series data (sensor values and the like). For example, missing values such as NaN (Not a Number, non-numeric data) are defined in advance. If the data corresponding to the definition is included, the preprocessing unit 181 removes the data as a missing value. In addition, an abnormal value is defined using an average and a standard deviation for each time-series sensor value of normal data and diagnostic data. When the preprocessing unit 181 detects data corresponding to the definition, the preprocessing unit 181 removes the data as an abnormal value. The definition of outliers using the mean and standard deviation is explained below. The preprocessing unit 181 treats data that does not satisfy the following expression as an abnormal value.

Here, x _i is the value of the i-th sensor. Μ _i is an average value of the values of the i-th sensor. Also, δ _i is the standard deviation of the value of the i-th sensor. N is a parameter. Note that n is a positive value. The average value μ _i and the standard deviation δ _i are calculated in advance based on, for example, past statistical data. Alternatively, the preprocessing unit 181 may calculate the average value μ _i and the standard deviation δ _i as needed based on the sensor value in the most recent period (normal period).

The index calculation unit 201 uses the minimum variance index value as an example of the performance index value. The minimum dispersion index value η (τ) is a value of 0 or more and 1 or less, and the closer to 1, the higher the control performance, and the closer to 0, the lower the control performance. The index calculation unit 201 calculates the minimum variance index value in normal data and the minimum variance index value in diagnostic data in each system.
The index calculation unit 201 calculates the minimum dispersion index value η (τ) by calculating the following equation (2).

  In equation (2), the numerator on the right side is the minimum variance of the measurement value y. The denominator on the right side is the actual variance of the measurement value y. Τ is a dead time.

Note that the index calculation unit 201 may calculate the T ² statistic as a performance index using principal component analysis used in process management. Sensor data can be expressed by the following equation (3) by principal component analysis.

In Expression (3), X is a data matrix of m rows and n columns. That is, XεR ^{m × n} . T is a score matrix of m rows and p columns. That is, TεR ^{m × p} . P is a loading matrix of n rows and p columns. That is, PεR ^{n × p} . T on the right shoulder of the matrix represents transposition.
Note that n is the number of time series data. M is the number of sensors. P is the number of principal components.

Then, the index calculation unit 201 calculates a T ² statistic that is an index indicating how much variation is on the principal component plane, using a loading matrix composed of the calculated principal components. Specifically, the index calculation unit 201 calculates the T ² statistic according to the following equation (4).

In equation (4), x (t) is a vector composed of n variables at time t. That is, x (t) εR ⁿ . Moreover, lambda is a diagonal matrix with the eigenvalues of the matrix XX ^T diagonal elements.

Index calculating unit 201 for each line is calculated and T ² statistic in normal data, and T ² statistic in diagnostic data. At this time, the index calculation unit 201 calculates the T ² statistic in the diagnostic data using the loading matrix in the normal data.
Note that the index calculation unit 201 may use a Q statistic indicating how far the principal component plane is peeled instead of the T ² statistic. In addition, the index calculation unit 201 may use ^two statistics, that is, a T2 statistic and a Q statistic.

The deterioration degree calculation unit 221 calculates the degree of deterioration of control performance in each system using the performance index in normal data and the performance index in diagnosis data calculated by the index calculation unit 201. For example, when the index calculation unit 201 calculates the minimum variance index value as the performance index, the minimum variance index value is a scalar value, and thus the deterioration level calculation unit 221 calculates the deterioration level using the following equation (5). .
The deterioration degree calculation unit 221 calculates the minimum dispersion index value in the diagnostic data normalized by the minimum dispersion index value in normal data as the deterioration degree. This can reduce the possibility of misdiagnosis due to different operating conditions in each system.

In Expression (5), W _j is the degree of deterioration of the system j. Further, η _base is a minimum dispersion index value in normal data of this system j. Further, η _test is a minimum dispersion index value in the diagnosis data of the system j.

In the case of the T ² statistic is computed by principal component analysis and performance index, since T ² statistic is a one-dimensional vector, deterioration calculator 221 calculates the deterioration degree of the equation below (6) To do.
Again, deterioration degree calculation section 221, a T ² statistic in the diagnosis data normalized by T ² statistic in normal data, is calculated as a deterioration degree. As a result, as in the case of using the minimum variance index value, it is possible to reduce the possibility of misdiagnosis due to different operating conditions in each system.

In Expression (6), W _j is the degree of deterioration of the system j. T ² _base (t) is a T ² statistic in normal data of this series j. T ² _test (t) is a T ² statistic in the diagnostic data of this series j. Mean () is a function for obtaining an average value of vector elements.

In equation (6), the degree of deterioration is calculated using the average value of the vector elements, but instead,
The deterioration degree calculation unit 221 may calculate the deterioration degree using the median value, maximum value, or minimum value of the vector elements. For example, when using the median value, the deterioration degree calculating unit 221 normalizes the median value of the T ² statistic element in the diagnostic data by dividing the median value of the T ² statistic element in the normal data. Calculate the degree of deterioration. Similarly, when using the maximum value, deterioration degree calculation unit 221, normalized by the maximum value of T ² statistic elements in the diagnosis data is divided by the maximum value of T ² statistic elements in normal data And the degree of deterioration is calculated. Similarly, when a minimum value, deterioration degree calculation unit 221, normalized by the minimum value of T ² statistic elements in the diagnosis data is divided by the minimum value of T ² statistic elements in normal data And the degree of deterioration is calculated.
Further, the deterioration degree calculation unit 221 only has to calculate the deterioration degree on the basis of the performance index of normal data, and the normalization method is not limited to division. The deterioration degree calculation unit 221 may obtain the degree of deterioration by normalizing the performance index of the diagnostic data by an operation other than division.

  The threshold value determination unit 261 determines the threshold value for the deterioration level calculated by the deterioration level calculation unit 221. The threshold value determined by the threshold value determination unit 261 is for determining a system having a high possibility of abnormality, that is, a system to be maintained and inspected. Note that the threshold value determination unit 261 according to the present embodiment determines the threshold value acquired by the threshold value input unit 241 as a threshold value regarding the degree of deterioration as it is.

Next, a processing procedure of the abnormal value diagnosis apparatus according to the present embodiment will be described.
FIG. 2 is a flowchart showing a procedure of processing performed by the abnormal value diagnosis apparatus 1. Hereinafter, the contents of the processing will be described with reference to this flowchart.

  First, in step S <b> 11, the system information in each system that is a target of diagnosis by the abnormal value diagnosis apparatus 1 is stored by the operation of the system information setting unit 141. Specifically, the system information stored in this step is information on devices in each system (information identifying a crisis, etc.), information on a date range in a normal period, and information on sensor data.

  In step S12, the threshold value input unit 241 acquires a threshold value to be inspected. This threshold value is a threshold value related to the degree of deterioration. This threshold value is input by the administrator of the abnormal value diagnosis apparatus 1, for example.

  In step S <b> 13, the current value acquisition unit 101 acquires the current value of each system to be diagnosed. The current values acquired by the current value acquisition unit 101 are the valve opening, the indoor temperature setting value, and the indoor temperature. Furthermore, the current value acquisition unit 101 may acquire the current value of indoor humidity.

  In step S14, the current value acquisition unit 101 collectively stores the current values of all the systems acquired in step S13 in the database unit 121. The database unit 121 stores the current value data passed from the current value acquisition unit 101.

  In the following, the abnormal value diagnosis apparatus 1 executes the processes from step S15 to S18 for each system. Then, in S19, it is determined whether or not processing for all systems has been completed. If an incomplete system remains, the abnormal value diagnosis apparatus 1 returns to step S15 and repeats the process. That is, the abnormal value diagnosis apparatus 1 repeats the processing from steps S15 to S18 for the number of systems.

  In step S15, the diagnosis data extraction unit 161 extracts time-series data of the normal period and the diagnosis period of each system from the database unit 121. At this time, the diagnostic data extraction unit 161 refers to information on the date range of the normal period set for each system. Further, the diagnostic data extraction unit 161 refers to, for example, information on the date range of the diagnostic period. Alternatively, the diagnosis data extraction unit 161 sets a range retroactive to the date and time of a predetermined length from the current time as the diagnosis period. The diagnostic data extraction unit 161 passes the extracted normal period and diagnosis period data to the preprocessing unit 181.

  In step S <b> 16, the preprocessing unit 181 performs preprocessing on the data extracted by the diagnostic data extraction unit 161. Specifically, the preprocessing unit 181 removes missing value data and abnormal value data from each time-series data. The preprocessing unit 181 passes the preprocessed data to the index calculation unit 201.

  In step S17, the index calculation unit 201 calculates a performance index for each of the normal period data and the diagnosis period data of each system. The index calculation unit 201 passes the obtained performance index value to the deterioration degree calculation unit 221.

  In step S18, the deterioration degree calculation unit 221 calculates the deterioration degree of each system. Specifically, the deterioration degree calculation unit 221 calculates the deterioration degree for each system based on the performance index for the normal period and the performance index for the diagnosis period. More specifically, the deterioration degree calculation unit 221 obtains a numerical value of the deterioration degree for each system by normalizing the performance index for the diagnosis period using the performance index for the normal period.

  In step S19, the abnormal value diagnosis apparatus 1 determines whether or not the calculation of the performance index and the deterioration degree has been completed in all the systems. If completed (step S19: YES), the process proceeds to the next step S20. If not completed (step S19: NO), the process returns to step S15 to perform processing for another system.

  In step S20, the inspection determination unit 281 rearranges the systems in the descending order of the degree of deterioration (that is, the order in which the possibility of abnormality is high) based on the calculated deterioration degree of each system. In other words, the inspection determination unit 281 sorts the systems using the numerical value of the deterioration degree as a key.

  In step S21, the threshold value determination unit 261 determines a threshold value related to the degree of deterioration. The threshold value determination unit 261 passes the determined threshold value to the inspection determination unit 281.

  In step S <b> 22, the inspection determination unit 281 determines a system to be inspected based on the threshold value received from the threshold value determination unit 261. Specifically, the inspection determination unit 281 selects a system whose degree of deterioration is higher (more deteriorated) than the threshold value as an inspection / maintenance target. Further, the inspection determination unit 281 does not select a system that has not deteriorated as much as the threshold value as an inspection / maintenance target at the present time.

In step S23, the inspection determination unit 281 outputs the determination result as diagnosis result data. Specifically, the inspection determination unit 281 outputs deterioration degree data for each system. At this time, the inspection determination unit 281 may output data of performance indexes corresponding to the degree of deterioration (performance indexes related to normal data and performance indexes related to diagnostic data) together. Further, the inspection determination unit 281 may arrange the systems in the order of the degree of deterioration (the order of the results sorted in step S20) and output the data.
Above, the process of the whole flowchart is complete | finished.

Next, a configuration example of the system information set by the system information setting unit 141 will be described.
FIG. 3 is a schematic diagram illustrating a configuration example of the system information. As shown in the figure, the system information can be expressed as tabular data. The system information has information for each system to be diagnosed, and has data items such as equipment, sensor value, and normal period. In the illustrated example, the system information includes information on each system of system 1, system 2, system 3,... (Hereinafter, continued). Regarding system 1, the device is “air conditioner 1”, the sensor values are {indoor temperature 1, indoor humidity 1, indoor temperature set value 1, valve opening 1}, and the normal period is “2016/10/1 ~ 2016/10/7 ". That is, the range of the date when the system 1 was normal was from October 1, 2016 to October 7, 2016.
Similarly, system information is held for system 2, system 3, and the like.

  As described above, according to the first embodiment, in each system, the degree of deterioration is calculated using the performance index of normal data and the performance index of diagnostic data. Thereby, the system | strain (apparatus) peculiar threshold value is not required but the system | strain with high possibility of abnormality can be determined. In addition, the abnormal value diagnosis apparatus according to the present embodiment diagnoses a plurality of systems at the same time (in other words, collectively), and rearranges them in descending order of degree of deterioration (or in ascending order). As a result, the manager or the inspector of the facility can easily determine the inspection order and the inspection plan. In addition, based on the diagnosis result data output from the abnormal value diagnosis apparatus of the present embodiment, only systems with a high degree of deterioration (in other words, systems with a high possibility of abnormality) can be inspected. That is, it is not necessary to inspect all the systems, and the inspection cost can be reduced.

  A first modification of the first embodiment will be described. In the first embodiment, the threshold value determination unit 261 determines a threshold value related to the degree of deterioration. As a modification, the threshold value determination unit 261 may determine a threshold value related to rank instead of a threshold value related to the degree of deterioration. For example, when the number of systems is 10, the threshold value determination unit 261 determines a threshold value such as “threshold value up to the top 5” or “up to the top 50%”. The inspection determination unit 281 receives the determined threshold value from the threshold value determination unit 261. Then, the inspection determination unit 281 arranges the systems up to the top 5 (or up to the top 50%) when the systems are rearranged in the order of the degree of deterioration (in other words, the order of the possibility of abnormality). Determine the system to be checked.

  Next, a second modification of the first embodiment will be described. In the first embodiment, the threshold value determination unit 261 determines the threshold value by using the value (value regarding the degree of deterioration) input in the threshold value input unit 241 as it is. Instead, the threshold value determination unit 261 may determine the threshold value based on the deterioration level value of each system calculated by the deterioration level calculation unit 221. For example, it is assumed that the set of deterioration degree values in each system has a normal distribution (or a distribution close to a normal distribution). And the threshold value determination part 261 performs calculation by the following formula | equation (7), and determines a threshold value.

In equation (7), ρ is a threshold value to be obtained. Further, mu _w is the mean value of the deterioration degree of each system. Σ _w is a standard deviation of the degree of deterioration of each system. N is a parameter and is a real number determined as appropriate. As described above, by determining the threshold value based on the actually calculated deterioration degree statistics, it is possible to reduce the effort of the administrator or the like to input the threshold value.

  Next, a third modification of the first embodiment will be described. In the first embodiment, the inspection determination unit 281 determines a system to be inspected based on the deterioration levels of all the systems for which information is set by the system information setting unit 141. In this modification, after dividing the system into a plurality of groups, the inspection determination unit 281 determines the system to be inspected.

In this third modification, the system information set includes information defining a group to which each system belongs.
FIG. 4 is a schematic diagram illustrating a configuration example of the system information set by the system information setting unit 141 in the third modification of the first embodiment. In the example shown here, system information of system 1, system 2, system 3, system 4, system 5,... Is included. And in addition to the system | strain information shown in FIG. 3, the system | strain information of a 3rd modification has a data item called "group." This “group” is information for identifying a group to which each system belongs. In the illustrated example, the devices of the system 1, the system 2, and the system 3 are the air conditioner 1, the air conditioner 2, and the air conditioner 3, respectively, and these systems belong to the group “1”. Further, the devices of the system 4 and the system 5 are the heat source device 1 and the heat source device 2, respectively, and these systems belong to the group “2”. Thus, for example, it is possible to configure a group of series for each type of device.

And in this 3rd modification, the inspection determination part 281 determines the series which are the object of inspection for every group. Specifically, it is as follows.
For example, when the threshold value determination unit 261 determines the threshold value “one system in each group is subject to inspection”, the inspection determination unit 281 selects one system selected from the group “1” and the group “2”. One system to be checked is determined as an inspection target. Also in this case, the inspection determination unit 281 preferentially selects a system with a high degree of deterioration to be calculated.
Alternatively, for example, when the threshold value determination unit 261 determines a threshold value of “60% of the entire system in each group (however, rounded off after the decimal point) is an inspection target”, the inspection determination unit 281 is as follows: Determine the inspection target. That is, the inspection determination unit 281 determines two systems (3 × 60% = 1.8) selected from the group “1” including three systems as inspection targets. The inspection determination unit 281 determines one system (2 × 60% = 1.2) selected from the group “2” including two systems as an inspection target. In this case, the inspection determination unit 281 preferentially selects a system with a high degree of deterioration calculated.

(Second Embodiment)
Next, a second embodiment will be described. It should be noted that here, description of matters common to the already described embodiment may be omitted, and the description will be focused on matters specific to the present embodiment.

FIG. 5 is a block diagram showing a schematic functional configuration of the abnormal value diagnosis apparatus according to the second embodiment. As illustrated, the abnormal value diagnosis apparatus 2 includes a current value acquisition unit 101, a database unit 121, a system information setting unit 141, a diagnostic data extraction unit 161, a preprocessing unit 181, an index calculation unit 201, A deterioration degree calculation unit 221, a threshold value determination unit 262, an inspection determination unit 281, and an abnormal data storage unit 302 are provided.
That is, the abnormal value diagnosis apparatus 2 does not include the threshold value input unit 241 included in the abnormal value diagnosis apparatus 1 of the first embodiment, but includes an abnormal data storage unit 302 instead.
Moreover, the abnormal value diagnosis apparatus 2 includes a threshold value determination unit 262 instead of the threshold value determination unit 261 included in the abnormal value diagnosis apparatus 1 of the first embodiment.

  The current value acquisition unit 101, the database unit 121, the system information setting unit 141, the diagnostic data extraction unit 161, the preprocessing unit 181, the index calculation unit 201, the deterioration degree calculation unit 221, and the inspection determination unit The functions of each part 281 are the same as those functions in the first embodiment.

The abnormal data storage unit 302 stores the degree of deterioration calculated from the data of the system that actually has an abnormality as a past record. Here, the abnormal data storage unit 302 stores all the deterioration degrees at the time of abnormality calculated each time an abnormality occurs.
That is, the abnormality data storage unit 302 stores the degree of deterioration calculated based on the data in the period when there was an abnormality for the system that was actually abnormal.
Note that the abnormal data storage unit 302 may store the numerical value of the degree of deterioration in association with information for identifying a system having an abnormality. However, since the degree of deterioration is calculated as a normalized numerical value as already described in the first embodiment, the meaning of the value itself does not change depending on the system.

  The threshold value determination unit 262 determines the threshold value using the deterioration degree data stored in the abnormal data storage unit 302. The abnormal data storage unit 302 stores a plurality of deterioration levels. Therefore, the threshold value determination unit 262 may use, for example, an average value or a median value of the plurality of deterioration degree values as the threshold value.

As described above, according to the second embodiment, the same effect as in the first embodiment can be obtained.
Further, according to the second embodiment, the threshold value is determined by using the degree of deterioration calculated from data when an abnormality has actually occurred in the past. Thereby, the abnormal value diagnostic apparatus which outputs a more accurate diagnostic result can be obtained.

Here, a first modification of the second embodiment will be described. In the second embodiment, the abnormal data storage unit 302 stores the deterioration degree calculated from the data of the system that actually had an abnormality as the past performance. As a modification thereof, the abnormality data storage unit 302 may store the degree of deterioration calculated from the operation data that simulates the abnormality.
Further, the abnormality data storage unit 302 may store the degree of deterioration calculated from the operation data when the abnormality is simulated instead of or in addition to the actual abnormality. .
Note that it may be possible to perform an operation that simulates an abnormality before actual operation of an air conditioner, a heat source device, or the like, but it is difficult to simulate the abnormality during actual operation of these devices. Therefore, it is desirable to calculate the degree of deterioration by performing the operation with the most deteriorated control performance within the range of normal operation. By storing the degree of deterioration in this way, the administrator can interactively determine a threshold value for determining an abnormality.

(Third embodiment)
Next, a third embodiment will be described. It should be noted that here, description of matters common to the already described embodiment may be omitted, and the description will be focused on matters specific to the present embodiment.

FIG. 6 is a block diagram showing a schematic functional configuration of the abnormal value diagnosis apparatus according to the third embodiment. As illustrated, the abnormal value diagnosis apparatus 3 includes a current value acquisition unit 101, a database unit 121, a system information setting unit 141, a diagnostic data extraction unit 161, a preprocessing unit 181, an index calculation unit 201, The deterioration degree calculation part 221, the threshold value input part 241, the threshold value determination part 261, the inspection determination part 283, and the deterioration degree correction | amendment part 323 are provided.
That is, the abnormal value diagnosis apparatus 3 of the present embodiment is characterized by having the deterioration degree correction unit 323. Further, the abnormal value diagnosis apparatus 3 includes an inspection determination unit 283 instead of the inspection determination unit 281 included in the abnormal value diagnosis apparatus 1 of the first embodiment.

  The current value acquisition unit 101, the database unit 121, the system information setting unit 141, the diagnostic data extraction unit 161, the preprocessing unit 181, the index calculation unit 201, the deterioration degree calculation unit 221, and a threshold value input unit The functions of the units 241 and the threshold value determination unit 261 are the same as those in the first embodiment.

The deterioration degree correction unit 323 corrects the deterioration degree calculated by the deterioration degree calculation unit 221. The deterioration degree correction unit 323 does not uniformly correct the deterioration degree in all the systems, but corrects the deterioration degree of the system that has not been inspected for a predetermined period (for example, two years), for example. Specifically, for example, the deterioration degree correction unit 323 changes the value of the deterioration degree of the system in which the predetermined period has passed without being inspected to be equal to the maximum value of the deterioration degree in all the systems. And the deterioration degree correction | amendment part 323 outputs the deterioration degree computed by the deterioration degree calculation part 221 as it is about the system | strain in which the said predetermined period has not passed.
And the inspection determination part 283 determines the system | strain which should be inspected using the value of the deterioration degree which the deterioration degree correction | amendment part 323 correct | amended instead of directly using the value of the deterioration degree which the deterioration degree calculation part 221 calculated. To do.

  That is, the deterioration degree correction unit 323 corrects the deterioration degree of the system that has passed without a check for a predetermined period to a maximum value (that is, a value that indicates the highest possibility of abnormality). If such correction is not performed, depending on the threshold value, there may be a system that is not subject to inspection for a long time. A system that has passed for a long time while maintaining a state where the degree of deterioration is not so high is considered to be less likely to be an actual abnormality, but on the other hand, for example, at least every predetermined period for quality assurance in a factory etc. It is conceivable that regular inspections must be performed. That is, when there is a system that has not been inspected for a certain period, the deterioration degree correction unit 323 corrects the deterioration degree so as to increase. Thereby, the inspection determination part 283 selects the system of the non-inspection state which passed for the predetermined period as an inspection object.

  As described above, according to the third embodiment, the same effect as in the first embodiment can be obtained. In addition, according to the third embodiment, when there is a system that has not been inspected for a certain period of time, the deterioration determination unit 323 corrects the deterioration degree of the system, so that the inspection determination unit 283 The system must be selected for inspection. That is, even if the deterioration degree calculated by the deterioration degree calculation unit 221 is low, the system is always subject to inspection every predetermined period. Thereby, abnormality of the system | strain which is inspection object can be avoided.

Here, a first modification of the third embodiment will be described. In 3rd Embodiment, the deterioration degree correction | amendment part 323 correct | amended the deterioration degree so that it might become the maximum value in all the systems regarding the system | strain which the predetermined period passed in the non-inspection state.
In the first modification, the deterioration degree correction unit 323 corrects the deterioration degree data by a different method. Specifically, in the first modification, flag information indicating whether or not the inspection object is a compulsory inspection object can be attached to the numerical data indicating the degree of deterioration. And the deterioration degree correction | amendment part 323 sets this flag information to "true" about the system | strain which passed more than the predetermined period in the state of no inspection. On the other hand, the deterioration degree correction unit 323 sets the flag information to “false” for a system that has not been inspected for a predetermined period or longer. And the deterioration degree correction | amendment part 323 does not change the numerical value itself of the deterioration degree which the deterioration degree calculation part 221 calculated about each system | strain.
Moreover, in this 1st modification, the inspection determination part 283 refers to said flag information showing whether it is a forced inspection object first about each system | strain. When the flag information is “true”, the inspection determination unit 283 selects the system as an inspection target regardless of the numerical value of the deterioration degree of the system. Then, the inspection determination unit 283 sets the systems whose flag information is “true” to be inspected, and among the systems whose flag information is “false”, when there are systems whose deterioration degree exceeds the threshold, Are also selected for inspection.
Even when this first modification is used, the abnormal value diagnosis apparatus 3 can select a system that has been inspected for a predetermined period or longer as an inspection target.

(Fourth embodiment)
Next, a fourth embodiment will be described. It should be noted that here, description of matters common to the already described embodiment may be omitted, and the description will be focused on matters specific to the present embodiment.

FIG. 7 is a block diagram showing a schematic functional configuration of the abnormal value diagnosis apparatus according to the fourth embodiment. As illustrated, the abnormal value diagnosis apparatus 4 includes a current value acquisition unit 101, a database unit 121, a system information setting unit 141, a diagnostic data extraction unit 161, a preprocessing unit 181, an index calculation unit 201, The deterioration degree calculation unit 221, the threshold value input unit 241, the threshold value determination unit 261, the inspection determination unit 281, the inspection plan planning unit 344, and the display unit 364 are provided.
That is, the abnormal value diagnosis apparatus 4 of the present embodiment has a feature that it includes an inspection plan drafting unit 344 and a display unit 364 in addition to the configuration of the abnormal value diagnosis apparatus 1 of the first embodiment.

  The current value acquisition unit 101, the database unit 121, the system information setting unit 141, the diagnostic data extraction unit 161, the preprocessing unit 181, the index calculation unit 201, the deterioration degree calculation unit 221, and a threshold value input unit The functions of the units 241, the threshold determination unit 261, and the inspection determination unit 281 are the same as those in the first embodiment.

  The inspection plan drafting unit 344 drafts an inspection plan for actually inspecting the system to be inspected determined by the inspection judging unit 281. The inspection plan planning unit 344 prepares an inspection plan for each system so as to minimize the travel distance (or travel time) of the person in charge of inspection using, for example, a layout diagram of each system in order to formulate an inspection plan. decide. This planning process is performed when a plurality of systems to be inspected are determined, such as a cost for circulating these systems (for example, a distance cost, a time cost, an economic cost converted into an amount, etc. ) Is reduced to the problem of optimization. In other words, it is a problem of finding a cost-optimal route as in the so-called “traveling salesman problem”. The inspection planning unit 344 can use a method based on an existing technique as a solution for the optimization problem. The method may be, for example, a genetic algorithm, an annealing method, or other methods. In other words, the inspection planning unit 344 inspects the systems based on the set of systems output by the inspection determination unit 281 and information indicating the cost between the systems (for example, the layout of each system). An order (which may be a full order or a partial order in some cases) is obtained and output. Further, the inspection plan drafting unit 344 may output route information for patrol the system to be inspected.

The display unit 364 displays the diagnosis result and the inspection plan on the screen of the abnormal value diagnosis device 4. Here, the data of the diagnosis result is a list of systems arranged in descending order of the degree of deterioration and a value of the degree of deterioration of each system. The display unit 364 displays the arrangement of those systems in a graphical form, for example. At this time, the display unit 364 may highlight only the system selected as the inspection target. As a result, it is easy to visually recognize which system is determined as the inspection target and what the positional relationship of these systems is. In addition, the display unit 364 displays a moving route for the inspector on the diagram showing the arrangement of each system based on the order and route information obtained by the inspection planning unit 344.
An example of display by the display unit 364 will be described below.

  FIG. 8 is a schematic diagram illustrating an example of a screen on which the display unit 364 according to the fourth embodiment displays a diagnosis result. As shown in the figure, the display unit 364 displays all the systems diagnosed by the abnormal value diagnosis device 4 in order from the higher value side (left side of the screen) to the lower side (right side of the screen). In the illustrated example, these systems are a system 5, a system 2, a system 1, a system 9, a system 3, a system 7, (omitted), and a system 10 in descending order of deterioration. In addition, the display unit 364 indicates a system selected as an inspection target among these systems as a “inspection target” surrounded by a frame. In the illustrated example, systems 5 to 6 are systems to be inspected. In addition, the display unit 364 shows the degree of deterioration calculated for each system by a graph named “Possibility of occurrence of abnormality”. Further, the threshold value is indicated by a broken line in the graph of the degree of deterioration.

  FIG. 9 is a schematic diagram illustrating an example of a screen on which the display unit 364 according to the fourth embodiment displays an inspection plan. As shown in the figure, the display unit 364 shows the layout of each system in a sketch. In the illustrated example, air conditioners 1 to 10 corresponding to systems 1 to 10 are arranged on each floor from 1F to 3F of a building. For example, air conditioners 1 to 4 are arranged on 1F. Air conditioners 5 to 8 are arranged on 2F. Air conditioners 9 and 10 are arranged on 3F. Among these series, the equipment of the series selected as the inspection object is air conditioners 1, 2, 3, 5, 6, and 9 (as also shown in FIG. 8). And the display part 364 has shown the path | route which circulates these air conditioners calculated | required by the inspection plan planning part 344 with the thick arrow line. In the example shown in the drawing, the inspection route circulates in the order of air conditioner 1 → air conditioner 3 → air conditioner 2 → air conditioner 6 → air conditioner 5 → air conditioner 9.

  As described above, according to the fourth embodiment, the same effect as in the first embodiment can be obtained. In addition, according to the fourth embodiment, the abnormal value diagnosis apparatus has an inspection plan that minimizes not only the diagnosis result indicating the system to be inspected but also the moving cost (for example, the cost such as the moving distance and the moving time). Make a plan. Thereby, a more economical inspection can be performed. Moreover, according to this embodiment, the abnormal value diagnostic apparatus can display the diagnostic result showing the system | strain which should be checked with a graphical user interface. In addition, according to the present embodiment, the abnormal value diagnosis apparatus can display the obtained inspection plan with a graphical user interface. That is, the manager of the facility can visually grasp the diagnosis result and the inspection plan.

  So far, each embodiment and its modifications have been described. The features of the first embodiment, the features of the second embodiment, the features of the third embodiment, and the features of the fourth embodiment In addition, an abnormal value diagnosis apparatus that combines the features of the modifications of each of these embodiments may be implemented.

In each of the above embodiments, the index calculation unit 201 calculates the performance index based on the minimum variance index value or the T ² statistic. Instead, the index calculation unit 201 may calculate a performance index by another method. For example, the index calculation unit 201 may calculate a performance index based on a p value (significance probability).

  In each of the above embodiments, no particular mention is made of a scheduling function including automatic execution of diagnostic processing. For example, the abnormal value diagnosis apparatus may include a schedule management unit. In that case, the schedule management unit performs control so that a series of diagnostic processing functions provided in the abnormal value diagnosis apparatus is automatically executed according to a predetermined schedule. The diagnosis processing referred to here has already been described, including acquisition of data from sensors, calculation of the degree of deterioration based on those time-series data, and determination of a system to be inspected based on the calculated degree of deterioration. It is a series of processes.

  Moreover, in each said embodiment, although the air conditioner and the heat source machine were made into the object of a diagnosis, the apparatus which an abnormal value diagnosis object makes a diagnosis object is not restricted to these. Each embodiment may be applied to other devices in order to determine the necessity of maintenance based on the performance index of the devices that require regular maintenance.

  According to at least one embodiment described above, a performance index representing the control performance in the normal period based on the extracted normal period data, and a performance index representing the control performance in the diagnosis period based on the extracted diagnostic period data And an index calculation unit for calculating for each system. In addition, based on the calculated performance index in the normal period and the calculated performance index in the diagnosis period, a deterioration degree calculation that calculates the degree of deterioration representing the degree of deterioration of the performance index in the diagnosis period relative to the normal period for each system Have a part. In addition, a threshold determination unit that determines a threshold for inspection determination is provided. Moreover, it has the inspection determination part which determines the system | system | group which should be made inspection object among all the systems based on the said deterioration degree calculated by the deterioration degree calculation part and the said threshold value determined by the threshold value determination part. With such a configuration, it is possible to specify a series to be inspected without setting an individual threshold value for each system. In addition, since the determination is performed using the data of the degree of deterioration normalized based on the performance index during the normal period, the determination based on a numerical value having a property common to each system is possible. In addition, since only systems that are estimated to have a high degree of deterioration are subject to inspection, inspection costs can be reduced.

  In addition, you may make it implement | achieve the function of the abnormal value diagnostic apparatus in embodiment mentioned above with a computer. In that case, a program for realizing this function may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be read into a computer system and executed. Here, the “computer system” includes an OS and hardware such as peripheral devices. The “computer-readable recording medium” is a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, a DVD-ROM, a USB memory, or a storage device such as a hard disk built in a computer system. That means. Furthermore, a “computer-readable recording medium” dynamically holds a program for a short time, like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. In this case, a volatile memory inside a computer system serving as a server or a client in that case may be included and a program held for a certain period of time. The program may be a program for realizing a part of the above-described functions, or may be a program that can realize the above-described functions in combination with a program already recorded in a computer system.

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 1 ... Abnormal value diagnostic apparatus, 2 ... Abnormal value diagnostic apparatus, 3 ... Abnormal value diagnostic apparatus, 4 ... Abnormal value diagnostic apparatus, 101 ... Current value acquisition part (data acquisition part), 121 ... Database part (database), 141 ... System information setting unit, 161 ... diagnostic data extraction unit (data extraction unit), 181 ... pre-processing unit, 201 ... index calculation unit, 221 ... deterioration degree calculation unit, 241 ... threshold value input unit, 261 ... threshold value determination unit, 262 ... Threshold value determination unit, 281 ... inspection determination unit, 283 ... inspection determination unit, 302 ... abnormal data storage unit, 323 ... deterioration degree correction unit, 344 ... inspection plan planning unit, 364 ... display unit

Claims (13)

A data acquisition unit for acquiring data from sensors for each system;
A database that stores the data acquired by the data acquisition unit as a series for each of the systems and for each of the sensors;
A system information setting unit that sets information of a normal period, which is a period in which the system is in a normal state for each system, as system information;
The information of the normal period of the system is obtained by referring to the system information, the data of the normal period of the system is extracted from the database as normal period data, and the system of the diagnosis period which is the period to be diagnosed A data extraction unit that extracts data from the database as diagnosis period data;
A performance index representing control performance in the normal period based on the extracted normal period data and a performance index representing control performance in the diagnosis period based on the extracted diagnosis period data are calculated for each system. An index calculation unit to
Based on the calculated performance index in the normal period and the calculated performance index in the diagnosis period, a degree of deterioration indicating the degree of deterioration of the performance index in the diagnosis period with respect to the normal period is calculated for each system. A deterioration degree calculation unit;
A threshold value determination unit for determining a threshold value for inspection determination;
Based on the degree of deterioration calculated by the degree-of-deterioration calculating unit and the threshold value determined by the threshold value determining unit, an inspection determination unit that determines a system to be inspected among all systems,
An abnormal value diagnostic apparatus comprising: The threshold value determination unit determines a value input from the outside in advance as the threshold value.
The abnormal value diagnosis apparatus according to claim 1. The threshold value determination unit determines the threshold value based on statistics of the deterioration level for each of the systems calculated by the deterioration level calculation unit.
The abnormal value diagnosis apparatus according to claim 1. An abnormal data storage unit that stores the degree of deterioration calculated based on data in a period in which there was an abnormality for a system that actually had an abnormality,
Further comprising
The threshold value determination unit determines the threshold value based on the degree of deterioration when there is an actual abnormality stored in the abnormality data storage unit.
The abnormal value diagnosis apparatus according to claim 1. The abnormal data storage unit stores the degree of deterioration calculated from the operation data when simulating an abnormality, instead of an actual abnormality or in addition to an actual abnormality,
The abnormal value diagnosis apparatus according to claim 4. The degree of deterioration for correcting the degree of deterioration calculated by the degree-of-deterioration calculating unit so as to be preferentially selected as an inspection object in the current determination for the system that has not been selected as an inspection object for a predetermined period of time or longer. Correction section,
Further comprising
The inspection determination unit performs a determination based on the corrected deterioration degree for the system in which the deterioration degree is corrected by the deterioration degree correction unit.
The abnormal value diagnostic apparatus according to any one of claims 1 to 5. With respect to a plurality of the systems that are determined to be inspected by the inspection determination unit, between the systems at the time of inspection based at least on the movement cost related to at least one of the movement distance between the systems or the movement cost An inspection plan planning department that develops an inspection plan including information on the order of
The abnormal value diagnosis apparatus according to any one of claims 1 to 6, further comprising: A preprocessing unit for performing a preprocessing for removing at least one of missing value data or abnormal value data from the series of the normal period data and the diagnosis period data extracted by the data extraction unit;
Further comprising
The index calculation unit calculates the performance index related to the normal period and the performance index related to the diagnosis period, respectively, based on the normal period data and the diagnosis period data preprocessed by the preprocessing unit. ,
The abnormal value diagnostic apparatus according to any one of claims 1 to 7. Information on the plurality of systems arranged in the order of the calculated degree of deterioration, information indicating the degree of deterioration for each of the systems, and a system to be inspected by the inspection determination unit among all the systems A display unit that displays information indicating which of the systems determined to be present as diagnostic result information;
The abnormal value diagnosis apparatus according to any one of claims 1 to 8, further comprising: The system information setting unit also sets information for identifying a group to which the system belongs for each system,
The inspection determination unit determines, for each group, a system that should be an inspection target among systems belonging to the group.
The abnormal value diagnostic apparatus according to any one of claims 1 to 9. Execute diagnosis processing including acquisition of the data from the sensor, calculation of the degree of deterioration based on the data, and determination of the system to be inspected based on the calculated degree of deterioration according to a predetermined schedule Schedule management department,
The abnormal value diagnosis apparatus according to any one of claims 1 to 10, further comprising: A data acquisition process for acquiring data from sensors for each system,
A database storage process for storing the data acquired in the data acquisition process in a database as a series for each of the systems and for each sensor;
A system information setting process for setting information of a normal period, which is a period in which the system is in a normal state for each system, as system information,
The information of the normal period of the system is obtained by referring to the system information, the data of the normal period of the system is extracted from the database as normal period data, and the system of the diagnosis period which is the period to be diagnosed A data extraction process for extracting data from the database as diagnostic period data;
A performance index representing control performance in the normal period based on the extracted normal period data and a performance index representing control performance in the diagnosis period based on the extracted diagnosis period data are calculated for each system. Index calculation process to
Based on the calculated performance index in the normal period and the calculated performance index in the diagnosis period, a degree of deterioration indicating the degree of deterioration of the performance index in the diagnosis period with respect to the normal period is calculated for each system. Deterioration degree calculation process,
A threshold value determination process for determining a threshold value for inspection determination;
Based on the degree of deterioration calculated in the degree of deterioration calculation process and the threshold value determined in the threshold value determination process, an inspection determination process for determining a system to be inspected among all systems,
An abnormal value diagnosis method including: Computer
The abnormal value diagnostic apparatus according to any one of claims 1 to 11,
Program to function as.

Images