Device abnormality determination device, abnormality determination system, abnormality determination method, method for controlling abnormality determination system, and abnormality determination program

Abstract

This abnormality determination device determines an abnormality in a device that performs timing control for sequentially implementing a plurality of events. The apparatus calculates an evaluation index of similarity with respect to normal data for data to be analyzed including an event interval between a first event and a second event and at least one piece of processing data. Whether the equipment is abnormal or not is judged by comparing the evaluation index with a preset threshold value.

Description

Technical Field

[0001] The present invention relates to an abnormality determination device, an abnormality determination system, an abnormality determination method, a control method of the abnormality determination system and an abnormality determination program of equipment.

[0002] This application claims priority based on Japanese Patent Application No. 2022-085339 filed with the Japan Patent Office on May 25, 2022, and the contents of which are incorporated herein by reference. Background Art

[0003] For example, equipment such as thermal power boilers or coal gasification combined power generation equipment (IGCC) is responsible for the supply of electricity to social infrastructure, and the impact caused by the reduction in operating rate due to poor conditions is large. Therefore, it is necessary to determine the poor conditions of such equipment as abnormalities at an early stage. For example, Patent Document 1 discloses the following content, that is, based on normal learning data selected from data obtained by sensors installed in the equipment, the existence range of normal cases is determined, and based on the deviation between the newly obtained observation data and the existence range of normal cases, the abnormality of the equipment is determined.

[0004] Previous technical literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent No. 5538597 Summary of the invention

[0007] Technical issues to be solved by the invention

[0008] Generally speaking, the control of the equipment is performed for each event. In the above-mentioned Patent Document 1, the observation data obtained by the sensor installed in the equipment is judged as abnormal according to the degree of deviation from the existence range of normal cases, so that the abnormality judgment when the event occurs can be performed. However, in the equipment that performs the timing control of sequentially issuing multiple events for making the equipment perform a series of actions, multiple events are carried out in coordination, and there are examples in which the failure caused by the front-end event affects the back-end event. In the above-mentioned Patent Document 1, since the influence of the events issued in this timing is not considered, it is not easy to judge the abnormality in the equipment implementing the timing control with sufficient accuracy.

[0009] At least one embodiment of the present invention is completed in view of the above situation, and its purpose is to provide an abnormality determination device, an abnormality determination system, an abnormality determination method, a control method of the abnormality determination system and an abnormality determination program for an equipment that can appropriately determine abnormalities that may occur in the equipment.

[0010] Means for solving technical problems

[0011] At least one embodiment of the present invention is an abnormality determination device for a device that issues a plurality of events in sequence in order to solve the above-mentioned problems, and includes:

[0012] An event interval acquisition unit, configured to acquire an event interval for a first event and a second event included in the plurality of events;

[0013] A processing data acquisition unit, used for acquiring at least one processing data of the device;

[0014] an evaluation index calculation unit for calculating an evaluation index of a similarity with respect to normal data for the analysis target data including the event interval and the processed data; and

[0015] The abnormality determination unit is used to determine whether the device has an abnormality by comparing the evaluation index with a preset threshold value.

[0016] At least one embodiment of the present invention is a method for determining an abnormality of a device that sequentially issues a plurality of events in order to solve the above-mentioned problem, and includes:

[0017] A step of obtaining an event interval for a first event and a second event included in the plurality of events;

[0018] A step of obtaining at least one processing data of the device;

[0019] a step of calculating an evaluation index of similarity with respect to normal data for the analysis target data including the event interval and the processed data; and

[0020] The step of determining whether the device has an abnormality by comparing the evaluation index with a preset threshold value.

[0021] At least one embodiment of the present invention is a device abnormality determination program that sequentially issues a plurality of events in order to solve the above-mentioned problems.

[0022] He can use a computer to perform the following steps:

[0023] A step of obtaining an event interval for a first event and a second event included in the plurality of events;

[0024] A step of obtaining at least one processing data of the device;

[0025] a step of calculating an evaluation index of similarity with respect to normal data for the analysis target data including the event interval and the processed data; and

[0026] The step of determining whether the device has an abnormality by comparing the evaluation index with a preset threshold value.

[0027] An abnormality determination system for equipment according to at least one embodiment of the present invention is an abnormality determination system composed of an abnormality determination device for equipment that sequentially issues a plurality of events and an information processing device in order to solve the above-mentioned problems, and comprises:

[0028] an event interval acquisition unit configured to acquire an event interval for a first event and a second event included in the plurality of events in accordance with a request from the information processing device;

[0029] A processing data acquisition unit, used for acquiring at least one processing data of the device;

[0030] an evaluation index calculation unit for calculating an evaluation index of a similarity with respect to normal data for the analysis target data including the event interval and the processed data; and

[0031] The abnormality determination unit is used to determine whether the device has an abnormality by comparing the evaluation index with a preset threshold value.

[0032] A control method for an abnormality determination system of a device according to at least one embodiment of the present invention is a control method for an abnormality determination system composed of an abnormality determination device for a device that sequentially issues a plurality of events and an information processing device in order to solve the above-mentioned problem.

[0033] The abnormality determination device performs the following steps:

[0034] an event interval obtaining step for obtaining an event interval for a first event and a second event included in the plurality of events in accordance with a request from the information processing device;

[0035] a processing data acquisition step, for acquiring at least one processing data of the device;

[0036] an evaluation index calculation step for calculating an evaluation index of similarity with respect to normal data for the analysis target data including the event interval and the processed data; and

[0037] The abnormality determination step is used to determine whether the device has an abnormality by comparing the evaluation index with a preset threshold value.

[0038] Effects of the Invention

[0039] According to at least one embodiment of the present invention, it is possible to provide an abnormality determination device, an abnormality determination system, an abnormality determination method, a control method of an abnormality determination system, and an abnormality determination program for a device that can appropriately determine an abnormality that may occur in the device. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1This is a diagram schematically showing the configuration of a coal gasification furnace power generation facility according to one embodiment.

[0041] Figure 2 It is a flowchart showing the timing control of one embodiment.

[0042] Figure 3 It is aimed at Figure 2 The corresponding time chart shows the time-dependent change in the opening of the nitrogen shut-off valve sealing the delivery pipe and the shut-off valve at the hopper outlet.

[0043] Figure 4 is Figure 3 This is a timing chart showing when a failure occurs in the first limit switch.

[0044] Figure 5 This is a block diagram of the structure of an abnormality determination device according to an embodiment.

[0045] Fig. 6A This is a flowchart showing an abnormality determination method according to one embodiment.

[0046] Figure 6B This is a flowchart showing an abnormality determination method according to one embodiment.

[0047] Figure 7 It means by Figure 5 A time chart of an example of the Mahalanobis distance calculated by the evaluation index calculation unit, the event interval acquired by the event interval acquisition unit 102, and each processing data acquired by the processing data acquisition unit.

[0048] Figure 8 This is an example of calculating the SN ratio.

[0049] Fig. 9 This is a block diagram of the structure of an abnormality determination system according to an embodiment. DETAILED DESCRIPTION

[0050] Hereinafter, several embodiments of the present invention will be described with reference to the drawings. However, the structures described as embodiments or shown in the drawings do not limit the scope of the present invention to these, but are merely illustrative examples.

[0051] First, a device that is a determination target of the abnormality determination device according to at least one embodiment will be described. The device may broadly include devices that perform time sequence control for sequentially executing a plurality of events, and a coal gasification furnace power generation device 1 is exemplified as one embodiment.

[0052] Figure 1 1 is a diagram schematically showing the configuration of a coal gasification furnace power plant 1 according to one embodiment. The coal gasification furnace power plant 1 includes a pulverized fuel supply device 2 , a gasification furnace 4 , a gas turbine 6 , and a generator 8 .

[0053] The pulverized fuel supply device 2 is a device for supplying pulverized coal 10 as pulverized fuel to the gasifier 4. In the present embodiment, the pulverized fuel supply device 2 is configured to be able to supply the pulverized coal 10 supplied from the outside to the gasifier 4 via a plurality of supply systems (supply system A, supply system B, ...) from a pulverized coal bunker 11 that can temporarily store the pulverized coal 10 supplied from the outside.

[0054] A plurality of discharge pipes 12-1, 12-2, ... corresponding to the respective supply systems A, B, ... are connected to the pulverized coal bunker 11. Each discharge pipe 12-1, 12-2, ... can be opened and closed independently, and is respectively connected to the upstream side of a plurality of hoppers 14-1, 14-2, ... corresponding to the respective supply systems A, B, .... A plurality of conveying pipes 16-1, 16-2, ... are connected to the downstream side of the plurality of hoppers 14-1, 14-2, .... The plurality of conveying pipes 16-1, 16-2, ... merge with each other on the downstream side and are connected to the gasifier 4.

[0055] Nitrogen supply pipes 17-1, 17-2, ... for supplying nitrogen from the outside to remove the pulverized coal 10 remaining in each of the transport pipes 16-1, 16-2, ... are connected to the plurality of transport pipes 16-1, 16-2, .... Transport pipe sealed nitrogen shutoff valves 18-1, 18-2, ... for switching the supply / stop of nitrogen are provided to the plurality of nitrogen supply pipes 17-1, 17-2, .... The transport pipe sealed nitrogen shutoff valves 18-1, 18-2, ... are control valves whose opening and closing states can be controlled by the timing control described later, and are provided with first limit switches 20-1, 20-2, ... for detecting the opening and closing states. The first limit switches 20-1, 20-2, ... are mechanical switches as follows: a limit switch for detecting the fully open state of the delivery pipe sealed nitrogen shut-off valves 18-1, 18-2, ... (hereinafter referred to as the first limit switch fully open) is grouped with a limit switch for detecting the fully closed state (hereinafter referred to as the first limit switch fully closed), and are respectively arranged on the respective delivery pipe sealed nitrogen shut-off valves 18-1, 18-2, ..., and can output a signal corresponding to the fully open state / fully closed state by switching ON (open) / OFF (closed).

[0056] Furthermore, hopper outlet shutoff valves 22-1, 22-2, ... are provided on the plurality of conveying pipes 16-1, 16-2, ... for switching the open and closed states of the conveying pipes 16-1, 16-2, .... The hopper outlet shutoff valves 22-1, 22-2, ... are control valves capable of controlling the open and closed states by a timing control described later, and second limit switches 24-1, 24-2, ... are provided for detecting the open and closed states. The second limit switches 24-1, 24-2, ... are mechanical switches in which a limit switch for detecting the fully open state of the hopper outlet shutoff valves 22-1, 22-2, ... (hereinafter referred to as the second limit switch fully open) and a limit switch for detecting the fully closed state (hereinafter referred to as the second limit switch fully closed) are combined and provided on the respective hopper outlet shutoff valves 22-1, 22-2, ..., and can output a signal corresponding to the fully open state / fully closed state by switching ON / OFF.

[0057] In the pulverized fuel supply device 2 having the above-mentioned structure, in the plurality of supply systems A, B, ..., it is possible to replenish the pulverized coal 10 to the plurality of hoppers 14-1, 14-2, ..., or to supply the pulverized coal 10 to the gasifier 4 from the plurality of hoppers 14-1, 14-2, .... For example, when the hopper 14-1 corresponding to the supply system A is replenished with the pulverized coal 10, by setting the discharge pipe 12-1 corresponding to the supply system A to an open state, the pulverized coal bunker 11 and the hopper 14-1 are both pressurized at normal pressure, and the pulverized coal 10 is discharged to the hopper 14-1 through the discharge pipe 12-1. At this time, the discharge pipes 12-2, 12-3, ... corresponding to the other supply systems B, C, ... are closed. When the pulverized coal 10 is supplied to the gasifier 4 from the hopper 14-2 corresponding to the supply system B, the conveying pipe 16-2 corresponding to the hopper 14-2 is opened, and the other conveying pipes 16-1, 16-3, ... corresponding to the other supply systems A, C, ... are closed.

[0058] The gasifier 4 partially burns the pulverized coal 10 supplied from the pulverized fuel supply device 2 having such a structure to gasify the pulverized coal 10 and generate the generated gas. The generated gas generated in the gasifier 4 is subjected to a gas purification device (not shown) to remove impurities to produce gas fuel. The gas turbine 6 is connected to a generator 8, and the gas turbine 6 is driven by the gas fuel, thereby generating electricity through the generator 8.

[0059] Next, the timing control performed in the coal gasifier power plant 1 having the above-mentioned structure will be described. Here, as an example of the timing control, the timing control of the transport pipe sealing nitrogen shutoff valve 18-1 and the hopper outlet shutoff valve 22-1 constituting the supply system A in the pulverized fuel supply device 2 provided in the coal gasifier power plant 1 will be described as the control object.

[0060] In the following description, the sequence control of the supply system A among the plurality of supply systems A, B, ... provided in the pulverized fuel supply device 2 is described, but the sequence control of the other supply systems B, ... is the same unless otherwise specified.

[0061] In the present embodiment, a timing control is described for switching the conveying pipe sealing nitrogen shutoff valve 18-1 from a fully closed state to a fully open state when the supply of pulverized coal 10 from the hopper 14-1 to the gasifier 4 via the conveying pipe 16-1 is stopped in the supply system A, thereby removing the pulverized coal 10 remaining in the conveying pipe 16-1 by nitrogen from the nitrogen supply pipe 17-1, and then switching the hopper outlet shutoff valve 22-1 from a fully open state to a fully closed state.

[0062] Figure 2 is a flowchart showing a timing control of one embodiment, Figure 3 It is aimed at Figure 2 The time chart of the opening of the corresponding pipe sealing nitrogen shutoff valve 18-1 and the hopper outlet shutoff valve 22-1 shows the time change. In this time sequence control, in the initial state, the pipe sealing nitrogen shutoff valve 18-1 is fully closed, and the hopper outlet shutoff valve 22-1 is fully open.

[0063] When the timing control is started, the delivery pipe sealing nitrogen shutoff valve 18-1 in the fully closed state is first issued an opening action instruction (event) to fully open the delivery pipe sealing nitrogen shutoff valve 18-1 (step S100). When the delivery pipe sealing nitrogen shutoff valve 18-1 receives the opening action instruction issued in step S100 at time tn1, Figure 3 As shown, the opening degree increases from the fully closed state (opening degree 0%), and after a predetermined operation period Tn, it becomes the fully open state (opening degree 100%) at time tn2. The operation period Tn is the period required for the delivery pipe sealing nitrogen gas shutoff valve 18-1 to become the fully open state from the fully closed state, and depends on the specifications of the delivery pipe sealing nitrogen gas shutoff valve 18-1. Until the delivery pipe sealing nitrogen gas shutoff valve 18-1 becomes the fully open state by the instruction of the opening operation issued in step S100 as described above, the first limit switch fully closed 20-1 is switched from the ON state to the OFF state, and the first limit switch fully open 20-1 is switched from the OFF state to the ON state, thereby detecting that the first event E1 has been issued.

[0064] Next, a closing action instruction (event) is issued to the hopper outlet shutoff valve 22-1 in a fully open state so as to be fully closed (step S101). If the hopper outlet shutoff valve 22-1 receives the closing action instruction issued in step S101, Figure 3As shown, at time th1, the opening starts to decrease from the fully open state (opening 100%), and after a predetermined action period Th, it becomes a fully closed state (opening 0%) at time th2. The action period Th is the period required for the hopper outlet shut-off valve 22-1 to become a fully closed state from the fully open state, and is a parameter that depends on the specifications of the hopper outlet shut-off valve 22-1. Until the hopper outlet shut-off valve 22-1 becomes a fully closed state by the instruction of the closing action issued in step S101 as described above, the second limit switch fully open 24-1 is switched from the ON state to the OFF state, and the second limit switch fully closed 24-1 is switched from the OFF state to the ON state, thereby detecting that the second event E2 has been issued.

[0065] Next, based on the first event E1 detected by the first limit switch fully opened 20-1 and the first limit switch fully closed 20-1 in step S100, and the second event E2 detected by the second limit switch fully opened 24-1 and the second limit switch fully closed 24-1 in step S101, the first event interval EI1 is issued (step S102). The first event interval EI1 is the interval between two events occurring in different machines. In the present embodiment, it is the interval between the first event E1 about the nitrogen shutoff valve 18-1 of the pipe seal detected by the first limit switch fully opened 20-1 and the first limit switch fully closed 20-1, and the second event E2 about the hopper outlet shutoff valve 22-1 detected by the second limit switch fully opened 24-1 and the second limit switch fully closed 24-1 in step S101. Specifically, it is obtained by the following formula.

[0066] EI1=E2-E1(th2-tn2)

[0067] In this way, the states of the delivery pipe sealing nitrogen shutoff valve 18-1 and the hopper outlet shutoff valve 22-1, which are the control objects of the timing control, are transmitted to the controller as the first event E1 and the second event E2 detected by the first limit switch fully open 20-1, the first limit switch fully closed 20-1, the second limit switch fully open 24-1, and the second limit switch fully closed 24-1, and the first event interval EI1 obtained based on these. In the controller, each structure of the coal gasification furnace power generation equipment 1 is controlled according to the received result. Here, when the first limit switch fully open 20-1, the first limit switch fully closed 20-1, the second limit switch fully open 24-1, and the second limit switch fully closed 24-1 are operating normally, the controller appropriately recognizes Figure 3The operation of the delivery pipe sealing nitrogen shutoff valve 18 and the hopper outlet shutoff valve 22-1 is controlled to stably control the coal gasifier power generation equipment 1. However, in the first limit switch fully open 20-1, the first limit switch fully closed 20-1, the second limit switch fully open 24-1, and the second limit switch fully closed 24-1, due to a malfunction or the like, a malfunction may occur in which the switching operation is not normally performed with the opening and closing of the delivery pipe sealing nitrogen shutoff valve 18-1 and the hopper outlet shutoff valve 22-1.

[0068] For example, due to a malfunction in the first limit switch fully open 20-1, even if the delivery pipe sealing nitrogen gas shut-off valve 18 changes from a fully closed state to a fully open state due to the first event E1, the output of the first limit switch fully open 20-1 does not normally switch from an OFF signal to an ON signal (remains as an OFF signal), resulting in a situation where the signal corresponding to the fully open state of the delivery pipe sealing nitrogen gas shut-off valve 18 is not normally output.

[0069] At this time, it is assumed that the controller receives the first event E1, the second event E2, and the first event interval EI1 based on the outputs of the first limit switch fully open 20-1, the first limit switch fully closed 20-1, the second limit switch fully open 24-1, and the second limit switch fully closed 24-1, and in the timing control, after issuing an opening instruction to the delivery pipe sealing nitrogen shut-off valve 18 in a fully open state, issues a closing instruction to the hopper outlet shut-off valve 22-1 in a fully closed state, as shown in FIG. Figure 3 As shown, after the first limit switch fully closed 20-1 switches from the ON state to the OFF state, the first limit switch fully opened 20-1 switches from the OFF state to the ON state, and then, after the second limit switch fully opened 24-1 switches from the ON state to the OFF state, the second limit switch fully closed 24-1 switches from the OFF state to the ON state.

[0070] However, if a malfunction occurs when the first limit switch is fully opened 20-1, the command for the opening action of the conveying pipe sealing nitrogen shut-off valve 18 and the command for the closing action of the hopper outlet shut-off valve 22-1 are issued in sequence through timing control, but there is a possibility that the first limit switch fully opened 20-1 remains in the OFF state and the second limit switch fully closed 24-1 becomes the ON state, and the first event E1 is performed after the second event E2 (that is, the value of the first event interval EI1, which should be positive, is recognized as a negative value).

[0071] Although in Figure 4An example of this phenomenon is shown in FIG. 2 , but since a malfunction occurs in the first limit switch fully open 20-1, at time tn3 (corresponding to time tn1), the first limit switch fully closed 20-1 is switched from the ON state to the OFF state. At time tn4 (corresponding to time tn2), even if the delivery pipe sealing nitrogen shut-off valve 18 is in the fully open state, the first limit switch fully open 20-1 remains in the OFF state. At time tn5 (a time after time tn4), the first limit switch fully open 20-1 gradually becomes the ON state.

[0072] Assuming that no malfunction occurs in the first limit switch fully open 20-1, the first limit switch fully open 20-1 switches from the OFF state to the ON state at time tn4 as the original process. That is, as described above, since the first limit switch fully open 20-1 becomes the ON state at time tn5, the first event E1 is detected at this timing. On the other hand, since the hopper outlet shut-off valve 22-1 is in a normal state, the second limit switch fully open 24-1 switches from the ON state to the OFF state at time th3 (corresponding to time th1), and further the second limit switch fully closed 24-1 switches from the OFF state to the ON state at time th4 (corresponding to time th2), so the second event E2 is detected.

[0073] As described above, by EI1=E2-E1, the time tn5 is subtracted from the time th4, so the first event interval EI1 becomes a negative value. As a reason for this phenomenon, for example, although the ON / OFF of the first limit switch fully open 20-1 and the second limit switch fully closed 24-1 is an event issued as a sequence, it is not an event that should be issued as a sequence control, so this phenomenon can occur. If this phenomenon occurs, the controller performs an operation for maintaining the stable operation of the coal gasifier power generation equipment 1, which may cause an abnormality in the output of the generator 8, etc. This abnormality can be appropriately determined by the abnormality determination device 100 described below.

[0074] Figure 5This is a block diagram of the structure of an abnormality determination device 100 of an embodiment. The abnormality determination device 100 is composed of, for example, a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and a computer-readable storage medium. Moreover, as an example, a series of processes for realizing various functions are stored in a storage medium in the form of a program, and the CPU reads the program into the RAM and executes information processing / calculation processing to realize various functions. In addition, the program can also be applied in a manner of being pre-installed in a ROM or other storage medium, provided in a state stored in a computer-readable storage medium, transmitted via a wired or wireless communication means, etc. The computer-readable storage medium is a magnetic disk, an optical magnetic disk, a CD-ROM, a DVD-ROM, a semiconductor memory, etc.

[0075] like Figure 5 As shown, the abnormality determination device 100 includes an event registration unit 101 , an evaluation standard storage unit 102 , an evaluation index calculation unit 103 , an evaluation unit 104 , and a display unit 105 .

[0076] The event registration unit 101 is a structure for registering information about events that include multiple events of the timing control and are monitored. Generally speaking, the timing control includes multiple events that are released in sequence, and the event registration unit 101 can also register the events of the timing control in advance (of course, it is also possible to register events that are not controlled by the timing control). The registration of events of the timing control can be performed as follows: for example, the operator confirms the multiple events included in the timing control and inputs them through an interface such as a keyboard or mouse, or the multiple events included in the timing control are automatically determined by analyzing the timing control.

[0077] The evaluation criterion storage unit 102 is a structure for calculating and storing an evaluation criterion (threshold value MDref) for comparison with the evaluation index calculated by the evaluation index calculation unit 103 in order to determine abnormality in the evaluation unit 104 , and includes an event interval acquisition unit 106 and a processed data acquisition unit 107 .

[0078] The event interval acquisition unit 106 is configured to acquire the first event interval EI1 based on the outputs of the first limit switch fully open 20-1, the first limit switch fully closed 20-1, the second limit switch fully open 24-1, and the second limit switch fully closed 24-1 during the normal operation of the coal gasifier power generation equipment 1.

[0079] The processing data acquisition unit 107 is a structure for acquiring at least one processing data Pd of the coal gasifier power generation equipment 1 as a determination object during the normal operation of the coal gasifier power generation equipment 1. The processing data Pd may be a physical quantity that can be detected by various sensors installed in the coal gasifier power generation equipment 1, or a control signal processed by the controller. In this embodiment, the processing data acquisition unit 107 acquires the output Pd1 of the generator 8 and the air ratio Pd2 detected by the air ratio sensor 23 installed at the inlet of the gasifier 4 as the processing data Pd.

[0080] The evaluation criterion storage unit 102 obtains an evaluation criterion based on the event interval (first event interval EI1) acquired by the event interval acquisition unit 106 and the processed data Pd acquired by the processed data acquisition unit 107, and stores the evaluation criterion in a readable manner.

[0081] The evaluation index calculation unit 103 is a configuration for calculating an evaluation index for abnormality determination, and includes an event interval acquisition unit 108 and a processed data acquisition unit 109 .

[0082] The event interval acquisition unit 108 is configured to acquire the first event interval EI1 based on the outputs of the first limit switch fully open 20-1, the first limit switch fully closed 20-1, the second limit switch fully open 24-1, and the second limit switch fully closed 24-1, similarly to the event interval acquisition unit 106 included in the evaluation reference storage unit 102. The processed data acquisition unit 109 is configured to acquire processed data Pd, similarly to the processed data acquisition unit 107 included in the evaluation reference storage unit 102.

[0083] The evaluation index is calculated as an index indicating the similarity with respect to normal data based on the first event interval EI1 acquired by the event interval acquisition unit 108 and the processed data Pd acquired by the processed data acquisition unit 109 during the abnormality determination period in which it is unclear whether the coal gasifier power generation equipment 1 is operating normally. In the present embodiment, the Mahalanobis distance MD obtained by the Mahalanobis-Taguchi method (MT method) is calculated as an example of such an evaluation index.

[0084] The evaluation unit 104 is a configuration for performing evaluation by comparing the evaluation index calculated by the evaluation index calculation unit 103 with the evaluation criteria stored in the evaluation criteria storage unit 102 , and includes an abnormality determination unit 110 and a cause identification unit 111 .

[0085] The abnormality determination unit 110 is a structure for determining the presence or absence of an abnormality by comparing the evaluation index with a threshold value serving as an evaluation criterion. In this embodiment, the Mahalanobis distance MD calculated by the evaluation index calculation unit 103 is compared with the threshold value MDref serving as an evaluation criterion read from the evaluation criterion storage unit 102, and an abnormality determination is performed when the Mahalanobis distance MD exceeds the threshold value MDref.

[0086] The cause determination unit 111 is a structure for determining the cause of the abnormality based on the contribution to the evaluation index when the abnormality determination unit 110 determines that there is an abnormality. When the Mahalanobis distance MD is calculated as the evaluation index as described above, the SN ratio indicating the contribution to the Mahalanobis distance MD is calculated, thereby determining the cause based on the SN ratio.

[0087] The display unit 105 is a structure for displaying the first event interval EI1, the processed data Pd, and the Mahalanobis distance MD as an evaluation index. The display unit 105 displays these in a manner recognizable by the operator (for example, a display), thereby allowing the operator to easily grasp various information related to abnormality determination.

[0088] Next, an abnormality determination method implemented by the abnormality determination device 100 having the above-described configuration will be described. Fig. 6A and Figure 6B is a flowchart showing an abnormality determination method according to one embodiment. Figure 7 It means by Figure 5 A time chart of an example of the Mahalanobis distance MD calculated by the evaluation index calculation unit 103 , the first event interval EI1 acquired by the event interval acquisition unit 106 or the event interval acquisition unit 108 , and each processed data Pd acquired by the processed data acquisition unit 107 or the processed data acquisition unit 109 .

[0089] First, refer to Fig. 6A , explaining how to calculate the evaluation criterion (MDref) stored in the evaluation criterion storage unit 102. The calculation of the evaluation criterion based on the evaluation criterion storage unit 102 is performed using normal data acquired during the period when the coal gasifier power generation equipment 1 is operating normally. The period during which the normal data is acquired is the period during which the operation of the coal gasifier power generation equipment 1 is obviously normal. Figure 7 The normal data acquisition period is from time t1 to time t2.

[0090] First, information about events including multiple events of the timing control and being monitored objects is logged in through the event logging unit 101 (step S200). Specifically, multiple events included in the timing control are determined, and the determined multiple events are logged in respectively. This logging can be performed manually by an operator or by automatically identifying multiple events included in the timing control.

[0091] Next, the event interval acquisition unit 106 of the evaluation reference storage unit 102 determines the first event E1 and the second event E2 according to the switching states of the first limit switch fully open 20-1, the first limit switch fully closed 20-1, the second limit switch fully open 24-1, and the second limit switch fully closed 24-1 during the acquisition of the normal data (step S201), and acquires the first event interval EI1 according to the first event E1 and the second event E2 (step S202). In the present embodiment, the first event E1 is determined by the first limit switch fully closed 20-1 corresponding to the delivery pipe sealing nitrogen shut-off valve 18 switching from ON to OFF and the first limit switch fully open 20-1 switching from OFF to ON, and the second event E2 is determined by the second limit switch fully open 24-1 corresponding to the hopper outlet shut-off valve 22 switching from ON to OFF and the second limit switch fully closed 24-1 switching from ON to OFF, and the interval between the two is acquired as the first event interval EI1.

[0092] exist Figure 7 2 shows the temporal change of the first event interval EI1 successively acquired in step S202. During the acquisition of normal data (between time t1 and time t2), the first limit switch fully open 20-1, the first limit switch fully closed 20-1, the second limit switch fully open 24-1, and the second limit switch fully closed 24-1 all operate normally, so the first event interval EI1 shows a positive substantially constant value.

[0093] Next, the processing data acquisition unit 107 of the evaluation standard storage unit 102 acquires at least one processing data Pd of the coal gasification furnace power generation equipment 1 as the determination object during the acquisition of normal data as normal data (step S203). In this embodiment, the output Pd1 of the generator 8 and the air ratio Pd2 detected by the air ratio sensor 23 installed at the inlet of the gasification furnace 4 are acquired as the processing data Pd, and the output Pd1 of the generator 8 and the air ratio Pd2 detected by the air ratio sensor 23 installed at the inlet of the gasification furnace 4 are respectively displayed. Figure 7 The temporal changes shown.

[0094] Next, the evaluation criterion storage unit 102 calculates a threshold value MDref of the Mahalanobis distance MD, which is an evaluation criterion for abnormality determination, based on the first event interval EI1 acquired in step S202 and the processed data Pd acquired in step S203 during the acquisition of normal data (step S204). The threshold value MDref calculated in this way is readable and stored in the evaluation criterion storage unit 102 as an evaluation criterion.

[0095] Next, refer to Figure 6B , indicating the use of Fig. 6A This abnormality determination is performed during a period in which it is unclear whether the operation of the coal gasification furnace power generation facility 1 is normal.

[0096] First, the event interval acquisition unit 108 included in the evaluation index calculation unit 103 acquires the first event interval EI1 based on the event registered by the event registration unit 101 in step S200 (step S300). Figure 7 1 shows the temporal change of the first event interval EI1 successively acquired in step S300. During the acquisition of normal data (between time t1 and time t2), as described above, the first event interval EI1 shows a positive substantially constant value, but during the abnormality determination period (after time t2), when the first limit switch is fully opened 20-1 and a failure occurs, the first event interval EI1 is temporarily confused and shows a negative value.

[0097] Next, the processing data acquisition unit 109 of the evaluation index calculation unit 103 acquires at least one processing data Pd of the coal gasification furnace power generation equipment 1 as the determination object (step S301). In this embodiment, the output Pd1 of the generator 8 and the air ratio Pd2 detected by the air ratio sensor 23 installed at the inlet of the gasification furnace 4 are acquired as the processing data Pd, and the processing data Pd is displayed respectively. Figure 7 The temporal changes shown.

[0098] Next, the evaluation index calculation unit 103 obtains the sample mean and sample covariance for the first event interval EI1 acquired in step S300 and the processed data Pd acquired in step S301 , and calculates the Mahalanobis distance MD for the analysis data using each sample (step S302 ).

[0099] Next, the abnormality determination unit 110 compares the Mahalanobis distance MD as the evaluation index calculated in step S302 with the threshold MDref as the evaluation standard stored in the evaluation standard storage unit 102, thereby determining whether there is an abnormality (step S303). The threshold MDref is a reference Fig. 6AAs described above, it is calculated in advance using normal data by the evaluation standard storage unit 102. As a result, when the Mahalanobis distance MD exceeds the threshold MDref (step S303: Yes), it is determined that "abnormality exists" (step S304).

[0100] exist Figure 7 2 shows that after time t8, each of the processed data Pd and the Mahalanobis distance MD changes, thereby causing a clear abnormality in the coal gasifier power generation equipment 1, but at the previous stage, i.e., time t5, t6, and t7, the Mahalanobis distance MD exceeds the threshold MDref, thereby making an abnormality determination in the precursor stage. This means that by using the Mahalanobis distance MD calculated from the first event interval EI1 and the processed data Pd as an evaluation index, an abnormality that is difficult to determine using the processed data Pd can be determined at an early stage.

[0101] In addition, Figure 7 In the example, from time t3 to t4, the Mahalanobis distance MD shows an action that greatly exceeds the threshold MDref, but during this period, the output of the generator 8 in the processing data Pd shows an action that temporarily decreases, so it is excluded from the judgment object. When the processing data Pd shows an action that exceeds the allowable range and is in a state that is not suitable for judgment as shown above, the abnormality judgment may not be performed.

[0102] Next, the cause identification unit 111 identifies the cause of the abnormality by calculating the SN ratio indicating the degree of contribution to the Mahalanobis distance MD (step S305 ). Figure 8 This is an example of calculating the SN ratio. Figure 8 In FIG. 1 , the contribution rate to the Mahalanobis distance MD is shown for each of the first event interval EI1 and the processed data Pd included in the analysis data. When abnormality determination is performed by referring to the SN ratio calculated in this way, the cause of abnormality can be determined based on the contribution to the Mahalanobis distance MD for each of the first event interval EI1 or the processed data Pd included in the analysis data.

[0103] When the Mahalanobis distance MD is equal to or smaller than the threshold value MDref (step S303: No), it is determined as "no abnormality" (step S306).

[0104] Next, the display unit 105 displays the abnormality determination result (step S307). In step S307, in addition to the abnormality determination result, Figure 7 As shown, the Mahalanobis distance MD, the first event interval EI1 included in the analysis data, or the processed data Pd are displayed as various information serving as a basis for abnormality determination, so that the operator can easily grasp the status of the coal gasifier power generation facility 1 .

[0105] Furthermore, in the above-described embodiment, in step S302, the evaluation index is calculated using the first event interval EI1 of a group of first events E1 and second events E2 selected from the plurality of events registered in step S200. In contrast, in other embodiments, in step S302, the evaluation index may be calculated using the plurality of first event intervals EI1 of a plurality of groups of first events E1 and second events E2 selected from the plurality of events registered in step S200.

[0106] In this case, the evaluation index calculation unit 103 may calculate the Mahalanobis distance MD as an evaluation index based on the plurality of first event intervals EI1, and the cause determination unit 111 may calculate the SN ratio indicating the contribution to the Mahalanobis distance MD. Thus, the cause determination unit 111 determines which event interval EI has a large contribution to the Mahalanobis distance MD during abnormality determination, thereby determining the event that is the cause of the abnormality with respect to the majority of events issued as a time series.

[0107] In addition, in the above-mentioned embodiment, among the multiple supply systems A, B, ... of the coal gasifier power generation equipment 1, the abnormality determination is performed on the timing control of the single supply system A, but the abnormality determination may be performed on the timing control of two or more supply systems. In this case, the multiple events of the multiple timing controls of each supply system may be divided into groups for each supply system and registered, and the first event interval EI1 obtained for the first event E1 and the second event E2 included in each registered group is processed as analysis data together with the processed data Pd.

[0108] Furthermore, in the above embodiment, the event issued from a device different from the transfer pipe sealing nitrogen shutoff valve 18-1 and the hopper outlet shutoff valve 22-1 is described, but a plurality of events issued from the same device may be targeted.

[0109] As described above, according to the above-mentioned embodiments, at least one processing data is acquired for the device as the abnormality determination object, and the event interval is acquired for the first event and the second event among the multiple events sequentially implemented in the timing control implemented on the device. At least one processing data and the event interval are processed as analysis object data and used to calculate the evaluation index of the similarity with respect to normal data. By comparing the evaluation index calculated according to the event interval together with the processing data with the threshold value, it is possible to perform abnormality determination taking into account the influence between the events issued as a timing.

[0110] In addition, the abnormality determination system 200 (see FIG. 1 ) can be configured such that each process executed by the abnormality determination device 100 in the above-mentioned embodiment is executed by the information processing device 202 which can communicate with the abnormality determination device 100. Fig. 9 ). In this case, the information processing device 202 includes at least one of the event logging unit 101 and the display unit 105, and the abnormality determination device 100 includes: an evaluation reference unit 102, an evaluation calculation unit 104, and an evaluation unit 104. Figure 6B In the abnormality determination method shown, the processing in the standard index calculation unit 103 is executed in response to a request made by the abnormality determination device 100 .

[0111] Furthermore, constituent elements in the above-described embodiments may be appropriately replaced with well-known constituent elements within a scope not departing from the spirit of the present invention, and the above-described embodiments may be appropriately combined.

[0112] The contents described in the above-mentioned embodiments can be understood, for example, as follows.

[0113] (1) An abnormality determination device for a device according to one embodiment is an abnormality determination device for a device (100) for sequentially issuing a plurality of events, comprising:

[0114] An event interval acquisition unit (108) is used to acquire an event interval (EI1) for a first event (E1) and a second event (E2) included in the plurality of events;

[0115] A processing data acquisition unit (109), used for acquiring at least one processing data (Pd) of the device;

[0116] An evaluation index calculation unit (103) for calculating an evaluation index (MD) of similarity with respect to normal data for the analysis target data including the event interval and the processing data; and

[0117] The abnormality determination unit (110) is used to determine whether the device has an abnormality by comparing the evaluation index with a preset threshold value (MDref).

[0118] According to the above (1) method, at least one processing data is acquired for a device that is an abnormality determination target, and an event interval is acquired for a first event and a second event among a plurality of events that are sequentially released about the device. At least one processing data and the event interval are processed as analysis target data and used to calculate an evaluation index of similarity to normal data. By comparing the evaluation index calculated based on the event interval together with the processing data with a threshold value, an abnormality determination can be made that takes into account the influence between events released as a time series.

[0119] (2) In another embodiment, in the embodiment of (1) above,

[0120] The abnormality determination device for the equipment further comprises an event registration unit (101) for registering the plurality of events.

[0121] The event interval acquisition unit acquires the event interval for the first event and the second event included in the plurality of events registered by the event registration unit.

[0122] According to the above-mentioned method (2), a plurality of events are registered, and the event intervals determined for the first event and the second event arbitrarily selected from the plurality of registered events are used together with the processing data for calculating the evaluation index. By calculating the evaluation index based on the event intervals obtained from each event registered in this way, it is possible to appropriately perform abnormality determination in the device.

[0123] Furthermore, even when some events are omitted from the event registration unit, evaluation indicators are obtained from event intervals and processing data regarding the registered events, thereby making it possible to perform abnormality determination taking into account the potential impact of the omitted events.

[0124] (3) In another embodiment, in the above embodiment (1) or (2),

[0125] The normal data includes the event interval and the processing data acquired during a period in which the device is operating normally.

[0126] According to the above-mentioned aspect (3), it is possible to acquire normal data for calculating the evaluation index with high reliability.

[0127] (4) In another embodiment, in any one of the above (1) to (3),

[0128] The analysis target data does not include a period in which the processed data deviates from a preset allowable range.

[0129] According to the above aspect (4), the accuracy of abnormality determination can be improved by excluding the period outside the permissible range from the analysis target data used for calculation of the evaluation index.

[0130] (5) In another embodiment, in any one of the above (1) to (4),

[0131] The event interval acquisition unit acquires a plurality of event intervals corresponding to different combinations of the first event and the second event.

[0132] The analysis target data includes the plurality of event intervals.

[0133] According to the above-mentioned mode (5), a plurality of sets of first events and second events are selected from a plurality of events, thereby including a plurality of event intervals in the analysis target data. Thus, by considering the influence between a plurality of sets of events cooperating in the timing control, a more accurate abnormality determination can be performed.

[0134] (6) In another embodiment, in the above embodiment (5),

[0135] The abnormality determination device for equipment further comprises: a cause determination unit (111) for determining the cause of the abnormality based on the contribution to the evaluation index when the abnormality determination unit determines that the abnormality exists.

[0136] According to the above aspect (6), by obtaining the contribution to the evaluation index for a plurality of event intervals included in the analysis target data, the cause of the abnormality can be appropriately identified based on the event interval with a high contribution.

[0137] (7) In another embodiment, in any one of the above (1) to (6),

[0138] The second event is published on the condition that a predetermined period of time has passed after the first event is published.

[0139] According to the above-mentioned aspect (7), it is possible to appropriately determine that an abnormality exists in a device that issues a second event after a predetermined period of time has passed from a first event.

[0140] (8) In another embodiment, in any one of the above (1) to (7),

[0141] The abnormality determination device for equipment further comprises: a display unit (105) for displaying the event interval, the at least one processing data, and the evaluation index.

[0142] According to the above-mentioned aspect (8), by displaying the event interval, processing data and evaluation index related to the abnormality determination, the operator of the present device can easily understand the status of the equipment.

[0143] (9) In another embodiment, in any one of the above (1) to (8),

[0144] The event interval is obtained based on the first event and the second event respectively acquired from different devices included in the equipment.

[0145] According to the above-mentioned aspect (9), the event interval is obtained based on the first event and the second event respectively acquired from different devices included in the equipment to be determined as abnormality.

[0146] (10) A device abnormality determination method according to one aspect is a device abnormality determination method for sequentially issuing a plurality of events, comprising:

[0147] A step of obtaining an event interval (EI1) for a first event (E1) and a second event (E2) included in the plurality of events;

[0148] A step of obtaining at least one processing data (Pd) of the device;

[0149] A step of calculating an evaluation index (MD) of similarity with respect to normal data for the analysis target data including the event interval and the processing data; and

[0150] The step of comparing the evaluation index with a preset threshold value (MDref) to determine whether the device has an abnormality.

[0151] According to the above-mentioned method (10), at least one processing data is obtained for the device as the abnormality determination object, and the event interval is obtained for the first event and the second event among the multiple events sequentially released about the device. At least one processing data and the event interval are processed as analysis object data and used to calculate an evaluation index of similarity with respect to normal data. The evaluation index calculated based on the event interval together with the processing data is compared with the threshold value, thereby making it possible to perform abnormality determination taking into account the influence between the events released as a time series.

[0152] (11) An abnormality determination program for a device according to one embodiment is an abnormality determination program for a device that issues a plurality of events in sequence.

[0153] He can use a computer to perform the following steps:

[0154] A step of obtaining an event interval (EI1) for a first event (E1) and a second event (E2) included in the plurality of events;

[0155] A step of obtaining at least one processing data (Pd) of the device;

[0156] A step of calculating an evaluation index (MD) of similarity with respect to normal data for the analysis target data including the event interval and the processing data; and

[0157] The step of comparing the evaluation index with a preset threshold value (MDref) to determine whether the device has an abnormality.

[0158] According to the above (11) method, at least one processing data is obtained for the device as the abnormality determination object, and the event interval is obtained for the first event and the second event among the multiple events sequentially released about the device. At least one processing data and the event interval are processed as analysis object data and used to calculate an evaluation index of similarity with respect to normal data. The evaluation index calculated based on the event interval together with the processing data is compared with the threshold value, thereby making it possible to perform abnormality determination taking into account the influence between the events released as a time series.

[0159] (12) An abnormality determination system for a device according to one embodiment is an abnormality determination system comprising an abnormality determination device for a device that sequentially issues a plurality of events, and an information processing device, and comprising:

[0160] an event interval acquisition unit configured to acquire an event interval for a first event and a second event included in the plurality of events in accordance with a request from the information processing device;

[0161] A processing data acquisition unit, used for acquiring at least one processing data of the device;

[0162] an evaluation index calculation unit for calculating an evaluation index of a similarity with respect to normal data for the analysis target data including the event interval and the processed data; and

[0163] The abnormality determination unit is used to determine whether the device has an abnormality by comparing the evaluation index with a preset threshold value.

[0164] According to the above-mentioned method (12), at least one processing data is obtained for a device that is an abnormality determination target according to a request from an information processing device, and an event interval is obtained for a first event and a second event among a plurality of events that are sequentially published about the device. At least one processing data and the event interval are processed as analysis target data and used to calculate an evaluation index of similarity to normal data. By comparing the evaluation index calculated based on the event interval together with the processing data with a threshold value, an abnormality determination can be made that takes into account the influence between events published as a time series.

[0165] (13) A method for controlling an abnormality determination system of a device according to one embodiment is a method for controlling an abnormality determination system comprising an abnormality determination device of a device that sequentially issues a plurality of events, and an information processing device, wherein:

[0166] The abnormality determination device performs the following steps:

[0167] an event interval obtaining step for obtaining an event interval for a first event and a second event included in the plurality of events in accordance with a request from the information processing device;

[0168] a processing data acquisition step, for acquiring at least one processing data of the device;

[0169] an evaluation index calculation step for calculating an evaluation index of similarity with respect to normal data for the analysis target data including the event interval and the processed data; and

[0170] The abnormality determination step is used to determine whether the device has an abnormality by comparing the evaluation index with a preset threshold value.

[0171] According to the above-mentioned method (13), at least one processing data is obtained for a device that is an abnormality determination target according to a request from an information processing device, and an event interval is obtained for a first event and a second event among a plurality of events that are sequentially published about the device. At least one processing data and the event interval are processed as analysis target data and used to calculate an evaluation index of similarity to normal data. By comparing the evaluation index calculated based on the event interval together with the processing data with a threshold value, an abnormality determination can be made that takes into account the influence between events published as a time series.

[0172] Explanation of symbols

[0173] 1- coal gasifier power generation equipment, 2- pulverized fuel supply device, 4- gasifier, 6- gas turbine, 8- generator, 10- pulverized coal, 11- pulverized coal bunker, 12-1, 12-2, ... discharge pipe, 14-1, 14-2, ... hopper, 16-1, 16-2, ... delivery pipe, 17-1, 17-2, ... nitrogen supply pipe, 18-1, 18-2, ... delivery pipe sealing nitrogen shut-off valve, 20-1, 20-2, ... first limit switch, 22-1, 22-2, ... hopper outlet shut-off valve, 24-1, 24-2, ... second limit switch, 100- abnormality determination device, 101-event logging unit, 102-evaluation benchmark storage unit, 103-evaluation index calculation unit, 104-evaluation unit, 105-display unit, 106-event interval acquisition unit, 107-processing data acquisition unit, 108-event interval acquisition unit, 109-processing data acquisition unit, 110-abnormality determination unit, 111-cause determination unit, 200-abnormality determination system, 202-information processing device, A, B,... supply system, E1-first event, E2-second event, EI1-first event interval, MD-Mahalanobis distance, MDref-threshold, Pd-processing data.

Claims

1. An abnormality determination device for a device, which is an abnormality determination device for a device that publishes multiple events in sequence, and comprises: An event interval acquisition unit, configured to acquire an event interval for a first event and a second event included in the plurality of events; A processing data acquisition unit, used for acquiring at least one processing data of the device; an evaluation index calculation unit for calculating an evaluation index of a similarity with respect to normal data for the analysis target data including the event interval and the processed data; and The abnormality determination unit is used to determine whether the device has an abnormality by comparing the evaluation index with a preset threshold value.

2. The abnormality determination device for equipment according to claim 1, further comprising: An event logging unit, used for logging the plurality of events, The event interval acquisition unit acquires the event interval for the first event and the second event included in the plurality of events registered by the event registration unit.

3. The abnormality determination device for equipment according to claim 1 or 2, wherein: The normal data includes the event interval and the processing data acquired during a period in which the device is operating normally.

4. The abnormality determination device for equipment according to claim 1 or 2, wherein: The analysis target data does not include a period in which the processed data deviates from a preset allowable range.

5. The abnormality determination device for equipment according to claim 1 or 2, wherein: The event interval acquisition unit acquires a plurality of event intervals corresponding to different combinations of the first event and the second event. The analysis target data includes the plurality of event intervals.

6. The abnormality determination device for equipment according to claim 5, further comprising: A cause identification unit is configured to identify a cause of the abnormality based on a contribution to the evaluation index when the abnormality determination unit determines that the abnormality exists.

7. The abnormality determination device for equipment according to claim 1 or 2, wherein: The second event is published on the condition that a predetermined period of time has passed after the first event is published.

8. The abnormality determination device for equipment according to claim 1 or 2, comprising: The display unit is used to display the event interval, the at least one processing data and the evaluation index.

9. The abnormality determination device for equipment according to claim 1 or 2, wherein: The event interval is obtained based on the first event and the second event respectively acquired by different devices included in the equipment.

10. A method for determining an abnormality of a device, which is a method for determining an abnormality of a device that performs a timing control for sequentially issuing a plurality of events, comprising: A step of obtaining an event interval for a first event and a second event included in the plurality of events; A step of obtaining at least one processing data of the device; A step of calculating an evaluation index of similarity with respect to normal data for the analysis target data including the event interval and the processed data; and The step of determining whether the device has an abnormality by comparing the evaluation index with a preset threshold value.

11. An abnormality determination program of a device, which is an abnormality determination program of a device that issues a plurality of events in sequence, He can use a computer to perform the following steps: A step of obtaining an event interval for a first event and a second event included in the plurality of events; A step of obtaining at least one processing data of the device; A step of calculating an evaluation index of similarity with respect to normal data for the analysis target data including the event interval and the processed data; and The step of determining whether the device has an abnormality by comparing the evaluation index with a preset threshold value.

12. A device abnormality determination system, which is a device abnormality determination system composed of a device abnormality determination device and an information processing device that publishes multiple events in sequence, characterized in that have: an event interval acquisition unit configured to acquire an event interval for a first event and a second event included in the plurality of events in accordance with a request from the information processing device; A processing data acquisition unit, used for acquiring at least one processing data of the device; An evaluation index calculation unit for calculating an evaluation index of a similarity with respect to normal data for the analysis target data including the event interval and the processed data; and The abnormality determination unit is used to determine whether the device has an abnormality by comparing the evaluation index with a preset threshold value.

13. A method for controlling an abnormality determination system of a device, the method comprising an abnormality determination device of the device and an information processing device that sequentially issues a plurality of events, characterized in that: The abnormality determination device performs the following steps: an event interval obtaining step for obtaining an event interval for a first event and a second event included in the plurality of events in accordance with a request from the information processing device; a processing data acquisition step, for acquiring at least one processing data of the device; An evaluation index calculation step for calculating an evaluation index of similarity with respect to normal data for the analysis target data including the event interval and the processed data; and The abnormality determination step is used to determine whether the device has an abnormality by comparing the evaluation index with a preset threshold value.

Images