Monitoring control device and monitoring control system

Abstract

A monitoring control device that can increase the accuracy of achieving a monitored goal includes: a relevance definition unit for defining the relevance between monitored elements in a plurality of monitored elements; a relevant model management database for managing relevant models using relevant monitored elements of the monitored elements; a monitoring data management database for managing monitoring data of the monitored elements; and a causal relation calculator for calculating—when a fault occurrence is detected from at least one of the monitored elements—which monitored element of the relevant models the fault occurrence has a causal relation with, based on the relevant models managed by the relevant model management database and the monitoring data managed by the monitoring data management database.and a cause identifier to identify a cause for the occurrence of the error based on the monitored element with the causal relation.

Description

Technical field

[0001] The present invention relates to a monitoring control device, and in particular to a monitoring control device that monitors a plant, such as a water treatment plant. State of the art

[0002] Recently, mechanisms have been developed for monitoring systems in various types of plants, including water treatment plants. When a fault occurs, each of these mechanisms estimates the causal relationship of the fault and provides information to an operator within the plant.

[0003] For example, patent document 1 discloses a technology that defines the relevance between identifiers of a field device and other field devices, such as an upstream and downstream position relation of a line network arranged in a plant, so that a possible causal relation is defined, and builds a causal model using processed data and a Bayesian network.

[0004] An analyst converts the constructed causal model into a form corresponding to a quality management matrix (QMM) and presents the result. State of the art patent document

[0005] Patent document 1: Japanese patent application disclosure JP 2022 - 115 643 A Summary Problem to be solved with the invention

[0006] Patent document 1 merely defines the relevance between identifiers of the field device and the other field devices, such as an upstream and downstream position relation of a power grid, for identifying the causal relation, and has a problem in that the identifiable causal relation is limited.

[0007] The present invention was conceived to solve such a problem, and its objective is to provide a monitoring control device that can improve the accuracy of obtaining a monitored target without limiting the causal relationship between monitored elements that can be identified. Ways to solve the problem

[0008] A monitoring control device according to the present invention is a monitoring control device that monitors a plurality of monitored elements, which are monitored targets, wherein the monitoring control device comprises: a relevance definition unit for defining the relevance between monitored elements in the plurality of monitored elements; a relevant model management database for managing a plurality of relevant models using relevant monitored elements of the plurality of monitored elements, wherein the relevant monitored elements are defined by the relevance definition unit; a monitoring data management database for managing monitoring data of the plurality of monitored elements;a causal relation calculator for calculating – when the occurrence of an error is detected from at least one of the plurality of monitored elements – with which monitored element of the plurality of relevant models the occurrence of the error has a causal relation, based on the plurality of relevant models managed by the relevant model management database and the monitoring data managed by the monitoring data management database; and a cause identifier for identifying a cause for the occurrence of the error based on the monitored element with the causal relation calculated by the causal relation calculator, wherein the causal relation calculator determines the monitored element that has the causal relation with the occurrence of the error based on a chronological change in the monitoring data before and after the occurrence of the error. Effects of the invention

[0009] Since the monitoring control device according to the present invention calculates with which monitored element of a plurality of relevant models of a fault occurrence has a causal relation, based on the plurality of relevant models and the monitoring data, and identifies a reason for the fault occurrence based on the monitored element that has the causal relation, the monitoring control device can increase the accuracy for obtaining a monitored target without limiting a causal relation between monitored elements that can be identified. Brief description of the drawings Fig. Figure 1 is a block diagram showing a configuration of a monitoring control device according to the present invention. Fig. Figure 2 is a diagram showing an exemplary management database for information on monitored elements of the monitoring control device according to the present invention. Fig. Figure 3 is a diagram showing an exemplary management database for monitoring data of the monitoring control device according to the present invention. Fig. Figure 4 is a diagram showing an exemplary relevant model of the monitored elements defined by a monitored element relevance definition unit in the monitoring control device according to the present invention. Fig. Figure 5 is a diagram that schematically shows a case in which the monitoring control device according to the present invention has detected an error occurrence. Fig. Figure 6 is a diagram showing a current value and a last value of output values ​​of a sensor in a monitoring data management database in the monitoring control device according to the present invention. Fig. Figure 7 is a diagram showing an exemplary calculation of a causal relationship with the occurrence of errors in the monitoring control device according to embodiment 1 of the present invention. Fig. Figure 8 is a diagram showing chronological changes in the monitoring data before and after the occurrence of the fault in the monitoring data management database of the monitoring control device according to embodiment 1 of the present invention. Fig. Figure 9 is a diagram showing the identification of a reason for the occurrence of the fault using a cause identifier in the monitoring control device according to embodiment 1 of the present invention. Fig. Figure 10 is a diagram showing chronological changes in the monitoring data before and after the fault occurred in the monitoring data management database of the monitoring control device according to embodiment 1 of the present invention. Fig. Figure 11 is a diagram showing an exemplary display on a display of the monitoring control device according to embodiment 1 of the present invention. Fig. Figure 12 is a diagram showing an exemplary calculation of a causal relation using a causal relation calculator in the monitoring control device according to embodiment 2 of the present invention. Fig. Figure 13 is a diagram showing chronological changes in the monitoring data before and after the occurrence of the fault in the monitoring data management database of the monitoring control device according to embodiment 2 of the present invention. Fig. Figure 14 is a diagram showing an exemplary calculation of a causal relation using the causal relation calculator in the monitoring control device according to embodiment 3 of the present invention. Fig. Figure 15 is a diagram showing chronological changes in the monitoring data before and after the fault occurred in the monitoring data management database of the monitoring control device according to embodiment 3 of the present invention. Fig. Figure 16 is a diagram showing an exemplary calculation of a causal relation using the causal relation calculator in the monitoring control device according to embodiment 4 of the present invention. Fig. Figure 17 is a diagram showing chronological changes in the monitoring data before and after the occurrence of the fault in the monitoring data management database of the monitoring control device according to embodiment 4 of the present invention. Fig. Figure 18 is a diagram showing an exemplary calculation of a causal relation using the causal relation calculator in the monitoring control device according to embodiment 5 of the present invention. Fig. Figure 19 is a diagram showing chronological changes in the monitoring data before and after the fault occurred in the monitoring data management database of the monitoring control device according to embodiment 5 of the present invention. Fig. Figure 20 is a diagram showing an exemplary calculation of a causal relation using the causal relation calculator in the monitoring control device according to embodiment 6 of the present invention. Fig. Figure 21 is a diagram showing chronological changes in the monitoring data before and after the occurrence of the fault in the monitoring data management database of the monitoring control device according to embodiment 6 of the present invention. Fig. Figure 22 is a block diagram showing a configuration of a monitoring control device according to a modification of the present invention. Fig. Figure 23 is a diagram showing a hardware configuration that illustrates the monitoring control devices according to embodiments 1 to 6 of the present invention. Fig. Figure 24 is a diagram showing a hardware configuration that illustrates the monitoring control devices according to embodiments 1 to 6 of the present invention. Fig. Figure 25 is a block diagram showing a configuration when the monitoring control devices according to embodiments 1 to 6 are arranged in a server computer 300 that configures a cloud environment. Fig. Figure 26 is a block diagram showing a configuration of a monitoring and control system according to embodiment 7 of the present invention. Description of embodiments Embodiment 1

[0010] Fig. Figure 1 is a block diagram showing a configuration of a monitoring control device 100 according to embodiment 1 of the present invention. As shown in Fig. As shown in Figure 1, the monitoring control device 100 comprises: a management database (DB) 10 for information for monitored elements, a relevance definition unit 20 for monitored elements, a management database (DB) 30 for relevant models, a management database (DB) 40 for monitoring data, a causal relation calculator 50, a cause identifier 60, and a display 70.

[0011] The administrative database 10 for monitored element information manages information for the monitored elements, such as equipment, facility and sensor information, which is to be managed through monitoring data. Fig. Figure 2 is a diagram showing an example of the management database 10 for information on monitored items in tabular format. Fig. Figure 2 shows, as monitored elements, plant, operating plant, equipment, facility and sensor names, and shows "processing center" as plant, "water treatment plant" as operating plant, "primary sedimentation equipment" and "reaction tank equipment" as equipment, "facility A", "facility B" and "facility C" as facilities, and sensors A to E as sensor names.

[0012] The relevance definition unit 20 for monitored elements defines the relevance between monitored elements. Relevance between monitored elements refers to physical relevance, such as equipment relevance, which indicates an equipment configuration of, for example, a plant, equipment, facility, and sensor; wiring relevance, for example, between power lines in an electrical energy system; treatment procedure relevance, for example, in a processing plant and a wastewater treatment plant; or correlation relevance, such as a correlation between the experience of a monitoring expert and a correlation calculated from a statistical analysis.

[0013] The relevance between monitored elements also includes system relevance between systems, such as a monitoring and control system and an equipment management system. Furthermore, the relevance between monitored elements encompasses the relevance of information from another sensor and information from a data source about an identical object, such as a target object like equipment, a facility, a road, a bridge, or a tunnel. This includes, for example, inspection information, point group information, water level sensor information, vibration sensor information, camera image information, and weather information. This can broaden the relevance between monitored elements. Defining the relevance between monitored elements means mapping the monitored elements to one another.

[0014] The Relevant Model Management Database 30 manages relevant watched items, defined by the Relevance Definition Unit 20 for watched items, as a relevant model. This management uses a graph database (DB), and the relevant model is managed as a tree structure. More specifically, the relevant model is managed using a labeled property graph. Alternatively, the relevant model can be managed using a relational database (RDB). A specific example of the Relevant Model Management Database 30 is described later.

[0015] The monitoring data management database 40 manages the monitoring data output chronologically by a monitoring device, such as a sensor. This monitoring data is fed to the causal relation computer 50 along with the relevant model, which is managed by the relevant models management database 30. Fig. Figure 3 is a diagram showing an example of the administrative database 40 for monitoring data in tabular format. Fig. 3 indicates the current and last values ​​of the respective output values ​​in sensors A to E.

[0016] When a plurality of sensors SC, arranged in respective parts of a plant, detect fault occurrences, the causal relation calculator 50 calculates with which monitored element of which relevant model a fault occurrence has a causal relation, based on the relevant models and a chronological change in the monitoring data before and after the fault occurrence.

[0017] The cause identifier 60 identifies a monitored element that is said to be a cause of the error occurrence, based on the monitored element that has the causal relation calculated by the causal relation calculator 50.

[0018] Display 70 is a display that shows the monitored element that has a causal relationship with the fault occurrence identified by cause identifier 60, and the reason for the fault occurrence.

[0019] Next, operations of the parts of the monitoring control device 100 are described using the management database 30 for relevant models when the labeled property graph is used.

[0020] Fig. Figure 4 illustrates an exemplary relevant model of the monitored elements, defined by the relevance definition unit 20 for monitored elements, and illustrates an example of defining the relevant model based on the management database 10 for information on monitored elements in Fig. 2.

[0021] In Fig. Figure 4 shows a plurality of monitored elements developed in a tree diagram. A treatment center, as a plant, comprises primary sedimentation equipment and reaction vessel equipment as part of a water treatment plant. The primary sedimentation equipment includes devices A and B, and the reaction vessel equipment includes device C. Device A includes sensor A, device B includes sensors B and C, and device C includes sensors D and E.

[0022] Each monitored element is surrounded by a solid line, and a relevant model indicating equipment relevance is defined by connecting the monitored elements with a solid line. A relevant model indicating a water treatment procedure is shown by a closely spaced dashed arrow leading from facility B to facility C. A relevant model indicating correlation relevance is defined by a bidirectional dashed arrow with single dots and dashes between facility A and facility C. A relevant model indicating power lines with the same wiring is defined by a widely spaced dashed arrow leading from facility A to facility B.

[0023] An example, if a sensor SC ( Fig. 1) if an error has been detected in the relevant model of the monitored elements in the system, which are defined in this way, with reference to Fig. 5 described. Fig. Figure 5 schematically shows a case in which sensor D has detected an error. Fig. 6 specifies a current value and a last value of an output value from sensor D in the management database 40 for monitoring data. As in Fig. As shown in Figure 6, the current value of the output value of sensor D is 20.0, and its last value is 18.5.

[0024] If the difference between the current value and the last value of the sensor's output is greater than the current value, if the current value exceeds an existing threshold, or if previous statistical results indicate that the current value is anomalous, then an error notification is triggered. Since the difference between the current value and the last value in the example is... Fig. If 6 has a value of 1.5, which is large, it was notified of the error.

[0025] Next, an example calculation of a causal relation using the causal relation calculator 50 will be presented with reference to Fig. 7 described. During Fig. Figure 5 schematically shows the case in which sensor D detected the occurrence of the fault. Fig. Figure 7 schematically shows the probability that a fault will also occur in sensor E.

[0026] A sensor that is relevant to a sensor that has detected a fault, i.e., a sensor that has a causal relationship with the occurrence of the fault, is highly likely to have a chronological change before and after the occurrence of the fault, similar to the sensor that detected the occurrence of the fault. Fig. Figure 8 illustrates - as a change value - a calculation result of the chronological change in the monitoring data before and after the fault occurrence of the current value and the last value of the output value of each of the sensors A to E in the management database 40 for monitoring data.

[0027] In Fig. 8 is the change in the output value of sensor D from the last value of 1.5. Since the last output value of sensor E is 8.2 and the current value is 10.0, the change in the output value from the last value is 1.8, which is greater than the change in sensor D. Consequently, it is clear that sensor E has a causal relationship with the occurrence of the error, similar to sensor D. Each of the change values ​​of sensors A through C from their last values ​​is less than 0.5.

[0028] Therefore, calculating the change value in each sensor from the last recorded value allows us to identify a sensor with a chronological change similar to the chronological change in the sensor that detected the fault, and thus a sensor with a causal relationship to the fault. This causal relationship can be calculated by determining the chronological change in the monitoring data using a statistical method based on the Euclidean distance. When the Euclidean distance is used, the sensor that is closest to the sensor that detected the fault, in terms of the two-point measurement, is considered to have the causal relationship to the fault.

[0029] Next, the cause of the error will be identified using the cause identifier 60 with reference to Fig. 9 described. Fig. Figure 9 illustrates an example for identifying – if there is a probability that a fault will also occur in sensor E – sensor E as a sensor that has the causal relation with the occurrence of the fault, similar to sensor D, which has detected the occurrence of the fault.

[0030] Fig. Figure 10 illustrates – as a change value – a calculation result of the chronological change in the monitoring data before and after the fault occurrence, based on the current value and the last value of the output value of each of the sensors A to E in the management database 40 for monitoring data. Fig. 10 is the change value in the output value of sensor D from the last value of 1.5, and the change value in the output value of sensor E from the last value is 1.8. Consequently, both change values ​​are higher than 0.5.

[0031] If a sensor that has a causal relationship with the sensor that detected the fault is detected according to the magnitude of the change value, the sensor can be identified by whether the change value exceeds a threshold. For example, if the change value threshold is 0.5, then the change value in sensor E exceeds the threshold of 0.5. Since the relevant model, which specifies equipment relevance, shows that sensor D and sensor E are sensors relevant to equipment C, sensor E is identified as the one that has the causal relationship with the fault, similar to sensor D.

[0032] Furthermore, sensor D and sensor E are connected to facility C via a common solid line, and facility C is identified as a reason for the detection of the fault by sensor D. Here, the solid line connecting sensor D and sensor E to facility C is an edge in the labeled property graph. A name can be defined for the edge. Fig. 9 is "EDGE 1", the name of the edge, and "GROUND-FLAG 1" is a property (an attribute). Here, EDGE 1 is a name used to define "equipment relevance".

[0033] Consequently, the cause identifier 60 identifies that sensor D and sensor E have a causal relationship with a fault occurrence based on the detection of the fault by sensor D. This can lead to the identification of device C as a cause of the fault occurrence. This result is displayed on the display 70 and presented to the user of the monitoring control device 100.

[0034] Fig. Figure 11 illustrates an example display on display 70. In Fig. 11. The detection of the fault occurrence by sensor D is indicated by a light, a causal relationship with the fault occurrence is represented by the connection of sensor D and sensor E with the device C by means of solid lines, and arrows from sensor D and sensor E to the device C indicate that the device C is a cause of the fault occurrence.

[0035] As described above, the monitoring control device 100 according to embodiment 1 defines the relationship between monitored elements, such as the equipment and sensors, which belong to a unit smaller than the component elements of the equipment and are associated with the equipment, so that a causal relationship between the monitored elements and a fault occurrence can be identified, and a device causing the fault can be identified. This can reduce the cost of maintaining a plant more than the cost of replacing all relevant equipment without determining the cause of a fault.

[0036] Furthermore, defining system relevance between systems, such as a monitoring and control system and an equipment management system, allows monitoring and control between the systems and monitoring and control over an entire plant.

[0037] Furthermore, calculating a change value in each sensor from the last recorded value allows for the identification of a sensor with a chronological change similar to that in the sensor that detected a fault, as a sensor that has a causal relationship with the fault. Consequently, the accuracy for identifying the device causing the fault can be increased. Design 2

[0038] Next, a monitoring control device 100A according to embodiment 2 of the present invention is described. A block diagram showing a configuration of the monitoring control device 100A is identical to that of the monitoring control device 100 according to embodiment 1 in Figure 1. Fig. 1. In the monitoring control device 100A according to embodiment 2, processes in the causal relation computer 50 and in the cause identifier 60 are different from those of the monitoring control device 100.

[0039] An exemplary calculation of a causal relation by the causal relation calculator 50 in the monitoring control device 100A is described with reference to Fig. 12 described. Fig. Figure 12 schematically shows a case in which sensor D has detected an error. Fig. Figure 13 illustrates weights assigned to sensors A to E, as well as the calculation result of the chronological change in the monitoring data before and after the fault occurrence from the current value and the last value of the output value of each of the sensors A to E in the management database 40 for monitoring data as the change value.

[0040] In other words, if a relevant model with a higher causal relationship to the fault occurrence is known in advance among the relevant models defined by the relevance definition unit 20 for monitored elements, then the causal relationship calculator 50 assigns a weight to the relevant model. Here, the monitored elements, such as facility A, are nodes in the labeled property graph, and the names of the sensors with higher causal relationships can be defined for the nodes.

[0041] In Fig. In section 12, sensor A is defined as a sensor with a higher causal relationship to facility A, and a weight of 1.0 for sensor A is defined as a property of facility A. Similarly, the names of sensors with higher causal relationships for facility B and facility C are defined, and the weights of the respective sensors are specified. Here, a weight is defined as a value that is to be multiplied by, for example, a current value. If the weight is 2.0, the current value is multiplied by 2.0.

[0042] Fig. 13 denotes weights assigned to sensors A to E, and each of the weights for sensors B and C is 2.0. Fig. 13 is the change value in the output value of sensor D from the last value of 1.5, and the change value in the output value of sensor E from the last value is 1.8. Consequently, both change values ​​are higher than 0.5.

[0043] Assuming the threshold of the change value is 0.5, then the change value in each of sensors D and E exceeds the threshold of 0.5. Since the relevant model, which indicates the equipment relevance, shows that sensor D and sensor E are sensors that are relevant to the facility C, sensors D and E are identified as those that have the causal relationship with the occurrence of the fault.

[0044] Since it is known in advance that facility B has a higher causal relationship with sensors B and C, and each of the weights of sensors B and C is 2.0, the current value is multiplied by 2.0. The change in the output value from the last value exceeds the threshold of 0.5. Consequently, sensors B and C are identified as having the causal relationship with the occurrence of the error.

[0045] Since the relevant model that specifies the water treatment procedure shows that facilities B and C are part of the relevant model for the water treatment procedure, and it is clear that facility B also causes the error to occur, facility B and facility C can be identified as causing the error to occur.

[0046] As described above, the following applies: Since the monitoring control device 100A according to embodiment 2 identifies a device causing a fault occurrence using not only the relevant model indicating equipment relevance defined by physical connections between the equipment and the devices, but also the relevant models defined by procedures, such as the water treatment procedure, the monitoring control device 100A can indirectly identify the cause of the fault occurrence in a device. Consequently, the cause of a fault occurrence can be obtained from a broader range, and the accuracy of obtaining a system can be increased.

[0047] If a relevant model with a higher causal relationship to the occurrence of an error is known in advance, a weight is assigned to that relevant model. Consequently, the relevant model with the higher causal relationship to the occurrence of the error can be reliably identified. embodiment 3

[0048] Next, a monitoring control device 100B according to embodiment 3 of the present invention is described. A block diagram showing a configuration of the monitoring control device 100B is identical to that of the monitoring control device 100 according to embodiment 1 in Figure 1. Fig. 1. In the monitoring control device 100B according to embodiment 3, processes in the causal relation computer 50 and in the cause identifier 60 are different from those of the monitoring control device 100.

[0049] An exemplary calculation of a causal relation by the causal relation calculator 50 in the monitoring control device 100B is described with reference to Fig. 14 described. Fig. Figure 14 schematically shows a case in which sensor D has detected an error. Fig. Figure 15 illustrates – as a change value – a calculation result of the chronological change in the monitoring data before and after the fault occurrence, based on the current value and the last value of the output value of each of the sensors A to E in the monitoring data management database 40. Fig. 15 is the change in the output value of sensor D from the last value of 1.5. Since the last output value of sensor A was 11.0 and the current value is 11.5, the change from the last value is 0.5. The change in each of sensors B, C, and E from the last value is 0.

[0050] Assuming the change value threshold is 0.5, then the change value in both sensors A and D exceeds this threshold. Since the relevant model, which specifies equipment relevance, shows that sensor A is relevant to facility A and sensor D is relevant to facility C, and the change value in sensor A exceeds the threshold of 0.5, it is clear that sensor A has a causal relationship with the fault occurrence. Furthermore, since the change value in sensor D also exceeds the threshold of 0.5, it is clear that sensor D has a causal relationship with the fault occurrence.

[0051] Since the relevant model indicating the correlation relevance shows that facility C correlates with facility A, facility A can be identified as a cause of the error occurrence detected by sensor D.

[0052] As described above, the following applies: Since the monitoring control device 100B according to embodiment 3 identifies a device causing a fault occurrence using not only the relevant model indicating equipment relevance, defined by physical connections between the equipment and devices, but also the relevant model indicating correlation relevance, the monitoring control device 100B can identify a cause for the fault occurrence by considering the correlation between the experience of a monitoring expert and a correlation calculated from a statistical analysis. Consequently, the monitoring control device 100B can identify a cause for the fault occurrence that cannot be identified using the other relevant models. Design 4

[0053] Next, a monitoring control device 100C according to embodiment 4 of the present invention is described. A block diagram showing a configuration of the monitoring control device 100C is identical to that of the monitoring control device 100 according to embodiment 1 in Figure 1. Fig. 1. In the monitoring control device 100C according to embodiment 4, processes in the causal relation computer 50 and in the cause identifier 60 are different from those of the monitoring control device 100.

[0054] An exemplary calculation of a causal relation by the causal relation calculator 50 in the monitoring control device 100C is described with reference to Fig. 16 described. Fig. Figure 16 schematically shows a case in which sensor D has detected an error. Fig. Figure 17 illustrates - as a change value - a calculation result of the chronological change in the monitoring data before and after the fault occurrence of the current value and the last value of the output value of each of the sensors A to E in the management database 40 for monitoring data.

[0055] In Fig. 17 is the change in the output value of sensor D from the last value of 1.5. Since the last output value of sensor A was 11.0 and the current value is 11.5, the change from the last value is 0.5. Since the last output value of sensor B was 3.0 and the current value is 3.6, the change from the last value is 0.6. Since the last output value of sensor C was 5.0 and the current value is 5.9, the change from the last value is 0.9. The change in sensor E from the last value is 0.

[0056] Assuming the change value threshold is 0.5, the following holds true: Since the change value in each of sensors A, B, C, and D exceeds the threshold of 0.5, it is clear that sensors A, B, C, and D have a causal relationship with the fault occurrence. Here, the relevant model indicating equipment relevance shows that sensor D is a sensor relevant to facility C, and the relevant model indicating correlation relevance shows that facility C is correlated with facility A. Since the relevant model indicating power lines shows that facility A is a facility relevant to facility B, facility A can be identified as a cause of the fault occurrence detected by sensor D.

[0057] As described above, the monitoring control device 100C according to embodiment 4 can identify a device causing a fault by using a combination of the relevant model indicating equipment relevance, defined by physical connections between the equipment and the devices; the relevant model indicating correlation relevance; and the relevant model indicating power lines. Consequently, the accuracy in identifying the cause of a fault can be increased, and the accuracy in maintaining the plant can be improved. Design 5

[0058] Next, a monitoring control device 100D according to embodiment 5 of the present invention is described. A block diagram showing a configuration of the monitoring control device 100D is identical to that of the monitoring control device 100 according to embodiment 1 in Figure 1. Fig. 1. In the monitoring control device 100D according to embodiment 5, processes in the causal relation computer 50 and the cause identifier 60 are different from those of the monitoring control device 100.

[0059] An exemplary calculation of a causal relation by the causal relation calculator 50 in the monitoring control device 100D is described with reference to Fig. 18 described.

[0060] Fig. Figure 18 schematically shows a case in which sensor D has detected an error. Fig. Figure 19 illustrates - as a change value - a calculation result of the chronological change in the monitoring data based on the current value, the last value and the second-to-last value of the output value of each of the sensors A to E in the management database 40 for monitoring data.

[0061] In Fig. 19 is the change value, a value calculated at a time with a chronological change, the last value of the output value of sensor C is 5.9, its second-to-last value is 5.0, and the current value is 5.9.

[0062] Consequently, it shows Fig. 19, that a chronological change occurs between the last time and the second-to-last time. Meanwhile, the last value of the output value of sensor D is 18.5, its second-to-last value is 18.5, and the current value is 20.0. Consequently, this shows Fig. 19, that a chronological change occurs between the present time and the last time.

[0063] These results show that sensors C and D have a causal relationship with the occurrence of the fault. Furthermore, it is possible to determine that a chronological change in facility B preceded the time of the fault occurrence in facility C. Here, the relevant model, which indicates equipment relevance, shows that sensor D is a sensor that is relevant to facility C.

[0064] Furthermore, the relevant model specifying the water treatment procedure shows that facilities B and C are part of the relevant model of the water treatment procedure, and the chronological change in the monitoring data of sensor C precedes the chronological change in the monitoring data of sensor D. Consequently, it is possible to identify facility B as a cause of the fault occurrence detected by sensor D.

[0065] As described above, the monitoring control device 100D according to embodiment 5 can identify a device causing a fault occurrence, taking into account the time of occurrence with a chronological change in the monitoring data, as well as using the relevant model that specifies the equipment relevance defined by physical connections between the equipment and the devices, and the relevant model that specifies the water treatment procedure. Consequently, the accuracy in identifying the cause of a fault occurrence can be increased, and the accuracy in maintaining the plant can be increased. Design 6

[0066] Next, a monitoring control device 100E according to embodiment 6 of the present invention is described. A block diagram showing a configuration of the monitoring control device 100E is identical to that of the monitoring control device 100 according to embodiment 1 in Figure 1. Fig. 1. In the monitoring control device 100E according to embodiment 6, processes in the causal relation computer 50 and the cause identifier 60 are different from those of the monitoring control device 100.

[0067] An exemplary calculation of a causal relation by the causal relation calculator 50 in the monitoring control device 100E is described with reference to Fig. 20 described. Fig. Figure 20 schematically shows a case in which sensor D has detected an error. Fig. Figure 21 illustrates - as a change value - a calculation result of the chronological change in the monitoring data based on the current value, the last value, the second-to-last value and the third-to-last value of the output value of each of the sensors A to E in the management database 40 for monitoring data.

[0068] In Fig. 21 is the change value, a value calculated at a time with a chronological change; each of the current value, the last value, and the second-to-last value of the output value of sensor A is 11.5, and its third-to-last value is 11.0. Consequently, it shows Fig. 21, that a chronological change occurs between the second-to-last and the third-to-last time. However, each of the current value and the last value of sensor C is 5.9, and each of the second-to-last and the third-to-last values ​​is 5.0. Consequently, it shows Fig. 21, that a chronological change occurs between the last time and the second-to-last time. However, each of the current value, the second-to-last value, and the third-to-last value of sensor D is 18.5, and its current value is 20.0. Consequently, it shows Fig. 21, that a chronological change occurs between the present time and the last time.

[0069] These results show that sensors C and D have a causal relationship with the occurrence of the fault. Furthermore, it is possible to determine that a chronological change in facility B preceded the time of the fault occurrence in facility C, and a chronological change in facility A preceded it further. Here, the relevant model, which indicates equipment relevance, shows that sensor D is a sensor relevant to facility C.

[0070] Furthermore, the relevant model specifying the water treatment procedure shows that facilities B and C are part of the relevant model for the water treatment procedure, and the relevant model specifying power lines shows that facility A is a facility relevant to facility B. Additionally, the chronological change in the monitoring data of sensor C precedes the chronological change in the monitoring data of sensor D, and the chronological change in the monitoring data of sensor A precedes the chronological change in the monitoring data of sensor C. Consequently, it is possible to identify facilities B and A as the causes of the fault occurrence detected by sensor D.

[0071] As described above, the monitoring control device 100E according to embodiment 6 can identify facilities that cause a fault to occur, taking into account the time of occurrence with chronological changes in the monitoring data, and using a combination of the relevant model indicating equipment relevance defined by physical connections between the equipment and the facilities, the relevant model indicating the water treatment procedure, and the relevant model indicating power lines. Consequently, the accuracy in identifying the cause of a fault can be increased, and the accuracy in maintaining the plant can be improved. modification

[0072] Correlation relevance, such as a correlation between the experience of a monitoring expert and a correlation calculated from a statistical analysis, is described as the relevance between monitored elements as defined by the relevance definition unit 20 for monitored elements in the monitoring control devices 100 to 100E according to embodiments 1 to 6. When defined, the experience of the monitoring expert and statistical values ​​can be treated as input information, the relevance of the input information can be calculated by machine learning using artificial intelligence (AI), and the calculation results can be fed to the relevance definition unit 20 for monitored elements.

[0073] Fig. Figure 22 illustrates the monitoring control devices 100 to 100E, in which information about the relevance between monitored elements, calculated by an AI 200, is fed to the relevance definition unit 20 for monitored elements.

[0074] Since the use of such a configuration can define the correlation relevance between the monitored elements from vague information, such as the experience of the monitoring expert, the processes in the relevance definition unit 20 for monitored elements can be simplified. Hardware configuration

[0075] Each of the component elements of the monitoring control devices 100 to 100E according to embodiments 1 to 6 described above can be configured using a computer and is implemented by causing the computer to execute a program. In other words, the monitoring control devices 100 to 100E are implemented, for example, by means of a processing circuit 1000, which is located in Fig. Figure 23 shows that a processor, such as a central processing unit (CPU) or a digital signal processor (DSP), is applied to the processing circuit 1000. The processing circuit 1000 causes the program, which is stored in memory, to implement the functions of each of the units.

[0076] The processing circuit 1000 can be dedicated hardware. If the processing circuit 1000 is dedicated hardware, then the processing circuit 1000 is, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination of these.

[0077] Each function of the component elements of the monitoring control devices 100 to 100E can be implemented by a separate processing circuit, or the functions can be implemented together by a single processing circuit.

[0078] Fig. Figure 24 illustrates a hardware configuration when the processing circuit 1000 is configured using a processor. In this case, the functions of the monitoring control units 100 to 100E are implemented by any combination of software, etc. (software, firmware, or software and firmware). The software, etc., is described as a program and stored in a memory 1002.

[0079] A processor 1001, acting as the processing circuit 1000, implements the functions of each of the units by reading and executing a program stored in memory 1002. In other words, this program causes a computer to execute procedures and processes of the operations of the component elements of the monitoring control devices 100 to 100E.

[0080] Examples of memory 1002 include: non-volatile or volatile semiconductor memories, such as RAM, ROM, flash memory, erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), a hard disk drive (HDD), a magnetic disk, a flexible disk, an optical disk, a compact disc, a minidisc, a digital versatile disc (DVD), an associated drive unit, and furthermore any storage medium that may be used in the future.

[0081] The above describes a configuration that allows hardware and software, etc., to implement the functions of each of the component elements of the monitoring control devices 100 to 100E. However, the configuration is not limited to this; some of the component elements of the monitoring control devices 100 to 100E can be implemented using dedicated hardware, and others can be implemented using software, etc.

[0082] For example, the processing circuit 1000, which acts as the dedicated hardware, can implement the functions of one part of the component elements, and the processing circuit 1000, which acts as processor 1001, can implement the functions of another part of the component elements by reading and executing a program stored in memory 1002.

[0083] As described above, the monitoring control devices 100 to 100E can implement any of the functions in hardware, software, etc., or any combination thereof. Other applications

[0084] The monitoring control devices 100 to 100E according to embodiments 1 to 6 described above can be arranged in a server computer that configures a cloud environment, so that monitoring data can be supplied to the server computer in the cloud through a communication network, and the server computer can identify a causal relationship that can be displayed on a display connected to the communication network.

[0085] Fig. Figure 25 is a block diagram showing a configuration when the monitoring control devices 100 to 100E are arranged in a server computer 300 that configures a cloud environment.

[0086] As in Fig. As shown in Figure 25, the server computer 300 comprises the following: the management database 10 for information on monitored elements, the relevance definition unit 20 for monitored elements, the management database 30 for relevant models, the management database 40 for monitoring data, the causal relation computer 50, and the cause identifier 60. The functions of these are identical to those of the component elements of the monitoring control devices 100 to 100E.

[0087] The monitoring data, detected by a plurality of sensors SC arranged in the respective parts of a plant, are fed to the server computer 300 through a communication network, and the server computer 300 identifies a causal relation which is displayed on a display DP which is connected to the communication network.

[0088] A plant manager receives and manages the plant using the causal relationship displayed on the DP screen. The monitoring control devices 100 to 100E, located in the server computer 300 that configures the cloud environment, can be accessed from anywhere, thus improving user convenience. Model 7

[0089] Fig. Figure 26 is a block diagram showing a configuration of a monitoring and control system 400 according to embodiment 7 of the present invention. As shown in Fig. As shown in Figure 26, the monitoring control system 400 comprises the following: the management database 10 for information for monitored elements, the relevance definition unit 20 for monitored elements, the management database 30 for relevant models, the management database 40 for monitoring data, the causal relation calculator 50, the cause identifier 60, the display 70, and sensors 80.

[0090] The monitoring control system 400 has the majority of sensors 80 arranged in the respective parts of a plant, in addition to the component elements of the monitoring control devices 100 to 100E.

[0091] The structure of the monitoring and control system 400 can identify a causal relationship of an error occurrence and identify a reason for the error occurrence that is detected by the sensors 80.

[0092] While the present invention is described in detail, the above description is illustrative in all aspects and does not limit the invention. It is understood that numerous modifications not described by way of example can be devised without deviating from the scope of the invention.

[0093] Embodiments of the present invention can be freely combined or appropriately modified and features omitted, within the scope of the invention. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] JP 2022 - 115 643 A

[0005]

Claims

[1] Monitoring control device which monitors a plurality of monitored elements which are monitored targets, wherein the monitoring control device comprises: - a relevance definition unit to define the relevance between monitored elements in the plurality of monitored elements; - a management database for relevant models for managing a plurality of relevant models using relevant watched elements of the plurality of watched elements, wherein the relevant watched elements are defined by the relevance definition unit; - an administrative database for monitoring data to manage monitoring data of the majority of monitored elements; - a causal relation calculator for calculating - when the error occurrence is detected from at least one of the plurality of monitored elements - with which monitored element of the plurality of relevant models the error occurrence has a causal relation, based on the plurality of relevant models managed by the relevant model management database and the monitoring data managed by the monitoring data management database; and - a cause identifier to identify a cause for the occurrence of the error based on the monitored element with the causal relation calculated by the causal relation calculator, wherein the causal relation calculator determines the monitored element that has the causal relation with the occurrence of the error based on a chronological change in the monitoring data before and after the occurrence of the error. [2] Monitoring control device according to claim 1, wherein the relevance definition unit defines as follows the relevance between the monitored elements: Equipment relevance, which specifies an equipment configuration, including at least one operating plant, equipment, facility and sensor; Wiring relevance, including a power line, at least in an electrical energy system; and Relevance to treatment procedures includes at least one treatment procedure in a processing plant and a sewage treatment plant. [3] Monitoring control device according to claim 2, wherein the relevance definition unit defines as the following: correlation relevance including a correlation between experience values ​​of a monitoring expert and a correlation calculated from a statistical analysis. [4] Monitoring control device according to claim 1, wherein the relevance definition unit defines as the following: system relevance between the monitored elements: system relevance between a monitoring control system and an equipment management system. [5] Monitoring control device according to claim 1, wherein the relevance definition unit defines as the following: relevance in the information from another sensor and the information from a data source on an identical object. [6] Monitoring control device according to claim 1, wherein the causal relation computer - as the monitored element which has the causal relation with the occurrence of the fault - calculates a monitored element which has a similar chronological change in the monitoring data before and after the occurrence of the fault to the monitored element from which the occurrence of the fault was detected. [7] Monitoring control device according to claim 6, wherein the causal relation computer in the monitoring data of the plurality of monitored elements managed by the monitoring data management database determines a monitored element that has a similar change value between the last detection value and the current detection value of the monitored element from which the fault occurrence was detected, to the monitored element that has the causal relation with the fault occurrence. [8] Monitoring control device according to claim 1, wherein, if a relevant model with a higher causal relation with the occurrence of the error in the plurality of relevant models defined by the relevance definition unit is known in advance, the causal relation calculator assigns weights to the monitored elements of the relevant model and calculates the monitored elements that have the causal relation with the occurrence of the error taking into account the weights. [9] Monitoring control device according to claim 3, wherein the cause identifier identifies a cause of the fault occurrence using a combination of equipment relevance, wiring relevance, treatment procedure relevance and correlation relevance. [10] Monitoring control device according to claim 3, wherein the relevance definition unit defines the correlation relevance using a calculation result of the relevance between the monitored elements as input, wherein the calculation is performed by machine learning using artificial intelligence, wherein the machine learning is performed using the experience of the monitoring expert and a statistical value as input information. [11] Monitoring control device according to claim 1, comprising the following: A display showing the monitored element that has a causal relationship with the occurrence of the error, as well as the reason for the occurrence of the error. [12] Monitoring and control system that monitors a plurality of monitored elements which are the monitored targets, wherein the monitoring and control system comprises: - a relevance definition unit to define the relevance between monitored elements in the plurality of monitored elements; - a management database for relevant models for managing a plurality of relevant models using relevant watched elements of the plurality of watched elements, wherein the relevant watched elements are defined by the relevance definition unit; - an administrative database for monitoring data to manage monitoring data of the majority of monitored elements; - a causal relation calculator for calculating - when the error occurrence is detected from at least one of the plurality of monitored elements - with which monitored element of the plurality of relevant models the error occurrence has a causal relation, based on the plurality of relevant models managed by the relevant model management database and the monitoring data managed by the monitoring data management database; and - a cause identifier to identify a cause for the occurrence of the error based on the monitored element with the causal relation calculated by the causal relation calculator, where the majority of monitored elements have a majority of sensors, and where the causal relation calculator calculates the monitored element that has the causal relation with the occurrence of the error, based on a chronological change in the monitoring data before and after the occurrence of the error. [13] Monitoring and control system according to claim 12, comprising: A display showing the monitored element that has a causal relationship with the occurrence of the error, as well as the reason for the occurrence of the error.

Images