Plant design assistance system and plant design support method
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Filing Date
- 2025-12-10
- Publication Date
- 2026-03-13
Abstract
Description
Plant design support system and plant design support method
[0001] The technology disclosed in this specification relates to a plant design support technology.
[0002] In recent years, with the advancement of plant automation, the scope of monitoring operations per operator has tended to increase, and operations in which multiple units are monitored and operated simultaneously (multi-unit operation) have been introduced.
[0003] It is important that human-machine interfaces (HMIs) for plants be designed to reduce human error and improve operator performance, and verification is required for scenarios that assume responses to various events. However, in cases where there are diverse scenarios, the order of occurrence of events is uncertain, and branching probabilities are time-dependent, such as when events occur during multi-unit operation, verifying all scenarios is difficult in terms of time and human workload.
[0004] For example, according to the technology described in Patent Document 1, first, the plant design details, environmental conditions, and operators are represented in a virtual space. Then, operator actions based on the operator's characteristics and predetermined operator behavior patterns are reproduced, and the plant situation is reproduced using a plant simulator in chronological order. As a result, by evaluating whether the actions were performed according to the patterns or within the time limit, it becomes possible to evaluate the plant design in a virtual space, thereby reducing the cost and time required for verifying the plant mockup or the operators.
[0005] Furthermore, for example, according to the technology described in Patent Literature 2, a trained model is generated by machine learning of an event to occur and the worker and equipment states at the time of the event occurrence, based on worker information and equipment information at the time of the event occurrence and worker information and equipment information acquired during a certain period before the event occurs.The trained model can then be used to predict an event occurrence scenario according to the worker and equipment states.
[0006] International Publication No. 2018 / 189849 International Publication No. 2021 / 176663
[0007] Conventional techniques generate scenarios that replicate operator behavior based on predetermined behavior patterns or trained models, and then evaluate plant design and other aspects based on these scenarios.
[0008] Therefore, in order to perform the above evaluations using a variety of scenarios that take into account the uncertainty of the order in which events occur, the time dependency of branching probabilities, etc., it was necessary to create a huge number of combinations of scenarios in advance.In addition, because operator information was generated from static operator characteristic information such as body type, voice volume, and voice quality, it was not possible to consider the mutual influence between the operator status and the plant status, which change from moment to moment.
[0009] The technology disclosed in this specification has been made in consideration of the problems described above, and is a technology for supporting plant design and scenario generation by taking into account the ever-changing operator status and plant status.
[0010] A plant design support system that is a first aspect of the technology disclosed in the present specification includes an operator state identification unit that identifies an operator state, which is an index indicating the state of the operator, based on the operator's behavior history for each unit of time; a transition probability calculation unit that calculates a transition probability of the operator's behavior based on the plant state, which is an index indicating the state of the plant, and the operator state; an operator behavior identification unit that identifies the operator's behavior based on the transition probability; and a plant state identification unit that identifies the plant state in a later unit of time from the plant state in a previous unit of time based on the behavior of the plant caused by the operator's behavior.
[0011] According to at least a first aspect of the technology disclosed in the present specification, by identifying the plant state at a later time unit from the plant state at a previous time unit based on the plant behavior caused by the actions of the operators, it is possible to support plant design and the generation of highly accurate scenarios by taking into account the mutual influence between the operator state and the plant state, which changes from moment to moment.
[0012] Furthermore, objects, features, aspects, and advantages associated with the technology disclosed herein will become more apparent from the detailed description and accompanying drawings set forth below.
[0013] 1 is a diagram illustrating an example of a hardware configuration of a plant design support system; 2 is a diagram conceptually illustrating an example of a configuration of a plant design support system according to an embodiment; 3 is a flowchart illustrating an example of an operation of a plant design support system according to an embodiment; 4 is a diagram illustrating an example of an environmental state held by an environmental state specifying unit according to an embodiment; 5 is a diagram illustrating an example of an operator state held by an operator state specifying unit according to an embodiment; 6 is a diagram illustrating an example of an action transition probability P based on an environmental state and an operator state stored in a transition information storage unit according to an embodiment; ｏｐｅｒａｔｏｒ 1 is an example of the behavior transition probability P based on the plant state stored in the transition information storage unit according to the embodiment. ｐｌａｎｔ 1 is a diagram illustrating an example of a transition of operator behavior according to an operator model related to an embodiment. FIG. 2 is a diagram illustrating an example of a plant event held by a plant event identification unit related to an embodiment. FIG. 3 is an example of a plant state held by a plant dynamics analysis unit related to an embodiment. FIG. 4 is a diagram illustrating an example of HMI design information stored in an HMI design information storage unit. FIG. 5 is a diagram illustrating an example of a plant state presentation by an HMI information presentation unit. FIG. 6 is a diagram illustrating examples of four scenarios generated in an embodiment. FIG. 7 is a diagram conceptually illustrating an example of a configuration of a plant design support system related to an embodiment. FIG. 8 is a flowchart illustrating an example of an operation of the plant design support system related to an embodiment. FIG. 9 is a diagram illustrating an example of an event response failure criterion based on a plant state, which is scenario evaluation information stored in a scenario evaluation information storage unit of an embodiment. FIG. 10 is a diagram illustrating an example of an operator performance evaluation criterion based on operator behavior, which is scenario evaluation information stored in a scenario evaluation information storage unit of an embodiment. FIG. 11 is a diagram illustrating an example of a scenario evaluation held by a scenario evaluation unit of an embodiment. FIG. 12 is a diagram illustrating an example of a scenario aggregation result presented by a scenario aggregation unit of an embodiment.
[0014] Hereinafter, embodiments will be described with reference to the accompanying drawings. In the following embodiments, detailed features are shown for the purpose of explaining the technology, but these are merely examples and are not necessarily essential features for enabling the embodiments to be implemented.
[0015] The drawings are schematic, and for the sake of convenience, components may be omitted or simplified as appropriate. The relative sizes and positions of components shown in different drawings are not necessarily accurately depicted and may be changed as appropriate. Hatching may also be used in drawings such as plan views that are not cross-sectional views to facilitate understanding of the embodiments.
[0016] In the following description, the same components are denoted by the same reference numerals, and their names and functions are also the same. Therefore, detailed descriptions of them may be omitted to avoid duplication.
[0017] Furthermore, in the description given in this specification, when a certain component is described as "comprising," "including," or "having," unless otherwise specified, this is not an exclusive expression that excludes the presence of other components.
[0018] Furthermore, in the description of this specification, even if ordinal numbers such as "first" or "second" are used, these terms are used for convenience to make it easier to understand the contents of the embodiments, and the contents of the embodiments are not limited to the order that may result from these ordinal numbers.
[0019] First Embodiment A plant design support system and a plant design support method according to this embodiment will be described below.
[0020] <Configuration of Plant Design Support System> Fig. 1 is a diagram showing an example of the hardware configuration of a plant design support system. As shown in the example of Fig. 1, the plant design support system includes a processor 1 (processing circuit) that performs calculations, a memory 2 and a hard disk 3 that can store information, an input device 4 that can input information such as a mouse, a keyboard, a touch panel, or various switches, an output device 5 that can output information such as a display, a liquid crystal display device, or a lamp, and a system bus 6 that connects these components via communication or the like.
[0021] The memory 2 and the hard disk 3 may be, for example, a hard disk drive (HDD), a random access memory (RAM), a read only memory (ROM), a flash memory, an erasable programmable read only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or other volatile or non-volatile semiconductor memory, a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, or a DVD (recording medium), or any recording medium that will be used in the future.
[0022] The processor 1 may execute a program stored in the memory 2 and hard disk 3, an external CD-ROM, an external DVD-ROM, an external flash memory, etc. That is, the processor 1 may be, for example, a central processing unit (CPU), a microprocessor, a microcomputer, or a digital signal processor (DSP).
[0023] Note that the hardware configuration illustrated in FIG. 1 may not match the numbers and other details of the configuration illustrated in FIG. 2, etc. This is because the configuration illustrated in FIG. 2, etc., represents conceptual units.
[0024] 2 is a diagram conceptually illustrating an example of the configuration of a plant design support system according to this embodiment. As illustrated in the example of FIG. 2, the plant design support system includes an environmental state identification unit 111, an operator state identification unit 112, a transition probability calculation unit 113, an operator behavior identification unit 114, a plant event identification unit 115, a plant dynamics analysis unit 116, and an HMI information presentation unit 117.
[0025] These functional units are realized by the processor 1 executing a program stored in the memory 2 or hard disk 3, an external CD-ROM, an external DVD-ROM, an external flash memory, etc. Note that the above functions may be realized by a plurality of processors 1, a plurality of memories 2, and a plurality of hard disks 3 working together.
[0026] As shown in the example of FIG. 2 , the environmental state identification unit 111 is connected to the environmental information storage unit 101, the operator state identification unit 112 is connected to the action history storage unit 102, the transition probability calculation unit 113 is connected to the transition information storage unit 103, the operator action identification unit 114 is connected to the operator model 104, the plant event identification unit 115 is connected to the plant event information storage unit 105, the plant dynamic characteristics analysis unit 116 is connected to the plant model 106, and the HMI information presentation unit 117 is connected to the HMI model 107 and the HMI design information storage unit 108.
[0027] <Operation of Plant Design Support System> Fig. 3 is a flowchart showing an example of operation of the plant design support system according to this embodiment. Hereinafter, with reference to Fig. 3, specific operations of the plant design support system will be described, in which identification of operator actions based on the environmental state, operator states, and plant states, identification of plant events based on the plant states, analysis of plant dynamic characteristics based on operator actions and plant events, and dynamic generation of scenarios by repeatedly reproducing HMI information presentation.
[0028] First, the environmental state identification unit 111 identifies the environmental state at time t from the environmental information stored in the environmental information storage unit 101 (step ST101). The environmental information stored in the environmental information storage unit 101 is external factors that affect the behavior of the operator, and includes information about the operator's working environment, such as temperature, humidity, or lighting, or information about the situation on the site due to a disaster, such as a fire or earthquake. The environmental state is an index that indicates the state of the environment identified based on the environmental information.
[0029] 4 is a diagram showing an example of an environmental state held by the environmental state determination unit 111 according to this embodiment. As shown in the example in Fig. 4, the environmental state 400 includes temperature and humidity 401, lighting 402, site conditions 403, and environmental state 404, each of which indicates the current state in three levels: "good," "fair," and "poor."
[0030] In this embodiment, the environmental condition identification unit 111 identifies the environmental condition as a comprehensive evaluation from the temperature and humidity 401, lighting 402, and site conditions 403 of the environmental conditions 400 stored in the environmental information storage unit 101. The environmental condition identification unit 111 identifies the environmental condition as "fair" based on the results that the temperature and humidity 401 are "good," the lighting 402 is "fair," and the site conditions 403 are "bad."
[0031] Next, the operator state identification unit 112 identifies the operator state (an index showing the physical or mental state of the operator) at time t from the operator's behavior history for each time unit stored in the behavior history storage unit 102 (step ST102). The behavior history storage unit 102 stores the history of operator behavior up to the previous time unit identified by the operator behavior identification unit 114, i.e., up to time t-1, as the operator behavior history for each time unit.
[0032] 5 is a diagram showing an example of an operator status stored in the operator status identification unit 112 according to this embodiment. As shown in the example in Fig. 5, an operator status 500 includes a behavior transition interval 501, a load situation 502, and an operator status 503, each of which indicates the current status in three stages: "good," "passable," or "poor."
[0033] In this embodiment, the operator state identification unit 112 identifies the operator state as a comprehensive evaluation based on the behavior transition interval 501 and the workload status 502 of the operator state 500 stored in the behavior history storage unit 102. The behavior transition interval 501 is the time interval from the time of the last operator action to the current time. The workload status 502 is an index indicating the degree of physical and cognitive workload on the operator due to the last operator action. The operator state identification unit 112 identifies the operator state as "fair" based on the results that the behavior transition interval 501 is "good" and the workload status 502 is "fair."
[0034] Next, the transition probability calculation unit 113 calculates a transition probability of an operator's action at time t from the environmental state at time t identified by the environmental state identification unit 111, the operator's state identified by the operator's state identification unit 112, the operator's action transition information stored in the transition information storage unit 103, and the plant state at the previous time, i.e., time t-1, analyzed by the plant dynamics analysis unit 116 (step ST103). Note that the transition probability of an operator's action at time t can be calculated from at least the plant state and the operator state.
[0035] FIG. 6 shows the behavior transition probability P based on the environmental state and the operator state stored in the transition information storage unit 103 according to this embodiment. ｏｐｅｒａｔｏｒ 600. The transition probability calculation unit 113 determines the behavior transition probability from the operator state 601 and the environmental state 602. In the example of this embodiment, the environmental state is "OK" and the operator state is "OK", so the behavior transition probability P ｏｐｅｒａｔｏｒ 600 is 0.6.
[0036] In this embodiment, a method for calculating the behavior transition probability from the environmental state and the operator state has been shown, but the behavior transition probability may also be calculated using the environmental state and the operator state individually without combining them.
[0037] FIG. 7 shows the behavior transition probability P based on the plant state stored in the transition information storage unit 103 according to this embodiment. ｐｌａｎｔ 7 shows an example of the plant temperature 701, and the vertical axis shows the behavior transition probability P ｐｌａｎｔ 702 is shown.
[0038] When the plant temperature is within the appropriate range, the transition probability is high because operators are likely to be able to act smoothly. On the other hand, as the plant temperature deviates from the appropriate range, operators are more likely to fail to act due to impatience or tension, so the transition probability is low.
[0039] When the plant temperature at the previous time, i.e., time t-1, is 310° C. from the plant state analyzed by the plant dynamics analysis unit 116, the behavior transition probability P ｐｌａｎｔ is 0.6 from the graph shown in FIG.
[0040] From these results, the behavior transition probability P(t) of the operator behavior at time t is the product of the respective transition probabilities, and is therefore given as follows:
[0041]
[0042] In this embodiment, a calculation method based on the plant temperature is shown, but the method is not limited to this. The behavior transition probability P(t) of the operator behavior at time t may be calculated based on other plant states or plant states of adjacent plants, or a combination thereof.
[0043] Furthermore, in this embodiment, a method for calculating the transition probability as the product of the action transition probability based on the environmental state and the operator state and the action transition probability based on the plant temperature has been described, but the method is not limited to this, and the action transition probability P(t) of the operator action at time t may be calculated based on the sum or combination of these. For example, the transition probability of the operator action at time t may be calculated based on the transition probability based on the plant state and the transition probability based on the operator state.
[0044] Next, the operator behavior identification unit 114 identifies the operator behavior at time t based on the operator behavior transition probability calculated by the transition probability calculation unit 113, the operator model 104, and the plant state display presented by the HMI information presentation unit 117 (step ST104). The operator model 104 simulates the behavior of an operator, and receives information from the HMI information presentation unit 117 and inputs information to the HMI information presentation unit 117.
[0045] 8 is a diagram showing an example of the transition of operator actions by the operator model 104 according to this embodiment. The operator model 800 includes, as operator actions, monitoring 801, in which the operator checks the plant status display presented by the HMI information presentation unit 117 from the top left to the bottom right of the screen to confirm that there are no abnormalities in the plant status, recognition 802, in which the operator determines the cause of the abnormality based on the plant status and a response procedure corresponding to that cause, and operation 804, including screen transitions, for executing the response procedure based on the determination result. Transitions from one action to another in the operator model 800 occur in the direction of the arrows, with a probability calculated using the action transition probability P(t) at time t.
[0046] In this embodiment, if the operator's behavior at the previous time, i.e., time t-1, was "awareness" shown by the solid line, the probability of transitioning to "judgment" shown by the dotted line at time t and the probability of maintaining "awareness" are respectively shown as follows:
[0047]
[0048]
[0049] In this embodiment, an example of a simple behavioral transition in which "monitoring," "recognition," "judgment," and "operation" progress in one direction is shown, but the present invention is not limited to this. For example, the model can become more complex depending on the plant state, such as recognizing that another plant event has occurred while making a judgment about an occurred plant event, or completing an operation to respond to a plant event and transitioning to monitoring.
[0050] In this embodiment, an example in which the process transitions from "cognition" to "judgment" will be described.
[0051] Next, the plant event identification unit 115 identifies a plant event at time t from the plant information stored in the plant event information storage unit 105 and the plant state at the previous time analyzed by the plant dynamics analysis unit 116, i.e., at time t-1 (step ST105). Examples of plant events include a failure of plant equipment, a piping rupture, or a power loss. Plant equipment refers to equipment that constitutes a plant and varies depending on the type of plant, but may include, for example, a motor, a pump, or a valve. Plant states refer to the operating states or process values (including plant temperature) of various plant equipment. The occurrence of a plant event is influenced by the plant state; for example, the higher the plant temperature, the more likely plant equipment is to fail, and the higher the flow rate in a piping, the more likely a piping rupture occurs.
[0052] 9 is a diagram showing an example of a plant event held by the plant event identification unit 115 according to this embodiment. As shown in the example in Fig. 9, a plant event 900 includes a pump A state 901, a valve B state 902, a plant temperature 903, and a pipe flow rate 904, which are plant states at the previous time, i.e., time t-1, identified by the plant dynamics analysis unit 116, and a plant event 905 at time t identified from the results of these.
[0053] In this embodiment, at time t-1, pump A state 901 is in operation, valve B state 902 is open, plant temperature 903 is 300° C., and pipe flow rate 904 is 10.0 m ３ / h, the plant event identifying unit 115 identifies the occurrence of a failure in pump A as the plant event at time t.
[0054] Next, the plant dynamics analysis unit 116 performs dynamics analysis by simulating (reproducing) the behavior of the plant caused by the operator behavior identified by the operator behavior identification unit 114 and the plant event identified by the plant event identification unit 115, in accordance with the plant model 106. Then, the plant dynamics analysis unit 116 identifies the plant state at time t from the plant state at time t-1 based on the analysis result of the plant behavior (step ST106).
[0055] 10 shows an example of a plant state held by the plant dynamics analysis unit 116 in this embodiment. A plant state 1000 includes an operator action 1001 at time t, a plant event 1002, a pump A state 1003, a valve B state 1004, a plant temperature 1005, and a pipe flow rate 1006.
[0056] In this embodiment, the operator makes a decision, and as a result of the occurrence of a pump A failure as a plant event, the plant state at time t is as follows: Pump A is stopped, Valve B is open, the plant temperature is 310°C, and the pipe flow rate is 5.0 m ３ / h.
[0057] Next, the HMI information presenting unit 117 presents the plant state analyzed by the plant dynamics analyzing unit 116 in accordance with the HMI model 107 and the HMI design information stored in the HMI design information storing unit 108 (step ST107). Examples of the HMI design information stored in the HMI design information storing unit 108 include screen elements such as the shape, size, arrangement, or color of graphics or characters, or operation methods of screen elements such as blinking cycles or screen transitions.
[0058] Fig. 11 is a diagram showing an example of HMI design information stored in the HMI design information storage unit 108. As shown in the example in Fig. 11, HMI design information 1100 includes a pump status display 1101, a valve status display 1102, and a process value display 1103.
[0059] 12 is a diagram showing an example of a plant state presentation 1200 by the HMI information presentation unit 117. As shown in the example in Fig. 12, the pump A state, valve B state, plant temperature, and pipe flow rate analyzed by the plant dynamic characteristics analysis unit 116 are displayed according to the HMI design information 1100.
[0060] After the operations from step ST101 to step ST107 are performed for time t, the end of the scenario is determined (step ST108). The end of the scenario is determined based on whether the elapsed time from the start of the scenario or the plant state has reached a designated value.
[0061] If the scenario termination condition is not satisfied, steps ST101 to ST107 are performed for time t+1. In this case, the plant state at time t+1 is dynamically determined by reflecting the results of the operator behavior determination at time t, the plant dynamics analysis, and the HMI information presentation. The operations from step ST101 to step ST106 are repeatedly performed until the scenario termination condition is satisfied (step ST108), and a scenario is dynamically generated.
[0062] The above series of scenario generation operations are then repeated until the designated number of scenarios are generated (step ST109).
[0063] Fig. 13 is a diagram showing examples of four scenarios 1300 generated in this embodiment. In Fig. 13, operator actions identified by the operator action identification unit 114 in Fig. 2 are shown by solid lines, and plant events identified by the plant event identification unit 115 in Fig. 2 are shown by dotted lines, along an elapsed time 1301.
[0064] Although four scenarios are shown as an example in FIG. 13, in practice, a large number of scenarios are generated to increase the comprehensiveness of the scenarios.
[0065] As described above, the plant design support system according to this embodiment can generate highly accurate scenarios that can reproduce operator actions and plant behavior by using the results from the previous time to identify operator actions or plant events for the next time, taking into account the mutual influence between the operator status and the plant status, which change from moment to moment.
[0066] Furthermore, because the operator's actions are identified using transition probabilities according to the current state, a different scenario is generated each time a trial is performed, making it possible to efficiently generate scenarios that progress in various patterns.
[0067] Second Embodiment A plant design support system and a plant design support method according to this embodiment will be described. In the following description, components similar to those described in the above-described embodiment will be denoted by the same reference numerals, and detailed descriptions thereof will be omitted as appropriate.
[0068] <Configuration of Plant Design Support System> The hardware configuration of the plant design support system according to this embodiment is the same as the configuration shown in FIG.
[0069] 14 is a diagram conceptually illustrating an example of the configuration of a plant design support system according to this embodiment. As illustrated in the example of FIG. 14, the plant design support system includes an environmental state identification unit 111, an operator state identification unit 112, a transition probability calculation unit 113, an operator behavior identification unit 114, a plant event identification unit 115, a plant dynamics analysis unit 116, an HMI information presentation unit 117, a scenario evaluation unit 211, and a scenario aggregation unit 212.
[0070] These functional units are realized by the processor 1 executing a program stored in the memory 2 or hard disk 3, an external CD-ROM, an external DVD-ROM, an external flash memory, etc. Note that the above functions may be realized by a plurality of processors 1, a plurality of memories 2, and a plurality of hard disks 3 working together.
[0071] As shown in the example of FIG. 14, a scenario evaluation unit 211 is connected to the scenario evaluation information storage unit 201 .
[0072] <Regarding Operation of Plant Design Support System> Fig. 15 is a flowchart showing an example of operation of the plant design support system according to this embodiment. Hereinafter, the operation of the plant design support system will be specifically described with reference to Fig. 15 .
[0073] First, as shown in FIG. 3, steps ST101 to ST109 are carried out.
[0074] Next, the generated scenarios are evaluated, and the evaluation results are compiled for each scenario feature.
[0075] Specifically, the scenario evaluation unit 211 evaluates each scenario generated up to step ST109 in accordance with scenario evaluation information indicating predetermined criteria for evaluating scenarios stored in the scenario evaluation information storage unit 201 (step ST201).
[0076] FIG. 16 is a diagram showing an example of event response failure criteria based on the plant state, which is scenario evaluation information stored in the scenario evaluation information storage unit 201 of this embodiment.
[0077] As shown in an example in Figure 16, the event response failure criteria 1600 include criteria for event response failure based on the operating status or process values of plant equipment, and if any of the conditions is met, the scenario evaluation unit 211 determines that the event response has failed.
[0078] FIG. 17 is a diagram showing an example of operator performance evaluation criteria based on operator behavior, which is scenario evaluation information stored in the scenario evaluation information storage unit 201 of this embodiment.
[0079] As shown in an example in Fig. 17, an operator performance evaluation criterion 1700 is identified from the results of a load 1701 and a number of actions 1702. The load 1701 is a comprehensive index of the physical and cognitive load of the operator actions performed in a series of scenarios, and is shown in three levels: "good," "fair," and "poor." The number of actions 1702 is the total number of operator actions performed in a series of scenarios, and is shown in three levels: "low," "medium," and "high." The higher the value shown in Fig. 17, the better the operator performance.
[0080] 18 is a diagram showing an example of a scenario evaluation held by the scenario evaluation unit 211 in this embodiment. In the scenario evaluation 1800, for each scenario, a plant temperature 1801, a pipe flow rate 1802, a load 1803, a number of actions 1804, an event response success / failure 1805, and an operator performance 1806 are provided.
[0081] In scenario 1 in FIG. 18, the plant temperature 1801 is 300° C., and the pipe flow rate 1802 is 10.0 m ３ / h, neither of which meets the event response failure criteria in Fig. 16. Therefore, the event response success or failure is evaluated as "success." Furthermore, in scenario 1 in Fig. 18, the load 1803 is "good" and the number of actions 1804 is "few," so the operator performance 1806 is evaluated as "80."
[0082] Similarly, in scenario 2 in FIG. 18, the plant temperature 1801 is 315° C., and the pipe flow rate 1802 is 8.0 m ３ 18, the load 1803 is "unacceptable" and the number of actions 1804 is "medium", so the operator performance 1806 is evaluated as "60".
[0083] Similarly, in scenario 3 in FIG. 18, the plant temperature 1801 is 310° C., and the pipe flow rate 1802 is 9.0 m ３ / h, and the success or failure of the event response is evaluated as "success." Furthermore, in scenario 3 in Fig. 18, the load 1803 is "unacceptable" and the number of actions 1804 is "high," so the operator performance 1806 is evaluated as "100."
[0084] Similarly, in scenario 4 in FIG. 18, the plant temperature 1801 is 360° C., and the pipe flow rate 1802 is 0.3 m ３ / h, and the success or failure of the event response is evaluated as "failed." Furthermore, in scenario 4 in Fig. 18, the load 1803 is "fair" and the number of actions 1804 is "medium," so the operator performance 1806 is evaluated as "40."
[0085] Next, the scenario tallying unit 212 tally and presents the evaluation results of each scenario evaluated by the scenario evaluating unit 211 for each feature of the scenario (step ST202).
[0086] 19 is a diagram showing an example of a scenario aggregation result presented by the scenario aggregation unit 212 in this embodiment. As shown in the example in Fig. 19, a scenario aggregation result 1900 includes scenario characteristics 1901, an event response success probability 1902, and an operator performance average 1903.
[0087] In the example of FIG. 19, the scenario features 1901 are compiled for "Occurrence of a failure in pump A," "Occurrence of a failure in valve B," and "Failure in operation B when a failure in valve B occurs."
[0088] "Failure of pump A occurs" refers to a scenario in which a failure of pump A occurs, and corresponds to all of scenarios 1, 2, 3, and 4 in Fig. 18. Of these, the scenarios in which the event was successfully handled are scenarios 1, 2, and 3, and the probability of success of handling the event 1902 is calculated as follows:
[0089]
[0090] In this case, the operator performances are 80, 60, 100, and 40 for scenario 1, scenario 2, scenario 3, and scenario 4, respectively, and the operator performance average 1903 is calculated as follows:
[0091]
[0092] "Valve B failure occurs" refers to a scenario in which a failure of valve B occurs, and corresponds to three scenarios, Scenario 2, Scenario 3, and Scenario 4 in Fig. 18. Of these, the scenarios in which the event was successfully handled are Scenario 2 and Scenario 3, and the probability of event handling success 1902 is calculated as follows:
[0093]
[0094] In this case, the operator performances are 60, 100, and 40 for scenario 2, scenario 3, and scenario 4, respectively, so the operator performance average 1903 is calculated as follows:
[0095]
[0096] "Operation B fails when valve B fails" refers to a scenario in which operation B is not performed among the scenarios in which valve B fails, and corresponds to scenario 4 in Fig. 18. Therefore, the event response success probability 1902 is calculated as follows:
[0097]
[0098] In this case, the operator performance is 40 in scenario 4, so the operator performance average 1903 is calculated as follows:
[0099]
[0100] Although four scenarios are tallied in this embodiment, highly reliable statistical results can be obtained by generating, evaluating, and tallying a large number of scenarios.
[0101] Furthermore, in this embodiment, an example has been described in which scenarios are compiled based on whether or not a plant event has occurred or whether or not an operator has taken action, but this is not limited to this, and scenarios may also be compiled based on time, such as the time it takes for an operator to recognize a plant event after it has occurred.
[0102] As described above, according to the plant design support system of this embodiment, the generated scenarios are evaluated from the perspective of the success or failure of event response and operator performance, and the evaluation results are compiled for each scenario feature, making it easy to analyze scenarios, such as which scenarios affect the success or failure of event response or operator performance.
[0103] Furthermore, based on the above analysis results, it is possible to modify and evaluate HMI design information to support event response or operator performance, and compare the results, enabling efficient HMI design evaluation at the design stage.
[0104] <Regarding the Effects Produced by the Multiple Embodiments Described Above> Next, examples of the effects produced by the multiple embodiments described above will be described. Note that in the following description, the effects will be described based on the specific configurations exemplified in the multiple embodiments described above, but these may be replaced with other specific configurations exemplified in the present specification to the extent that similar effects are produced. In other words, for convenience, only one of the associated specific configurations may be described as a representative below, but the representatively described specific configuration may be replaced with another associated specific configuration.
[0105] Furthermore, the replacement may be made across multiple embodiments, i.e., configurations illustrated in different embodiments may be combined to produce the same effect.
[0106] According to the embodiment described above, the plant design support system includes an operator state identification unit 112, a transition probability calculation unit 113, an operator action identification unit 114, and a plant state identification unit. Here, the plant state identification unit corresponds to, for example, the plant dynamics analysis unit 116. The operator state identification unit 112 identifies an operator state, which is an index indicating the state of the operator, based on the operator's action history for each unit of time. The transition probability calculation unit 113 calculates a transition probability of the operator's action based on the plant state, which is an index indicating the state of the plant, and the operator state. The operator action identification unit 114 identifies the operator's action based on the transition probability. The plant dynamics analysis unit 116 identifies the plant state for a later unit of time from the plant state for a previous unit of time based on the behavior of the plant caused by the operator's action.
[0107] Furthermore, according to the embodiment described above, the plant design support system includes a processor 1 (processing circuit) that executes a program and a memory 2 that stores the program to be executed. The processor 1 executes the program to realize the following operations.
[0108] That is, an operator state, which is an index indicating the state of an operator, is identified based on the operator's behavior history for each unit of time. Then, a transition probability of the operator's behavior is calculated based on the plant state, which is an index indicating the state of the plant, and the operator state. Then, the operator's behavior is identified based on the transition probability. Then, the plant state for a later unit of time is identified from the plant state for a previous unit of time based on the behavior of the plant caused by the operator's behavior.
[0109] With this configuration, by identifying the plant state at a later time unit from the plant state at a previous time unit based on the plant behavior caused by the operator's actions, it is possible to support plant design and the generation of highly accurate scenarios by taking into consideration the mutual influence between the operator's state and the plant state, which change from moment to moment. Furthermore, because the transition probability is controlled according to the current operator's state and the plant state and the operator's actions are identified, a different scenario is generated for each trial, making it possible to efficiently generate comprehensive scenarios.
[0110] Furthermore, even if other configurations shown as examples in this specification are appropriately added to the above configuration, that is, even if other configurations in this specification that were not mentioned as the above configuration are appropriately added, the same effect can be achieved.
[0111] Furthermore, according to the embodiment described above, the plant design support system includes an HMI information presentation unit 117 for presenting the plant status based on the HMI design information. With this configuration, the plant status can be presented using screen elements such as graphics or characters, making it easier for operators to understand the plant status.
[0112] Furthermore, according to the embodiment described above, the plant dynamics analysis unit 116 dynamically identifies the plant state at a later time unit in accordance with the operator actions identified over time by the operator action identification unit 114. With this configuration, it is possible to generate a highly accurate scenario that reproduces the plant behavior resulting from the operator actions, taking into consideration the mutual influence between the operator state and the plant state, which change from moment to moment.
[0113] Furthermore, according to the embodiment described above, the plant design support system includes a plant event identification unit 115 for identifying a plant event based on the plant state. The plant dynamics analysis unit 116 then identifies the plant state for a later time unit from the plant state for a previous time unit based on the plant behavior resulting from the operator actions and the plant event. This configuration makes it possible to support the generation of highly accurate scenarios by identifying the plant state for a later time unit from the plant state for a previous time unit based on the plant behavior resulting from the operator actions and the plant event, thereby taking into account the mutual influence between the operator state and the plant state, which changes from moment to moment.
[0114] Furthermore, according to the embodiment described above, the plant design support system includes an environmental state identification unit 111 for identifying an environmental state based on environmental information. The transition probability calculation unit 113 calculates transition probabilities based on the plant state, the operator state, and the environmental state. This configuration identifies the plant state at a later time unit from the plant state at a previous time unit based on the plant behavior resulting from the operator actions identified while taking the environmental state into consideration. This makes it possible to support the generation of highly accurate scenarios by taking into consideration the mutual influence between the operator states and the plant state, which change from moment to moment.
[0115] Furthermore, according to the embodiment described above, the plant design support system includes a scenario evaluation unit 211 and a scenario aggregation unit 212. The scenario evaluation unit 211 outputs evaluation results for scenarios generated based on the plant state and operator state for each time unit after the identification based on predetermined evaluation criteria. The scenario aggregation unit 212 aggregates the evaluation results for each scenario feature. With this configuration, by evaluating the generated scenarios and aggregating the evaluation results for each scenario feature, it is possible to efficiently perform scenario analysis for various scenarios, such as determining which scenarios result in a worsening evaluation or which HMI elements should be improved, or comparing evaluation results when changing the HMI design.
[0116] According to the embodiment described above, in the plant design support method, an operator state, which is an index indicating the state of an operator, is identified based on the operator's behavior history for each unit of time. Then, a transition probability of the operator's behavior is calculated based on the plant state, which is an index indicating the state of the plant, and the operator state. Then, the operator's behavior is identified based on the transition probability. Then, a plant state for a later unit of time is identified from the plant state for a previous unit of time based on the behavior of the plant caused by the operator's behavior.
[0117] With this configuration, by identifying the plant state at a later time unit from the plant state at a previous time unit based on the plant behavior caused by the operator's actions, it is possible to support plant design and the generation of highly accurate scenarios by taking into consideration the mutual influence between the operator's state and the plant state, which change from moment to moment. Furthermore, because the transition probability is controlled according to the current operator's state and the plant state and the operator's actions are identified, a different scenario is generated for each trial, making it possible to efficiently generate comprehensive scenarios.
[0118] Unless otherwise specified, the order in which the processes are performed can be changed.
[0119] Furthermore, even if other configurations shown as examples in this specification are appropriately added to the above configuration, that is, even if other configurations in this specification that were not mentioned as the above configuration are appropriately added, the same effect can be achieved.
[0120] <Regarding Modifications of the Multiple Embodiments Described Above> In the multiple embodiments described above, the dimensions, shapes, relative positional relationships, and implementation conditions of each component may be described, but these are merely examples in all aspects and are not limiting.
[0121] Therefore, countless modifications and equivalents not shown as examples are contemplated within the scope of the technology disclosed in the present specification, including, for example, modifying, adding, or omitting at least one component, and further, extracting at least one component from at least one embodiment and combining it with a component from another embodiment.
[0122] Furthermore, the descriptions in this specification are incorporated by reference for all purposes related to the present technology, and none of them are admitted to be prior art.
[0123] Furthermore, each component described in the above-described embodiments is envisioned as software or firmware, as well as corresponding hardware, and as software it is referred to as, for example, a "unit," and as hardware it is referred to as, for example, a "processing circuitry."
[0124] Furthermore, the technology disclosed in this specification may also be in a form in which each component is distributed across multiple devices, that is, in a form such as a system that is a combination of multiple devices.
[0125] 1 Processor, 2 Memory, 3 Hard disk, 4 Input device, 5 Output device, 6 System bus, 101 Environmental information storage unit, 102 Action history storage unit, 103 Transition information storage unit, 104 Operator model, 105 Plant event information storage unit, 106 Plant model, 107 HMI model, 108 HMI design information storage unit, 111 Environmental state identification unit, 112 Operator state identification unit, 113 Transition probability calculation unit, 114 Operator action identification unit, 115 Plant event identification unit, 116 Plant dynamic characteristics analysis unit, 117 HMI information presentation unit, 201 Scenario evaluation information storage unit, 211 Scenario evaluation unit, 212 Scenario aggregation unit, 400 Environmental state, 401 Humidity, 402 Lighting, 403 Site situation, 404 Environmental state, 500 Operator state, 501 Action transition interval, 502 Load status, 503 Operator status, 601 Operator status, 602 Environmental status, 701 Plant temperature, 800 Operator model, 801 Monitoring, 802 Recognition, 803 Decision, 804 Operation, 900 Plant event, 901 Pump A status, 902 Valve B status, 903 Plant temperature, 904 Pipe flow rate, 905 Plant event, 1000 Plant status, 1001 Operator action, 1002 Plant event, 1003 Pump A status, 1004 Valve B status, 1005 Plant temperature, 1006 Pipe flow rate, 1100 HMI design information, 1101 Pump status display, 1102 Valve status display, 1103 Process value display, 1200 Plant status presentation, 1300 Scenario, 1301 Elapsed time, 1701 Load, 1702 Number of actions, 1800 Scenario evaluation, 1801 Plant temperature, 1802 Pipe flow rate, 1803 Load, 1804 Number of actions, 1805 Success or failure of event response, 1806 Operator performance, 1900 Scenario aggregation result, 1901 Characteristics, 1902 Event response success probability, 1903 Operator performance average.
Claims
1. An operator status identification unit for identifying the operator status, which is an indicator showing the operator's state based on the operator's activity history for each time unit, A transition probability calculation unit for calculating the transition probability of the operator's actions based on the plant state, which is an indicator of the plant's condition, and the operator's condition, An operator behavior identification unit for identifying the operator's actions based on the transition probability, The system includes a plant state identification unit for identifying the plant state in a later time unit from the plant state in a earlier time unit, based on the behavior of the plant caused by the operator's actions. Plant design support system.
2. The plant design support system according to claim 1, The system further includes an HMI information display unit for displaying the plant status based on HMI design information. Plant design support system.
3. A plant design support system according to claim 1 or 2, In accordance with the operator's actions identified sequentially over time by the operator action identification unit, the plant state identification unit dynamically identifies the plant state for subsequent time units. Plant design support system.
4. A plant design support system according to claim 1 or 2, The plant event identification unit further comprises a plant event identification unit for identifying plant events based on the aforementioned plant state, The plant state identification unit identifies the plant state for a later time unit from the plant state for a earlier time unit, based on the behavior of the plant caused by the operator's actions and the plant events. Plant design support system.
5. A plant design support system according to claim 1 or 2, It further includes an environmental condition identification unit for identifying environmental conditions based on environmental information, The transition probability calculation unit calculates the transition probability based on the plant state, the operator state, and the environmental state. Plant design support system.
6. A plant design support system according to claim 1 or 2, A scenario evaluation unit for outputting evaluation results based on predetermined evaluation criteria for a scenario generated based on the plant state and operator state at specified time units, The system further includes a scenario aggregation unit for aggregating the evaluation results for each characteristic of the aforementioned scenarios. Plant design support system.
7. Based on the driver's activity history for each time unit, the driver's status, which is an indicator showing the driver's state, is identified. Based on the plant status, which is an indicator of the plant's condition, and the operator status, the probability of transitions in the operator's actions is calculated. Based on the transition probability, the operator's actions are identified. Based on the behavior of the plant resulting from the operator's actions, the plant state for a later time unit is determined from the plant state for a earlier time unit. Plant design support methods.