JP7252703B2 - Operation support device, operation support method, operation support program, and recording medium - Google Patents

Description

The present invention relates to an operation assistance device, an operation assistance method, an operation assistance program, and a recording medium.

2. Description of the Related Art Conventionally, at a production site or the like, there are cases where the details of operations when operating a plant or the like differ depending on the skill level of a worker (operator). For example, when an event such as an alarm, data anomaly, or equipment or environmental anomaly occurs in a plant, the operator should understand the content of the anomaly that caused the event and take appropriate action according to the degree of urgency. It is necessary to operate the plant (abnormal operation). Prediction of event occurrence, discovery of anomaly after event occurrence, content of operation at the time of anomaly, time to recover from anomaly, etc. may vary depending on the operator's skill level.

Also, in a plant, the behavior (trend) of a batch differs depending on the production method such as the operation method of the plant, and as a result, the quality and production volume of the produced product may differ. A batch that produces the best results in terms of product quality, etc. is called a golden batch, and an optimal production method is required to obtain the golden batch results.

Further, improvement of skills in plant operating methods and the like is achieved by educating and training operators. Training of workers is performed, for example, by having a highly skilled worker (veteran) who is skilled in the work accompany and guide a worker with a low skill level. For example, a skilled worker conveys the points in the work to the worker, and also guides the worker by coordinating the operation of the plant and the content of the work (operation content) to improve the skill (know-how) of the skilled worker. ) is handed down. In order to improve the skill level, in addition to the operation method when the plant is in a normal state, it is necessary to teach the operation method when the plant is in an abnormal state with a low frequency of occurrence, so it takes a lot of time. In a production site where the number of skilled workers is decreasing, it becomes difficult to implement worker training, and the labor costs of skilled workers may increase.

Production sites include, for example, industrial plants such as chemicals, plants that manage and control wellheads such as gas fields and oil fields and their surroundings, plants that manage and control power generation such as hydropower, thermal power, and nuclear power, and solar and wind power plants. There are plants such as plants that manage and control energy harvesting, plants that manage and control water supply, sewage, dams, etc., and factories for machines, electric machines, automobiles, etc. (hereinafter collectively referred to as "plants"). In a plant, a distributed control system (DCS) is constructed in which field devices such as measuring instruments or operating devices called field devices and controllers for controlling these devices are connected via communication means. , a highly automated operation is realized. In a plant with such a distributed control system, the logic that collects information from various devices such as sensors to improve safety, detects anomalies based on the collected information, and generates alarms is the system. registered with

However, the logic when an alarm occurs does not necessarily correspond to the operation in an abnormal state when an alarm occurs, and there are cases where the operator cannot be instructed to take appropriate measures only with the occurrence of an alarm. Skilled workers collect various information through their five senses at the production site. Some of this information is difficult to collect by sensors installed in the plant. The information collected by skilled workers may include, for example, information on the quality of manufactured products, information on production plans, weather information, operating conditions of facilities, and the like. Based on the collected information, a skilled worker may perform an appropriate operation based on a judgment different from the logic for generating an event such as an alarm.

Meanwhile, at production sites, there are techniques for controlling plants and the like based on various information (see Patent Documents 1 to 6, for example). In recent years, in particular, in addition to information acquired from sensors installed in plants, it has been proposed to acquire information from all sorts of things by IoT (Internet of Things). For example, Industry 4.0 and the Industrial Internet are concepts that aim to improve the efficiency of production sites by using various information acquired by IoT. These concepts include analyzing the vast amount of data (big data) that accumulates information obtained by IoT and utilizing it at the production site. Big data collects and accumulates a wide range of data that could not be collected by a single device or system until now. By using big data, it becomes possible to analyze a wide range of data that could not be analyzed until now, and it may be possible to analyze many events whose causal relationships were unknown. Therefore, IoT is expected to be applied to various industrial fields.

In addition, as a data analysis method, a method using machine learning such as deep learning, which generates a model of the relationship between input and output using a multi-layered neural network, and a support vector machine, which generates a pattern recognition model based on training data, is proposed. It is

JP-A-11-287400 Japanese Patent Application Laid-Open No. 2003-177818 JP 2003-167624 A Japanese Patent Application Laid-Open No. 2003-140741 WO 02/35302 JP 2006-185120 A

However, in order to analyze the acquired data, in what kind of situation (event in the plant), which data should be analyzed, and what kind of judgment (decision of operation to be performed) should be made. Although it is necessary to decide in advance, the judgment criteria are handed down to workers by skilled workers through coordination, and workers cannot inherit the skills of skilled workers only by collecting data by IoT. there was a case.

In addition, when the judgment criteria are generated by machine learning based on the data collected by IoT, it becomes difficult to learn the events and phenomena that occur infrequently in the actual plant. In addition, it takes time to collect data on events and events with a high frequency of occurrence, and in some cases it is not possible to transfer skills related to work corresponding to events that occur infrequently in plant operations.

In addition, in order to find out the judgment criteria of skilled workers through interviews and formulate the logic by expressing them in a program or the like, it is necessary to interview skilled workers for each work decision. For example, if an interview is conducted every time an event occurs, the burden on a skilled worker may increase. On the other hand, when multiple events were interviewed collectively, the skilled workers' memory faded with the passage of time, and there were cases in which they were unable to find out accurate judgment criteria. In addition, when a skilled worker passes down the judgment criteria, it is possible to convey what is in the skilled worker's consciousness, but the judgment is made unconsciously (because of the skill, it is tacit knowledge). It was sometimes difficult to convey

In addition, when skilled workers accompany and train workers, it takes a considerable amount of time to hand down the judgment criteria for all events that occur in the plant, and automation and labor saving at the production site are progressing. In particular, it has become difficult to secure skilled workers, and in some cases it has become difficult to pass on the skills of skilled workers.

In addition, since the skills of workers who receive skill transfer differ for each worker, training tools such as manuals that are standardized based on the judgment criteria of experts make it difficult to transfer skills according to the skill level of workers. There was a case.

The present invention has been made in view of the above circumstances, and an operation support device, an operation support method, an operation support program and a The purpose is to provide a recording medium.

(1) In order to solve the above problems, the operation support device of the present invention includes an event information acquisition unit that acquires event information related to an event that occurs during operation of a plant; An event information acquisition unit that acquires information, a perception information acquisition unit that acquires perception information related to perception recognized by a worker, and an operation information acquisition unit that acquires operation information related to the operation of the plant performed by the worker. an extraction unit for extracting a relationship between the event information and at least one of the event information, the perceptual information, or the operation information; and supporting the operation of the plant based on the extracted relationship. and a solution generation unit that generates a solution for

(2) In addition, in the operation support device of the present invention, the extraction unit compares the event information with at least one of the event information, the perceptual information, and the operation information in chronological order. Extract relationships.

(3) In addition, in the operation support device of the present invention, the event information acquisition unit acquires an alarm occurrence event that notifies the abnormality of the plant, and the operation information acquisition unit acquires the operation at the time of abnormality performed by the operator. to acquire the operation information related to

(4) Further, in the operation support device of the present invention, the event information acquisition unit acquires a batch occurrence event indicating the occurrence of a batch exhibiting a predetermined behavior in the plant, and the operation information acquisition unit acquires a Acquire operation information related to the performed model operation for generating the batch generation event.

(5) Further, in the operation support device of the present invention, the extraction unit extracts the relationships according to the skill level of the worker, and the solution generation unit generates the solutions according to the skill level. do.

(6) In addition, the operation support device of the present invention further includes a solution providing unit that provides the generated solution according to occurrence of the event during operation of the plant.

(7) In addition, in the operation support device of the present invention, the sensory information acquisition unit acquires information of an external device connected via a network.

(8) Further, in the operation support device of the present invention, the sensory information acquisition unit acquires information on at least one of vibrations, sounds, and alarms emitted by equipment.

(9) Further, in the operation support device of the present invention, the sensory information acquisition unit acquires visual information visually recognized by the worker from a line-of-sight detection device that detects the line of sight of the worker.

(10) In addition, the operation support device of the present invention further includes a determination criterion acquisition unit that acquires a determination criterion of the worker who has performed the operation, and the extraction unit further includes the acquired determination criterion. Extract relationships.

(11) In order to solve the above problems, the operation support method of the present invention includes an event information acquisition step of acquiring event information related to an event that occurs during operation of a plant, and an event information related to an event detected in the plant. An event information acquisition step of acquiring information, a perception information acquisition step of acquiring perception information related to the perception recognized by the worker, and an operation information acquisition step of acquiring operation information related to the operation of the plant performed by the worker. an extracting step of extracting a relationship between the event information and at least one of the event information, the perceptual information, or the operation information; and supporting the operation of the plant based on the extracted relationship. and a solution generation step of generating a solution for

(12) In order to solve the above problems, the operation support program of the present invention includes an event information acquisition process for acquiring event information related to an event that occurs during operation of a plant, and an event information related to an event detected in the plant. Event information acquisition processing for acquiring information, perception information acquisition processing for acquiring perception information related to perception recognized by a worker, and operation information acquisition processing for acquiring operation information related to the operation of the plant performed by the worker. an extraction process for extracting a relationship between the event information and at least one of the event information, the perceptual information, or the operation information; and supporting the operation of the plant based on the extracted relationship. The computer is caused to execute a solution generation process for generating a solution to be used.

(13) In order to solve the above problems, the recording medium of the present invention includes an event information acquisition process for acquiring event information relating to an event that occurs during operation of a plant, and an event information acquisition process relating to an event detected in the plant. An event information acquisition process for acquiring a perceptual information acquisition process for acquiring sensory information related to the perception recognized by the worker, and an operation information acquisition process for acquiring operation information related to the operation of the plant performed by the worker. an extraction process for extracting a relationship between the event information and at least one of the event information, the perceptual information, or the operation information; and supporting the operation of the plant based on the extracted relationship. An operation support program for causing a computer to execute a solution generation process for generating a solution is recorded.

Advantageous Effects of Invention According to the present invention, it is possible to provide an operation support device, an operation support method, an operation support program, and a recording medium capable of facilitating transfer of skills related to work corresponding to an event that occurs in plant operation. .

It is a block diagram showing an example of a system configuration containing an operation support device in an embodiment. It is a block diagram which shows an example of the software configuration of the operation assistance apparatus in embodiment. 6 is a flow chart showing an example of event information acquisition operation of the operation support device in the embodiment. 4 is a flow chart showing an example of an event information acquisition operation of the operation support device in the embodiment; 4 is a flow chart showing an example of perceptual information acquisition operation of the operation support device in the embodiment. 4 is a flow chart showing an example of operation information acquisition operation of the operation support device in the embodiment. 7 is a flow chart showing an example of a skill-based solution generation operation of the operation support device in the embodiment. 7 is a flow chart showing an example of a relationship extraction operation of the operation support device according to the embodiment; 9 is a flowchart showing an example of a level-based solution providing operation of the operation support device in the embodiment; 7 is a flow chart showing an example of a feedback operation for work assistance of the operation assistance device in the embodiment; It is a block diagram which shows an example of the hardware constitutions of the operation assistance apparatus in embodiment.

Hereinafter, an operation assistance device, an operation assistance method, an operation assistance program, and a recording medium according to an embodiment of the present invention will be described in detail with reference to the drawings.

First, the outline of the operation support device will be described with reference to FIG. FIG. 1 is a block diagram showing an example of a system configuration including an operation support device according to an embodiment.

In FIG. 1, an operation support system 100 has an operation support device 1, a sales database 2A, a quality database 2B, a field database 2C, a skill-based solution database 3, and an execution environment 4. FIG.

The sales database 2A, the quality database 2B, and the field database 2C exemplify sources of big data acquired by IoT. In the concept of IoT, it is proposed to connect all kinds of things to networks and works such as the Internet and acquire a wide range of information (big data) obtained from all kinds of things. Sources of big data acquired by IoT are not limited to these databases, and any object that can be connected to a network and can provide data is assumed. The big data may include data such as measurement data acquired from a measuring instrument, image data acquired from a camera, voice data acquired from a microphone, mail data, and communication data, for example. The acquired big data is accumulated by the operation support device 1 . Acquisition of big data is performed by selecting from available data. Acquisition of big data may be selected on a case-by-case basis, for example, for analysis or the like, or pre-selected data may be acquired regularly or irregularly.

The business database 2A is a database having customer information, distribution information, sales information, and the like. The business database 2A is implemented by, for example, a core business system that manages the management resources of the entire company. The core business system may have, for example, accounting information, production management information, sales management information, inventory management information, etc., in addition to these pieces of information. As the core business system, for example, an ERP (Enterprise Resource Planning) system for the process manufacturing industry can be used.

In addition, the core business system may include a maintenance management system that manages plant operating status information, plant maintenance and repair work information, and the like. The information in the core business system may be referred to when a worker determines maintenance details or plans a schedule. For example, a worker refers to the required specifications included in the customer information and operates the plant so as to produce products that meet the required specifications. A skilled worker may produce a product that meets the customer's needs by referring to the past order information included in the customer information together with the required specifications. Also, skilled workers may decide to adjust the production speed of a product or to switch the production type while confirming the distribution information and production management information. In other words, the worker may change the operation of the plant based on the acquired information by acquiring and perceiving such information from the business database 2A.

The quality database 2B is a database having inspection data, analysis data, and the like. The quality database 2B can also be implemented by, for example, a core business system that manages the management resources of the entire company. For example, the operator can refer to the inspection data and analysis data to confirm the pass/fail of the product quality and determine whether the product shipping work can be carried out. A skilled worker may fine-tune process setting values and the like by referring to temporal changes (trends) in inspection data and analysis data.

Note that the quality database 2B may be realized by, for example, an MES (Manufacturing Execution System). MES is a system that monitors or manages plant conditions. The MES may have functions such as a PIMS (Plant Information Management System) or a CMMS (Computerized Maintenance Management System). PIMS functions as a plant data management system that collects and records plant data. The PIMS, for example, collects information on inspection devices and analysis devices and records it as inspection data and analysis data. The CMMS may also record the maintenance history of inspection devices and analysis devices. In other words, the operator may change the operation of the plant based on the acquired information by acquiring and perceiving such information from the quality database 2B.

The site database 2C is a database having normal event information, abnormal event information, plant data, control data, monitor information, sensor data, operator data, operation history, perception information, and the like. The field database 2C can also be implemented in the MES.

Normal event information is event information that occurs during normal operation of the plant, and includes, for example, a start notification event that notifies the start of a process, a status notification event that notifies that a process has reached a predetermined state, and an end of a process. is an end notification event or the like that notifies the For example, a worker normally starts a given work in response to the occurrence of an event.

Abnormal event information is event information that occurs when an abnormality occurs in a plant. For example, a process alarm event for notifying a process abnormality automatically detected by a sensor or the like, an operation error, or an abnormal operation is notified. It is an operation alarm etc. for A worker who recognizes these alarm events, for example, implements an operation at the time of anomaly, which is an operation for the process for resolving the anomaly, according to the cause of the alarm event.

Plant data is information related to plant operations, such as plant operating conditions and production quantities. Control data is control data for controlling the plant. The control data can be acquired as a control program or the like from a control device that controls the plant, for example. The monitor information is a monitor image obtained from a monitor camera installed in the plant. Sensor data is data obtained from sensors placed in the plant. Operator data is data for identifying a worker, and may include information on the skill (knowledge or skill) level of the worker, for example. The operation history is history information in which details of operations performed by the operator are recorded. Perceptual information is information obtained by measuring and recording visual, auditory, olfactory, tactile, and gustatory senses that can be perceived by a worker.

The operator gradually operates the gas concentration adjustment valve while referring to gas concentration data output from a gas concentration sensor arranged in the plant as sensor data, for example. A skilled worker may determine the operation amount of the gas concentration adjustment valve while referring to, for example, the gas concentration trend, the composition data of the reactor outlet, and monitor information in addition to the gas concentration data. In other words, the worker may change the operation of the plant based on the acquired information by acquiring and perceiving the information from the field database 2C.

The operation support device 1 collects and accumulates information selected from big data provided by the sales database 2A, the quality database 2B, and the site database 2C. The operation support device 1 searches and reads out the accumulated information, and generates judgment logic for judging the operation content when the operator performs the operation. The operation support device 1 generates and provides a skill-based solution according to the skill level of the worker from the generated judgment logic. Details of the operation support device 1 will be described later.

The skill-based solution database 3 stores skill-based solutions provided from the operation support device 1 . A skill-specific solution is a solution that is used according to the skill level of the worker. In the present embodiment, abnormal operation information, model operation information, automation information, and e-learning information are exemplified as solutions by skill.

<Operation information at abnormal time>
The abnormal operation information is information for supporting the abnormal operation of the plant (abnormal operation) when an abnormality occurs in the plant. The abnormal operation information supports appropriate abnormal operation according to the skill level of the worker. The plant operations that a worker can perform depend on the worker's skill level. For example, a new worker can set a device such as a valve to a specified state (for example, valve opening for a valve). On the other hand, a veteran operator can adjust multiple devices while checking the gauges. The abnormal operation information provides operators with operation instructions according to their skill level, thereby making it possible to take accurate (error-free) and prompt measures in response to an abnormality that has occurred in the plant. The abnormal operation information is generated according to the skill level of the worker who operates the plant based on the operation performed by the worker with the high skill level. The abnormal operation information is provided to the operator in association with the alarm event. The abnormal operation information is provided, for example, by displaying characters, still images, moving images, etc. on a display device, or by notifying with sound.

<Example operation information>
The model operation information is information for supporting the model operation of the plant (model operation) for generating the golden batch. The model operation information supports appropriate operation according to the skill level of the operator, similar to the abnormal operation. For example, a new worker can operate equipment while checking the numerical value of one gauge. On the other hand, a seasoned operator can see the trend of multiple gauges while adjusting the equipment. The model operation information allows the operator to perform an operation to generate a trend similar to the golden batch by giving an operation instruction according to the skill level of the worker. The model operation information is generated based on operations performed by a worker with a high skill level, and operation details corresponding to the skill level of the worker who operates the plant. The abnormal operation information is provided to the operator in association with the normal event. The model operation information is provided, for example, by displaying characters, still images, moving images, etc. on the display device, or by notifying with sound.

<Automation information>
Automation information is information for operating a control device that controls a plant. The automation information is, for example, a program that can be executed in a DCS (Distributed Control System) controller or a PLC (Programmable Logic Controller) controller that automatically controls a plant. Automation information is generated in executable form in each controller. The automation information may be the program itself executable in the control device, or may be, for example, a linkable library. Further, the automation information may represent the code of the program, or may give a predetermined parameter to the control device.

<e-learning information>
The e-learning information is e-learning learning content that can be provided to the worker. Operation information in abnormal conditions and model operation information are so-called online contents that provide information such as operation instructions when a worker operates a plant at a production site. It is a so-called offline content for improving knowledge. The e-learning information includes learning content corresponding to the skill level of the plant operating workers based on the operations performed by the workers with high skill levels. In e-learning, the skill level may be selected appropriately. Also, skill level certification and level-up certification may be performed according to the learning result. The e-learning information is provided, for example, by displaying characters, still images, moving images, etc. on a display device, or by notifying with sound. The e-learning information may provide, for example, a plant operating environment using VR (Virtual Reality) goggles.

The solution provided by the operation support device 1 may be a skill-based manual or the like generated according to the skill level. A manual is a content in which work procedures are documented, and workers can refer to the content when they carry out work.

The execution environment 4 is an environment that uses skill-based solutions. A skill-based solution can be used, for example, implemented in a plant controller. The execution environment 4 can support abnormal operations in response to alarm events and model operations by executing skill-specific solutions in plant control devices. Also, automation information can be used for plant control.

In addition, the execution environment 4 is an environment (waigaya environment) in which, for example, a skilled worker instructs a new worker on the operation details, confirms and guides the new worker on the operation details, or discusses the work method. can also be used. Skills of skilled workers can be handed down to workers in such a Waigaya environment. In this embodiment, for example, by executing the e-learning information in the Waigaya environment, it becomes possible to receive guidance from a skilled worker as well, and the learning effect can be improved. Also, the execution environment 4 may be the VR goggles described above, a computer installed at the production site, or a handheld computer or smartphone carried by the worker.

The execution environment 4 may acquire the skill-specific solution generated in the skill-specific solution database 3 by, for example, requesting the skill-specific solution database 3 from the execution environment 4 .

The execution environment 4 provides the operation support device 1 with feedback information for skill-based solutions. The feedback information is, for example, information that feeds back these information to the operation support device 1 when there is an error or additional information in the instruction content or the like in the operation of the plant. The operation support device 1 provided with the feedback information can correct the skill-based solution. Note that the execution environment 4 may be implemented as part of the operation support device 1 . In the present embodiment, a case where the execution environment 4 is implemented as a function of the operation support device 1 will be exemplified and described later.

By the way, a skilled worker perceives various information and makes a decision in plant operation. For example, a skilled worker can see documents and equipment visually (visual), hear what other people are saying and sounds from equipment (auditory), feel temperature (tactile), and smell odors. (smell) to perceive information. When things are connected to the network by IoT, it becomes possible to collect and format this information that was perceived by workers by computer. In addition, by clarifying which formalized information the worker has made a decision based on, it is possible to convert know-how into formal knowledge. Collecting information perceived by skilled workers by means of IoT makes it possible to formalize work in various fields. The perceptual information that can be used in this embodiment is exemplified below.

<Scheduler information>
The scheduler information is information for managing the worker's schedule. The scheduler information includes, for example, various information such as date and time, location, action schedule, contact information, and order list. By collecting scheduler information by means of IoT, it is possible to collect information on what the worker used as an information source, how he acted, or how he made a plan. The scheduler information may be a constraint condition that restricts actions in the worker's decision making. By formalizing the scheduler information, it is possible to improve the prediction accuracy of the determined intention.

<Order information>
The order information is, for example, information about who the salesperson (worker) contacted by phone or e-mail, what kind of negotiations were conducted, and what kind of order was received. By collecting order information through IoT, it becomes possible to clarify sales processes that were previously thought to be individual skills, and for example, to improve the accuracy of standardized manuals.

<Design/manufacturing information>
For example, in the production of a product such as a mold, a high-performance (high-precision) product can be produced by a skilled worker adding know-how to the information described in the design drawing. The design/manufacturing information includes who the worker had a meeting with and what (customer requests, etc.), what part of the design drawing was confirmed or changed, and what part of the manufactured mold was palpated. It is information such as whether it was judged as such. By collecting design/manufacturing information through IoT, the product design process and manufacturing process of skilled workers can be formalized.

<Quality control information>
Quality control information is information on quality control work, and includes, for example, work procedures in analysis work, contents of referred instructions, analysis results, and the like. Collecting quality control information by means of IoT makes it possible to convert the details of operations in quality control work into explicit knowledge, thereby improving work efficiency or improving the accuracy of manuals.

<On-site information>
Site information is information on work at the production site, such as site location, work date and time, worker flow line (moving route), perceived information (sound, temperature, etc.), and referenced work instructions. It is information such as contents and operation contents. By collecting on-site information by IoT, it is possible to convert the details of operations in on-site work into explicit knowledge, improve work efficiency, or improve the accuracy of manuals.

<Educational information>
Educational information is the information when worker education is conducted for workers at the production site, etc. For example, the skill level of the trainee, events at the site (measurement values, equipment status, elapsed time, etc., production information such as what an on-site worker can recognize), what kind of judgment was made, what kind of work was instructed and how. By collecting training information by IoT, it becomes possible to formalize the training contents in worker training and implement pseudo-OJT (On-the-Job Training).

It should be noted that the collection and effect of information by IoT as described above is an example of this embodiment, and does not limit the field of application of this embodiment.

Next, the software configuration of the operation support device 1 will be described with reference to FIG. FIG. 2 is a block diagram showing an example of the software configuration of the operation support device 1 according to the embodiment.

In FIG. 2 , the operation support device 1 has functions of an information acquisition unit 11 , a solution analysis unit 12 , a solution provision unit 13 and a solution correction unit 14 . The information acquisition unit 11 has functions of an event information acquisition unit 111 , an event information acquisition unit 112 , a sensory information acquisition unit 113 , an operation information acquisition unit 114 and a criterion acquisition unit 115 . The solution analysis unit 12 has functions of an extraction unit 121 and a solution generation unit 122 . The solution providing unit 13 has functions of an abnormal operation information providing unit 131 , a model operation information providing unit 132 , an automation information providing unit 133 and an e-learning information providing unit 134 . Also, the solution correction unit 14 has the functions of a feedback information acquisition unit 141 and a feedback information application unit 142 .

Each function of the operation support device 1 according to the present embodiment will be described as a function module realized by an operation support program (software) that controls the operation support device 1 .

The information acquisition unit 11 acquires information from outside or inside the operation support device 1 . Information collected by the information acquisition unit 11 may include information collected by the above-described IoT.

The event information acquisition unit 111 acquires event information related to events that occur during plant operation. Event information generated during plant operation is information generated when a predetermined condition is satisfied during plant operation, and is generated from a process control device or the like, for example. The event information is, for example, abnormal event information that occurs under predetermined conditions when the plant is abnormal, or normal events that occur under predetermined conditions when the process is normal.

Abnormal events include, for example, alarm events for reporting abnormalities. Abnormal events are non-regular and non-steady events that occur when an abnormality occurs, and may occur infrequently and may be experienced by workers infrequently.

On the other hand, normal events include, for example, batch events that indicate the batch status of processes. A batch event is an event that notifies that a batch has reached a predetermined state. A batch event occurs, for example, when trend data indicating batch behavior reaches a predetermined state. Batch events are, for example, a batch start event that indicates that a process has started, a threshold event that indicates that trend data has reached a predetermined threshold, and a batch end event that indicates that a process has ended.

The trend data may also include trend data that resulted in the golden batch that yielded the best results in product quality and yield in the process. Whether or not the batch result is the golden batch depends on the plant operation, and there may be a time lag with the plant operation. When trend data with golden batch results is detected, it is possible to notify that golden batch results have been obtained by generating a batch event. The operation of the plant is acquired and recorded by the operation information acquisition unit 114, which will be described later. By acquiring the batch event, the event information acquisition unit 111 notifies the solution analysis unit 12, which will be described later, that the result of the golden batch has been acquired, and causes the batch event acquired by the operation information acquisition unit 114 to occur. allows for a correlation with the operation of the plant when

The event information acquisition unit 111 records the acquired event information together with predetermined accompanying information. Accompanying information includes, for example, the date and time of acquisition of event information (or the date and time of occurrence of the event), the place of occurrence of the event (part in the plant), the identification information of the worker at the time of occurrence of the event, and the like. Accompanying information is information used to associate event information with event information, perceptual information, or operational information.

The event information acquisition unit 112 acquires event information related to events detected in the plant. Event information is information related to all events that can be detected in a plant, such as numerical values of measuring instruments installed in the plant, sensor status, equipment status, etc., including event information and perceptual information. good too. The event information acquisition unit 112 may acquire information input to a control device such as a PLC that controls the plant, for example.

The event information acquisition unit 112 records the acquired event information together with predetermined incidental information. Accompanying information includes, for example, the date and time when the event information was obtained (or the date and time when the event occurred), the place where the event occurred (site in the plant, sensor ID, field device ID, etc.), identification information of the worker at the time of the event occurrence, etc. be. Accompanying information is information used to associate event information with event information, perceptual information, or operational information.

The perceptual information acquisition unit 113 acquires perceptual information related to perception recognized by the operator. Perceptual information includes, for example, visual information that can be recognized by the worker's sense of sight, auditory information that can be recognized by the worker's hearing, tactile information that can be recognized by the worker's sense of touch, olfactory information that can be recognized by the worker's sense of smell, and information that can be recognized by the worker's sense of smell. It includes taste information that can be recognized by taste. Perceptual information is acquired to analyze what the workers who operate the plant perceive and act on. Therefore, events that the operator cannot actually recognize are not included.

The visual information is, for example, numerical values of gauges, information contained in the display screen of the display device, printed matter information, visual information such as the state of the product, and information obtained from the actions of other workers. The visual information is, for example, information about what the worker was visually recognizing (gazing at) based on the worker's line of sight detected by a line of sight detection device capable of analyzing the movement of the person's eyeballs.

Auditory information is, for example, information obtained from alarm sounds, abnormal sounds generated from equipment, auscultatory sounds using a stethoscope, received sounds from transceivers or intercoms, and the like. Auditory information is, for example, audio data collected by a microphone attached to the worker.

The tactile information is, for example, information obtained from abnormal vibration, abnormal temperature, and the like when the facility is palpated. The olfactory information is information obtained from odors generated from equipment, odors generated from products, and the like. Taste information is information obtained from the taste of a product such as food.

The sensory information acquisition unit 113 acquires information of an external device connected via a network such as the Internet. An external device is a device outside the operation support device 1 connected to the operation support device 1 via a network. For example, the perceptual information acquisition unit 113 may acquire information on field devices connected by field communication, and may also acquire big data using IoT. The sensory information acquisition unit 113 may select and acquire information to be acquired from big data using IoT. The perceptual information acquisition unit 113 records the acquired perceptual information together with predetermined accompanying information. The accompanying information includes, for example, the acquisition date and time of the sensory information (or the worker's perception date and time), the worker's identification information, the worker's perceived location, the worker's movement route, the worker's skill level, and the like. Accompanying information is information used for associating sensory information with operation information or the like. That is, the perceptual information includes who (identification information of the worker), when (acquisition date and time), and what was perceived. The sensory information acquisition unit 113 records the acquired sensory information in chronological order for each worker, for example.

The perceptual information acquisition unit 113 may acquire only the perceptual information of a specific worker. Workers include workers with various skill levels, such as new workers, mid-career workers, and skilled workers. For example, when recording an action of a skilled worker, the sensory information acquisition unit 113 may acquire the sensory information of the skilled worker to be recorded. Until now, it has been difficult to record sensory information, unlike sensing data from sensors used in plant control. In this embodiment, by utilizing perceptual information among the various information that can be acquired by IoT, it is possible to analyze the behavior of workers, which has been difficult to analyze until now.

Note that the sensory information acquisition unit 113 may acquire the sensory information based on the operator's explicit instruction, for example, when the operator determines the operation content. Further, the perceptual information acquisition unit 113 may acquire the perceptual information according to the worker's selection when the worker confirms the work record at a later date. Further, the perceptual information acquisition unit 113 may acquire perceptual information based on an interview with the worker at a later date.

The operation information acquisition unit 114 acquires operation information related to plant operations performed by workers. Plant operations include all operations performed by workers at the production site, such as operating switches on the control panel, operating valves, feeding raw materials, collecting products, inspecting equipment, cleaning equipment, etc. may contain. The operation information acquisition unit 114 may acquire the operation information and the information on the worker and time in association with each other. That is, the operation information includes who (which worker) operated what and when. The operation information acquisition unit 114 records the acquired sensory information in chronological order for each worker, for example.

Similarly to the perceptual information acquisition unit 113, the operation information acquisition unit 114 may acquire operation information performed by a skilled worker when recording work information of a skilled worker, for example. That is, the operation information acquired by the operation information acquiring unit 114 can be acquired in association with the sensory information acquired by the sensory information acquiring unit 113 .

The operation information acquired by the operation information acquiring unit 114 may include an operation target and an operation method for the operation target. For example, if the operation target is a switch, the operation information may include information such as button press, long press, repeated press, and the like. Further, when the operation target is a valve, the operation information may include information such as fully open, fully closed, rotation direction, rotation angle, rotation speed, or opening degree of the valve. Further, when the object to be operated is a moving item, the operation information may include information such as the item's moving route, moving speed, moving order, and the like. In addition, when there are multiple operation targets, the operation information may include information such as the operation order and the timing of the multiple operations. The operation information acquisition unit 114 acquires an operation performed by the worker, for example, an image captured by a television camera capable of capturing the appearance of the worker, a signal indicating that an operation has been performed from the operation target, and a signal recorded by the operator. obtained from work records, etc. For example, the operation information acquisition unit 114 starts acquisition of the operation information when the operator inputs the start of work, and ends the acquisition when the operator inputs the end of work. The operation information acquisition unit 114 may start acquisition of operation information upon occurrence of an event such as an alarm event, and end acquisition upon cancellation of the event. The operation information acquisition unit 114 records the acquired operation information together with predetermined accompanying information. The accompanying information includes, for example, the date and time when the operation information was obtained (or the date and time when the work was performed), the identification information of the worker, the place where the work was performed, the skill level of the worker, and the like. Accompanying information is information used to associate operation information and perceptual information.

The operation information acquired by the operation information acquisition unit 114 includes the operation information of the plant when the result of the golden batch was obtained.

The criterion acquisition unit 115 acquires the criterion of the worker who operated the plant. The determination criterion acquisition unit 115 acquires the determination criterion of the worker who has operated the plant related to the operation information acquired by the operation information acquisition unit 114 . Since perceptual information is used to determine the details and timing of operations, it is desirable not to include information that does not affect the judgment of operations even if the information is actually perceived by the operator. . Therefore, it may not be possible to determine whether or not the worker actually referred to the perceptual information based only on the incidental information. In addition, only the worker himself/herself knows what kind of thinking (logic) the worker used to determine the execution of the operation content based on the perceptual information. The determination criterion acquisition unit 115 acquires from the operator a determination criterion as to what kind of logic the operation content is determined based on which perceptual information. The criterion acquisition unit 115 may acquire the criterion based on an interview with the worker at a later date. The criterion acquisition unit 115 records the acquired criterion.

For example, the operator may decide to perform one operation based on one perceptual information, and perceive the result as perceptual information to determine the next operation. In addition, the operator may acquire a plurality of perceptual information and determine one operation content based on the acquired plurality of information. For example, in order to acquire which sensory information the worker perceives, the criterion acquisition unit 115 selectably displays the sensory information acquired by the sensory information acquisition unit 113 to allow the worker to select. Acquire the judgment criteria by Further, for example, the determination criterion acquisition unit 115 displays the perceptual information in a selectable manner, further displays the determination conditions for how the determination is made based on the selected perceptual information so as to be input, and acquires the determination criterion. You may

The solution analysis unit 12 has the functions of an extraction unit 121 and a solution generation unit 122, and analyzes the information acquired by the information acquisition unit 11 to generate solutions.

The extraction unit 121 extracts the event information acquired by the event information acquisition unit 111, at least the event information acquired by the event information acquisition unit 112, the sensory information acquired by the sensory information acquisition unit 113, or the operation information acquisition unit 114. Extract the relevance with any one of the operation information acquired in . Relevance extraction is performed based on, for example, the accompanying information described above.

For example, the extraction unit 121 sorts each piece of information in chronological order using date and time information included in incident information of event information and date and time information included in incident information, perception information, or operation information as a search condition. Then, the event information whose date and time are close to each other and at least one of the event information, the perception information, and the operation information are regarded as relevant information, and the relevance is extracted. The relevance of these pieces of information may be extracted as search conditions for extracting, for example, plant name, worker name, skill level, etc., in addition to date and time information.

For example, the extracting unit 121 extracts perceptual information and judgment criteria information from among operations performed by the worker (the worker to be extracted) within a predetermined period, using the operation content as a search condition. Criteria for performing operations can be associated and extracted (analyzed). Further, the extracting unit 121 extracts the judgment criteria and the operation information using the perceptual information as a search condition, for example, so as to associate what kind of operation should be performed under what kind of condition of the perceptual information. can be extracted.

In addition, the extraction unit 121 can extract the determination criteria according to the worker to be extracted (including the skill level of the worker). For example, if the worker to be extracted is a specific skilled worker, the extraction unit 121 extracts the association of judgment criteria for various operations (work) performed by the skilled worker at the production site. can be done. In addition, when there are multiple workers to be extracted, it is possible to extract the average and variation of judgment criteria among the plurality of workers, and to collect majority opinions. Moreover, when the workers to be extracted are a plurality of workers with different skill levels, it is possible to extract the difference in the judgment criteria for each skill level.

The extraction unit 121 can generate a worker's judgment logic based on the extracted relevance. Judgment logic is a formalized knowledge of an operator's thinking, and for example, an output result corresponding to operation information with perceptual information and judgment criteria as input conditions. The decision logic can be formalized by expressing it in a stylized way such as a flow chart, a sequence diagram, a collaboration diagram (UML), or the like. By generating the judgment logic, it is possible to visualize the judgment logic, which is the content of the worker's thoughts, and to make the skill explicit and pass it on.

For example, the extraction unit 121 extracts an alarm occurrence event based on an alarm occurrence event notifying a plant abnormality acquired by the event information acquisition unit 111 and operation information related to the operation during an abnormality acquired by the operation information acquisition unit 111. Extract the relevance to the abnormal operation.

In addition, the extraction unit 121 extracts the batch occurrence event indicating the occurrence of a batch exhibiting a predetermined behavior in the plant acquired by the event information acquisition unit 111 and the batch occurrence event performed by the worker acquired by the operation information acquisition unit. A relationship between the golden batch and the model operation may be extracted based on operation information related to the model operation to be generated.

Further, the extracting unit 121 makes it possible to generate e-learning information according to the skill level of the worker by performing the relevance described above according to the skill level of the worker.

By the way, among the skilled workers, there is a worker called "Takumi". A craftsman is a person who has advanced knowledge and skills in the field in which he or she works, and who should serve as a role model for other workers. Skilled workers who have a certain amount of skill can aim to further improve their skills by receiving the succession of the skills of the craftsman while receiving guidance from the craftsman. The extracting unit 121 extracts the judgment criteria of the craftsman-level worker, and extracts information (for example, solutions by skill) that is useful for improving the skills of other skilled workers or workers whose skill level is lower than that of skilled workers. It becomes possible to obtain information for generating. Similarly, the judgment criteria for experienced workers are useful information for mid-level workers or workers with lower skill levels than mid-level workers, and the judgment standards for mid-level workers are useful information for new workers. can do.

The solution generation unit 122 generates a solution that supports plant operation by associating event information with at least one of event information, perception information, and operation information based on the relationships extracted by the extraction unit 121. .

For example, the solution generation unit 122 generates a solution related to abnormal operation information that associates an alarm occurrence event with an abnormal operation. Further, the solution generation unit 122 may generate a solution related to model manipulation information that associates a golden batch with a model manipulation. Further, the solution generation unit 122 may generate a solution related to automation information in which event information and operation information are associated. In addition, the solution generation unit 122 may generate a solution related to e-learning that associates event information, perception information, and operation information. The solution generator 122 may generate these solutions based on the judgment logic generated by the extractor 121 .

The solution providing unit 13 has the functions of an abnormal operation information providing unit 131, a model operation information providing unit 132, an automation information providing unit 133, and an e-learning information providing unit 134, and provides solutions generated by the solution generating unit 122. offer.

The abnormal operation information providing unit 131 provides a solution related to the abnormal operation information according to an alarm generation event that occurs during plant operation. The abnormal operation information is, as described above, information for supporting the abnormal operation of the plant when an abnormality occurs in the plant. The abnormal operation information provision unit 131 supports appropriate abnormal operation according to the skill level of the operator. For example, the abnormal operation information provision unit 131 acquires the skill level of a worker who is working in the plant from the event information, and selects and provides operation information according to the skill level of the worker. The abnormal operation information providing unit 131 grasps the progress of the abnormal operation based on the operation information acquired by the operation information acquiring unit 114, and instructs the operator to perform an appropriate operation at an appropriate timing. good too. As a result, it is possible to instruct an appropriate and prompt response to the abnormality that is the cause of the alarm generation. The abnormal operation information provision unit 131 provides the abnormal operation information by, for example, displaying characters, still images, moving images, etc. on a display device, or by notifying with sound.

The model operation information providing unit 132 provides solutions related to model operation information so as to obtain golden batch results. The exemplary operation information is, as described above, information for supporting exemplary plant operations (exemplary operations) for generating the golden batch. Like the abnormal operation information providing unit 131, the model operation information providing unit 132 acquires the skill level of the worker from the event information, and selects and provides model operation information according to the skill level of the operator. . Since the golden batch has a predetermined trend over time, the model operation information providing unit 132 provides an appropriate operation method at an appropriate timing. The model operation information providing unit 132 provides the model operation information by, for example, displaying characters, still images, moving images, etc. on a display device, or by notifying with sound. Note that the model operation information providing unit 132 may generate feedback information for the solution correction unit 14 described later by collecting operation results and process trend results performed on the provided model operation information. .

The automation information providing unit 133 provides a solution related to automation information to the control device of the plant. Automation information is information for operating the control apparatus which controls a plant as mentioned above. The automation information providing unit 133 provides the DCS or PLC with, for example, a program executable by the DCS or PLC. For example, when the operation to be performed for a predetermined event is simple enough to be performed in a simple judgment, the automation information providing unit 133 provides the automation information to make the judgment made into explicit knowledge to the control device. It becomes possible to incorporate it as a program.

The e-learning information providing unit 134 provides e-learning information related to e-learning as a solution according to the level of the worker who is to be educated. The e-learning information is e-learning learning content that can be provided to the worker. The educated person is a worker whose skill is to be improved. The e-learning information is content for improving the skill level of workers, such as manual documents, work training and work support programs.

For example, the e-learning information providing unit 134 may provide e-learning information for each skill level of the trainee. For example, the e-learning information providing unit 134 makes it possible to select the worker's skill level, and provides e-learning information according to the selected skill level. The skill level may be a plurality of levels such as, for example, newcomer level, intermediate level level, and skilled level level. The e-learning information may be, for example, a program for operation training. A program for operation training, for example, presents a task simulating an event corresponding to a work site to the trainee and asks the trainee to answer, thereby certifying the level of the trainee and It may improve the knowledge level of the trainee. The operation training program may utilize the motions and sensitivities of intermediate level or skilled level workers collected by IoT. In addition, the operation training program may certify the skill level of the trainee based on the accuracy rate of the trainee, and may provide higher-level tasks. The operation training program may adjust the difficulty of questions according to the skill level of the trainee. The operation training program is a formalization of the knowledge of workers with a high skill level, making it possible to pass on the skills.

Note that the e-learning information providing unit 134 may be, for example, a web server capable of providing a web page to a web browser. By providing e-learning information on a web browser, it can be easily used from a wide range of devices such as PCs and smartphones.

The solution correction unit 14 has the functions of the feedback information acquisition unit 141 and the feedback information application unit 142, and provides information for correcting the content of the solution provided by the solution provision unit 13. FIG.

When content to be corrected is found in the content of the solution generated by the solution generation unit 122, the feedback information acquisition unit 141 acquires the content to be corrected as feedback information. Modification of content includes deletion, addition or change of part or all of the content. The feedback information acquisition unit 141 acquires, for example, correction information corrected by the worker who used the content. Modification of contents is performed, for example, when there is a change in the operation target or a change in the set value of the device due to a change in the process of the plant.

The feedback information application unit 142 applies the feedback information acquired by the feedback information acquisition unit 141 to the contents. The feedback information applying unit 142, for example, provides feedback information to the extracting unit 121 and causes it to regenerate the judgment logic. Also, the feedback information application unit 142 may directly modify the content generated by the solution generation unit 122 based on the feedback information.

The information acquisition unit 11, the solution analysis unit 12, the solution provision unit 13, the solution correction unit 14, the event information acquisition unit 111, the event information acquisition unit 112, the sensory information acquisition unit 113, and the operation information acquisition included in the operation support device 1 Unit 114, criteria acquisition unit 115, extraction unit 121, solution generation unit 122, abnormal operation information provision unit 131, model operation information provision unit 132, automation information provision unit 133, e-learning information provision unit 134, feedback information acquisition unit 141 and the feedback information application unit 142 have been described as being implemented by software, as described above. However, at least one or more of the functions of the operation support device 1 may be realized by hardware.

Further, one of the above functions of the operation support device 1 may be implemented by dividing one function into a plurality of functions. Further, the operation support device 1 may integrate any two or more of the functions described above into one function.

Further, the operation support device 1 may be a device realized by one housing, or may be a system realized by a plurality of devices connected via a network or the like. For example, the operation support device 1 may be a virtual device such as a cloud service provided by a cloud computing system. Further, the operation support device 1 may be a general-purpose computer such as a server device, or may be a dedicated device with limited functions.

At least one or more of the above functions of the operation support device 1 may be realized by another device. That is, the operation support device 1 does not need to have all the functions described above, and may have some of the functions.

Next, the event information acquisition operation will be described with reference to FIG. FIG. 3 is a flowchart showing an example of the event information acquisition operation of the operation support device 1 according to the embodiment. The operations of the operation support device 1 described below are executed by an operation support program executed by the operation support device 1. FIG. Therefore, the subject of the operation will be described as the operation support device 1 .

In FIG. 3, the operation support device 1 determines whether event information has been acquired (step S11). Whether the event information has been acquired can be determined, for example, by whether the event information acquisition unit 111 has acquired the event information. When it is determined that the event information has not been acquired (step S11: NO), the operation support device 1 repeats the operation of step S11 and waits for the acquisition of event information.

On the other hand, when determining that the event information has been acquired (step S11: YES), the operation support device 1 records the acquired event information (step S12). The event information recorded in step S12 can include, for example, incidental information for association with perceptual information or the like. As described above, the accompanying information includes, for example, the date and time when the event information was obtained (or the date and time when the event occurred), the place where the event occurred (part in the plant), the identification information of the worker when the event occurred, and the like.

After executing the process of step S12, the operation support device 1 determines whether or not to end the recording (step S13). Whether or not to end recording can be determined, for example, by whether or not an instruction to end recording has been received from the operator. A worker may sequentially perform operations on a plurality of events. One event is recorded as one piece of event information (step S12), and when performing an operation for continuous events, the recording operation may be continued until an instruction to end recording is given. If it is determined not to end the recording (step S13: NO), the operation support device 1 returns to the process of step S11 and repeats the event information acquisition operation. On the other hand, when it is determined that the recording is finished (step S13: YES), the operation support device 1 finishes the operation shown in the flowchart.

Next, the event information acquisition operation will be described with reference to FIG. FIG. 4 is a flowchart showing an example of an event information acquisition operation of the operation support device 1 according to the embodiment.

In FIG. 4, the operation support device 1 determines whether event information has been acquired (step S21). Whether or not the event information has been acquired can be determined by, for example, whether or not the event information acquisition unit 112 has acquired the event information. If it is determined that event information has not been acquired (step S21: NO), the operation support device 1 repeats the operation of step S21 and waits for event information to be acquired.

On the other hand, when determining that event information has been acquired (step S21: YES), the operation support device 1 records the acquired event information (step S22). The event information recorded in step S22 can include, for example, incidental information for association with perceptual information or the like. As described above, the accompanying information includes, for example, the acquisition date and time of the event information (or the date and time of the occurrence of the event), the place where the event occurred (site in the plant, sensor ID, field device ID, etc.), the worker at the time of the event occurrence identification information and the like.

After executing the process of step S22, the operation support device 1 determines whether or not to end the recording (step S23). Whether or not to end recording can be determined, for example, by whether or not an instruction to end recording has been received from the operator. A worker may sequentially perform operations for a plurality of events. One event is recorded as one piece of event information (step S22), and when performing operations for continuous events, the recording operation may be continued until an instruction to end recording is given. When it is determined not to finish recording (step S23: NO), the operation support device 1 returns to the process of step S21 and repeats the event information acquisition operation. On the other hand, if it is determined that the recording is finished (step S23: YES), the operation support device 1 finishes the operation shown in the flowchart.

Next, the perceptual information acquisition operation will be described with reference to FIG. FIG. 5 is a flowchart showing an example of the perceptual information acquisition operation of the operation support device 1 according to the embodiment.

In FIG. 5, the operation support device 1 determines whether or not sensory information has been acquired (step S31). Whether or not the perceptual information has been acquired can be determined by, for example, whether or not the perceptual information acquisition unit 113 has acquired the perceptual information. When it is determined that the sensory information has not been acquired (step S31: NO), the operation support device 1 repeats the operation of step S31 and waits for the sensory information to be acquired.

On the other hand, when determining that the sensory information has been acquired (step S31: YES), the operation support device 1 records the acquired sensory information (step S32). The perceptual information recorded in step S32 can include, for example, accompanying information for association with event information. As described above, the accompanying information includes, for example, the acquisition date and time of the sensory information (or the worker's perception date and time), the worker's identification information, the worker's perceived location, the worker's movement route, the worker's skill level, etc. be.

After executing the process of step S32, the operation support device 1 determines whether or not to end the recording (step S33). Whether or not to end recording can be determined, for example, by whether or not an instruction to end recording has been received from the operator. A worker may sequentially perform operations on multiple perceptions. When one perception is recorded as one perception information (step S32) and an operation for continuous perception is performed, the recording operation may be continued until an instruction to end recording is given. If it is determined not to finish recording (step S33: NO), the operation support device 1 returns to the process of step S31 and repeats the perceptual information acquisition operation. On the other hand, when it is determined that the recording is finished (step S33: YES), the operation support device 1 finishes the operation shown in the flowchart.

Next, operation information acquisition operation will be described with reference to FIG. FIG. 6 is a flow chart showing an example of operation information acquisition operation of the operation support device 1 in the embodiment.

In FIG. 6, the operation support device 1 determines whether or not operation information has been acquired (step S41). Whether or not the operation information has been acquired can be determined, for example, by whether or not the operation information acquisition unit 114 has acquired the operation information. When it is determined that the operation information has not been acquired (step S41: NO), the operation support device 1 repeats the operation of step S41 and waits for acquisition of the operation information.

On the other hand, when determining that the operation information has been acquired (step S41: YES), the operation support device 1 records the acquired operation information (step S42). The perceptual information recorded in step S42 can include, for example, accompanying information for associating with event information. As described above, the accompanying information includes, for example, the acquisition date and time of the operation information (or the date and time when the work was performed), the identification information of the worker, the place where the work was performed, the skill level of the worker, and the like.

After executing the process of step S42, the operation support device 1 determines whether or not to end the recording (step S43). Whether or not to end recording can be determined, for example, by whether or not an instruction to end recording has been received from the operator. A worker may sequentially perform a plurality of operations. One operation is recorded as one piece of operation information (step S42), and when continuous operations are performed, the recording operation may be continued until an instruction to end recording is given. If it is determined not to finish recording (step S43: NO), the operation support device 1 returns to the process of step S41 and repeats the operation of acquiring operation information. On the other hand, when it is determined that the recording is finished (step S43: YES), the operation support device 1 finishes the operation shown in the flowchart.

Next, the skill-based solution generation operation will be described with reference to FIG. FIG. 7 is a flowchart illustrating an example of a skill-based solution generation operation of the operation support device according to the embodiment.

In FIG. 7, the operation support device 1 acquires the event information recorded in the process of step S12 (step S51). After executing the process of step S51, the operation support device 1 acquires the event information recorded in the process of step S22 (step S52). After executing the process of step S52, the operation support device 1 acquires the sensory information recorded in the process of step S32 (step S53). After executing the process of step S53, the operation support device 1 acquires the operation information recorded in the process of step S42 (step S54). After executing the process of step S54, the operation support device 1 acquires a criterion (step S55). The criteria are acquired and recorded in the criteria acquiring unit 115 .

After executing the process of step S55, the operation support device 1 sorts the acquired event information, event information, perception information, and operation information in chronological order based on the time information of the incidental information, and sorts the time of each information. clear the appropriate order (step S56). For example, in a large-scale plant, a huge amount of event information and event information are acquired. By sorting and comparing the information in chronological order in the process of step S56, it is possible to organize a huge amount of information and facilitate extraction of the relationship between the information.

After executing the process of step S56, the operation support device 1 extracts the relevance (step S57). Details of the processing in step S57 will now be described with reference to FIG. FIG. 8 is a flowchart illustrating an example of a relationship extraction operation of the operation support device according to the embodiment;

In FIG. 8, the operation support device 1 extracts event information (step S571). Extraction of the event information can be executed by, for example, extracting the event information to be searched by the extraction unit 121 . For example, the extraction unit 121 extracts event information by searching recorded event information from identification information that identifies an alarm event that occurred in the past in a predetermined process and a skilled worker who performed an operation in response to the alarm event. do.

After executing the process of step S571, the operation support device 1 extracts event information corresponding to the event extracted in the process of step S571 (step S572). The event information is sorted in chronological order in the process of step S56, for example. For example, the operation support device 1 extracts events having close occurrence times from the recorded events. Extraction of events may be performed by using, for example, the plant or process in which the alarm event occurred as a search condition. Further, the operation support device 1 may search business data, quality data, and environmental data related to the plant in which the alarm event occurred as search conditions.

After executing the process of step S572, the operation support device 1 determines whether or not the extracted number of events n is one (n=1) (step S573). The number of events varies depending on the search conditions when extracted in the process of step S752. If the number of events n is one (step S573: YES), the operation support device 1 determines whether automation is possible (step S574). Whether or not automation is possible is when the relationship between the occurrence of an event and the phenomenon is clear, and the event can be considered as the condition for the occurrence of the event. Whether or not automation is possible can be executed, for example, in the determination program of the operation support device 1 . Further, whether or not automation is possible may be determined by inputting a judgment by the operation support device 1 and by inputting a judgment by a skilled worker or the like.

If it is determined that automation is possible (step S574: YES), the operation support device 1 extracts the relationship between the extracted event information and the event information (step S575), and performs relationship extraction processing (step S57). end the processing of

On the other hand, if it is determined that automation is not possible (step S574: NO), or if the number of events n is not 1 (step S573: NO), the operation support device 1 generates perceptual information corresponding to each event and Operation information is extracted (step S576). For example, when the number of events is n, the operation support device 1 extracts perceptual information and operation information corresponding to event (i) (where i is a natural number from 1 to n). In this embodiment, it is assumed that operation methods are extracted for each skill. For example, even with the same perceptual information for the same event, workers with different skill levels may perform different operations. For example, when the pressure measured by the pressure gauge installed in the reaction tank falls within a predetermined range, a new worker tightens the raw material input valve to be injected into the reaction tank, whereas an experienced worker The operator may increase the speed of the pump that discharges the product from the reaction tank. By extracting operation methods for each skill, it is possible to extract differences in operation according to skill levels.

After executing the process of step S576, the operation support device 1 extracts skill-specific logic (step S577). The skill-specific logic can be executed by, for example, extracting criteria corresponding to events, perceptual information, or operation information based on the criteria acquired by the criteria acquisition unit 115, and associating the respective information. . Information to be associated is not limited. For example, any two or more pieces of information among event information, phenomenon information, perception information, and operation information can be associated with each other. Information may be associated with different types of information such as event information and operation information, or may be associated with the same type of information such as event information with event information. good. The skill-specific logic may include one of these associations (relationships) or may be a combination of a plurality of associations. By extracting skill-specific logic, it is possible to specify skill-specific operations based on event information, event information, and perception information.

After executing the process of step S577, the operation support device 1 determines whether i=n (step S578). If it is determined that i is not equal to n (step S578: NO), i is incremented by 1 (step S579), and the processing of steps S576 to S578 is repeated. On the other hand, if it is determined that i=n (step S578: YES), the operation support device 1 ends the relationship extraction process (step S57). That is, in steps S576 to S579, a process of extracting skill-specific logic in each event (i) is executed.

Returning to the description of FIG. 7 again. After executing the processing of step S57, the operation support device 1 generates skill-specific logic (step S58). The skill-specific logic can be generated by integrating all events of the skill-specific logic extracted in the process of step S577. For example, skill-specific logic may be generated by selecting only highly relevant logic from among the extracted logics. The degree of relevance may be calculated, for example, by the frequency or number of cases of association. Also, the skill-specific logic may be generated including all the extracted logics.

After executing the process of step S58, the operation support device 1 generates a skill-based solution (step S59). Skill-based solutions are, for example, abnormal operation information, model operation information, automation information, or e-learning information. Which solution to generate may be specified by selection, for example.

After executing the process of step S59, the operation support device 1 records the solution by skill generated in the process of step S59 (step S60), and ends the operation shown in the flowchart. The recorded solution is read out from the solution providing unit 13 and provided.

Next, the solution providing operation will be described with reference to FIG. FIG. 9 is a flowchart illustrating an example of a level-based solution providing operation of the operation support device according to the embodiment.

In FIG. 9, the operation support device 1 acquires the skill level (step S61). For example, the skill level may be acquired by displaying the skill level in a selectable manner by the solution providing unit 13 and selecting it by the worker who receives the provision of the solution.

After executing the process of step S61, the operation support device 1 branches the process according to the selected skill level (step S62). The number of skill levels that can be branched in step 62 may be, for example, a predetermined number of levels from newcomer to master level. FIG. 9 shows the case where there are n levels of intermediate levels from LV1 to LVn.

If the selected skill level is the "newcomer level" (step S62: "newcomer level"), the operation support device 1 provides content corresponding to the newcomer level (step S63). If the selected skill level is "LV (level) 1 intermediate level" (step S62: "LV1 intermediate level"), the operation support device 1 provides content corresponding to LV1 intermediate level (step S64). . If the selected skill level is "LV2 middle level" (step S62: "LV2 middle level"), the operation support device 1 provides content corresponding to LV2 middle level (step S65). If the selected skill level is "LVn intermediate level" (step S62: "LVn intermediate level"), the operation support device 1 provides content corresponding to the LVn intermediate level (not shown).

Contents for each skill level are selected according to event information relating to events that occur during plant operation. For example, based on the event information corresponding to the event information, the operator may instruct the confirmation item (event) to be confirmed, and instruct the operation item (operation content) according to the contents of the confirmation item. Also, the contents for each skill level may indicate confirmation items (events) based on event information and operation information corresponding to the event information. Also, the content for each skill level may indicate an operation item (operation details) based on event information corresponding to the event information and perceptual information. These contents can be generated by associating the acquired event information, event information, perception information, and operation information, as described with reference to FIG. 7, for example.

Next, a feedback operation for work assistance will be described with reference to FIG. FIG. 10 is a flowchart showing an example of a feedback operation for work assistance of the operation assistance device 1 according to the embodiment.

In FIG. 10, the operation support device 1 determines whether feedback information has been acquired (step S71). Whether or not the feedback information has been acquired can be determined, for example, by whether or not the feedback information acquisition unit 141 has received information input from the worker. When determining that the feedback information has not been acquired (step S71: NO), the operation support device 1 repeats the process of step S71 and waits for the acquisition of the feedback information.

When determining that the feedback information has been acquired (step S71: YES), the operation support device 1 determines whether or not the content of the proposal was appropriate (step S72). Whether or not the content of the proposal is appropriate can be determined, for example, by determining whether or not the feedback information input by the worker to the feedback information acquisition unit 141 includes information requiring correction. If it is determined that the content of the proposal is appropriate (step S72: YES), the operation support device 1 ends the operation support feedback operation shown in the flowchart. That is, no modification of skill-specific solutions is performed. Incidentally, when the content of the proposal is appropriate, that fact may be recorded and used for generating the content of the proposal in the judgment logic.

On the other hand, if it is determined that the content of the proposal is not appropriate (step S72: NO), the operation support device 1 corrects the content of the proposed operation (step S73). The modification of the operation content may be modification of the skill-based logic extraction in step S577, or may be modification of only the solution content such as the work support program to be generated. After executing the process of step S73, the operation support device 1 regenerates the solution by skill (step S74), and ends the operation support feedback operation shown in the flowchart.

Note that machine learning may be used to correct the operation content. For example, by inputting the determination result in step S72 to the support vector machine, it is possible to learn appropriate proposal contents. In this embodiment, since it is possible to acquire big data other than the perceptual information used by the operator to determine the details of the operation, machine learning provides significant information to information that even skilled workers have not noticed. may be discovered.

Next, the hardware configuration of the operation support device 1 will be described with reference to FIG. 11 . FIG. 11 is a block diagram showing an example of the hardware configuration of the operation support device 1 according to the embodiment.

11, the operation support device 1 has a CPU 101, a RAM (Random Access Memory) 102, a ROM (Read Only Memory) 103, an HDD 104, a touch panel 105, a communication I/F (Interface) 106, and a communication I/F 107.

The operation support device 1 can be realized by, for example, a general-purpose device such as a notebook PC, a tablet PC, a PDA, or a smartphone, or a device dedicated to operation support. The operation support device 1 is a device that executes the operation support program described with reference to FIG.

The CPU 101 controls the operation assistance device 1 by executing an operation assistance program stored in the RAM 102, ROM 103, or HDD 104. FIG. The operation support program is, for example, acquired from a recording medium recording the operation support program or from a server that provides the operation support program via a network, installed in HDD 104 , and stored in RAM 102 so as to be readable by CPU 101 .

The touch panel 105 has an operation display function having an operation input function and a display function. A touch panel 105 displays a UI (User Interface) of a program executed by the operation support device 1 . The touch panel 105 enables the operator to perform operation input using a fingertip, a touch pen, or the like. The operation support device 1 in this embodiment uses the touch panel 105 having an operation display function, but the operation support device 1 has a display device having a display function and an operation input device having an operation input function. There may be. In that case, the display screen of the touch panel 105 can be implemented as the display screen of the display device, and the operation of the touch panel 105 can be implemented as the operation of the operation input device. Note that the touch panel 105 may be implemented in various forms such as a head-mounted type, glasses type, wristwatch type display, and the like.

The communication I/F 106 controls communications via the network 9 with other devices that use general-purpose communications such as wireless LAN communications, wired LAN communications, infrared communications, and short-range wireless communications. The other device is, for example, the cloud server 91, the device in which the external application 92 is arranged, or the like. The cloud server 91 is a server that provides cloud services through cloud computing. The cloud server 91 may primarily store big data, for example. The cloud server 91 can provide information common to the operation support device 1 to other operation support devices (not shown). The external application 92 is an application that operates outside the operation support device 1, and may be a part of the functions of the operation support device described with reference to FIG. 2, for example. The communication I/F 106 includes other operation support devices, field devices capable of general-purpose communication, a maintenance information management server that manages maintenance information, a DCS (Distributed Control System) control device, and a PLC (not shown). Programmable Logic Controller) or the like. FIG. 12 shows the case where the external application 92 is arranged in a device connected via a network. May be present in content.

Communication I/F 107 controls field communication with field device 2 using a communication protocol that can be used in the field device. Various field communication protocols are used in various field devices used in plants. The communication I/F 107 controls communication with field devices 2 that use field communication such as ISA100, HART (registered trademark), BRAIN (registered trademark), FOUNDATION Fieldbus, and PROFIBUS. The operation support device 1 may have a plurality of communication I/Fs 107 according to the type of field communication.

As described above, the operation support device according to the present embodiment includes an event information acquisition unit that acquires event information related to an event that occurs during operation of a plant, and an event information acquisition unit that acquires event information related to an event detected in the plant. An event information acquisition unit, a perception information acquisition unit that acquires perception information related to perception recognized by a worker, an operation information acquisition unit that acquires operation information related to the operation of the plant performed by the worker, and the event an extraction unit for extracting a relationship between information and at least one of the event information, the perceptual information, and the operation information; and generating a solution for supporting the operation of the plant based on the extracted relationship. and a solution generation unit. As a result, it is possible to facilitate transfer of skills related to work corresponding to events that occur during plant operation.

The operation support device described above may be a device having the functions described above. good. Further, the operation support device may be implemented as part of the functions of another device connected via a network.

Further, the operation support method in the present embodiment includes an event information acquisition step of acquiring event information related to an event that occurs during operation of a plant, and an event information acquisition step of acquiring event information related to an event detected in the plant. And, a perceptual information acquisition step of acquiring perceptual information related to the perception recognized by the worker, an operation information acquisition step of acquiring operation information related to the operation of the plant performed by the worker, the event information, and at least an extraction step of extracting a relationship with any one of the event information, the perceptual information, or the operation information; and a solution generation step of generating a solution for supporting the operation of the plant based on the extracted relationship. including. As a result, it is possible to facilitate transfer of skills related to work corresponding to events that occur during plant operation.

Note that the operation support method described above may be any method that includes the steps described above, and these steps may be performed in any order. That is, the execution timing of each step described above is arbitrary, and for example, any one of the steps described above may be executed several times before executing another step.

Also, a program for realizing the functions constituting the device described in the present embodiment may be recorded in a computer-readable recording medium, and the program recorded in the recording medium may be read and executed by a computer system. , the above-described various processes of the present embodiment may be performed. Note that the “computer system” referred to here may include hardware such as an OS and peripheral devices. The "computer system" also includes the home page providing environment (or display environment) if the WWW system is used. In addition, "computer-readable recording medium" means writable non-volatile memory such as flexible disk, magneto-optical disk, ROM, flash memory, etc., portable medium such as CD-ROM, hard disk built in computer system, etc. storage device.

Furthermore, "computer-readable recording medium" means a volatile memory (e.g., DRAM (Dynamic Random Access Memory)), which holds a program for a certain period of time. Further, the above program may be transmitted from a computer system storing this program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in a transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. Further, the program may be for realizing part of the functions described above. Furthermore, it may be a so-called difference file (difference program) that implements the above-described functions in combination with a program already recorded in the computer system.

Although the embodiments of the present invention have been described above with reference to the drawings, the specific configuration is not limited to these embodiments, and various modifications are possible without departing from the scope of the present invention. be

1 operation support device 11 information acquisition unit 100 operation support system 111 event information acquisition unit 112 event information acquisition unit 113 perceptual information acquisition unit 114 operation information acquisition unit 115 criterion acquisition unit 12 solution analysis unit 121 extraction unit 122 solution generation unit 13 solution Provision unit 131 Operation information provision unit at abnormal time 132 Exemplary operation information provision unit 133 Automation information provision unit 134 e-learning information provision unit 14 Solution correction unit 141 Feedback information acquisition unit 142 Feedback information application unit 2A Sales database 2B Quality database 2C On-site database 3 Solution database by skill 4 Execution environment 101 CPU
102 RAMs
103 ROMs
104 HDDs
105 touch panel 106 communication I/F
107 Communication I/F
5 field device 9 network 91 cloud server 92 external application

Claims (11)

an event information acquisition unit that acquires event information related to an event that occurs when a predetermined condition is satisfied during plant operation;
an event information acquisition unit that acquires event information related to an event detected in the plant;
A perceptual information acquisition unit that acquires perceptual information that is information related to perception recognized by a worker operating the plant and that is used for analyzing what the worker perceives and acts on;
An operation information acquisition unit that acquires operation information related to the operation of the plant performed by the worker;
a criterion acquisition unit that acquires the criterion of the worker who performed the operation;
an extraction unit that extracts a relationship between the event information and at least one of the event information, the perceptual information, and the operation information;
a solution generation unit that generates a solution that supports the operation of the plant based on the extracted relationships,
The determination criterion obtaining unit selectably displays the perceptual information obtained by the perceptual information obtaining unit, and allows the operator to make a selection, thereby determining which perceptual information the content of the operation was determined based on. Acquiring the criteria,
The extraction unit compares the event information with at least one of the event information, the perceptual information, or the operation information in chronological order, and extracts the relevance based on the acquired judgment criteria. do,
Operation support device. The event information acquisition unit acquires an alarm occurrence event that reports an abnormality in the plant,
2. The operation support device according to claim 1, wherein said operation information acquisition unit acquires said operation information related to an operation performed by said worker at an abnormal time. The event information acquisition unit acquires a batch occurrence event indicating the occurrence of a batch exhibiting a predetermined behavior in the plant,
The operation support device according to claim 1 or 2, wherein the operation information acquisition unit acquires operation information related to a model operation performed by a worker for generating the batch generation event. The extraction unit extracts the relevance according to the skill level of the worker,
The operation support device according to any one of claims 1 to 3, wherein said solution generation unit generates said solution according to said skill level. The operation support device according to any one of claims 1 to 4, further comprising a solution providing unit that provides the generated solution according to occurrence of the event during operation of the plant. The operation support device according to any one of claims 1 to 5, wherein the perceptual information acquisition unit acquires information of an external device connected via a network. The operation support device according to any one of claims 1 to 6, wherein the perceptual information acquisition unit acquires information on at least one of vibrations, sounds, and alarms emitted by equipment. The operation support device according to any one of claims 1 to 7, wherein the sensory information acquisition unit acquires visual information visually recognized by the worker from a line-of-sight detection device that detects the line of sight of the worker. an event information acquisition step of acquiring event information relating to an event that occurs when a predetermined condition is satisfied during plant operation;
an event information acquisition step of acquiring event information related to an event detected in the plant;
A perceptual information acquisition step of acquiring perceptual information related to perception recognized by a worker operating the plant, which is used for analyzing what the worker perceives and acts on;
an operation information acquisition step of acquiring operation information related to the operation of the plant performed by the worker;
a determination criterion acquisition step of acquiring a determination criterion of the worker who performed the operation;
an extracting step of extracting a relationship between the event information and at least one of the event information, the perceptual information, and the operation information;
a solution generation step of generating a solution that supports the operation of the plant based on the extracted relationships;
The determination criterion obtaining step selectably displays the perceptual information obtained in the perceptual information obtaining step and allows the worker to select the perceptual information, thereby determining the content of the operation based on which perceptual information. Acquiring the criteria,
The extracting step includes comparing the event information with at least one of the event information, the perceptual information, or the operation information in chronological order , and extracting the relevance based on the acquired criterion. do,
Operation support method. event information acquisition processing for acquiring event information relating to an event that occurs when a predetermined condition is satisfied during plant operation;
an event information acquisition process for acquiring event information related to an event detected in the plant;
Perceptual information acquisition processing for acquiring perceptual information used for analyzing what the worker perceives and acts, which is information related to perception recognized by the worker operating the plant;
An operation information acquisition process for acquiring operation information related to the operation of the plant performed by the worker;
a criterion acquisition process for acquiring the criterion of the worker who performed the operation;
an extraction process for extracting a relationship between the event information and at least one of the event information, the perceptual information, and the operation information;
causing a computer to execute a solution generation process for generating a solution for supporting the operation of the plant based on the extracted relationships;
The determination criterion obtaining process displays the perceptual information obtained in the perceptual information obtaining process in a selectable manner and allows the operator to make a selection, thereby determining which perceptual information the content of the operation was determined based on. Acquiring the criteria,
The extraction process compares the event information with at least one of the event information, the perceptual information, or the operation information in chronological order, and extracts the relevance based on the acquired judgment criteria. do,
Operation support program. event information acquisition processing for acquiring event information relating to an event that occurs when a predetermined condition is satisfied during plant operation;
an event information acquisition process for acquiring event information related to an event detected in the plant;
Perceptual information acquisition processing for acquiring perceptual information used for analyzing what the worker perceives and acts, which is information related to perception recognized by the worker operating the plant;
An operation information acquisition process for acquiring operation information related to the operation of the plant performed by the worker;
a criterion acquisition process for acquiring the criterion of the worker who performed the operation;
an extraction process for extracting a relationship between the event information and at least one of the event information, the perceptual information, and the operation information;
causing a computer to execute a solution generation process for generating a solution for supporting the operation of the plant based on the extracted relationships;
The determination criterion obtaining process displays the perceptual information obtained in the perceptual information obtaining process in a selectable manner and allows the operator to make a selection, thereby determining which perceptual information the content of the operation was determined based on. Acquiring the criteria,
The extraction process compares the event information with at least one of the event information, the perceptual information, or the operation information in chronological order, and extracts the relevance based on the acquired judgment criteria. do,
A recording medium in which an operation support program is recorded.

Images