Plant design support device, inference device, machine learning device, plant design support method, inference method, and machine learning method

Abstract

To provide a plant design support device that enables the simplification and increased efficiency of plant design work. [Solution] The plant design support device 4 comprises a data acquisition unit 400 that acquires input data, and a data generation unit 401 that generates output data for the input data acquired by the data acquisition unit 400 by inputting the input data into a learning model 14. The learning model 14 is a trained model that has learned the correlation between input data and output data by machine learning. The input data includes information 11A regarding the design requirements required for a newly constructed plant, and the output data includes plant design information 12 that can reproduce the three-dimensional shape of each component constituting the plant, and in which the processes realized by each component are recorded in relation to the three-dimensional shape.

Description

Technical Field

[0001] The present invention relates to a plant design support device, an inference device, a machine learning device, a plant design support method, an inference method, and a machine learning method.

Background Art

[0002] When a new plant is designed by a designer, specifications, arrangements, connection relationships, etc. of each component constituting the plant are determined so as to satisfy the design requirements required for the plant. As a result, for example, as described in Patent Document 1, various work products such as an equipment list, an instrument list, an arrangement drawing, a specification document, and a requirement document are created.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Since the design requirements required for a plant are defined as various requirements, designing a plant so as to satisfy all of them greatly depends on the knowledge and experience of a skilled designer and is a very difficult task. In addition, as work products of the design work, various work products are created, but since information is dispersed and recorded in the various work products, it is a very difficult task to search for necessary information.

[0005] The present invention has been made in view of the above problems, and an object thereof is to provide a plant design support device, an inference device, a machine learning device, a plant design support method, an inference method, and a machine learning method that can facilitate and streamline the plant design work.

Means for Solving the Problems

[0006] To achieve the above objective, a plant design support device according to one aspect of the present invention is: A data acquisition unit that acquires input data, The system includes a data generation unit that inputs the input data acquired by the data acquisition unit into a learning model and generates output data for said input data, The aforementioned learning model, This is a trained model that has learned the correlation between the input data and the output data using machine learning. The aforementioned input data is This includes information on the design requirements for newly constructed plants. The output data mentioned above is: Information capable of reproducing the three-dimensional shape of each component constituting the plant, including plant design information in which the processes realized by each component are recorded in relation to the three-dimensional shape. [Effects of the Invention]

[0007] According to one aspect of the present invention, a plant design support device is provided, which inputs input data containing information on the design requirements required for a newly constructed plant into a learning model, thereby providing output data that includes plant design information capable of reproducing the three-dimensional shape of each component constituting the plant, and in which the processes realized by each component are recorded in relation to the three-dimensional shape. This generates [something]. Therefore, it is possible to simplify and streamline the plant design work.

[0008] Other issues, configurations, and effects will be clarified in the embodiments for carrying out the invention described later. [Brief explanation of the drawing]

[0009] [Figure 1] This is an overall diagram showing an example of plant design support system 1 and plant 10. [Figure 2A] This is a data structure diagram showing an example of a plant management database 20. [Figure 2B]It is a data configuration diagram showing an example of the plant management database 20. [Figure 2C] It is a data configuration diagram showing an example of the plant management database 20. [Figure 3] It is a block diagram showing an example of the machine learning device 3. [Figure 4] It is an explanatory diagram showing an example of the learning data 13 and the learning model 14. [Figure 5] It is a block diagram showing an example of the plant design support device 4. [Figure 6] It is a functional explanatory diagram showing an example of the plant design support device 4. [Figure 7] It is a hardware configuration diagram showing an example of the computer 900. [Figure 8] It is a flowchart showing an example of the machine learning method by the machine learning device 3. [Figure 9] It is a flowchart showing an example of the plant design support method by the plant design support device 4. [Figure 10] It is a screen configuration diagram showing an example of the plant design support input screen 15. [Figure 11] It is a screen configuration diagram showing an example of the plant design support output screen 16.

Embodiments for Carrying Out the Invention

[0010] Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. In the following, the scope necessary for the description for achieving the object of the present invention is schematically shown, and mainly the scope necessary for the description of the relevant part of the present invention will be described, and the parts where the description is omitted shall be based on known techniques.

[0011] (Configuration of the Plant Design Support System 1) FIG. 1 is an overall view showing an example of a plant design support system 1 and a plant 10. The plant design support system 1 functions as a system for supporting the design of a newly constructed plant 10 and generating a design plan. The plant 10 is, for example, any plant such as a natural gas plant, an oil refining plant, a chemical treatment plant, a power generation plant, an ironmaking plant, etc., and is not limited to these examples.

[0012] The plant 10 is composed of a plurality of components, and each component performs a predetermined process. In the plant 10, as a plurality of components, for example, various devices 100 for treating any fluid such as gas, liquid, and fluidized particulate matter, pipes 101 that connect between the devices 100 to form a fluid flow path, various instruments (not shown) composed of a flow rate sensor, a pressure sensor, a temperature sensor, etc., and various controllers (not shown) composed of a valve, a pump, a compressor, a controller, etc. are installed. The devices 100 include, for example, tower tanks and heat exchangers for performing reactions, distillations, extractions, absorptions, washings, flow rate adjustments, temperature adjustments, etc.

[0013] The main components of the plant design support system 1 include a plant management device 2 for managing various information of the plant 10, a machine learning device 3 for performing machine learning of the learning model 14, a plant design support device 4 for supporting the design of the plant 10 using the learning model 14, an administrator's terminal device 5A used by the administrator of the plant design support system 1, and a designer's terminal device 5B used by the designer of the plant 10. Each of the devices 2 to 5B is composed of, for example, a general-purpose or dedicated computer (see FIG. 7 described later), and is connected to a wired or wireless network 6 and configured to be able to transmit and receive various data to and from each other. Note that the number of each of the devices 2 to 5B and the connection configuration of the network 6 are not limited to the example of FIG. 1 and may be appropriately changed.

[0014] The plant management device 2 is a device that manages various information about the plant 10 as a database, and is equipped with a plant management database 20. The plant management database 20 stores information 11 (details described later) regarding various requirements for the plant 10, as well as plant design information 12 (details described later) designed based on the information 11 regarding various requirements. The plant management database 20 registers and stores information 11 and 12 regarding multiple plants 10 that are to be managed, and the details of the data structure will be described later.

[0015] The machine learning device 3 is a device that operates as the main component of the machine learning learning phase. For example, the machine learning device 3 acquires training data 13 (details described later) from the plant management database 20 and generates a learning model 14 (details described later) used by the plant design support device 4 based on the training data 13 using an arbitrary learning algorithm. In this embodiment, the machine learning device 3 generates the learning model 14 in cooperation with the administrator terminal device 5A. The trained learning model 14 is provided to the plant design support device 4 via the network 6, recording media, etc.

[0016] The plant design support device 4 is a device that operates as the main component of the machine learning inference phase. The plant design support device 4 uses the learning model 14 generated by the machine learning device 3 to generate plant design information 12 used for the design and construction of the plant 10. In this embodiment, the plant design support device 4 works in conjunction with a terminal device 5B for designers to generate the plant design information 12. The plant design information 12 is provided to the terminal device 5B, for example, and presented to the designer. The plant design information 12 is also provided to the plant management device 2 and registered in the plant management database 20.

[0017] Terminal devices 5A and 5B are composed of, for example, stationary computers and portable computers, and are used by administrators and designers, respectively. Terminal devices 5A and 5B have programs such as applications and browsers installed, accept various input operations, and output various information via display screens and audio. By sending and receiving various data with the plant management device 2 and the plant design support device 4, terminal devices 5A and 5B support the design and construction of the plant 10 by, for example, displaying the contents of the plant management database 20 on the display screen, accepting various input operations on the display screen to register new data in the plant management database 20, or modifying registered data.

[0018] (Plant Management Database 20) Figures 2A to 2C are data configuration diagrams showing an example of a plant management database 20. The plant management database 20 is a database for storing information 11 regarding various requirements required for each plant 10 (in the example in Figure 2, plants A, B, ..., N) and plant design information 12 for each plant 10.

[0019] The plant management database 20 stores various types of requirement information 11, including information 11A regarding the design requirements of the plant 10, information 11B regarding the design philosophy, information 11C regarding external conditions, information 11D regarding environmental requirements, and information 11E regarding cost requirements, as shown in Figure 2A.

[0020] Information regarding design requirements 11A includes, as shown in Figure 2B, for example, the raw materials, products, ancillary equipment, process flow of plant 10, and standards applicable to plant 10. This includes information regarding standards, etc. Design requirements for raw materials include, for example, the composition / composition group of raw materials, the minimum / maximum raw material temperature, the minimum / maximum raw material pressure, and the minimum / maximum supply quantity. Large values, etc., are set. The composition of raw materials is set, for example, the name of the raw material, the composition ratio and component properties of the raw materials, although the composition ratio and component properties may be set as a range. A group of raw material compositions is set when multiple raw materials are selectively supplied, and the composition of each raw material is set. The minimum / maximum values ​​for each parameter may be set as a range by setting both, or average values ​​or median values ​​may be set instead of or in addition to the minimum / maximum values, and average values ​​or median values ​​may be set as a range. Design requirements for the product are set, for example, the composition of the product, the minimum / maximum values ​​of the product temperature, the minimum / maximum values ​​of the product pressure, the minimum / maximum values ​​of the production volume, etc. The composition of the product is set, for example, the name of the product, the product Composition ratios and component properties are set, but these may also be set as ranges. Product composition groups are set when multiple products are selectively produced, and the composition of each product is set. Minimum / maximum values ​​for each parameter may be set as a range by setting both, or average or intermediate values ​​may be set instead of or in addition to minimum / maximum values, and average or intermediate values ​​may be set as ranges. Design requirements for ancillary equipment include, for example, design requirements for fuel systems, electrical systems, water systems, air systems, nitrogen systems, flare systems, etc. Design requirements for process flow include, for example, process flow diagrams (PFDs), utility flow diagrams (UFDs), material selection diagrams (MSDs), piping and instrumentation diagrams (P&IDs), cause and effect (CEs), etc. Design requirements for standards or criteria are determined by laws, industry, company, etc., and for example, design requirements for heating equipment, piping, fire protection equipment, hazardous materials, high-pressure gases, etc.

[0021] Information regarding the design concept 11B includes, as shown in Figure 2B, information regarding the design life, design margin, safety factor, operating rate, and turndown requirements of the plant 10. The design life is set, for example, to 30 years. The operating rate is set, for example, to 95%. The turndown requirement is the lower limit at which the plant 10 can be operated stably even with reduced production, and is set, for example, to 60%.

[0022] External conditions information 11C includes, as shown in Figure 2B, information such as the site conditions, meteorological conditions, geological conditions of the construction site of plant 10, and marine conditions of the sea area surrounding plant 10. Site conditions include, for example, latitude, longitude, and altitude. Meteorological conditions include, for example, maximum / minimum / standard temperature, maximum / minimum / standard humidity, maximum / minimum atmospheric pressure, wind direction / wind speed, rainfall, and snowfall. Geological conditions include, for example, seismic resistance requirements, ground characteristics, and foundation type. Marine conditions of the surrounding sea area include, for example, seawater temperature, tides, currents, and waves.

[0023] Environmental requirements information 11D includes, as shown in Figure 2C, information on pollution control requirements, noise regulations, and explosion-proof requirements for plant 10. Pollution control requirements include, for example, the emission limits for gaseous emissions (NOx, CO, SO2, PM, VOCs, etc.), liquid emissions (sanitation, process, oily process, pollution, acid, corrosive agents, solvents, chemicals, etc.), and solid waste emissions (garbage, catalysts, lubricants, solvents, filters, medical waste, etc.). Noise regulations include, for example, the noise level in the workplace, the noise level at which workers are affected, the noise level at the fence boundary, and the maximum noise level in emergencies. Explosion-proof requirements include, for example, explosion resistance performance and blast load standards.

[0024] Cost requirements information 11E includes, as shown in Figure 2C, information such as the economic useful life of plant 10, the discount rate, capital investment, and operating costs. The economic useful life is set, for example, as the investment recovery period, such as 20 years. The discount rate is set, for example, as 10%. Operating costs include, for example, raw materials, fuel, steam, and water. Costs for chemicals, electricity, carbon tax, and product prices are set.

[0025] The plant design information 12 is information capable of reproducing the three-dimensional shape of each component (equipment 100, piping 101, instruments, controllers, etc.) that constitutes the plant 10, and is data in which the processes realized by each component are recorded in relation to the three-dimensional shape. Any data format can be used for the plant design information 12, but for example, data formats such as 3D CAD, BIM, and CIM may be used, or multiple data formats may be combined.

[0026] The plant design information 12 is information that, for example, when displayed on the display screens of terminal devices 5A and 5B, can display the three-dimensional shape of each component from any viewpoint and field of view. Therefore, it is possible to reproduce in three dimensions the dimensions and arrangement of each component of the plant 10 that will actually be constructed. Furthermore, regarding the processes realized in the plant 10, the plant design information 12 associates, for example, the behavior of fluids (composition, flow direction, flow rate, pressure, temperature, etc.) with each component, and it is possible to perform process simulations by simulating actual operations in a virtual space. In addition, if design requirements related to the process flow include process flow diagrams, piping and instrumentation diagrams, etc., the plant design information 12 is information that incorporates the process flow diagrams, piping and instrumentation diagrams, etc., and the physical location of each component is associated with the process flow diagrams, piping and instrumentation diagrams, etc.

[0027] Each component of the plant 10 is assigned identification information (numbers, letters, or combinations thereof), such as an identification number, identification code, identification name, or identification tag, and this information is used in the plant design information 12 to associate various types of information.

[0028] (Machine learning device 3) Figure 3 is a block diagram showing an example of a machine learning device 3. The machine learning device 3 comprises a control unit 30, a communication unit 31, and a storage unit 32.

[0029] The control unit 30 functions as a training data acquisition unit 300 and a machine learning unit 301 by executing a machine learning program 320 stored in the memory unit 32. The communication unit 31 is connected to an external device via the network 6 and functions as a communication interface for sending and receiving various types of data.

[0030] The training data acquisition unit 300 is connected to external devices (for example, the plant management device 2, the administrator's terminal device 5A, etc.) via the communication unit 31 and the network 6, and acquires training data 13 consisting of input data and output data. The training data 13 is used as training data, validation data, and test data in supervised learning. In addition, the output data of the training data 13 is used as correct answer data (hereinafter referred to as correct answer labels) in supervised learning.

[0031] The machine learning unit 301 uses each of the multiple sets of training data 13 stored in the memory unit 32 to perform machine learning on the learning model 14. When performing machine learning, the machine learning unit 301 can employ any method, such as online learning, batch learning, or mini-batch learning. The machine learning unit 301 may also perform predetermined preprocessing on the input data to be input to the learning model 14, or perform predetermined postprocessing on the output data output from the learning model 14.

[0032] In addition to the machine learning program 320, the memory unit 32 temporarily stores the training data 13 acquired by the training data acquisition unit 300, and also stores the trained learning model 14 (specifically, the adjusted weight parameter set) generated by the machine learning unit 301. The trained learning model 14 stored in the memory unit 32 is used by the network 6, recording media, etc. It is provided to the actual system (for example, the plant design support device 4) via this method.

[0033] Figure 4 is an explanatory diagram showing an example of training data 13 and a training model 14. The training data 13 used for machine learning of the training model 14 consists of input data containing information 11A regarding the design requirements of the plant 10 and output data containing plant design information 12.

[0034] The design requirements information 11A included in the input data is, as shown in Figure 2B, at least one of the following: information on the raw materials of plant 10, information on the products, information on ancillary equipment, information on the process flow, and information on standards or criteria applicable to plant 10. The details of each piece of information are the same as in Figure 2B, so a detailed explanation is omitted.

[0035] The input data constituting the training data 13 may further include, in addition to the information 11A regarding the design requirements of the plant 10, at least one of the following: information 11B regarding the design philosophy, information 11C regarding external conditions, information 11D regarding environmental requirements, and information 11E regarding cost requirements, as shown in Figure 4.

[0036] When the input data includes design philosophy information 11B, the design philosophy information 11B is at least one of the following, as shown in Figure 2B: information regarding the design life of plant 10, information regarding the design margin of plant 10, information regarding the safety factor of plant 10, information regarding the operating rate of plant 10, and information regarding the turndown requirements of plant 10. Details of each piece of information are the same as in Figure 2B, so their explanation is omitted.

[0037] When the input data includes information on external conditions 11C, the information on external conditions 11C is at least one of the following, as shown in Figure 2B: information on the site conditions of the construction site of plant 10, information on the meteorological conditions of the construction site of plant 10, information on the geological conditions of the construction site of plant 10, and information on the marine conditions of the sea area surrounding plant 10. The details of each piece of information are the same as in Figure 2B, so the explanation is omitted.

[0038] If the input data includes information on environmental requirements 11D, the information on environmental requirements 11D is at least one of the following, as shown in Figure 2C: information on pollution control requirements for plant 10, information on noise regulations for plant 10, and information on explosion-proof requirements for plant 10. The details of each piece of information are the same as in Figure 2C, so a detailed explanation is omitted.

[0039] If the input data includes cost requirement information 11E, the cost requirement information 11E is at least one of the following, as shown in Figure 2C: information on the economic useful life of plant 10, information on the discount rate of plant 10, information on the capital investment of plant 10, and information on the operating costs of plant 10. The details of each piece of information are the same as in Figure 2C, so we will omit further explanation.

[0040] The information 11A to 11E relating to the various requirements described above is set as, for example, a string (text), an image, a continuous variable, a discrete variable, or a categorical variable, depending on the definition and content of the information. Therefore, the machine learning unit 301 may perform preprocessing on the input data to be input to the learning model 14, such as feature engineering, encoding, scaling, or dimensionality reduction.

[0041] The plant design information 12 included in the output data, as described above, is information that can reproduce the three-dimensional shape of each component (equipment 100, piping 101, instruments, controllers, etc.) that constitutes the plant 10, and is data in which the processes realized by each component are recorded in relation to the three-dimensional shape. The plant design information 12 is that the plant 10 is designed to satisfy the various requirements information 11A to 11E included in the input data. If design changes are made later, or if modifications are carried out after construction, the plant design information 12 may reflect these design changes or modifications.

[0042] The learning data acquisition unit 300 acquires learning data 13 by referring to various types of information 11 and 12 registered in the plant management database 20, or by receiving input operations from the administrator's terminal device 5A. When the learning data acquisition unit 300 refers to the plant management database 20, for example, it acquires learning data 13 by acquiring various types of information 11 and 12 related to a specific plant 10.

[0043] The learning model 14 employs, for example, a neural network structure and comprises an input layer 140, a hidden layer 141, and an output layer 142. Synapses (not shown) connect each neuron between each layer, and each synapse is associated with a weight. The weight parameters, consisting of the weights of each synapse, are adjusted by machine learning.

[0044] The input layer 140 has a number of neurons corresponding to various requirements information 11 as input data, and each value of the input data is input to each neuron. The output layer 142 has a number of neurons corresponding to plant design information 12 as output data, and outputs output data (inference results) for the input data.

[0045] The machine learning unit 301 inputs multiple sets of training data 13 into the learning model 14 and generates a trained learning model 14 by having the learning model 14 learn the relationship between the input data (information about various requirements 11) and output data (plant design information 12) contained in the training data 13.

[0046] The machine learning unit 301 performs machine learning on the learning model 14 using multiple sets of training data 13 acquired by the training data acquisition unit 300. Then, the machine learning unit 301 generates a trained learning model 14 by having the learning model 14 learn the correlation between input data and output data.

[0047] In this embodiment, the data configuration of the training data 13 and the learning model 14 was described as being as shown in Figure 4. However, multiple data configurations with different conditions may be adopted, for example, differences in machine learning methods, input data, output data, etc. In that case, the training data acquisition unit 300 acquires multiple types of training data 13 corresponding to the multiple data configurations with different conditions, and the machine learning unit 301 performs machine learning for each learning model using these training data 13.

[0048] (Plant design support system 4) Figure 5 is a block diagram showing an example of the plant design support system 4. Figure 6 is a functional diagram illustrating an example of the plant design support system 4.

[0049] The plant design support device 4 comprises a control unit 40, a communication unit 41, and a storage unit 42. The control unit 40 functions as a data acquisition unit 400, a data generation unit 401, a simulation processing unit 402, and an output processing unit 403 by executing the plant design support program 420 stored in the storage unit 42. The communication unit 41 is connected to an external device via the network 6 and functions as a communication interface for sending and receiving various types of data.

[0050] The data acquisition unit 400 is connected to external devices (e.g., plant management device 2, designer terminal device 5B, etc.) via the communication unit 41 and network 6, and acquires input data including information 11 regarding various requirements for the newly constructed plant 10. The data acquisition unit 400 acquires input data by referring to information registered in the plant management database 20 or by receiving input operations for various requirements information 11 from the designer's terminal device 5B. At that time, the data acquisition unit 400 acquires input data according to the data structure of the learning model 14.

[0051] The data generation unit 401 inputs the input data (information 11 regarding various requirements) acquired by the data acquisition unit 400 into the learning model 14, thereby generating output data including plant design information 12 from the input data. The data generation unit 401 may perform predetermined preprocessing on the input data to be input to the learning model 14, similar to the machine learning unit 301, or it may perform predetermined postprocessing on the output data output from the learning model 14, similar to the machine learning unit 301.

[0052] Furthermore, the data generation unit 401 may generate multiple plant design information 12 as design proposals for the plant 10. For example, the data generation unit 401 may generate multiple output data, each containing plant design information 12, by inputting multiple input data, each containing partially different information 11 regarding various requirements, into the learning model 14. Alternatively, if the learning model 14 is configured to output output data containing multiple plant design information 12 for input data, the data generation unit 401 may generate output data containing multiple plant design information 12 by inputting input data into the learning model 14.

[0053] The simulation processing unit 402 executes a process simulation based on the plant design information 12 included in the output data generated by the data generation unit 401, and outputs the simulation results for the design proposal defined as the plant design information 12. For example, various evaluation indicators for when the process is executed with each component defined in the plant design information 12 are output as the simulation results. As evaluation indicators, for example, any indicators related to the productivity, safety, environmental performance, or cost of the plant 10 can be used, but they may also correspond to various requirements for the plant 10.

[0054] Furthermore, if the data generation unit 401 generates multiple plant design information 12 as multiple design proposals, the simulation processing unit 402 may execute process simulations based on the multiple plant design information 12 and output the simulation results for each design proposal.

[0055] The output processing unit 403 performs output processing to output output data (plant design information 12) generated by the data generation unit 401 and the results of the simulation performed by the simulation processing unit 402 (for example, various evaluation indicators). For example, the output processing unit 403 may transmit the plant design information 12 and display information for displaying the simulation results to the designer's terminal device 5B so that the information is displayed on the display screen of the terminal device 5B, or it may transmit the plant design information 12 to the plant management device 2 so that the plant design information 12 is registered in the plant management database 20.

[0056] The memory unit 42 stores the plant design support program 420 as well as the trained learning models 14 used by the data generation unit 401. The number of learning models 14 stored in the memory unit 42 is not limited to the example above; for example, multiple trained models with different conditions, such as differences in machine learning methods, input data, and output data, may be stored and used selectively or in parallel. The memory unit 42 may also be replaced by the memory unit of an external computer (which may be the plant management device 2 in this embodiment), in which case the data generation unit 401 only needs to access the external computer.

[0057] (Hardware configuration of each device) Figure 7 is a hardware configuration diagram showing an example of the computer 900 that constitutes each device. Each device 2 to 5B in the plant design support system 1 is composed of a general-purpose or dedicated computer 900.

[0058] As shown in Figure 7, the computer 900 comprises, as its main components, a bus 910, a processor 912, memory 914, an input device 916, an output device 917, a display device 918, a storage device 920, a communication interface unit 922, an external device interface unit 924, an I / O device interface unit 926, and a media input / output unit 928. Note that the above components may be omitted as appropriate depending on the intended use of the computer 900.

[0059] The processor 912 consists of one or more arithmetic processing units (CPU (Central Processing Unit), MPU (Micro-Processing Unit), DSP (Digital Signal Processor), GPU (Graphics Processing Unit), NPU (Neural Processing Unit), etc.) and operates as a control unit that oversees the entire computer 900. The memory 914 stores various data and programs 930 and consists of volatile memory (DRAM, SRAM, etc.) that functions as main memory, and non-volatile memory (ROM), flash memory, etc.

[0060] The input device 916 consists of, for example, a keyboard, mouse, numeric keypad, or electronic pen, and functions as an input unit. The output device 917 consists of, for example, a sound (voice) output device or a vibration device, and functions as an output unit. The display device 918 consists of, for example, a liquid crystal display, an organic EL display, electronic paper, or a projector, and functions as an output unit. The input device 916 and the display device 918 may be configured as an integrated unit, such as a touch panel display. The storage device 920 consists of, for example, an HDD or SSD, and functions as a storage unit. The storage device 920 stores various data necessary for the execution of the operating system and program 930.

[0061] The communication I / F unit 922 is connected by wire or wireless to a network 940 such as the Internet or an intranet (which may be the same as network 6 in Figure 1) and functions as a communication unit that sends and receives data with other computers according to a predetermined communication standard. The external device I / F unit 924 is connected by wire or wireless to external devices 950 such as cameras, printers, scanners, and reader / writers and functions as a communication unit that sends and receives data with external devices 950 according to a predetermined communication standard. The I / O device I / F unit 926 is connected to I / O devices 960 such as various sensors and actuators and functions as a communication unit that sends and receives various signals and data with the I / O devices 960, such as detection signals from sensors and control signals to actuators. The media input / output unit 928 consists of, for example, a drive device such as a DVD drive or CD drive, a memory card slot, and a USB connector, and reads and writes data to media (non-temporary storage media) 970 such as DVDs, CDs, memory cards, and USB memory.

[0062] In the computer 900 having the above configuration, the processor 912 calls and executes the program 930 stored in the storage device 920 from the memory 914, and controls various parts of the computer 900 via the bus 910. The program 930 may also be stored in the memory 914 instead of the storage device 920. The program 930 may be recorded on the media 970 in an installable or executable file format and provided to the computer 900 via the media input / output unit 928. The program 930 may also be provided to the computer 900 by downloading it via the network 940 through the communication interface unit 922. Furthermore, the computer 900 controls the processor 9 The various functions that 12 implements by executing program 930 may also be implemented using hardware such as FPGA (Field-Programmable Gate Array) or ASIC (Application Specific Integrated Circuit).

[0063] Computer 900 is an electronic device of any form, consisting of, for example, a stationary computer or a portable computer. Computer 900 may be a client computer, a server computer, a cloud computer, or an embedded computer such as a control panel or controller (including microcontrollers, programmable logic controllers, and sequencers).

[0064] (Operation of Plant Design Support System 1) The following describes a series of operations performed by the plant design support system 1. This series of operations is performed through the cooperation of each part of the plant design support device 4 (each step of the plant design support method executed by the plant design support program 420) and terminal devices 5A and 5B.

[0065] (Machine learning methods) Figure 8 is a flowchart illustrating an example of a machine learning method using the machine learning device 3. In the following explanation, it is assumed that the plant management database 20 stores information 11 and 12 about numerous plants 10.

[0066] First, in step S100, when the learning data acquisition unit 300 receives, for example, operation information from the administrator's terminal device 5A instructing it to start machine learning, it refers to the plant management database 20, acquires a desired number of learning data 13, and temporarily stores the acquired learning data 13 in the storage unit 32.

[0067] Next, in step S110, the machine learning unit 301 prepares a pre-training model 14 in order to start machine learning. The pre-training model 14 prepared here is, for example, a neural network model, in which the weights of each synapse are set to initial values.

[0068] Next, in step S120, the machine learning unit 301 randomly selects, for example, one set of training data 13 from the multiple sets of training data 13 stored in the memory unit 32.

[0069] Next, in step S130, the machine learning unit 301 inputs information 11 (input data) regarding various requirements contained in a set of training data 13 to the input layer 140 of the prepared pre-training (or training) learning model 14. As a result, plant design information 12 (output data) is output as an inference result from the output layer 142 of the learning model 14, but this output data is generated by the pre-training (or training) learning model 14. Therefore, in the pre-training (or training) state, the output data output as an inference result shows information different from the plant design information 12 (ground truth labels) contained in the training data 13.

[0070] Next, in step S140, the machine learning unit 301 compares the plant design information 12 (ground truth labels) included in the set of training data 13 acquired in step S120 with the plant design information 12 (output data) output as an inference result from the output layer 142 in step S130, and performs machine learning by adjusting the weight of each synapse (backpropagation). In this way, the machine learning unit 301 trains the learning model 14 on the correlation between the information 11 regarding various requirements and the plant design information 12.

[0071] Next, in step S150, the machine learning unit 301 determines whether predetermined learning termination conditions have been met, for example, based on the evaluation value of the error function, which is based on the plant design information 12 (ground truth labels) included in the training data 13 and the plant design information 12 (output data) output as an inference result, or based on the remaining number of untrained training data 13 stored in the storage unit 32.

[0072] In step S150, if the machine learning unit 301 determines that the learning termination condition has not been met and that machine learning should continue (No in step S150), it returns to step S120 and repeats steps S120 to S140 multiple times using the untrained training data 13 on the learning model 14 that is currently being trained. On the other hand, in step S150, if the machine learning unit 301 determines that the learning termination condition has been met and that machine learning should be terminated (Yes in step S150), it proceeds to step S160.

[0073] Then, in step S160, the machine learning unit 301 stores the trained model 14 (set weight parameters) generated by adjusting the weights associated with each synapse in the storage unit 32, and the series of machine learning methods shown in Figure 8 is completed. In the machine learning method, step S100 corresponds to the training data acquisition step, steps S110 to S150 are the machine learning steps, and step S160 is the trained model storage step.

[0074] As described above, the machine learning device 3 and machine learning method according to this embodiment can provide a learning model 14 that generates output data including plant design information 12 in response to input data including information 11 regarding various requirements required for a newly constructed plant 10.

[0075] (Plant design support methods) Figure 9 is a flowchart illustrating an example of a plant design support method using the plant design support device 4. In the following explanation, it is assumed that the memory unit 42 stores a trained model 14 that has been trained by the machine learning device 3.

[0076] First, in step S200, the designer's terminal device 5B displays the plant design support input screen 15 in cooperation with the plant design support device 4, and accepts input operations from the designer to input the requirements for the newly constructed plant 10 on the plant design support input screen 15. Then, the designer's terminal device 5B transmits input operation information indicating the result of the input operations performed by the designer to the plant design support device 4.

[0077] Figure 10 is a screen configuration diagram showing an example of a plant design support input screen 15. The plant design support input screen 15 includes requirement setting areas 150A to 150E for setting information 11A to 11E regarding various requirements for a new plant 10, and a design proposal generation button 152 for instructing the generation of plant design information 12.

[0078] In the requirements setting areas 150A to 150E, for example, pressing the setting buttons 151A to 151E displays the detailed settings screen (not shown), and the information 11A to 11E is entered into the detailed settings screen. Alternatively, the information 11A to 11E may be entered from the plant management database 20 if it is specified as being registered in the plant management database 20. When the design proposal generation button 152 is pressed, the information 11A to 11E regarding the various requirements entered on the detailed settings screen is sent to the plant design support device 4 as input operation information. At that time, the input operation information includes the information that has been checked for each checkbox in the requirements setting areas 150A to 150E (all information in the example in Figure 10).

[0079] Next, in step S210, when the data acquisition unit 400 receives the input operation information transmitted in step S200, it acquires input data to be input to the learning model 14 based on the input operation information. In the example of the plant design support input screen 15 shown in Figure 10, the data acquisition unit 400 acquires input data based on the operation results performed on the requirement setting areas 150A to 150E.

[0080] Next, in step S220, the data generation unit 401 inputs the input data (information on various requirements 11A to 11E) acquired in step S210 into the learning model 14, thereby generating output data (plant design information 12) from the input data.

[0081] Next, in step S230, the simulation processing unit 402 executes a process simulation based on the plant design information 12 included in the output data generated by the data generation unit 401, and outputs the simulation results for the design proposal defined as the plant design information 12.

[0082] Next, in step S240, the output processing unit 403 performs output processing to output the plant design information 12 included in the output data generated in step S220 and the simulation execution results output in step S230. For example, if the output processing unit 403 sends display information for displaying the plant design information 12 and the simulation execution results to the designer's terminal device 5B, the designer's terminal device 5B displays the plant design support output screen 16 based on that display information, thereby presenting the plant design information 12 and the simulation execution results to the designer.

[0083] Figure 11 is a screen configuration diagram showing an example of the plant design support output screen 16. The plant design support output screen 16 includes a plant design information display area 160 that displays plant design information 12 included in the output data generated by the data generation unit 401, a component display area 161 that displays information about the components, and a simulation execution result display area 162 that displays the results of the simulation.

[0084] The plant design information display area 160 displays the three-dimensional shape of any location in the plant 10, and operations such as moving, zooming in and out, rotating, and switching the display and hiding of individual components can be performed using the cursor 163. The component display area 161 displays information about the component selected by the cursor 163 in the plant design information display area 160 (for example, the pressure vessel in the example in Figure 11). The simulation execution result display area 162 displays the results of the simulation, for example, evaluation indicators.

[0085] As described above, the series of plant design support methods shown in Figure 9 is completed. In the above plant design support method, step S210 corresponds to the data acquisition process, step S220 to the data generation process, step S230 to the simulation processing process, and step S240 to the output processing process. The series of plant design support methods can be executed at any time as long as the plant design support device 4 can acquire input data. Alternatively, the plant design information 12 as the final design proposal may be generated by repeatedly changing some of the requirements for the plant 10.

[0086] As described above, according to the plant design support device 4 and plant design support method of this embodiment, by inputting input data including information 11 regarding various requirements for a newly constructed plant 10 into the learning model 14, output data including plant design information 12 is generated from the input data. Therefore, the design work for the plant 10 can be made easier and more efficient.

[0087] (Other embodiments) The present invention is not limited to the embodiments described above, and can be implemented with various modifications without departing from the spirit of the invention. All such modifications are included in the technical concept of the present invention.

[0088] In the above embodiment, the functions of each part of the machine learning device 3 and the plant design support device 4 were described as being realized by a single device, but the functions of each part may be distributed among multiple devices to be realized by multiple devices. Also, the control units of terminal devices 5A and 5B may function as the machine learning device 3 by executing the machine learning program 320, or as the plant design support device 4 by executing the plant design support program 420.

[0089] In the above embodiment, the case in which a neural network (including deep learning) is adopted as the learning model 14 was described, but other machine learning models may also be adopted. Examples of other machine learning models include VAE (Variational Autoencoder), GAN (Generative Adversarial Network), flow-based models, diffusion models, and other foundational models using other methods (e.g., large-scale language models) and multimodal models, but are not limited to these examples as long as they can generate output data (plant design information 12) from input data (information 11 regarding various requirements). Furthermore, the learning model 14 may also function as an explainable AI. In addition, as a machine learning algorithm, reinforcement learning, causal learning, meta-learning, etc. may be used in addition to supervised learning, and are not limited to these examples.

[0090] In the above embodiment, the case in which the plant design support system 1 operates according to the flowcharts shown in Figures 8 and 9 has been described. However, some parts of each process (each section) may be omitted, or other processes may be added. In the flowchart shown in Figure 9, for example, the simulation processing step (simulation processing unit 402) in step S230 may be omitted. In that case, the output processing step (output processing unit 403) in step S240 may be modified to omit outputting the simulation execution results.

[0091] (Inference device, inference method, and inference program) The present invention can be provided not only in the form of a plant design support device 4 (plant design support method or plant design support program 420) according to the above embodiment that generates plant design information 12, but also in the form of an inference device (inference method or inference program) used to infer plant design information 12. In that case, the inference device (inference method or inference program) may include a memory and a processor, the processor of which may execute a series of processes. The series of processes includes a data acquisition process (data acquisition step) that acquires input data including information 11 (for example, information 11A regarding design requirements) required for the plant 10, and an inference process (inference step) that, once the input data has been acquired by the data acquisition process, infers output data including plant design information 12 from the input data. [Explanation of symbols]

[0092] 1...Plant design support system, 2...Plant management device, 3...Machine learning device, 4...Plant design support device, 5A, 5B...Terminal device, 20...Plant management database, 30...Control unit, 31...Communication unit, 32...Storage unit, 40...Control unit, 41...Communication unit, 42...Storage unit, 300...Training data acquisition unit, 301...Machine learning unit, 320...Machine learning program, 400...Data acquisition unit, 401...Data generation unit, 402...Simulation processing unit, 403…Output processing unit, 420…Plant design support program

Claims

1. A data acquisition unit that acquires input data, The system includes a data generation unit that inputs the input data acquired by the data acquisition unit into a learning model and generates output data for said input data, The aforementioned learning model, This is a trained model that has learned the correlation between the input data and the output data using machine learning. The aforementioned input data is This includes information on the design requirements for newly constructed plants. The output data is, Information capable of reproducing the three-dimensional shape of each component constituting the plant, including plant design information in which the processes realized by each component are recorded in relation to the three-dimensional shape. Plant design support system.

2. The information regarding the design requirements included in the input data is: Information regarding the raw materials of the aforementioned plant, Information relating to the products of the aforementioned plant, Information relating to the ancillary facilities of the aforementioned plant Information regarding the process flow of the aforementioned process, and At least one piece of information relating to standards or criteria applicable to the aforementioned plant, The plant design support device according to claim 1.

3. The aforementioned input data is Information regarding the design concept of the aforementioned plant, Information regarding the external conditions of the aforementioned plant, Information regarding the environmental requirements of the aforementioned plant, and Further including at least one piece of information relating to the cost requirements of the plant, The plant design support device according to claim 1.

4. The information regarding the design concept included in the input data is, Information relating to the design life of the aforementioned plant, Information regarding the design margin of the aforementioned plant, Information regarding the safety factor of the aforementioned plant, Information regarding the operating rate of the aforementioned plant, and At least one piece of information relating to the turndown requirements of the aforementioned plant, The plant design support device according to claim 3.

5. The information regarding the external conditions included in the input data is, Information regarding the site conditions of the construction site of the aforementioned plant, Information regarding the weather conditions at the construction site of the aforementioned plant, Information regarding the geological conditions of the construction site of the aforementioned plant, and At least one piece of information relating to the marine conditions of the sea area surrounding the plant, The plant design support device according to claim 3.

6. The information regarding the environmental requirements included in the input data is: Information regarding the pollution control requirements of the aforementioned plant, Information regarding noise regulations for the aforementioned plant, and At least one piece of information relating to the explosion-proof requirements of the aforementioned plant, The plant design support device according to claim 3.

7. The information regarding the cost requirements included in the input data is: Information regarding the economic service life of the aforementioned plant Information regarding the discount rate of the aforementioned plant Information regarding capital investment in the aforementioned plant, and At least one piece of information relating to the operating costs of the aforementioned plant, The plant design support device according to claim 3.

8. The data generation unit, As a design proposal for the aforementioned plant, multiple pieces of plant design information are generated. A plant design support device according to any one of claims 1 to 7.

9. The system includes a simulation processing unit that performs simulations of the process based on multiple plant design information and outputs the simulation results for each of the design proposals. The plant design support device according to claim 8.

10. An inference device comprising memory and a processor, The aforementioned processor, A data acquisition process that obtains input data including information on the design requirements for a newly constructed plant, When the input data is acquired by the data acquisition process, an inference process is performed on the input data to infer output data which includes plant design information that is capable of reproducing the three-dimensional shape of each component constituting the plant, and in which the processes realized by each component are recorded in relation to the three-dimensional shape. Reasoning device.

11. A training data acquisition unit that acquires multiple sets of training data consisting of input data and output data, A machine learning unit uses multiple sets of the training data acquired by the training data acquisition unit to train a learning model on the correlation between the input data and the output data using machine learning. The machine learning unit has a memory unit that stores the learning model in which the correlation has been learned, The aforementioned input data is This includes information on the design requirements for newly constructed plants. The output data is, Information capable of reproducing the three-dimensional shape of each component constituting the plant, including plant design information in which the processes realized by each component are recorded in relation to the three-dimensional shape. Machine learning device.

12. A computer-based plant design support method, The data acquisition process involves obtaining input data, The system includes a data generation step which generates output data for the input data obtained in the data acquisition step by inputting the input data into a learning model, The aforementioned learning model, This is a trained model that has learned the correlation between the input data and the output data using machine learning. The aforementioned input data is This includes information on the design requirements for newly constructed plants. The output data is, Information capable of reproducing the three-dimensional shape of each component constituting the plant, including plant design information in which the processes realized by each component are recorded in relation to the three-dimensional shape. Plant design support methods.

13. An inference method performed by an inference device comprising memory and a processor, The aforementioned processor, A data acquisition process that obtains input data including information on the design requirements for a newly constructed plant, When the input data is acquired by the data acquisition process, an inference process is performed on the input data to infer output data which includes plant design information that is capable of reproducing the three-dimensional shape of each component constituting the plant, and in which the processes realized by each component are recorded in relation to the three-dimensional shape. Reasoning method.

14. A machine learning method performed by a computer, The training data acquisition process involves acquiring multiple sets of training data consisting of input data and output data, A machine learning step in which a learning model learns the correlation between the input data and the output data using machine learning, using multiple sets of the training data acquired in the training data acquisition step, The system includes a trained model storage step, which stores the trained model, which has learned the correlation relationship through the machine learning step, in a storage unit. The aforementioned input data is This includes information on the design requirements for newly constructed plants. The output data is, Information capable of reproducing the three-dimensional shape of each component constituting the plant, including plant design information in which the processes realized by each component are recorded in relation to the three-dimensional shape. Machine learning methods.

Images