Real-time simulation method, device and equipment for ship thermodynamic system based on cooperation of mechanism and data driving model and storage medium

Abstract

The invention discloses a mechanism and data driven model collaborative ship thermodynamic system real-time simulation method, which comprises the following steps of: verifying an initial mechanism model based on a fluid network theory by using operation data of a ship thermodynamic system, and constructing a mechanism model meeting the precision; generating simulation data according to the working condition set, the input and output parameter set and the boundary parameter range by utilizing the mechanism model meeting the precision; the operation data and the simulation data are combined to form a training sample, the initial data driving model is trained, and a data driving model meeting the precision is constructed; mechanism model simulation calculation and data driving model simulation calculation are simultaneously carried out in a ship thermodynamic system, a proper model is selected for simulation according to the real-time requirement, the calculation resource requirement and the model precision requirement, and collaborative real-time simulation of the mechanism model and the data driving model is achieved. According to the method provided by the invention, co-simulation of a mechanism and a data driving model can be realized, and the requirements of high precision and real-time performance of ship thermodynamic system simulation are met.

Description

Technical Field

[0001] The present invention relates to the field of digital testing and operation and maintenance technology for ship thermal systems, and in particular to a real-time simulation method, device, equipment and storage medium for ship thermal systems that collaborates with a mechanism and a data-driven model. Background Art

[0002] A ship's thermal system is a crucial component of a vessel, generating propulsion power and electrical energy. It features a complex system structure, a wide variety of equipment, and coupled transport mechanisms. Real-time simulation of a ship's thermal system, where computational evolution time is aligned with clock time, is used to test and evaluate the system's functionality and performance, optimizing its design, testing, operation, and maintenance. However, the complexity of current ship thermal systems and the precision required for simulation present significant challenges for real-time simulation. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a real-time simulation method, device, equipment and storage medium for ship thermal systems that collaborate with mechanism and data-driven models in response to the shortcomings of the existing technology. The method can achieve collaborative simulation of mechanism and data-driven models by constructing mechanism and data-driven model structures and modeling methods, and model collaborative operation rules, thereby meeting the high precision and real-time requirements of ship thermal system simulation.

[0004] To achieve the above objectives, according to one aspect of the present invention, a real-time simulation method for a ship thermal system using a coordinated mechanism and data-driven model is provided, comprising: Using the operating data of the ship's thermal system to verify the initial mechanism model based on the fluid network theory, and constructing a mechanism model that meets the accuracy requirements; using the mechanism model that meets the accuracy requirements to generate simulation data based on the operating condition set, input and output parameter sets, and boundary parameter ranges; combining the operating data and the simulation data to form training samples, training the initial data-driven model, and constructing a data-driven model that meets the accuracy requirements; The mechanism model simulation calculation and the data-driven model simulation calculation are carried out simultaneously in the ship thermal system, and a suitable model is selected for simulation according to the real-time requirements, computing resource requirements, and model accuracy requirements to achieve collaborative real-time simulation of the mechanism model and the data-driven model.

[0005] In the above scheme, the steps of selecting a suitable model based on real-time requirements, computing resource requirements, and model accuracy requirements are as follows: Determine whether the mechanism model meets the real-time requirements, and if not, use the data-driven model for simulation; Determine whether the computing resource margin of the mechanism model is greater than a set value, and if not, use the data-driven model to perform simulation; Determine whether the accuracy of the mechanism model is higher than that of the data-driven model. If not, use the data-driven model for simulation; if so, use the mechanism model for simulation.

[0006] In the above scheme, the initial data-driven model is established using a BP neural network with teacher learning, the learning rate adopts momentum-adaptive learning rate, and the learning rate selection range is between 0.01 and 0.8; the hidden layer activation function adopts the proportional coordination of tansig and logsig functions; the output layer activation function adopts purelin function, and the error adopts mean square error.

[0007] In the above scheme, the setting of the working condition set includes: clarifying the simulation working conditions, evaluating the complexity of the theoretical model and computing resources of each object in the ship thermal system, and clarifying the physical objects and simulation working conditions of the data-driven model.

[0008] In the above scheme, the setting of the input and output parameter set includes: for the equipment type data model, the input and output parameters are the interaction parameters between the fluid network and the equipment module and the equipment external boundary parameters; for the subsystem type data model, the input and output parameters are the interface parameters and external boundary parameters between the subsystem and the ship thermal system; for the system data model, the output parameters select the key parameters of the system, and the input parameters are the parameters that affect the output parameters.

[0009] In the above solution, the setting of the boundary parameter range includes: setting the boundary parameter range in the mechanism model, and the boundary parameter range envelops the upper and lower limits of the data-driven model simulation conditions.

[0010] In the above scheme, the accuracy of the mechanism model must meet the following requirements: when using the mechanism model to simulate known operating conditions, the relative error of the main steady-state parameters of the ship's thermal system shall not exceed 5%; The accuracy of the data-driven model meets the accuracy requirement: when comparing the output value of the data-driven model with the operating data, the relative error of the main steady-state parameters of the ship's thermal system is no more than 5%.

[0011] In the above scheme, the collaborative real-time simulation of the mechanism model and the data-driven model includes three structures, namely, a channel separation model structure, a parameter serial transfer model structure, and a complete data-driven real model structure.

[0012] In the above scheme, in the channel separation model structure, the mechanism model and the data-driven model are divided into a pressure flow channel and an enthalpy-temperature channel; wherein the pressure flow channel is solved as a whole using a fluid network method, and the enthalpy-temperature channel is solved independently with each equipment module as the basic unit.

[0013] In the above scheme, in the parameter serial transfer model structure, each module does not distinguish between channels, and the mechanism model and the data model each complete the complete mapping of object input and output; when the parameter serial transfer model structure model is running, each module is used as the basic unit for solution, and then the dynamic characteristics of the entire ship thermal system are characterized by parameter transfer.

[0014] In the above scheme, in the fully data-driven real model structure, the mechanism model does not participate in real-time calculations, and the data-driven model completes the calculations of all modules in the system; the mechanism model generates data-driven model simulation data based on the set working condition set, the input and output parameter set and the boundary parameter range and passes it to the data-driven model.

[0015] In addition, to achieve the above-mentioned purpose, the present invention also proposes a real-time simulation device for a ship thermal system that cooperates with a mechanism and a data-driven model, comprising: A construction module is used to verify an initial mechanism model based on fluid network theory using operating data of a ship thermal system, and construct a mechanism model that meets accuracy requirements; generate simulation data based on a set of operating conditions, a set of input and output parameters, and a boundary parameter range using the mechanism model that meets accuracy requirements; combine the operating data and the simulation data to form a training sample, train the initial data-driven model, and construct a data-driven model that meets accuracy requirements; The collaborative simulation module simultaneously carries out the simulation calculation of the mechanism model and the simulation calculation of the data-driven model in the ship thermal system, selects the appropriate model for simulation according to the real-time requirements, computing resource requirements, and model accuracy requirements, and realizes the collaborative real-time simulation of the mechanism model and the data-driven model.

[0016] In addition, to achieve the above-mentioned purpose, the present invention also proposes a real-time simulation device for a ship thermal system that cooperates with a mechanism and a data-driven model. The real-time simulation device for a ship thermal system that cooperates with a mechanism and a data-driven model includes: a memory, a processor, and a real-time simulation program for a ship thermal system that cooperates with a mechanism and a data-driven model stored on the memory and can be run on the processor. The real-time simulation program for a ship thermal system that cooperates with a mechanism and a data-driven model is configured to implement the real-time simulation method for a ship thermal system that cooperates with a mechanism and a data-driven model.

[0017] In addition, to achieve the above-mentioned purpose, the present invention also proposes a storage medium, on which is stored a real-time simulation program for a ship thermal system coordinated by a data-driven model. When the real-time simulation program for a ship thermal system coordinated by a mechanism and a data-driven model is executed by a processor, the real-time simulation method for a ship thermal system coordinated by a mechanism and a data-driven model is realized.

[0018] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art: The present invention provides a real-time simulation method for ship thermal systems that collaborates with mechanism-driven and data-driven models. This method can achieve rapid and high-precision simulation of ship thermal systems, lay the foundation for a large number of working condition calculations and ultra-real-time deductions of ship thermal systems, and effectively support the design optimization, test plan formulation, and operation and maintenance decision-making of ship thermal systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. Throughout the drawings, the same reference figures denote the same components. In the drawings: Figure 1 The figure is a flow chart of a method for real-time simulation of a ship thermal system in which a mechanism and a data-driven model are coordinated in the first embodiment of the present invention.

[0020] Figure 2 Schematic diagram of the process of training a data-driven model in Example 1 of the present invention.

[0021] Figure 3 This is a structural diagram of the channel separation model structure, the parameter serial transfer model structure, and the complete data-driven real model structure in the first embodiment of the present invention.

[0022] Figure 4 This is a flowchart of selecting a suitable model according to real-time requirements, computing resource requirements, and model accuracy requirements in the first embodiment of the present invention.

[0023] Figure 5 This is a schematic diagram of the functional modules of a real-time simulation device for a ship thermal system in which a mechanism and a data-driven model are coordinated in the second embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0025] It should be understood that the size of the serial numbers of the steps in the embodiment does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

[0026] It should be noted that the execution subject of the embodiments of this application can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, mobile phone, etc., or an electronic device capable of implementing the above functions, a real-time simulation device for a ship thermal system that collaborates with a mechanism and a data-driven model, etc. The following uses a real-time simulation device for a ship thermal system that collaborates with a mechanism and a data-driven model as an example to illustrate this embodiment and the following embodiments.

[0027] Example 1 The embodiment of the present application provides a real-time simulation method for a ship thermal system in which a mechanism and a data-driven model are coordinated. Figure 1 , Figure 1 Schematic diagram of a flow chart of a real-time simulation method for a ship thermal system in which a mechanism and a data-driven model are coordinated in an embodiment of the present application. A real-time simulation method for a ship thermal system in which a mechanism and a data-driven model are coordinated in an embodiment of the present application includes: S1. Use the operating data of the ship thermal system to verify the initial mechanism model based on fluid network theory and construct a mechanism model that meets the accuracy requirements; use the mechanism model that meets the accuracy requirements to generate simulation data according to the operating condition set, input and output parameter set and boundary parameter range; combine the operating data and simulation data to form a training sample, train the initial data-driven model, and construct a data-driven model that meets the accuracy requirements.

[0028] In the embodiments of the present application, it should be noted that the initial mechanism model is a traditional fluid network-based simulation model.

[0029] In an embodiment of the present application, the initial data-driven model is established using a BP neural network with teacher learning, the learning rate adopts momentum-adaptive learning rate, and the learning rate selection range is between 0.01 and 0.8. Specifically, in this embodiment, the learning rate is 0.1; the hidden layer activation function adopts the tansig and logsig functions in a proportional (the proportional value is adjustable) collaborative manner. Specifically, in this embodiment, the tansig function accounts for 0.5 and the logsig function accounts for 0.5; the activation function of the output layer uses the purelin function, and the error adopts the mean square error.

[0030] In the embodiment of the present application, it can be understood that the setting of the operating condition set includes: clarifying the simulation operating conditions, evaluating the complexity of the theoretical model and computing resources of each object in the ship thermal system, and clarifying the physical objects and simulation operating conditions of the data-driven model.

[0031] The settings of the input and output parameter sets include: for equipment data models, the input and output parameters are the interaction parameters between the fluid network and the equipment module and the external boundary parameters of the equipment; for subsystem data models, the input and output parameters are the interface parameters and external boundary parameters between the subsystem and the ship thermal system; for system data models, the output parameters select the key parameters of the system, and the input parameters are the parameters that affect the output parameters.

[0032] The setting of the boundary parameter range includes: setting the boundary parameter range in the mechanism model, and the boundary parameter range envelopes the upper and lower limits of the simulation conditions of the data-driven model.

[0033] Specifically, the data-driven model is trained by using the mechanism model that meets the accuracy requirements and generating data-driven model simulation training samples according to the working condition set, input and output parameter set and boundary parameter range. Figure 2 shown.

[0034] Specifically, in the embodiments of this application, the accuracy of the mechanism model must meet the following requirements: When using the mechanism model to simulate known operating conditions, the relative error of the main steady-state parameters of the ship's thermal system is no more than 5%. Simulations based on the mechanism model within the set boundary parameters should generate a sufficient number of training samples that cover the range of simulated operating conditions.

[0035] The accuracy of the data-driven model must meet the following requirements: when comparing the output value of the data-driven model with the operating data, the relative error of the main steady-state parameters of the ship's thermal system shall not exceed 5%.

[0036] Specifically, in the embodiment of the present application, the mechanism model and the data driven model collaborative real-time simulation includes three structures, such as Figure 3 As shown in the figure, they are channel separation model structure, parameter serial transfer model structure, and complete data driven real model structure respectively; In the channel separation model structure, the mechanism model and data-driven model are divided into pressure flow channel and enthalpy temperature channel. The pressure flow channel adopts the fluid network method for overall solution, while the enthalpy temperature channel is solved independently with each equipment module as the basic unit. In the parameter serial transfer model structure, each module does not distinguish between channels, and the mechanism model and data model each complete the complete mapping of object input and output. During operation, the parameter serial transfer model structure model uses each module as the basic unit for solution, and then characterizes the dynamic characteristics of the entire ship thermal system through parameter transfer. In the fully data-driven real model structure, the mechanism model does not participate in real-time calculations, and the data-driven model completes the calculations of all modules in the system; the mechanism model generates data-driven model simulation data based on the set working conditions, input and output parameter sets, and boundary parameter ranges and passes it to the data-driven model.

[0037] S2, simultaneously carry out mechanism model simulation calculations and data-driven model simulation calculations in the ship thermal system, select the appropriate model for simulation according to the real-time requirements, computing resource requirements, and model accuracy requirements, and realize the collaborative real-time simulation of the mechanism model and the data-driven model.

[0038] Specifically, in this embodiment, Figure 4 As shown in the figure, the steps to select a suitable model based on real-time requirements, computing resource requirements, and model accuracy requirements are as follows: Determine whether the mechanism model meets the real-time requirements. If not, use the data-driven model for simulation; Determine whether the computational resource margin of the mechanism model is greater than a set value (in this embodiment, the computational resource margin is defined as the ratio of available computer resources to computational resources, with a set value of 1.2). If not, use the data-driven model for simulation; Determine whether the accuracy of the mechanism model is higher than that of the data-driven model. If not, use the data-driven model for simulation; if so, use the mechanism model for simulation.

[0039] In summary, the embodiments of the present application provide a real-time simulation method for a ship thermal system that collaborates with a mechanism and a data-driven model. This method can achieve fast and high-precision simulation of a ship thermal system, lay the foundation for a large number of working condition calculations and ultra-real-time deductions of the ship thermal system, and effectively support the design optimization, test plan formulation, and operation and maintenance decision-making of the ship thermal system.

[0040] Example 2 The embodiment of the present application provides a real-time simulation device for a ship thermal system that cooperates with a mechanism and a data-driven model, such as Figure 5 Shown, including: Construction module 10 is used to verify the initial mechanism model based on fluid network theory using the operating data of the ship thermal system, and construct a mechanism model that meets the accuracy requirements; use the mechanism model that meets the accuracy requirements to generate simulation data based on the operating condition set, input and output parameter set, and boundary parameter range; combine the operating data and simulation data to form training samples, train the initial data-driven model, and construct a data-driven model that meets the accuracy requirements; The collaborative simulation module 20 simultaneously carries out mechanism model simulation calculations and data-driven model simulation calculations in the ship thermal system, selects appropriate models for simulation according to real-time requirements, computing resource requirements, and model accuracy requirements, and realizes collaborative real-time simulation of the mechanism model and the data-driven model.

[0041] Example 3 The present application also provides a real-time simulation device for a ship thermal system that collaborates with a mechanism and a data-driven model, such as a smartphone, tablet computer, laptop computer, desktop computer, rack server, blade server, tower server or cabinet server (including an independent server or a server cluster consisting of multiple servers) that can execute programs.

[0042] The real-time simulation equipment of the ship thermal system with the coordinated mechanism and data-driven model includes but is not limited to: a memory, a processor and a real-time simulation program of the ship thermal system with the coordinated mechanism and data-driven model stored in the memory and runnable on the processor. The real-time simulation program of the ship thermal system with the coordinated mechanism and data-driven model is configured to implement a real-time simulation method of the ship thermal system with the coordinated mechanism and data-driven model.

[0043] In this embodiment, the memory (i.e., readable storage medium) includes flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and programmable read-only memory (PROM). The memory may also be an external storage device of the real-time simulation device for a ship thermal system that utilizes a mechanism- and data-driven model, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash memory card, etc. Of course, the memory may also include both the internal storage unit and external storage device of the real-time simulation device for a ship thermal system that utilizes a mechanism- and data-driven model. In this embodiment, the memory is typically used to store the operating system and various application software installed in the real-time simulation device for a ship thermal system that utilizes a mechanism- and data-driven model, such as the program code of the real-time simulation device for a ship thermal system that utilizes a mechanism- and data-driven model in Example 2. Furthermore, the memory may also be used to temporarily store various data that has been output or is about to be output.

[0044] In some embodiments, the processor can be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. This processor is typically used to control the overall operation of a real-time simulation device for a ship thermal system that utilizes a coordinated mechanism and data-driven model. In this embodiment, the processor is used to execute program code stored in a memory or process data, for example, to execute the real-time simulation device for a ship thermal system that utilizes a coordinated mechanism and data-driven model, thereby implementing the real-time simulation method for a ship thermal system that utilizes a coordinated mechanism and data-driven model described in Example 1.

[0045] Example 4 The present application also provides a storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a disk, an optical disk, a server, an App store, etc., which stores a real-time simulation program for a ship thermal system that collaborates with a mechanism-driven model. When executed by a processor, the real-time simulation program for a ship thermal system that collaborates with a mechanism-driven model implements corresponding functions. The storage medium of this embodiment is used in a real-time simulation device for a ship thermal system that collaborates with a mechanism-driven model. When executed by a processor, the real-time simulation method for a ship thermal system that collaborates with a mechanism-driven model is implemented.

[0046] It should be pointed out that, according to the needs of implementation, the various steps described in this application can be split into more steps, or two or more steps or partial operations of the steps can be combined into new steps to achieve the purpose of the present invention.

[0047] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A real-time simulation method for ship thermal systems based on a coordinated mechanism and data-driven model, characterized in that: include: The initial mechanism model based on fluid network theory is verified using the operating data of the ship's thermal system to construct a mechanism model that meets the accuracy requirements; Generate simulation data based on the operating condition set, input and output parameter sets, and boundary parameter ranges using the mechanism model that meets the accuracy requirements; Combining the operating data and the simulation data to form a training sample, training the initial data-driven model, and constructing a data-driven model that meets the accuracy requirements; The mechanism model simulation calculation and the data-driven model simulation calculation are carried out simultaneously in the ship thermal system, and a suitable model is selected for simulation according to the real-time requirements, computing resource requirements, and model accuracy requirements to achieve collaborative real-time simulation of the mechanism model and the data-driven model.

2. The method for real-time simulation of a ship thermal system based on the coordination of mechanism and data-driven model according to claim 1 is characterized in that: The steps of selecting a suitable model based on real-time requirements, computing resource requirements, and model accuracy requirements are as follows: Determine whether the mechanism model meets the real-time requirements, and if not, use the data-driven model for simulation; Determine whether the computing resource margin of the mechanism model is greater than a set value, and if not, use the data-driven model to perform simulation; Determine whether the accuracy of the mechanism model is higher than that of the data-driven model. If not, use the data-driven model for simulation; if so, use the mechanism model for simulation.

3. The method for real-time simulation of a ship thermal system by coordinating a mechanism and a data-driven model according to claim 1 is characterized in that: The initial data-driven model is established using a BP neural network with teacher learning, the learning rate adopts momentum-adaptive learning rate, and the learning rate selection range is between 0.01 and 0.8; the hidden layer activation function adopts the proportional coordination of tansig and logsig functions; the output layer activation function adopts purelin function, and the error adopts mean square error.

4. The method for real-time simulation of a ship thermal system by coordinating a mechanism and a data-driven model according to claim 1 is characterized in that: The setting of the working condition set includes: clarifying the simulation working conditions, evaluating the complexity of the theoretical model and computing resources of each object in the ship thermal system, and clarifying the physical objects and simulation working conditions of the data-driven model.

5. The method for real-time simulation of a ship thermal system by coordinating a mechanism and a data-driven model according to claim 1 is characterized in that: The setting of the input and output parameter set includes: for equipment data models, the input and output parameters are the interaction parameters between the fluid network and the equipment module and the equipment external boundary parameters; for subsystem data models, the input and output parameters are the interface parameters and external boundary parameters between the subsystem and the ship thermal system; for system data models, the output parameters are system key parameters, and the input parameters are parameters that affect the output parameters; The setting of the boundary parameter range includes: setting the boundary parameter range in the mechanism model, and the boundary parameter range envelops the upper and lower limits of the data-driven model simulation working condition.

6. The method for real-time simulation of a ship thermal system by coordinating a mechanism and a data-driven model according to claim 1 is characterized in that: The accuracy of the mechanism model must meet the following requirements: when using the mechanism model to simulate known operating conditions, the relative error of the main steady-state parameters of the ship's thermal system shall not exceed 5%; The accuracy of the data-driven model meets the accuracy requirement: when comparing the output value of the data-driven model with the operating data, the relative error of the main steady-state parameters of the ship's thermal system is no more than 5%.

7. The method for real-time simulation of a ship thermal system by coordinating a mechanism and a data-driven model according to claim 1 is characterized in that: The collaborative real-time simulation of the mechanism model and the data-driven model includes three structures: a channel separation model structure, a parameter serial transfer model structure, and a complete data-driven real model structure; Wherein, in the channel separation model structure, the mechanism model and the data-driven model are divided into a pressure flow channel and an enthalpy temperature channel; The pressure flow channel is solved as a whole using the fluid network method, while the enthalpy temperature channel is solved independently using each equipment module as the basic unit; In the parameter serial transfer model structure, each module does not distinguish between channels, and the mechanism model and the data model each complete the complete mapping of object input and output; when the parameter serial transfer model structure model is run, it is solved with each module as the basic unit, and then the dynamic characteristics of the entire ship thermal system are characterized by parameter transfer; In the fully data-driven real model structure, the mechanism model does not participate in real-time calculations, and the data-driven model completes the calculations of all modules in the system; the mechanism model generates data-driven model simulation data based on the set operating condition set, the input and output parameter set, and the boundary parameter range, and passes it to the data-driven model.

8. A real-time simulation device for ship thermal systems that uses a coordinated mechanism and data-driven model, characterized in that: include: A construction module is used to verify the initial mechanism model based on the fluid network theory using the operating data of the ship thermal system, and to construct a mechanism model that meets the accuracy requirements; and to generate simulation data based on the operating condition set, the input and output parameter set, and the boundary parameter range using the mechanism model that meets the accuracy requirements; Combining the operating data and the simulation data to form a training sample, training the initial data-driven model, and constructing a data-driven model that meets the accuracy requirements; The collaborative simulation module simultaneously carries out the simulation calculation of the mechanism model and the simulation calculation of the data-driven model in the ship thermal system, selects the appropriate model for simulation according to the real-time requirements, computing resource requirements, and model accuracy requirements, and realizes the collaborative real-time simulation of the mechanism model and the data-driven model.

9. A real-time simulation device for ship thermal systems that uses a coordinated mechanism and data-driven model, characterized in that: The real-time simulation device for a ship thermal system in which the mechanism and data-driven model are coordinated includes: a memory, a processor, and a real-time simulation program for a ship thermal system in which the mechanism and data-driven model are coordinated, which is stored in the memory and can be run on the processor. The real-time simulation program for a ship thermal system in which the mechanism and data-driven model are coordinated is configured to implement a real-time simulation method for a ship thermal system in which the mechanism and data-driven model are coordinated as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a real-time simulation program for a ship thermal system that collaborates with a mechanism and a data-driven model. When the real-time simulation program for a ship thermal system that collaborates with a mechanism and a data-driven model is executed by a processor, a real-time simulation method for a ship thermal system that collaborates with a mechanism and a data-driven model as described in any one of claims 1 to 7 is implemented.

Images