Method and system for twin operation control of power supply of lunar nuclear reactor

Abstract

The invention discloses a lunar nuclear reactor power supply twinning operation control method and system, and relates to the technical field of space nuclear reactor power supply digital twinning. Measurement data of a power source sensor of a lunar nuclear reactor are received as input or correction conditions for simulation calculation; the operation state of the lunar nuclear reactor power supply is calculated according to the received sensor measurement data, fault early warning is realized through a data driving algorithm, and if a fault is found, a fault control instruction is automatically generated by adopting a neural network control algorithm and is sent to a pre-constructed simulation reactor power supply for verification; according to the verification condition of the simulated reactor power supply, a control instruction is sent to the lunar nuclear reactor power supply through an uplink communication link, or control instruction optimization is executed on the ground; according to the invention, operation, maintenance and management functions such as lunar nuclear reactor power supply state monitoring, fault early warning and control support are realized on the ground based on a digital twinning technology, so that full-life-cycle safety guarantee of a space nuclear reactor power supply is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twin of space nuclear reactor power supply, and specifically to a twin operation and control method of lunar nuclear reactor power supply. Background Art

[0002] As a highly efficient and long-lasting energy supply technology, space nuclear reactor power supply has the characteristics of high energy density, long service life, strong environmental adaptability, and autonomous operation. It can achieve power output from kilowatts to megawatts. It is particularly suitable for space missions that are difficult to obtain solar energy or have instantaneous high-power energy requirements. It is the preferred power source for building a lunar base.

[0003] Current research mainly focuses on the development and optimization of hardware prototypes of space reactor power supplies, and research on operation and maintenance systems is relatively limited. Traditional nuclear reactor operation and maintenance mainly rely on power station operators and distributed control systems (DCS). However, traditional methods are insufficient in environmental adaptability, prediction depth, and intelligence level for the special environment of the lunar surface. Moreover, it is difficult for the lunar base to achieve long-term manned operation, and it is also difficult to build a complex DCS system on the lunar surface. For this reason, the present invention proposes a twin operation and control method and system for lunar nuclear reactor power supplies. Summary of the invention

[0004] The purpose of the present invention is to provide a twin operation and control method and system for a lunar nuclear reactor power supply, and to realize real-time status monitoring, fault warning and control support of the lunar nuclear reactor power supply based on digital twin technology.

[0005] According to the second aspect of the present invention, in order to achieve the above-mentioned purpose, the present invention provides the following technical solution: a lunar nuclear reactor power supply twin operation and control method, comprising the following steps:

[0006] Receiving lunar nuclear reactor power sensor measurement data as input or correction conditions for simulation calculations;

[0007] The operating status of the lunar nuclear reactor power supply is calculated based on the received sensor measurement data, and fault warning is realized through data-driven algorithms. If a fault is found, a neural network control algorithm is used to automatically generate fault control instructions and send them to the pre-built simulated reactor power supply for verification;

[0008] According to the verification result of the simulated reactor power supply, it is judged whether the fault control instruction is feasible. If it is feasible, the fault control instruction is sent to the lunar nuclear reactor power supply through the uplink communication link. If it is not feasible, the fault control instruction is optimized by using the control algorithm according to the verification result;

[0009] Construct a visualization platform and use it to observe the calculation of the lunar nuclear reactor power supply, the generation of fault control instructions, the verification results of fault control instructions and the optimization process.

[0010] Furthermore, the lunar nuclear reactor power sensor measures data including core temperature, heat pipe cold / hot end temperature, temperature of temperature difference / Stirling thermoelectric converter cold / hot end temperature, radiation heat sink surface temperature, output electrical power, etc.

[0011] Furthermore, the lunar nuclear reactor power sensor measurement data is received as input or correction conditions for simulation calculation, as follows:

[0012] (31) Sensors are deployed in the lunar nuclear reactor power system using a combination of invasive and non-invasive methods, and their deployment locations are tested and optimized using a pre-built simulated reactor power supply;

[0013] (32) The data collected by the sensor is transmitted wirelessly to the lunar surface information processing unit, and is filtered, compressed and stored in the data storage chip;

[0014] (33) The stored data is sent to the lunar orbit communication relay satellite via the lunar surface communication base station, and then sent to the ground twin operation and control platform. After analysis, the original measurement data can be obtained.

[0015] Furthermore, the operating status of the lunar nuclear reactor power supply is calculated based on the received sensor measurement data, and fault warning is realized through a data-driven algorithm. If a fault is found, a neural network control algorithm is used to automatically generate a fault control instruction and send it to the simulated reactor power supply for verification, as follows:

[0016] (41) preprocessing the received measurement data, and using the preprocessed measurement data as input conditions for a pre-built “digital stack power supply”;

[0017] (42) Perform multi-physics coupling calculations on the space nuclear reactor power supply, and combine the real-time data transmitted back to provide the actual operation calculation status of the lunar nuclear reactor power supply;

[0018] (43) Combined with real-time feedback data, perform real-time status calculation and fault warning, and provide input conditions for the control module;

[0019] (44) Based on the safety diagnosis input, a control command is generated using a neural network-based control method, and the control command is executed in a fully automatic or semi-automatic manner.

[0020] Furthermore, the multi-physics field coupling calculation of the space nuclear reactor power supply is performed, and combined with the real-time data transmitted back, the actual operation calculation of the lunar nuclear reactor power supply is given, as follows:

[0021] (51) Simplifying the physical design to obtain a geometric model or geometric information suitable for multi-physical field coupling calculation;

[0022] (52) Considering the integrated coupling relationship of multiple components and multiple physical fields of the space nuclear reactor power supply, a multi-physics field coupling model including neutron physics, core heat transfer, structural mechanics, heat pipe / coolant heat transfer, thermoelectric conversion, and radiation heat dissipation model is constructed;

[0023] (53) Based on fully implicit internal coupling, C++ or Python programming language, and MPI and OpenMP hybrid distributed parallel computing, refined multi-physics field coupling calculations of space heat pipe reactor power supplies are carried out.

[0024] Furthermore, combined with the real-time feedback data, real-time status calculation and fault warning are performed to provide input conditions for the control module, as follows:

[0025] (61) The main features are extracted from the preprocessed data, including the mean, standard deviation, maximum value, minimum value of field parameters such as core temperature, strain, reactivity, and the change rate of core thermal power and output electric power. The principal component analysis method is used to reduce the dimension of the features to form the training set and validation set of the model. The calculation method of PCA is shown in formula (1):

[0026] Z=XW (1)

[0027] Among them, X is the input centralized data matrix; W is the principal component matrix; Z is the data matrix after dimensionality reduction;

[0028] (62) An artificial neural network was selected to train the model, and K-fold cross validation was used to evaluate the model to ensure its generalization ability:

[0029]

[0030] Among them, K is the number of mutually exclusive subsets split from the data set; E k is the mean square error of the k-fold model; E is the average error of K-fold cross validation;

[0031] (63) Inputting the real-time feedback data into the trained model to calculate whether the non-sensor measurement area or non-measurement parameter is within the normal range in the current and future period, and performing real-time status monitoring and prediction;

[0032] (64) During the state prediction process, the prediction results are compared with the normal operating threshold or safety range set by the system. If a prediction result exceeds the normal operating threshold or safety range, it is marked as abnormal;

[0033] (65) Multi-parameter analysis is performed on abnormal conditions, and a combination of Bayesian networks and expert system rules is used to carry out fault diagnosis, determine the fault type and possible causes, and provide input for the control support module.

[0034] Furthermore, according to the verification result of the simulated reactor power supply, it is determined whether the fault control instruction is feasible. If it is feasible, the fault control instruction is sent to the lunar nuclear reactor power supply through the uplink communication link. If it is not feasible, the control algorithm is used to optimize the fault control instruction according to the verification result, as follows:

[0035] (51) Transmitting control instructions via a ground optical fiber dedicated line to a ground simulated nuclear reactor power supply to verify the control effect and determine whether the control instructions are effective based on whether the core thermal power, reactivity, and output electrical power parameters meet expectations;

[0036] (52) If the control effect meets expectations, the execution command upload is performed, the control command is converted into a binary command, and sent from the ground communication station to the lunar orbit relay satellite, and then the command is transmitted to the lunar surface communication base station, and finally reaches the lunar surface nuclear reactor power control unit to execute the corresponding control command;

[0037] (53) If the control effect does not meet expectations, the ground twin operation and control platform analyzes the cause of the control effect deviation and optimizes the fault control instructions based on the analysis results, iterating until the control effect meets expectations.

[0038] Furthermore, 6. a visualization platform is constructed to observe the calculation of lunar nuclear reactor power supply, generation of fault control instructions, verification results of fault control instructions and optimization process, as follows:

[0039] (61) Rendering the geometric structure and scene of the lunar nuclear reactor power supply and the ground simulation reactor power supply according to the design scheme, and displaying them in multiple forms such as large screen, virtual reality and augmented reality;

[0040] (62) After receiving the measurement data stream, it is stored in the ground data center and then read by the visualization platform for real-time rendering, presenting the interactive status of software-hardware and ground-space data, dynamically displaying the operating status of the lunar nuclear reactor power supply, and providing an interface for operation and maintenance personnel to intervene in the above process;

[0041] (63) In case of a fault warning, an alarm will be given in screen + sound mode, and control command generation and optimization work will be automatically executed.

[0042] According to a second aspect of the present invention, the present invention provides a lunar nuclear reactor power supply twin operation and control system, which is used to implement the above-mentioned lunar nuclear reactor power supply twin operation and control method, including a lunar nuclear reactor power supply and a ground twin operation and control platform;

[0043] The ground twin operation and control platform includes:

[0044] The analog reactor power supply subsystem is used to receive the fault control instruction generated by the digital reactor power supply subsystem and verify the control effect of the fault control instruction;

[0045] The digital reactor power subsystem is used to calculate the operating status of the lunar nuclear reactor power supply based on the received sensor measurement data, and to implement fault warning through data-driven algorithms. If a fault is found, a neural network control algorithm is used to automatically generate fault control instructions. Specifically, it includes the following four subsystems:

[0046] A simulation calculation subsystem, used for performing simulation calculation according to the received measurement data;

[0047] A data acquisition and transmission subsystem, used to receive lunar nuclear reactor power sensor measurement data as input or correction conditions;

[0048] The control and decision support subsystem is used to determine whether the fault control instruction is feasible according to the verification result of the simulated reactor power supply. If it is feasible, the fault control instruction is sent to the lunar nuclear reactor power supply through the uplink communication link. If it is not feasible, the control algorithm is used to optimize the fault control instruction according to the verification result.

[0049] The visualization subsystem is used to build a visualization platform, which is used to observe the calculation of the lunar nuclear reactor power supply, the generation of fault control instructions, the verification results of fault control instructions and the optimization process.

[0050] Furthermore, the data acquisition and transmission subsystem includes a data acquisition module, a data processing and storage module, and a data transmission module, wherein the data acquisition module is configured as a plurality of sensors and is arranged in the lunar nuclear reactor power supply;

[0051] The simulation computing subsystem includes a numerical computing module based on physical models, a fast computing module based on data drive, a software interface module combining software and hardware, and a digital twin module for measuring data correction.

[0052] The simulated reactor power supply subsystem includes a ground reactor power supply module, an environmental simulation module and a load simulation module;

[0053] The control and decision support subsystem includes intelligent prediction and diagnosis module, control strategy optimization module and execution module;

[0054] The visualization subsystem includes data management module, user interface module and human-computer interaction module.

[0055] The present invention has at least the following beneficial effects:

[0056] The present invention provides a method for operating, maintaining and managing a lunar nuclear reactor power supply on the ground based on digital twin technology, providing status monitoring, fault warning and decision support for space reactor safety management, and providing full life cycle safety protection for the reactor power supply control system. The ground twin operation and control system and the autonomous control system of the lunar nuclear reactor power supply involved in the present invention are combined with each other, which can provide double protection for the engineering application of nuclear reactor power supply and assist the first nuclear reactor power supply engineering application of the International Lunar Research Station. In actual engineering applications, compared with the existing technology, the present invention still needs to overcome technical difficulties such as high-precision multi-scale multi-physics coupling simulation, complex multi-physics field coupling real-time calculation, advanced sensor and multi-source data fusion algorithm, and dynamic update and adaptation of digital twin models.

[0057] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a schematic diagram of the flow of the operation control method of the present invention;

[0059] Figure 2 It is a schematic diagram of the operation control principle of the operation control method of the present invention;

[0060] Figure 3 It is a structural schematic diagram of the operation and control system of the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0062] See also Figure 1 The present invention provides a technical solution: a lunar nuclear reactor power supply twin operation and control method, comprising the following steps:

[0063] S1. Receive the lunar nuclear reactor power sensor measurement data as input or correction conditions for simulation calculation, as follows:

[0064] S11. The sensors are deployed in the lunar nuclear reactor power system in a combination of invasive and non-invasive ways. The deployment location needs to be tested and optimized by the ground twin operation and control platform to ensure that the weight and volume of the entire power system are as small as possible;

[0065] S12. The collected data is wirelessly transmitted to the lunar intelligent information processing unit, and is filtered, compressed and stored in the data storage chip;

[0066] S13. The stored data is sent to the lunar orbit communication relay satellite via the lunar surface communication base station, and then sent to the ground receiving station. After analysis, the original measurement data can be obtained;

[0067] It should be noted that the measured data include core temperature, core deformation, cold / hot end temperature of heat pipe, cold / hot end temperature of temperature difference / Stirling type thermoelectric converter, surface temperature of radiation heat sink, output electric power, etc.

[0068] S2. Calculate the operating status of the lunar nuclear reactor power supply based on the received sensor measurement data, and implement fault warning through data-driven algorithms. If a fault is found, a neural network control algorithm is used to automatically generate fault control instructions and send them to the pre-built simulated reactor power supply for verification, as follows:

[0069] S21. Preprocess the received measurement data, including data cleaning, classification, time series processing, normalization and standardization, and use the preprocessed measurement data as input conditions for the pre-built "digital stack power supply";

[0070] S22. Perform multi-physics field coupling calculations on the space nuclear reactor power supply, and combine the real-time data sent back to provide the actual operation calculations of the lunar nuclear reactor power supply. The specific steps are as follows:

[0071] S221. Simplify the physical design to obtain a geometric model or geometric information suitable for multi-physics field coupling calculation;

[0072] S222. Consider the integrated coupling relationship of multiple components and multiple physical fields of space nuclear reactor power supply, and construct a multi-physics field coupling model including neutron physics, core heat transfer, structural mechanics, heat pipe / coolant heat transfer, thermoelectric conversion, radiation heat dissipation and other models;

[0073] S223. Based on fully implicit internal coupling, C++, Python and other languages ​​are used for programming, and MPI and OpenMP hybrid distributed parallel computing are used to carry out refined multi-physics coupling calculation of space heat pipe reactor power supply;

[0074] S23. Combined with the real-time feedback data, perform real-time status calculation and fault warning to provide input conditions for the control module. The specific steps are as follows:

[0075] S231. Extract the main features from the preprocessed data, including the mean, standard deviation, maximum value, minimum value of field parameters such as core temperature, strain, reactivity, and the rate of change of core thermal power and output power, etc. Use principal component analysis (PCA) and other methods to reduce the dimension of the features to form the training set and validation set of the model; the calculation method of PCA is shown in formula (1):

[0076] Z=XW (1)

[0077] Among them, X is the input centralized data matrix; W is the principal component matrix; Z is the data matrix after dimensionality reduction;

[0078] S232. Select an artificial neural network (ANN) to train the model and use K-fold cross validation (such as formula (2)) to evaluate the model to ensure its generalization ability:

[0079]

[0080] Among them, K is the number of mutually exclusive subsets split from the data set; E k is the mean square error of the k-fold model; E is the average error of K-fold cross validation;

[0081] S233. Input the real-time returned data into the trained model to calculate whether the non-sensor measurement area or non-measurement parameter is within the normal range in the current and future period, and perform real-time monitoring and prediction of the status;

[0082] S234. During the state prediction process, the prediction result is compared with the normal operation threshold or safety range set by the system. Once a prediction result exceeds the normal operation threshold or safety range, it is marked as abnormal;

[0083] S235. Perform multi-parameter analysis on abnormal conditions, use Bayesian network combined with expert system rules to carry out fault diagnosis, determine the fault type and possible cause, and provide input for the control support module;

[0084] S24. Generate control instructions based on safety diagnosis input using a neural network-based control method, and execute the control instructions in a fully automatic or semi-automatic manner;

[0085] S3. According to the verification result of the simulated reactor power supply, determine whether the fault control instruction is feasible. If it is feasible, send the fault control instruction to the lunar nuclear reactor power supply through the uplink communication link. If it is not feasible, optimize the fault control instruction using the control algorithm according to the verification result, as follows:

[0086] S31. Based on the output of the control algorithm and the experience of ground operation and maintenance personnel, the control action sequence is modified from the perspective of safety;

[0087] S32. Transmit the control command from the ground optical fiber dedicated line to the ground simulated nuclear reactor power supply to verify the control effect. The effectiveness of the control command is determined by whether the core thermal power, reactivity, and output electrical power parameters meet expectations. To ensure the timeliness of the control effect, the start-up time of the ground simulated nuclear reactor power supply should be earlier than that of the lunar nuclear reactor power supply.

[0088] S33. If the control effect meets expectations, the execution command upload is performed, the control command is converted into a binary command, and sent from the ground communication station to the lunar orbit relay satellite, and then the command is transmitted to the lunar surface communication base station, and finally reaches the lunar surface nuclear reactor power control unit to execute the corresponding control command;

[0089] S34. If the control effect does not meet expectations, the ground twin operation and control platform analyzes the cause of the control effect deviation, and optimizes the fault control instructions based on the analysis results, and iterates until the control effect meets expectations;

[0090] S4. Construct a visualization platform to observe the calculation of lunar nuclear reactor power supply, generation of fault control instructions, verification results of fault control instructions and optimization process, and provide an interface for operation and maintenance personnel to intervene in the specific process, as follows:

[0091] S41. Render the geometric structure and scene of the lunar nuclear reactor power supply and the ground simulation reactor power supply according to the design scheme, and display them in multiple forms such as large screen, virtual reality and augmented reality;

[0092] S42. After receiving the measurement data stream, it is stored in the ground data center, and then read by the visualization platform for real-time rendering, presenting the interactive status of software-hardware, ground-space data, and dynamically displaying the operating status of the lunar nuclear reactor power supply;

[0093] S43. In case of a fault warning, an alarm will be given in screen + sound mode, and the system will automatically connect to the ground twin operation and control platform to execute control instructions and their optimization, and allow operation and maintenance personnel to intervene in the above S1 to S4 processes.

[0094] Next, the technical solution of the present invention is further described in conjunction with the accompanying drawings:

[0095] The operation and control method described in this embodiment mainly involves the ground twin operation and control platform, the lunar nuclear reactor power supply and other auxiliary systems. The reactor power supply twin operation and control platform includes software and hardware parts. The software part mainly realizes the state prediction, fault warning and control instruction generation of the lunar nuclear reactor power supply through geometric and environmental modeling, numerical simulation, artificial intelligence and other means. It can also be called "digital reactor power supply"; the hardware part is mainly the ground simulation reactor power supply. By simulating its operating environment and load characteristics on the ground, a device that is almost the same as the lunar nuclear reactor power supply is created. It can also be called "companion" reactor power supply. Its startup time is earlier than that of the lunar nuclear reactor power supply. The purpose is to leave enough time for the verification process of the control instructions, thereby improving the timeliness of fault warning and control. Other auxiliary systems mainly include ground supercomputers, ground communication systems, lunar communication systems and other modules that perform data collection and transmission;

[0096] like Figure 2 As shown in the figure, the ground twin operation and control platform mainly includes the ground digital space reactor power supply and the ground analog reactor power supply. The measurement data of various sensors in the lunar nuclear reactor power supply system are sent to the ground twin operation and control platform through the downlink communication link, providing calculation input or correction conditions for the ground twin operation and control platform. The digital reactor power supply of the ground twin operation and control platform calculates the status of the lunar nuclear reactor power supply in real time, and realizes fault warning through intelligent algorithms. If a fault is found, a control instruction is automatically generated and sent to the analog reactor power supply of the ground twin operation and control platform for verification. If the control algorithm is confirmed to be feasible, the ground twin operation and control platform will generate a control instruction and send it to the lunar nuclear reactor power supply through the uplink communication link in an automatic or semi-automatic form. If it is not feasible, the ground twin operation and control platform will optimize the control algorithm according to the control result.

[0097] Embodiment 2:

[0098] like Figure 3 As shown, the present invention provides a lunar nuclear reactor power supply twin operation and control system, which is used to implement the lunar nuclear reactor power supply twin operation and control method described in Example 1, including a lunar nuclear reactor power supply and a ground twin operation and control platform;

[0099] The ground twin operation and control platform includes:

[0100] The analog reactor power supply subsystem is used to receive the fault control instruction generated by the digital reactor power supply subsystem and verify the control effect of the fault control instruction;

[0101] The digital reactor power subsystem is used to calculate the operating status of the lunar nuclear reactor power supply based on the received sensor measurement data, and to implement fault warning through intelligent algorithms. If a fault is found, the control algorithm is used to automatically generate fault control instructions. Specifically, it includes the following four subsystems:

[0102] A simulation calculation subsystem, used for performing simulation calculation according to the received measurement data;

[0103] A data acquisition and transmission subsystem, used to receive lunar nuclear reactor power sensor measurement data as input or correction conditions;

[0104] The control and decision support subsystem is used to determine whether the fault control instruction is feasible according to the verification result of the simulated reactor power supply. If it is feasible, the fault control instruction is sent to the lunar nuclear reactor power supply through the uplink communication link. If it is not feasible, the control algorithm is used to optimize the fault control instruction according to the verification result.

[0105] The visualization subsystem is used to build a visualization platform, which is used to observe the calculation of the lunar nuclear reactor power supply, the generation of fault control instructions, the verification results of fault control instructions and the optimization process, and provides an interface for operation and maintenance personnel to intervene in the above process.

[0106] Furthermore, the data acquisition and transmission subsystem includes a data acquisition module, a data processing and storage module, and a data sending module. It is a subsystem that executes the data transmission between the ground and the sky of the twin operation and control platform, and is the key to the state synchronization between the ground and the lunar surface. The data acquisition module is composed of multiple advanced sensors and is arranged in the reactor power system. It is necessary to consider the weight and volume limitations of the sensors to avoid adding more burdens to the payload weight, and to consider the arrangement method to avoid adding safety risks to the normal operation of the reactor power system, and try not to increase the sensor configuration of the original control system of the reactor power, and try to achieve non-invasive measurement.

[0107] The onboard data processing module mainly performs data compression tasks, and generally uses FPGA such as 690T, 325T; the onboard data storage module generally uses FLASH memory or AI chip, such as YULONG810A, which can perform compression and storage operations; the ground-side data processing mainly involves analysis, cleaning and other operations, and generally uses GPU servers; the onboard data transmission module refers to encoding, modulation, amplification, antenna transmission systems, etc.

[0108] The simulation computing subsystem includes a numerical computing module based on a physical model, a fast computing module based on data drive, a software interface module combining software and hardware, and a digital twin module for actual data correction. Its overall form is a set of software systems. The simulation computing subsystem is the key to the ground operation and maintenance system to achieve the "knowable" state of the reactor power supply. The so-called "knowable" means that the parameters related to the safety of reactor operation are mastered in real time, including temperature, power, reactivity, stress, strain and other parameters. This requires calculation and analysis of the reactor based on the existing mathematical and physical models, or solving some problems where the physical laws are unclear through data drive. At the same time, simulation calculations usually assume boundary conditions to analyze a problem, but the actual operation may not be completely consistent with the assumed boundary conditions. Therefore, some data needs to be sent back in real time to update the boundary conditions. In addition, the residual of the simulation calculation results and the actual operation status can also be used as the basis for data-driven algorithm learning, so as to continuously correct the data-driven algorithm and form a continuously developing and self-learning digital twin model.

[0109] The simulated reactor power subsystem includes a ground reactor power module, an environmental simulation module, and a load simulation module. It consists of six parts: a ground environment simulation device, a reactor body, a radiation shield, a thermoelectric conversion system, a waste heat discharge system, and an automatic control system. It is a device that is almost the same as the lunar nuclear reactor power supply by simulating its operating environment (including thermal environment, vacuum environment, etc.) and load characteristics on the ground. It can also be called a "companion flight" reactor power supply. Its startup time is earlier than that of the lunar nuclear reactor power supply. The purpose is to leave enough time for the verification of control instructions, thereby improving the timeliness of fault warning and control.

[0110] The control and decision support subsystem includes the intelligent prediction and diagnosis module, the control strategy optimization module and the execution module, which are the key to the twin operation and control platform to achieve reactor status prediction and safe control. The intelligent prediction and diagnosis module will carry out research on intelligent algorithms such as deep learning and machine learning based on the reactor power supply operation data, provide fault warning and safety diagnosis functions, and provide input conditions for the control module. The control strategy optimization and execution module mainly calculates the corresponding control strategy and execution instructions based on the safety diagnosis input, and executes the control instructions in a fully automatic or semi-automatic form;

[0111] The visualization subsystem includes a data management module, a user interface module, and a human-computer interaction module. The visualization subsystem displays the operating status of physical entities and virtual entities and the data transmission process in a visual form in the operation and maintenance control hall, which facilitates the operation and maintenance personnel to intuitively understand the system operation status and provides an interface for the operation and maintenance personnel to intervene in the above process. The system is mainly composed of a data management module, a user interface module, and a human-computer interaction module. It supports functions such as geometric modeling and physical field display in different dimensions, and has advanced visualization functions such as virtual reality (VR) and augmented reality (AR).

[0112] In summary, the present invention can realize real-time monitoring of the status of the lunar nuclear reactor power supply, fault warning and control support through the ground twin operation and control platform, the lunar nuclear reactor power supply and other auxiliary systems, which is an important means to ensure the success of the nuclear reactor engineering mission of the International Lunar Research Station.

[0113] Embodiment three:

[0114] Specifically, the above-mentioned data acquisition and transmission subsystem, simulation calculation subsystem, control and decision support subsystem and visualization subsystem can be partially embedded in the computer processing system. The computer calls the above-mentioned modules to complete the tasks of status monitoring, fault warning and control support for the lunar nuclear reactor power supply based on the above-mentioned twin operation and control method of the lunar nuclear reactor power supply; the above-mentioned data acquisition and transmission subsystem, simulation calculation subsystem, simulation reactor power supply subsystem, control and decision support subsystem and visualization subsystem can perform operations according to the specific steps given in the twin operation and control method of the lunar nuclear reactor power supply.

[0115] It should be noted that it should be understood that the division of the various modules of the above system is only the division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated, and these modules can all be implemented in the form of software calling through processing elements; they can also be all implemented in the form of hardware; some modules can also be implemented in the form of software calling through processing elements, and some modules can be implemented in the form of hardware. For example, the data acquisition and transmission subsystem can be a separately established processing element, or it can be integrated in a chip of the above device. In addition, it can also be stored in the memory of the above device in the form of program code, and the functions of the above signal processing module can be called and executed by a processing element of the above device. The implementation of other modules is similar. In addition, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module can be completed by the hardware integrated logic circuit in the processor element or the instructions in the form of software.

[0116] For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASIC), or one or more digital singnal processors (DSP), or one or more field programmable gate arrays (FPGA). For another example, when a module is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processors that can call program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0117] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0118] For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. When an element is referred to as being "assembled on", "installed on", "fixed on" or "set on" another element, it can be directly on the other element or there can also be a centered element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be a centered element at the same time. The terms "vertical", "horizontal", "up", "down", "left", "right" and similar expressions used herein are only for illustrative purposes and are not intended to be the only implementation method.

[0119] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0120] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

Claims

1. A lunar nuclear reactor power twin control method, characterized in that: The following steps are involved: Receiving lunar nuclear reactor power sensor measurement data as input or correction conditions for simulation calculations; The operating status of the lunar nuclear reactor power supply is calculated based on the received sensor measurement data, and fault warning is realized through data-driven algorithms. If a fault is found, a neural network control algorithm is used to automatically generate fault control instructions and send them to the pre-built simulated reactor power supply for verification; According to the verification result of the simulated reactor power supply, it is judged whether the fault control instruction is feasible. If it is feasible, the fault control instruction is sent to the lunar nuclear reactor power supply through the uplink communication link. If it is not feasible, the fault control instruction is optimized by using the control algorithm according to the verification result; Construct a visualization platform and use it to observe the calculation of the lunar nuclear reactor power supply, the generation of fault control instructions, the verification results of fault control instructions and the optimization process.

2. The method for controlling the twin power supply of a lunar nuclear reactor according to claim 1, characterized in that: The data measured by the lunar nuclear reactor power supply sensor include core temperature, core deformation, heat pipe cold / hot end temperature, temperature of the cold and hot ends of the temperature difference / Stirling type thermoelectric converter, radiation heat sink surface temperature, and output electric power.

3. The method for controlling the twin power supply of a lunar nuclear reactor according to claim 2, characterized in that: Receive the lunar nuclear reactor power sensor measurement data as input or correction conditions for simulation calculations, as follows: (31) Sensors are deployed in the lunar nuclear reactor power system using a combination of invasive and non-invasive methods, and their deployment locations are tested and optimized using a pre-built simulated reactor power supply; (32) The data collected by the sensor is transmitted wirelessly to the lunar surface information processing unit, and is filtered, compressed and stored in the data storage chip; (33) The stored data is sent to the lunar orbit communication relay satellite via the lunar surface communication base station, and then sent to the ground twin operation and control platform. After analysis, the original measurement data can be obtained.

4. The method for controlling the twin power supply of a lunar nuclear reactor according to claim 3, characterized in that: The operating status of the lunar nuclear reactor power supply is calculated based on the received sensor measurement data, and fault warning is realized through data-driven algorithms. If a fault is found, a neural network control algorithm is used to automatically generate fault control instructions and send them to the simulated reactor power supply for verification, as follows: (41) preprocessing the received measurement data, and using the preprocessed measurement data as input conditions for a pre-built "digital stack power supply"; (42) Perform multi-physics coupling calculations on the space nuclear reactor power supply, and combine the real-time data transmitted back to provide the actual operation calculation status of the lunar nuclear reactor power supply; (43) Combined with real-time feedback data, perform real-time status calculation and fault warning, and provide input conditions for the control module; (44) Based on the safety diagnosis input, a control command is generated using a neural network-based control method, and the control command is executed in a fully automatic or semi-automatic manner.

5. The method for controlling the twin power supply of a lunar nuclear reactor according to claim 4, characterized in that: Perform multi-physics coupling calculations of the space nuclear reactor power supply, and combine the real-time data sent back to give the actual operation calculation of the lunar nuclear reactor power supply, as follows: (51) Simplifying the physical design to obtain a geometric model or geometric information suitable for multi-physical field coupling calculation; (52) Considering the integrated coupling relationship of multiple components and multiple physical fields of the space nuclear reactor power supply, a multi-physics field coupling model including neutron physics, core heat transfer, structural mechanics, heat pipe / coolant heat transfer, thermoelectric conversion, and radiation heat dissipation model is constructed; (53) Based on fully implicit internal coupling, C++ or Python programming language, and MPI and OpenMP hybrid distributed parallel computing, refined multi-physics field coupling calculations of space heat pipe reactor power supplies are carried out.

6. The method for controlling the twin power supply of a lunar nuclear reactor according to claim 4, characterized in that: Combined with real-time feedback data, real-time status calculation and fault warning are performed to provide input conditions for the control module, as follows: (61) The main features are extracted from the preprocessed data, including the mean, standard deviation, maximum value, minimum value of field parameters such as core temperature, strain, reactivity, and the change rate of core thermal power and output electric power. The principal component analysis method is used to reduce the dimension of the features to form the training set and validation set of the model. The calculation method of PCA is shown in formula (1): Z=XW (1) Among them, X is the input centralized data matrix; W is the principal component matrix; Z is the data matrix after dimensionality reduction; (62) An artificial neural network was selected to train the model, and K-fold cross validation was used to evaluate the model to ensure its generalization ability: Among them, K is the number of mutually exclusive subsets split from the data set; E k is the mean square error of the k-fold model; E is the average error of K-fold cross validation; (63) Inputting the real-time feedback data into the trained model to calculate whether the non-sensor measurement area or non-measurement parameter is within the normal range in the current and future period, and performing real-time status monitoring and prediction; (64) During the state prediction process, the prediction results are compared with the normal operating threshold or safety range set by the system. If a prediction result exceeds the normal operating threshold or safety range, it is marked as abnormal; (65) Multi-parameter analysis is performed on abnormal conditions, and a combination of Bayesian networks and expert system rules is used to carry out fault diagnosis, determine the fault type and possible causes, and provide input for the control support module.

7. The method for controlling the twin power supply of a lunar nuclear reactor according to claim 6, characterized in that: According to the verification results of the simulated reactor power supply, it is judged whether the fault control instruction is feasible. If it is feasible, the fault control instruction is sent to the lunar nuclear reactor power supply through the uplink communication link. If it is not feasible, the control algorithm is used to optimize the fault control instruction according to the verification results, as follows: (51) Transmitting control instructions via a ground optical fiber dedicated line to a ground simulated nuclear reactor power supply to verify the control effect and determine whether the control instructions are effective based on whether the core thermal power, reactivity, and output electrical power parameters meet expectations; (52) If the control effect meets expectations, the execution command upload is performed, the control command is converted into a binary command, and sent from the ground communication station to the lunar orbit relay satellite, and then the command is transmitted to the lunar surface communication base station, and finally reaches the lunar surface nuclear reactor power control unit to execute the corresponding control command; (53) If the control effect does not meet expectations, the ground twin operation and control platform analyzes the cause of the control effect deviation and optimizes the fault control instructions based on the analysis results, iterating until the control effect meets expectations.

8. The method for controlling the twin power supply of a lunar nuclear reactor according to claim 7, characterized in that: Construct a visualization platform and use it to observe the calculation of lunar nuclear reactor power supply, generation of fault control instructions, verification results of fault control instructions and optimization process, as follows: (61) Rendering the geometric structure and scene of the lunar nuclear reactor power supply and the ground simulation reactor power supply according to the design scheme, and displaying them in multiple forms such as large screen, virtual reality and augmented reality; (62) After receiving the measurement data stream, it is stored in the ground data center and then read by the visualization platform for real-time rendering, presenting the interactive status of software-hardware and ground-space data, and dynamically displaying the operating status of the lunar nuclear reactor power supply; (63) In case of a fault warning, an alarm will be given in screen + sound mode, and the system will automatically connect to the ground twin operation and control platform to execute control instructions and their optimization, and allow operation and maintenance personnel to intervene in the above process.

9. A lunar nuclear reactor power supply twin operation and control system, used to implement the lunar nuclear reactor power supply twin operation and control method according to any one of claims 1 to 8, characterized in that: Including lunar nuclear reactor power supply and ground twin operation and control platform; The ground twin operation and control platform includes: The analog reactor power supply subsystem is used to receive the fault control instruction generated by the digital reactor power supply subsystem and verify the control effect of the fault control instruction; The digital reactor power subsystem is used to calculate the operating status of the lunar nuclear reactor power supply based on the received sensor measurement data, and to implement fault warning through intelligent algorithms. If a fault is found, the control algorithm is used to automatically generate fault control instructions. Specifically, it includes the following four subsystems: A simulation calculation subsystem, used for performing simulation calculation according to the received measurement data; A data acquisition and transmission subsystem, used to receive lunar nuclear reactor power sensor measurement data as input or correction conditions; The control and decision support subsystem is used to determine whether the fault control instruction is feasible according to the verification result of the simulated reactor power supply. If it is feasible, the fault control instruction is sent to the lunar nuclear reactor power supply through the uplink communication link. If it is not feasible, the control algorithm is used to optimize the fault control instruction according to the verification result. The visualization subsystem is used to build a visualization platform, which is used to observe the calculation of the lunar nuclear reactor power supply, the generation of fault control instructions, the verification results of fault control instructions and the optimization process.

10. The lunar nuclear reactor power supply twin operation and control system according to claim 9, characterized in that: The data acquisition and transmission subsystem includes a data acquisition module, a data processing and storage module, and a data transmission module, wherein the data acquisition module is configured as a plurality of sensors and arranged in the lunar nuclear reactor power supply; The simulation computing subsystem includes a numerical computing module based on physical models, a fast computing module based on data drive, a software interface module combining software and hardware, and a digital twin module for measuring data correction. The simulated reactor power supply subsystem includes a ground reactor power supply module, an environmental simulation module and a load simulation module; The control and decision support subsystem includes intelligent prediction and diagnosis module, control strategy optimization module and execution module; The visualization subsystem includes data management module, user interface module and human-computer interaction module.

Images