Optimization method and system for energy loss of a photo-thermal electrolysis hydrogen production system

Abstract

The present application relates to the technical field of data processing, and discloses a kind of photothermal electrolytic hydrogen production system energy loss optimization method and system, the method includes: based on the multi-source heterogeneous data of photothermal electrolytic hydrogen production system, the model initialization of photothermal electrolytic hydrogen production system is carried out, to construct digital twin model;The energy flow of photothermal electrolytic hydrogen production system is simulated by digital twin model, and energy flow distribution diagram is obtained;Energy loss point in energy flow distribution diagram is identified, and based on the root cause analysis report of energy loss point is analyzed to the energy balance equation and efficiency optimization criterion of photothermal electrolytic hydrogen production system;Optimization strategy generation is carried out to photothermal electrolytic hydrogen production system, and the correction parameter of optimization strategy is mapped to digital twin model in real time;When the energy loss of digital twin model reaches minimum loss constraint, the correction parameter corresponding to optimization strategy is output as target optimization strategy;The present application can improve the energy loss optimization efficiency of photothermal electrolytic hydrogen production system.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for optimizing energy loss in a photothermal electrolysis hydrogen production system. Background Technology

[0002] The lack of efficient integration solutions for the multi-source heterogeneous data generated by the operation of the photothermal electrolysis hydrogen production system makes it difficult to systematically extract dynamic behavioral characteristics closely related to energy loss. This leads to biases in the perception of the system's operating status and fails to provide comprehensive and accurate basic support for energy loss analysis, directly restricting the effectiveness of loss identification.

[0003] Existing energy loss optimization methods lack accurate system simulation and in-depth analysis capabilities, making it impossible to dynamically track energy transfer paths and locate core loss nodes. Furthermore, the optimization strategies fail to consider both energy efficiency improvement and operational stability, resulting in insufficient targeting of optimization measures. This makes it difficult to minimize system energy loss and may also cause system operational fluctuations, affecting the continuity and overall efficiency of the hydrogen production process. Therefore, how to improve the energy loss optimization efficiency of the photothermal electrolysis hydrogen production system has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and system for optimizing energy loss in a photothermal electrolysis hydrogen production system to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for optimizing energy loss in a photothermal electrolysis hydrogen production system, comprising:

[0006] S1. Collect real-time physical parameters and historical operating data of the photothermal electrolysis hydrogen production system through the Internet of Things platform to obtain multi-source heterogeneous data of the photothermal electrolysis hydrogen production system.

[0007] S2. Based on the multi-source heterogeneous data, the photothermal electrolysis hydrogen production system is initialized to construct a digital twin model of the photothermal electrolysis hydrogen production system.

[0008] S3. Simulate the energy flow of the photothermal electrolysis hydrogen production system using the digital twin model to obtain the energy flow distribution map of the photothermal electrolysis hydrogen production system;

[0009] S4. Identify the energy loss points in the energy flow distribution diagram, and analyze the root cause analysis report of the energy loss points based on the energy balance equation and efficiency optimization criteria of the photothermal electrolysis hydrogen production system;

[0010] S5. Based on the root cause analysis report, generate an optimization strategy for the photothermal electrolysis hydrogen production system, and map the correction parameters of the optimization strategy to the digital twin model in real time.

[0011] S6. When the energy loss of the digital twin model reaches the minimum loss constraint, the correction parameters corresponding to the optimization strategy are output as the target optimization strategy of the photothermal electrolysis hydrogen production system.

[0012] In a preferred embodiment, the step of collecting real-time physical parameters and historical operating data of the photothermal electrolysis hydrogen production system through an Internet of Things (IoT) platform to obtain multi-source heterogeneous data of the photothermal electrolysis hydrogen production system includes:

[0013] Distributed data integration of multi-source heterogeneous data is performed to obtain a unified data platform for the photothermal electrolysis hydrogen production system.

[0014] Through the unified data platform, the dynamic behavior characteristics of the photothermal electrolysis hydrogen production system that are highly related to energy loss are analyzed to obtain the feature description set of the photothermal electrolysis hydrogen production system.

[0015] The feature description set is loaded into a predefined model template and initial parameters are configured to construct a digital twin model of the photothermal electrolysis hydrogen production system.

[0016] In a preferred embodiment, simulating the energy flow of the photothermal electrolysis hydrogen production system using the digital twin model to obtain an energy flow distribution map of the photothermal electrolysis hydrogen production system includes:

[0017] After inputting the current system state parameters of the photothermal electrolysis hydrogen production system into the digital twin model, the energy transfer path in the digital twin model is dynamically tracked to obtain the energy flow trajectory data of the photothermal electrolysis hydrogen production system.

[0018] By identifying the energy conversion nodes and loss nodes in the energy flow trajectory data, the node energy distribution dataset of the photothermal electrolysis hydrogen production system is obtained.

[0019] Based on the energy flow direction and intensity gradient in the node energy distribution dataset, an energy flow distribution map of the photothermal electrolysis hydrogen production system is generated.

[0020] In a preferred embodiment, identifying energy conversion nodes and loss nodes in the energy flow trajectory data to obtain the node energy distribution dataset of the photothermal electrolysis hydrogen production system includes:

[0021] Extract the energy flow rate change rate and conversion efficiency index from the energy flow trajectory data to obtain the node feature set of the photothermal electrolysis hydrogen production system;

[0022] By comparing the input and output differences of node energy in the node feature set, the energy anomaly nodes of the photothermal electrolysis hydrogen production system are obtained.

[0023] The node type and corresponding energy value of the energy anomaly nodes are labeled to generate a node energy distribution dataset for the photothermal electrolysis hydrogen production system.

[0024] In a preferred embodiment, identifying energy loss points in the energy flow distribution map includes:

[0025] Based on the energy intensity gradient change, the energy flow distribution map is divided into multiple analysis regions to obtain the regional energy distribution feature set of the photothermal electrolysis hydrogen production system.

[0026] Based on a preset loss mode feature library, suspicious regions that conform to abnormal energy loss characteristics in the energy distribution feature set of the region are identified, and a set of suspicious regions of the photothermal electrolysis hydrogen production system is obtained.

[0027] Spatial correlation analysis was performed on the set of suspicious regions to obtain the energy loss points of the photothermal electrolysis hydrogen production system.

[0028] In a preferred embodiment, the root cause analysis report analyzing the energy loss point based on the energy balance equation and efficiency optimization criteria of the photothermal electrolysis hydrogen production system includes:

[0029] By traversing the spatial distribution information of the energy loss points and the topological relationship of the system components, a component association map of the energy loss points is constructed.

[0030] Multi-dimensional rule matching is performed on the component association graph, and key components and interaction relationships that cause energy loss are identified through parallel rule reasoning to generate the component interaction network of the energy loss points.

[0031] Based on the deviation between the real-time operating data and the baseline parameters at the energy loss points, the efficiency bottleneck of energy transfer between components is analyzed to obtain a root cause analysis report of the energy loss points.

[0032] In a preferred embodiment, the step of generating an optimization strategy for the photothermal electrolysis hydrogen production system based on the root cause analysis report, and mapping the correction parameters of the optimization strategy to the digital twin model in real time, includes:

[0033] Based on the efficiency bottleneck type in the root cause analysis report, the corresponding optimization strategy template is retrieved from the preset optimization strategy library to obtain the preliminary optimization scheme of the photothermal electrolysis hydrogen production system.

[0034] Based on the current operating status of the digital twin model, the parameters of the preliminary optimization scheme are adjusted for adaptation.

[0035] Under stability constraints, the parameters for parameter adaptation and adjustment are optimized using multiple objectives to obtain the modified parameter set of the photothermal electrolysis hydrogen production system;

[0036] The modified parameter set is dynamically loaded into the runtime environment of the digital twin model.

[0037] In a preferred embodiment, the parameters for parameter adaptation and adjustment under stability constraints are subjected to multi-objective optimization to obtain a set of corrected parameters for the photothermal electrolysis hydrogen production system, including:

[0038] The optimization framework is based on multiple objectives, including energy efficiency improvement and operational stability.

[0039] In the multi-objective optimization framework, the parameter space is explored for the parameters that are adapted and adjusted under stability constraints.

[0040] By identifying candidate parameter combinations that satisfy the multi-objective optimization framework through Pareto front analysis, the optimal parameter solution set of the photothermal electrolysis hydrogen production system is obtained.

[0041] The optimized parameter set that operates stably is used as the correction parameter set for the photothermal electrolysis hydrogen production system.

[0042] In a preferred embodiment, the multi-objective optimization is calculated as follows: ;

[0043] In the formula, For a multi-objective optimization function, The weighting factor corresponding to the energy efficiency improvement target. The weighting factor corresponding to the stable operation target. This is the baseline energy loss value for the digital twin model. The energy loss prediction value of the digital twin model after adapting and adjusting the parameters. for, The current system stability of the digital twin model after the parameters have been adapted and adjusted. The system baseline operational stability of the digital twin model is used as the benchmark. To address the above problems, the present invention also provides an energy loss optimization system for a photothermal electrolysis hydrogen production system, the system comprising:

[0044] The data acquisition and integration module is used to collect real-time physical parameters and historical operating data of the photothermal electrolysis hydrogen production system through the Internet of Things platform, so as to obtain multi-source heterogeneous data of the photothermal electrolysis hydrogen production system.

[0045] The digital twin modeling module is used to initialize the model of the photothermal electrolysis hydrogen production system based on the multi-source heterogeneous data, so as to construct a digital twin model of the photothermal electrolysis hydrogen production system.

[0046] The energy flow simulation module is used to simulate the energy flow of the photothermal electrolysis hydrogen production system through the digital twin model, and obtain the energy flow distribution map of the photothermal electrolysis hydrogen production system;

[0047] The energy loss point identification and root cause analysis module is used to identify energy loss points in the energy flow distribution diagram and analyze the root cause analysis report of the energy loss points based on the energy balance equation and efficiency optimization criteria of the photothermal electrolysis hydrogen production system.

[0048] The optimization strategy generation and parameter mapping module is used to generate an optimization strategy for the photothermal electrolysis hydrogen production system based on the root cause analysis report, and to map the correction parameters of the optimization strategy to the digital twin model in real time.

[0049] The target optimization strategy output module is used to output the correction parameters corresponding to the optimization strategy as the target optimization strategy of the photothermal electrolysis hydrogen production system when the energy loss of the digital twin model reaches the minimum loss constraint.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. This invention dynamically tracks energy transfer paths using a digital twin model, accurately locates loss points by combining a loss mode feature library with spatial correlation analysis, and delves into the root causes of loss by using component correlation maps, significantly improving the efficiency and accuracy of energy loss identification and providing a reliable basis for optimization.

[0052] 2. This invention uses a multi-objective optimization framework to balance energy efficiency and operational stability. The optimal correction parameters are selected through Pareto front analysis and mapped to the model in real time. This minimizes energy loss, avoids system operation fluctuations, ensures continuous hydrogen production, and significantly improves the overall operating efficiency of the system. Attached Figure Description

[0053] Figure 1 This is a schematic flowchart illustrating an energy loss optimization method for a photothermal electrolysis hydrogen production system according to an embodiment of the present invention.

[0054] Figure 2 A functional block diagram of an energy loss optimization system for a photothermal electrolysis hydrogen production system provided in an embodiment of the present invention;

[0055] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0056] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0057] This application provides a method for optimizing energy loss in a photothermal electrolysis hydrogen production system. The execution entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for optimizing energy loss in a photothermal electrolysis hydrogen production system can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0058] Reference Figure 1 The diagram shown is a flowchart illustrating a method for optimizing energy loss in a photothermal electrolysis hydrogen production system according to an embodiment of the present invention. In this embodiment, the method for optimizing energy loss in a photothermal electrolysis hydrogen production system includes:

[0059] S1. Collect real-time physical parameters and historical operating data of the photothermal electrolysis hydrogen production system through the Internet of Things platform to obtain multi-source heterogeneous data of the photothermal electrolysis hydrogen production system.

[0060] In this embodiment of the invention, the step of collecting real-time physical parameters and historical operating data of the photothermal electrolysis hydrogen production system through an Internet of Things platform to obtain multi-source heterogeneous data of the photothermal electrolysis hydrogen production system includes:

[0061] Distributed data integration of multi-source heterogeneous data is performed to obtain a unified data platform for the photothermal electrolysis hydrogen production system.

[0062] Through the unified data platform, the dynamic behavior characteristics of the photothermal electrolysis hydrogen production system that are highly related to energy loss are analyzed to obtain the feature description set of the photothermal electrolysis hydrogen production system.

[0063] The feature description set is loaded into a predefined model template and initial parameters are configured to construct a digital twin model of the photothermal electrolysis hydrogen production system.

[0064] Various data sources generated during the operation of the solar thermal electrolysis hydrogen production system were selected, including heterogeneous data such as temperature data from solar thermal acquisition equipment, solar irradiance data from photovoltaic panels, current and voltage data from electrolyzers, pressure data from hydrogen storage equipment, heat loss data from pipeline transmission, and environmental temperature and humidity data. A distributed file storage architecture was adopted, deploying each data source on independent storage nodes. Each storage node was equipped with a dedicated data read interface to obtain the raw data generated by the corresponding equipment in real time. The raw data in each storage node was cleaned. For missing values, three consecutive valid measurements before and after the missing data were selected, and the average trend of these valid measurements was calculated. Based on this trend, the missing values ​​were calculated and filled. For abnormal data that exceeded the reasonable range, the normal data under the same operating conditions was compared, and the normal data closest to the operating state of the abnormal data was selected for replacement. Subsequently, data standardization was performed. The historical maximum and minimum values ​​corresponding to each data type were selected. Each value in the data type was subtracted from the minimum value, and then divided by the difference between the maximum and minimum values ​​to ensure that all data were within the same order of magnitude range. Establish unified data mapping rules to map fields from different data sources to a pre-defined, unified standard. For example, convert the running time of all devices to the same time format and adjust the device numbers according to a unified naming convention to ensure consistency in field meaning and format across all data types. Through a distributed data synchronization protocol, processed data from each storage node is synchronized to the central storage unit in real time. The central storage unit then categorizes, archives, and indexes the synchronized data, forming a unified data platform for the solar thermal electrolysis hydrogen production system that enables centralized management, rapid querying, and retrieval.

[0065] Extract all historical and real-time operating data related to energy loss in the photothermal electrolysis hydrogen production system from the unified data platform, including energy output data under different light intensities during the photothermal conversion process, energy consumption data of the electrolyzer under different operating loads, temperature change data of different pipeline sections during heat transfer, and energy consumption data during equipment start-up and shutdown. Behavioral feature analysis was performed on the extracted data. For the photothermal conversion stage, the correlation between light intensity, ambient temperature, and photothermal conversion efficiency was continuously monitored over different time periods. Fluctuations in conversion efficiency were recorded when light intensity changed, identifying a significant decrease in conversion efficiency when light intensity fell below a certain level. For the electrolysis hydrogen production stage, the correlation between current and voltage changes in the electrolyzer and energy consumption was tracked. The impact of electrode material usage time on electrolysis efficiency was observed, summarizing the behavior that energy consumption increases when current and voltage fluctuations exceed a certain range. For the heat transfer stage, heat loss data under different pipe insulation aging degrees and transmission distances were compared, identifying that greater heat loss occurs with more severe insulation aging and longer transmission distances. For the equipment start-up and shutdown stage, the correlation between start-up and shutdown frequency, start-up and shutdown duration, and energy loss was analyzed, extracting the behavior that frequent start-ups and shutdowns lead to increased additional energy loss. These behavioral features obtained through data observation and correlation analysis were systematically organized, clarifying the specific manifestations, core influencing factors, and changing patterns of each feature, forming a complete feature description set for the photothermal electrolysis hydrogen production system.

[0066] The predefined model template is built based on the physical structure and operating principle of the photothermal electrolysis hydrogen production system. It includes functional units such as a photothermal acquisition module, an energy conversion module, an electrolysis hydrogen production module, a heat transfer module, and a hydrogen storage module. Each functional unit has a pre-defined structural mapping relationship with the actual system, which can simulate the operating logic of the actual components. Each feature in the feature description set is loaded into the corresponding functional unit of the model template according to the corresponding relationship. Features related to photothermal conversion efficiency are loaded into the photothermal acquisition module to clarify the dynamic correlation logic between light intensity, ambient temperature and conversion efficiency in this module; features related to electrolyzer energy consumption are loaded into the electrolysis hydrogen production module to define the interaction relationship between current and voltage changes and energy consumption and hydrogen production efficiency; features related to heat loss are loaded into the heat transfer module to determine the corresponding rules between pipe material, insulation layer status, transmission distance and heat loss; and features related to hydrogen storage pressure and storage capacity are loaded into the hydrogen storage module to clarify the matching relationship between pressure changes and storage capacity. Referring to the design parameters and initial operating state of the photothermal electrolysis hydrogen production system, and combining the feature information in the feature description set, the initial parameters of each functional unit are configured. Based on the feature description of the photothermal acquisition module, the conversion efficiency parameter corresponding to the initial light intensity is configured. The initial electrolysis voltage parameter is set according to the feature information of the electrolysis hydrogen production module, and the initial hydrogen storage pressure parameter is configured according to the feature rules of the hydrogen storage module, ensuring that each initial parameter is consistent with the initial operating state of the actual system. A dynamic correlation mechanism between the functional units is established, so that changes in the energy output of the photothermal acquisition module are transmitted to the electrolysis hydrogen production module through the energy conversion module. Changes in the operating state of the electrolysis hydrogen production module affect the pressure regulation of the hydrogen storage module. Each module interacts and responds dynamically according to the rules in the feature description set, ultimately constructing a digital twin model that can accurately simulate the actual operating state of the photothermal electrolysis hydrogen production system.

[0067] The beneficial effects include: achieving standardized integration and centralized management of multi-source heterogeneous data from the photothermal electrolysis hydrogen production system; deploying various data sources through a distributed storage architecture; cleaning and filling missing data, replacing abnormal data, standardizing and unifying the magnitude, and mapping rules to standardize fields; and then aggregating the data to the central storage unit through a synchronization protocol to ensure data integrity and consistency, forming a unified data platform that can be quickly queried and accessed; accurately identifying dynamic behavioral characteristics in the system that are highly correlated with energy loss; extracting various related data from the unified data platform; analyzing the correspondence between the operating status of each link and energy loss; clarifying the characteristic manifestations, core influencing factors, and change patterns; and forming a comprehensive and systematic feature description set; and based on a predefined model template of the system's physical structure and operating principles, loading the feature description set and configuring parameters adapted to the actual initial state, establishing a dynamic correlation mechanism for each functional unit, and constructing a digital twin model that can accurately simulate the actual operating state of the system, providing reliable data support and an accurate simulation foundation for system optimization, energy loss reduction, and improved operating efficiency.

[0068] In this embodiment of the invention, S2, based on the multi-source heterogeneous data, the photothermal electrolysis hydrogen production system is model initialized to construct a digital twin model of the photothermal electrolysis hydrogen production system;

[0069] Multi-source heterogeneous data from a solar thermal electrolysis hydrogen production system were collected, covering design parameters and historical operating data of solar thermal acquisition equipment, photovoltaic panels, electrolyzers, hydrogen storage equipment, and pipelines. This data was categorized and analyzed: design parameters were classified as static structural data to clarify the physical properties of the equipment and their hierarchical relationship within the system; operating data were classified as dynamic behavioral data to analyze the time-series correlations and energy interaction patterns between equipment; and environmental data were classified as external impact data to establish the correspondence between environmental factors and internal system parameters, providing a data foundation for model initialization.

[0070] Based on the system's physical composition, a digital twin model structure is constructed, dividing the system into modules for photothermal acquisition, energy conversion, hydrogen electrolysis, hydrogen storage, heat transfer, and environmental interaction. Each module corresponds to actual equipment or subsystems. The internal components of each module are detailed; for example, the photothermal acquisition module is divided into heat collectors and photovoltaic modules, simulating material properties and energy conversion mechanisms respectively. The hydrogen electrolysis module is divided into electrode, electrolyte, and tank components, simulating electrode reactions and energy consumption. According to the actual energy flow and mass transfer paths, the connections between modules are established: the energy output from the photothermal acquisition module flows to the energy conversion module; the electrical energy from the energy conversion module flows to the hydrogen electrolysis module and other electrical equipment; the hydrogen flow from the hydrogen electrolysis module flows to the hydrogen storage module; the heat transfer module connects each energy conversion stage; and the environmental interaction module inputs environmental parameters to the photothermal acquisition and hydrogen electrolysis modules, forming a complete system topology.

[0071] The classified and analyzed static structural data is loaded into the corresponding module components. For example, the heat collection area and material absorptivity of the photothermal acquisition module are loaded into the heat collection component, and the electrode material and tank volume of the electrolysis hydrogen production module are loaded into the corresponding component. Dynamic behavior data under typical operating conditions of the equipment are extracted, such as the conversion efficiency curve of the photothermal acquisition module under different illuminations and the energy consumption curve of the electrolysis hydrogen production module under different currents. These serve as the initial basis for the module's dynamic response, setting the module's input and output rules. For example, after receiving light intensity, the photothermal acquisition module outputs heat and electrical energy according to the curve. A mapping relationship between environmental data and system modules is established. For example, after light intensity is input into the environmental interaction module, it is transmitted to the photothermal acquisition module to trigger changes in conversion efficiency. At the same time, initial operating state parameters of the model are set, such as the ambient temperature at system startup and the initial energy storage of each module, so that the initial state of the model is consistent with the typical startup state of the actual system.

[0072] Historical operating data from the actual system was selected as the validation dataset. Environmental data was input into the environmental interaction module of the digital twin model, the model was run, and the output system operating data was recorded. The model output data was compared with the actual historical data one by one. If the output deviation of a certain module exceeded the acceptable range, such as a large difference between the simulated and actual values ​​of the conversion efficiency of the photothermal acquisition module, the parameter initialization process of that module was traced back to check the accuracy of the static structural data and whether the dynamic behavior data curves matched reality, and the corresponding parameters were adjusted. This process was repeated until the output deviation of all modules was within the acceptable range, ensuring that the digital twin model could accurately simulate the actual system operating state, completing the model initialization, and constructing the digital twin model of the photothermal electrolysis hydrogen production system.

[0073] The beneficial effects are as follows: by collecting and classifying heterogeneous data from multiple sources in the photothermal electrolysis hydrogen production system, a comprehensive data foundation covering static structure, dynamic behavior, and external influences is provided for model initialization; corresponding modules and components are built according to the physical composition of the system, and the connection relationship between modules is established according to the actual energy and material transport paths, so that the structure of the digital twin model is completely consistent with the physical topology of the actual system; static structural data, dynamic behavior data, environmental mapping relationships, and initial operating parameters are accurately loaded into each module and component of the model, and at the same time, through historical data verification and parameter adjustment, it is ensured that the initial state and dynamic response of the model are highly consistent with the actual system. Finally, a digital twin model that can accurately simulate the operating state of the photothermal electrolysis hydrogen production system is constructed, providing a reliable virtual mapping and analysis platform for subsequent system feature analysis, energy flow tracking, and optimization scheme formulation.

[0074] S3. Simulate the energy flow of the photothermal electrolysis hydrogen production system using the digital twin model to obtain the energy flow distribution map of the photothermal electrolysis hydrogen production system;

[0075] In this embodiment of the invention, the step of simulating the energy flow of the photothermal electrolysis hydrogen production system using the digital twin model to obtain the energy flow distribution map of the photothermal electrolysis hydrogen production system includes:

[0076] After inputting the current system state parameters of the photothermal electrolysis hydrogen production system into the digital twin model, the energy transfer path in the digital twin model is dynamically tracked to obtain the energy flow trajectory data of the photothermal electrolysis hydrogen production system.

[0077] By identifying the energy conversion nodes and loss nodes in the energy flow trajectory data, the node energy distribution dataset of the photothermal electrolysis hydrogen production system is obtained.

[0078] Based on the energy flow direction and intensity gradient in the node energy distribution dataset, an energy flow distribution map of the photothermal electrolysis hydrogen production system is generated.

[0079] The process of identifying energy conversion nodes and loss nodes in the energy flow trajectory data to obtain the node energy distribution dataset of the photothermal electrolysis hydrogen production system includes:

[0080] Extract the energy flow rate change rate and conversion efficiency index from the energy flow trajectory data to obtain the node feature set of the photothermal electrolysis hydrogen production system;

[0081] By comparing the input and output differences of node energy in the node feature set, the energy anomaly nodes of the photothermal electrolysis hydrogen production system are obtained.

[0082] The node type and corresponding energy value of the energy anomaly nodes are labeled to generate a node energy distribution dataset for the photothermal electrolysis hydrogen production system.

[0083] The system collects various real-time parameters of the current operating status of the photothermal electrolysis hydrogen production system, including the real-time light intensity received by the photothermal acquisition equipment, the real-time output power of the photovoltaic panels, the real-time current and voltage of the electrolyzer, the real-time internal pressure of the hydrogen storage equipment, the real-time wall temperature of the pipeline, the real-time temperature and humidity of the environment, and the real-time operating load of each piece of equipment. These parameters are then input one by one according to the correspondence between the functional units in the digital twin model. Photothermal acquisition-related parameters are input into the photothermal acquisition module, electrolysis operation-related parameters are input into the electrolysis hydrogen production module, and hydrogen storage status-related parameters are input into the hydrogen storage module, ensuring that each parameter accurately matches the corresponding simulation unit in the model. The dynamic tracking function of the digital twin model is activated, starting from the energy initiation node, the photothermal acquisition module, to record in real time the process data of light energy absorption and conversion into heat energy, including the conduction path, conduction speed, and temperature changes within the acquisition module. Subsequently, the entire process of heat energy transfer to the energy conversion module is tracked, recording the transfer path when heat energy is converted into electrical energy, the specific location of the conversion, and the energy state changes during the conversion process. Next, the path of electrical energy transfer to the electrolysis hydrogen production module is tracked, recording the distribution of electrical energy within the electrolyzer, the specific process and location of its reaction with the electrolyte and conversion into chemical energy. Simultaneously, the pipeline transmission paths involved in the energy transfer process are tracked, recording the heat flow trajectory within the pipelines and its transfer through the pipeline walls. Finally, the path of chemical energy storage to the hydrogen storage module and changes in storage state are tracked. This energy transfer information from different stages and paths is continuously recorded in chronological order and spatial location, forming energy flow trajectory data of the photothermal electrolysis hydrogen production system, containing complete information such as the starting position, nodes traversed, ending position, energy form changes, transfer speed, and energy value changes.

[0084] The energy flow trajectory data is analyzed segment by segment to identify locations where the energy form changes as energy conversion nodes. Examples include locations where light energy is converted to heat in the photothermal acquisition module, locations where heat energy is converted to electricity in the energy conversion module, and locations where electricity is converted to chemical energy in the electrolysis hydrogen production module. These locations will show a clear switch in energy form in the trajectory data. By identifying changes in the energy form field in the trajectory data, all such locations are identified, and the spatial coordinates, energy form before and after conversion, system operating status at the time of conversion, and energy values ​​before and after conversion are recorded for each conversion node. Simultaneously, locations where energy values ​​decrease without energy form conversion are identified as loss nodes. Examples include locations where heat is lost through the pipe walls during pipeline transmission, locations where energy is lost due to mechanical friction during equipment operation, and locations where energy is lost inside the electrolyzer that does not participate in the reaction. By comparing the energy values ​​of two adjacent recorded points in the trajectory data, when the energy value of the later recorded point is less than that of the earlier recorded point and the energy form has not changed, this location is a loss node. The spatial coordinates, equipment or component where loss occurs, energy values ​​before and after loss, environmental conditions at the time of loss, and equipment operating parameters are recorded for each loss node. The records of all identified energy conversion nodes and loss nodes are systematically organized and arranged in the order of energy transfer to form a node energy distribution dataset of the photothermal electrolysis hydrogen production system, which includes node type, location information, energy-related data, and operating status information.

[0085] Extract the spatial coordinates of all nodes in the node energy distribution dataset to clarify the actual location of each node in the physical structure of the photothermal electrolysis hydrogen production system. Based on the energy transfer sequence recorded in the dataset, determine the energy flow direction from the starting node to each conversion node, loss node, and finally to the termination node. Mark the connection relationship between each node in a planar or three-dimensional coordinate system to represent the energy transfer path. Calculate the energy difference between adjacent nodes and classify the intensity gradient levels according to the magnitude of the difference; the larger the energy difference, the higher the intensity gradient, and the smaller the energy difference, the lower the intensity gradient. Set a unique visual distinction standard for each gradient level. Draw a distribution map framework based on the physical structure layout of the photothermal electrolysis hydrogen production system. Mark all energy conversion nodes and loss nodes in the framework according to their actual locations, using different graphic symbols to distinguish between them: conversion nodes use circular symbols, and loss nodes use triangular symbols. Connect adjacent nodes with arrowed lines according to the determined energy flow direction; the arrows indicate the energy transfer direction. Select the corresponding line thickness or color depth according to the intensity gradient level; the higher the intensity gradient, the thicker or darker the line, and the lower the intensity gradient, the thinner or lighter the line. Next to each node, label the node name, the energy conversion type of the conversion node, and the loss value range of the loss node to ensure that the distribution map can clearly show the location distribution of each node, the specific direction of energy flow, and the energy intensity differences on different paths, and finally generate a complete and accurate energy flow distribution map of the photothermal electrolysis hydrogen production system.

[0086] Continuous energy transfer records corresponding to each node are extracted from the energy flow trajectory data. These records contain the energy input values, energy output values, and energy transfer timestamps for each node at different times. When extracting the energy flow rate of change, the energy output values ​​corresponding to two consecutive consecutive recorded times for each node are selected, the difference between these two values ​​is calculated, and the time span between these two recorded times is determined. The difference is divided by the corresponding time span to obtain the change in energy flow per unit time for that node, which is the energy flow rate of change for that node. When extracting the conversion efficiency index, for each energy conversion node, the total energy value transferred in through all input channels is summarized, and the total energy value transferred out through all output channels is summarized. The total output energy value is divided by the total input energy value to obtain the energy conversion efficiency index for that node. The name, spatial location information, corresponding energy flow rate of change, and conversion efficiency index of each node are recorded one by one. The relevant data of all nodes are organized and sorted according to the energy transfer order in the photothermal electrolysis hydrogen production system, forming a node feature set of the photothermal electrolysis hydrogen production system containing the core energy characteristic parameters of each node.

[0087] Historical node energy data of the solar thermal electrolysis hydrogen production system under stable operation conditions were collected. For each node, the energy input and output values ​​under different stable operating states were analyzed, and the energy difference between input and output under each state was calculated. The range of values ​​with the highest frequency among these differences was identified and defined as the normal energy input-output difference range for that node. The current energy input and output values ​​of each node were extracted from the node feature set, and the current energy difference between them was calculated. The current energy difference of each node was compared with the corresponding normal energy input-output difference range. If the current energy difference was greater than the upper limit of the normal difference range or less than the lower limit of the normal difference range, the node was identified as an energy anomaly node. The information of all identified nodes was recorded to obtain the energy anomaly nodes of the solar thermal electrolysis hydrogen production system.

[0088] Each energy anomaly node is categorized based on its role in the energy transfer process. If the node involves energy form conversion, it is classified as an energy conversion anomaly node; otherwise, it is classified as an energy loss anomaly node. Relevant energy values ​​for each anomaly node are extracted, including energy input, energy output, energy flow rate change, and conversion efficiency. The node's spatial coordinates within the photothermal electrolysis hydrogen production system, its associated equipment component, and its position within the energy transfer path are also supplemented. All anomaly node types, corresponding energy values, and supplementary information are systematically arranged according to the order of energy transfer, ensuring complete and orderly information for each anomaly node. This process ultimately generates a node energy distribution dataset for the photothermal electrolysis hydrogen production system.

[0089] The beneficial effects are as follows: By inputting the current operating status parameters of the photothermal electrolysis hydrogen production system into the digital twin model, the dynamic tracking function of the energy transfer path is accurately activated, and the location, form changes, speed, and numerical fluctuations of energy transfer throughout the entire process from light energy absorption and conversion to chemical energy storage are fully recorded. This forms comprehensive and clear energy flow trajectory data, providing a precise foundation for subsequent energy analysis. Through segmented analysis of the energy flow trajectory data, nodes where energy forms are converted and loss nodes where energy values ​​decrease without form conversion are accurately identified. The system records the spatial location of nodes, energy-related data, and operating status information, constructing a comprehensive and systematic node energy distribution dataset, providing a reliable basis for identifying key energy loss locations. Based on the differences in energy flow direction and intensity gradient in the node energy distribution dataset, and using the system's physical structure as a framework, node types are distinguished by symbols, and intensity gradients are represented by line features. Key node information is clearly marked, generating an energy flow distribution map that intuitively presents node distribution, energy flow direction, and intensity differences. This helps to quickly locate key energy loss links, providing intuitive and accurate decision support for energy optimization, loss control, and operational efficiency improvement of the photothermal electrolysis hydrogen production system.

[0090] S4. Identify the energy loss points in the energy flow distribution diagram, and analyze the root cause analysis report of the energy loss points based on the energy balance equation and efficiency optimization criteria of the photothermal electrolysis hydrogen production system;

[0091] In this embodiment of the invention, identifying energy loss points in the energy flow distribution map includes:

[0092] Based on the energy intensity gradient change, the energy flow distribution map is divided into multiple analysis regions to obtain the regional energy distribution feature set of the photothermal electrolysis hydrogen production system.

[0093] Based on a preset loss mode feature library, suspicious regions that conform to abnormal energy loss characteristics in the energy distribution feature set of the region are identified, and a set of suspicious regions of the photothermal electrolysis hydrogen production system is obtained.

[0094] Spatial correlation analysis was performed on the set of suspicious regions to obtain the energy loss points of the photothermal electrolysis hydrogen production system.

[0095] The root cause analysis report on the energy loss point based on the energy balance equation and efficiency optimization criteria of the photothermal electrolysis hydrogen production system includes:

[0096] By traversing the spatial distribution information of the energy loss points and the topological relationship of the system components, a component association map of the energy loss points is constructed.

[0097] Multi-dimensional rule matching is performed on the component association graph, and key components and interaction relationships that cause energy loss are identified through parallel rule reasoning to generate the component interaction network of the energy loss points.

[0098] Based on the deviation between the real-time operating data and the baseline parameters at the energy loss points, the efficiency bottleneck of energy transfer between components is analyzed to obtain a root cause analysis report of the energy loss points.

[0099] Observe the specific manifestation of the energy intensity gradient in the energy flow distribution map. Based on the gradient level corresponding to the line thickness or color depth, divide the entire distribution map into continuous regions in order of gradient from low to high. During the division process, the physical structure of the photothermal electrolysis hydrogen production system is used as the reference boundary to ensure that the spatial range of each region matches the actual distribution of equipment, pipelines, or components. This ensures that the energy intensity gradient within each divided region remains consistent or shows a continuous gradual trend, without any jumps across gradients. Determine the clear boundaries of each region, using physical structural features such as equipment outlines, pipeline segment nodes, and component connection locations as the basis for region boundary division. This ensures that the region division conforms to the energy gradient distribution law and corresponds to the actual physical layout of the system. Extract the core energy distribution characteristics of each region, including the spatial shape and range of the region, the specific types and numbers of energy conversion nodes and loss nodes within the region, the main direction of energy flow within the region, the overall average energy intensity level of the region, the changing trend of the energy intensity gradient within the region, and the distribution density and concentration location of loss nodes within the region. Organize and record these characteristic information of each region one by one, and arrange them according to the order of the region in the energy transfer path to form a complete set of regional energy distribution characteristics of the photothermal electrolysis hydrogen production system.

[0100] The pre-defined loss mode feature library is constructed by collecting various known cases of abnormal energy loss during the long-term operation of photothermal electrolysis hydrogen production systems. The library contains multiple clearly defined abnormal loss mode features, each of which details the corresponding energy intensity gradient change pattern, regional spatial distribution characteristics, and node type combination method. For example, the pipeline insulation failure mode corresponds to a rapid decrease in energy intensity gradient along the pipeline transmission direction, with the region extending in a long strip along the pipeline and containing multiple continuously distributed loss nodes; the electrolyzer electrode attenuation mode corresponds to a sudden drop in energy intensity gradient in the region where the conversion node is located, with the region concentrated in the internal space of the electrolyzer, containing energy conversion nodes, and the gradient change is abnormally correlated with the conversion efficiency; the equipment connection loosening mode corresponds to a disordered and irregular energy intensity gradient in a local small area, with the region distributed in a point-like or small patch-like pattern, containing loss nodes at the equipment connection points. The energy intensity gradient variation trend, spatial shape and distribution range, and node types and combinations contained in each region are extracted from the regional energy distribution feature set. These are then compared one by one with the features of each abnormal loss mode in the loss mode feature library. During the comparison process, all details of the regional features and the mode features are fully checked to ensure complete consistency. If all features of a region completely match the features of a certain abnormal loss mode in the library, the region is identified as a suspicious region. The spatial range, the corresponding abnormal loss mode type, and the location and features of the core nodes in each suspicious region are recorded. All identified suspicious regions are compiled and summarized to obtain the suspicious region set of the photothermal electrolysis hydrogen production system.

[0101] Detailed physical structure layout information of the photothermal electrolysis hydrogen production system was retrieved to clarify the spatial relationships and connection methods of each piece of equipment, pipeline, and component. Simultaneously, the precise spatial coordinates, boundary range, spatial positions of all nodes within the region, and connection paths between nodes were extracted from each suspicious region in the suspicious region set. Spatial correlation analysis was performed on all suspicious regions. First, it was determined whether there was a continuous physical connection between the suspicious regions, i.e., whether the boundary of one suspicious region was directly adjacent to or partially overlapped with the boundary of another suspicious region, forming a continuous anomalous region band. Next, it was analyzed whether the suspicious regions shared the same energy transfer path, i.e., whether the energy flow paths within different suspicious regions belonged to the same main line or branch line, and whether energy was directly transferred from one suspicious region to another. Then, it was determined whether the suspicious regions were concentrated around a specific piece of equipment or key component, i.e., whether the spatial centers of multiple suspicious regions pointed to the same main body or core component of the equipment, forming an anomalous region cluster centered on that equipment. For multiple suspicious areas that exhibit spatial continuity, share energy paths, or are clustered around the same equipment, these associated areas are determined to point to a common energy loss concentration location. For isolated suspicious areas without any spatial correlation, secondary verification is performed using historical regional energy data from normal system operation to eliminate misjudged areas caused by accidental fluctuations. All energy loss concentration locations identified through spatial correlation analysis are precisely located, clarifying their specific installation positions within the physical structure of the photothermal electrolysis hydrogen production system, the names of the corresponding equipment or components, and the energy transfer path links to which they belong, thus obtaining the energy loss points of the photothermal electrolysis hydrogen production system.

[0102] The spatial distribution information of energy loss points is extracted, including the precise physical location of each loss point in the photothermal electrolysis hydrogen production system, its associated equipment, the distribution of surrounding adjacent components, and its spatial distance relationship with other loss points. Simultaneously, complete component topology data is acquired, covering the model, specifications, installation location, hierarchical structure, and connection methods of all components. Each energy loss point is traversed one by one, precisely mapping its spatial location to the system component topology to identify the core component directly associated with each loss point. Starting from this core component, upstream and downstream related components, auxiliary support components, and control components are identified, clarifying the connection paths and interaction methods between these components and the core component. A graphical component relationship map is constructed, using nodes of different shapes to represent different types of components, solid lines to represent physical connections, and dashed lines to represent control signal transmission relationships. Component names, models, and their degree of association with loss points are labeled on nodes, while connection methods and energy transfer directions are labeled on lines, ensuring the map comprehensively presents all components related to energy loss points and their interconnections, forming a component relationship map of energy loss points.

[0103] The system includes a multi-dimensional rule set covering component performance degradation rules, connection reliability rules, component compatibility rules, environmental adaptability rules, and operating load rules. Among these, the component performance degradation rules clarify the correspondence between component operating time and energy transfer efficiency; the connection reliability rules define the correlation standard between the tightness of component connection parts and energy loss; the component compatibility rules specify the relationship between the technical parameter adaptation range of upstream and downstream components and energy transfer effect; the environmental adaptability rules explain the threshold of influence of environmental factors such as temperature, humidity, and dust concentration on component performance; and the operating load rules clarify the effect of the deviation range between the component's rated load and the actual operating load on energy loss. Each component in the component association graph, the connections between components, and the association information between components and loss points are compared one by one with all rules in the multi-dimensional rule set. Parallel rule reasoning is performed. For example, based on the component performance degradation rule, the actual runtime of associated components is compared with the degradation threshold set in the rule to determine whether energy loss is caused by component performance degradation. Based on the connection reliability rule, the energy loss data of component connections is checked against the tightening standards set in the rule to determine whether there is loss caused by loose connections. Through the component matching rule, the technical parameters of upstream and downstream components are verified to meet the adaptation requirements to determine whether energy transfer is hindered due to parameter mismatch. Through multi-dimensional parallel rule reasoning, key components that directly affect energy loss are screened out, and the interaction relationships between key components are clarified. For example, if the performance degradation of a component causes the operating load of downstream components to exceed the rated range, thus causing additional energy loss, key components are highlighted with bold nodes, and the interaction type is marked with lines of different colors, generating a component interaction network for energy loss points.

[0104] Real-time operational data of key components related to energy loss points is collected, including core data reflecting the component's operating status such as energy input, energy output, operating temperature, vibration frequency, working pressure, current, and voltage. Simultaneously, benchmark parameters of the photothermal electrolysis hydrogen production system under stable operation conditions are retrieved. These benchmark parameters are long-term validated standard data for each component during normal system operation, including standard energy transfer efficiency, rated operating temperature range, allowable vibration frequency range, rated working pressure, and standard current and voltage range. The real-time operational data of each key component is compared one by one with the corresponding benchmark parameters, and the deviation is recorded. For example, if the real-time energy transfer efficiency of a component is lower than the standard efficiency in the benchmark parameters, or if the operating temperature of a component exceeds the rated temperature range in the benchmark parameters, the efficiency bottleneck of energy transfer between components is analyzed based on these deviation data. By tracing the energy transfer path, the first component or the first connection point where the deviation occurs is identified. This location is the link with the lowest energy transfer efficiency. For example, if the real-time energy output value of an upstream component is lower than the benchmark parameter, the downstream component will receive insufficient energy. To meet the system's operating requirements, the downstream component is forced to operate under overload, thus increasing energy loss. This upstream component is the efficiency bottleneck. By conducting in-depth analysis of the root causes of efficiency bottlenecks and combining information such as component runtime, maintenance records, and environmental conditions, we determine whether the deviation is caused by component aging, untimely maintenance, loose connections, improper parameter settings, or design flaws. We systematically organize the location information of energy loss points, related key components, specific deviations between real-time data and benchmark parameters, the specific location of efficiency bottlenecks, the root cause analysis process, and relevant supporting data to form a complete and logically clear root cause analysis report of energy loss points.The beneficial effects include: scientifically dividing the energy flow distribution map based on the energy intensity gradient change and combined with the physical structural boundaries of the photothermal electrolysis hydrogen production system, ensuring that each region maintains a uniform or continuously gradually changing gradient characteristic; extracting core information such as the spatial range, number of node types, energy flow direction, intensity level, and distribution of loss nodes for each region, forming a comprehensive regional energy distribution feature set, providing a clear regional analysis basis for subsequent abnormal loss identification; and relying on a preset loss pattern feature library containing features of various known abnormal loss patterns, matching the gradient changes, spatial distribution, node combinations, and other features of each region in the regional energy distribution feature set with the patterns in the library one by one. Precise comparison efficiently identifies suspicious areas that match abnormal loss characteristics, forming a targeted set of suspicious areas and significantly improving the accuracy and efficiency of abnormal loss area identification. Combined with the detailed physical structure layout of the solar thermal electrolysis hydrogen production system, spatial correlation analysis is conducted on the set of suspicious areas. By judging the correlations such as spatial continuity, energy path sharing, and equipment concentration between areas, misjudged areas are eliminated and the location of concentrated energy loss is accurately located. The physical installation location, equipment, and energy transfer links corresponding to the loss points are clarified, providing accurate and reliable decision-making basis for the precise formulation of loss management plans and optimization of energy utilization efficiency of the solar thermal electrolysis hydrogen production system.

[0105] S5. Based on the root cause analysis report, generate an optimization strategy for the photothermal electrolysis hydrogen production system, and map the correction parameters of the optimization strategy to the digital twin model in real time.

[0106] In this embodiment of the invention, the step of generating an optimization strategy for the photothermal electrolysis hydrogen production system based on the root cause analysis report, and mapping the correction parameters of the optimization strategy to the digital twin model in real time, includes:

[0107] Based on the efficiency bottleneck type in the root cause analysis report, the corresponding optimization strategy template is retrieved from the preset optimization strategy library to obtain the preliminary optimization scheme of the photothermal electrolysis hydrogen production system.

[0108] Based on the current operating status of the digital twin model, the parameters of the preliminary optimization scheme are adjusted for adaptation.

[0109] Under stability constraints, the parameters for parameter adaptation and adjustment are optimized using multiple objectives to obtain the modified parameter set of the photothermal electrolysis hydrogen production system;

[0110] The modified parameter set is dynamically loaded into the runtime environment of the digital twin model.

[0111] The parameters for parameter adaptation and adjustment under stability constraints are subjected to multi-objective optimization to obtain a set of corrected parameters for the photothermal electrolysis hydrogen production system, including:

[0112] The optimization framework is based on multiple objectives, including energy efficiency improvement and operational stability.

[0113] In the multi-objective optimization framework, the parameter space is explored for the parameters that are adapted and adjusted under stability constraints.

[0114] By identifying candidate parameter combinations that satisfy the multi-objective optimization framework through Pareto front analysis, the optimal parameter solution set of the photothermal electrolysis hydrogen production system is obtained.

[0115] The optimized parameter set that operates stably is used as the correction parameter set for the photothermal electrolysis hydrogen production system.

[0116] The calculation method for the multi-objective optimization is as follows: ;

[0117] In the formula, For a multi-objective optimization function, The weighting factor corresponding to the energy efficiency improvement target. The weighting factor corresponding to the stable operation target. This is the baseline energy loss value for the digital twin model. The energy loss prediction value of the digital twin model after adapting and adjusting the parameters. for, The current system stability of the digital twin model after the parameters have been adapted and adjusted. The system baseline operational stability of the digital twin model is determined. Specific types of efficiency bottlenecks are extracted from the root cause analysis report. These types include decreased energy transfer efficiency due to component performance aging, energy leakage caused by loose connections between components, energy transfer obstruction due to mismatched upstream and downstream component parameters, additional energy loss due to operating load exceeding the rated range, and component performance fluctuations caused by insufficient environmental adaptability. A pre-defined optimization strategy library is structured and stored according to efficiency bottleneck type. Each bottleneck type corresponds to a dedicated optimization strategy template, which includes specific optimization directions, operation steps, applicable scenarios, and initial reference parameters for that type of bottleneck. For example, the template for component performance aging explicitly adopts the optimization direction of component maintenance or replacement, detailing the specific maintenance process or the selection criteria for replacement component models; the template for loose connections explicitly adopts the optimization strategy of tightening connections or replacing seals, detailing the use of tightening tools, the control method of tightening force, and the material selection requirements for seals. Extract the core characteristic keywords of efficiency bottleneck types from the root cause analysis report, and compare these keywords with the classification tags of each template in the optimization strategy library one by one. When a keyword completely matches the classification tag of a template, the template is determined to be the matching optimization strategy template. Integrate the optimization direction, operation steps, and initial reference parameters in the template, and combine them with the specific location of energy loss points and related component information to form a preliminary optimization plan for the photothermal electrolysis hydrogen production system.

[0118] Real-time acquisition of current operating status data from the digital twin model is performed, including core information reflecting the system's operating status such as real-time operating temperature, working pressure, energy input and output values, conversion efficiency, vibration frequency, and control signal response speed of each component in the model. Simultaneously, the simulation environment conditions and current operating load level of the model are recorded. The suitability of the initial reference parameters in the preliminary optimization scheme with the model's current operating status is analyzed. If the control parameters for component operating temperature in the preliminary scheme are higher than the model's actual current operating temperature, and the system's energy loss is mainly due to insufficient conversion efficiency caused by low temperature, then the temperature control parameters are adjusted to a reasonable range that matches the current actual temperature and can gradually improve conversion efficiency. If the flow control parameters for energy transfer paths in the preliminary scheme do not match the energy demand under the model's current operating load, resulting in excess or insufficient energy supply to some components, then the flow control parameters are adjusted according to the energy demand corresponding to the current load to make energy transfer more closely match the actual demand. During the adjustment process, the model's operational feedback is monitored in real time. If the energy loss of the corresponding component in the model decreases after adjusting a certain parameter without affecting the normal operation of other components, the adjusted parameter is retained. If the parameter adjustment causes fluctuations in system stability, the parameter is fine-tuned in reverse until the model operates stably and the loss is reduced, thus completing the parameter adaptation adjustment of the initial optimization scheme.

[0119] The stability constraints of the solar thermal electrolysis hydrogen production system are clearly defined. These constraints include maintaining the system operating temperature within a safe range to prevent component damage due to excessive heat or reduced conversion efficiency due to insufficient heat; controlling the system operating pressure within the rated range to prevent safety hazards caused by excessive pressure or energy transfer obstruction due to insufficient pressure; keeping energy output fluctuations within allowable limits to ensure stable hydrogen production; and ensuring that the component operating vibration frequency is below the critical threshold to prevent loose component connections or structural damage due to excessive vibration. The core objectives of the multi-objective optimization are determined: reducing system energy loss, improving energy conversion efficiency, extending component lifespan, and ensuring stable system operation. Based on the parameters after parameter adaptation and adjustment, optimization is refined step-by-step according to the optimization objectives. First, parameters directly related to energy loss are adjusted, and the change in loss values ​​at energy loss points in the model is observed. If the loss value decreases and the stability constraints are met, the parameter is optimized further until the loss is minimized. Next, parameters affecting conversion efficiency are adjusted to improve energy conversion efficiency while maintaining low energy loss levels, ensuring that the stability constraints are not exceeded during the adjustment process. For parameters that have mutual influence, a step-by-step fine-tuning method is adopted. First, one parameter is fixed, and the other parameter is adjusted to the optimal state. Then, the parameter is fixed again, and the first parameter is fine-tuned in the opposite direction until both parameters reach the optimization goal and meet the constraints. Through repeated verification and adjustment, the corrected parameter set of the photothermal electrolysis hydrogen production system that takes into account multiple goals and meets stability constraints is finally obtained.

[0120] A dedicated interface and data transmission path for parameter loading in the digital twin model are determined. This interface establishes a one-to-one correspondence with each functional module of the model, ensuring that parameters are accurately transmitted to the target module. Before loading, the completeness of the modified parameter set is verified to check whether it contains all parameters required by each functional module of the model, with no missing or omitted key parameters. Simultaneously, the compatibility of the parameter format with the model interface is verified to ensure that the parameter format meets the model's data reception requirements, avoiding loading failures due to format mismatches. According to the division of model functional modules, the corresponding parameters in the modified parameter set are transmitted one by one to the parameter configuration unit of each module. For example, the temperature control parameters of the photothermal acquisition module are transmitted to the temperature regulation unit of that module, and the current and voltage control parameters of the electrolysis hydrogen production module are transmitted to the power regulation unit of that module. During parameter transmission, the data transmission status is monitored in real time to ensure that each parameter is successfully transmitted to the target unit without data loss or errors. Once the parameters are loaded, the parameter update command for the model is triggered. Each functional module adjusts its operating logic according to the newly received corrected parameters, updates its internal operating parameter configuration, and monitors the operating status of the model in real time after loading. This confirms that each module can operate normally according to the corrected parameters, and that indicators such as energy transfer path, conversion efficiency, and system stability meet expectations. This completes the operation of dynamically loading the corrected parameter set into the digital twin model's operating environment.

[0121] The energy efficiency improvement target and the operational stability target are clearly defined as the core components of the multi-objective optimization framework. The energy efficiency improvement target focuses on increasing the energy conversion efficiency of the photothermal electrolysis hydrogen production system and reducing energy losses in each stage. Specifically, this is reflected in improving the light energy utilization rate of the photothermal acquisition module, the efficiency of converting electrical energy to chemical energy in the electrolysis hydrogen production module, and the heat energy retention rate of the heat transfer module, while reducing ineffective energy consumption during equipment operation and pipeline transmission. The operational stability target focuses on maintaining the stable operating status of each component of the system. Specifically, this is reflected in ensuring that key indicators such as operating temperature, working pressure, vibration frequency, current, and voltage of each component remain within safe ranges, avoiding sudden parameter changes, component failures, shutdowns, and large fluctuations in energy output. The equal priority relationship between the two targets is clearly defined, and a target-related logic is constructed. That is, any optimization operation must simultaneously consider energy efficiency improvement and operational stability, and the performance of one target should not be sacrificed for the performance of the other. This forms a multi-objective optimization framework with dual-objective synergistic optimization as the core and mutual constraints and balances between targets, providing clear directional guidance for subsequent parameter optimization.

[0122] Based on the parameter range after parameter adaptation and adjustment, the physical feasible boundaries of each parameter are defined. These boundaries are determined based on the hardware performance limits, design specifications, and actual operating experience of the photothermal electrolysis hydrogen production system components, ensuring that parameter combinations do not exceed the safe operating range of the components. All parameters involved in optimization are ranked according to their importance in affecting energy efficiency and stability, prioritizing the exploration of core parameters that have a more significant impact on these two objectives. During the parameter space exploration, a step-by-step expansion traversal method is adopted. First, secondary parameters are fixed as the baseline values ​​after adaptation and adjustment, and core parameters are adjusted one by one at reasonable intervals within their feasible boundaries. The energy efficiency performance and stability state of the system under different values ​​of each core parameter are recorded. Then, the core parameters are fixed as the initially selected optimal values, and secondary parameters are adjusted in the same way, recording the corresponding system operating data. During the exploration process, stability constraints are strictly followed. If a parameter combination causes the system operating temperature to exceed the safe range, the operating pressure to exceed the rated range, or the vibration frequency to exceed the standard, the test of that combination is immediately terminated and marked as infeasible. Only parameter combinations that meet the stability constraints and their corresponding energy efficiency data are retained, ensuring that the parameter space exploration is always carried out within a safe and feasible range.

[0123] During the parameter space exploration process, all parameter combinations that satisfy stability constraints are collected, along with the corresponding energy efficiency improvement and operational stability performance for each combination. Energy efficiency improvement is reflected by indicators such as the reduction rate of total system energy loss and the increase in conversion efficiency of key components. Operational stability performance is reflected by indicators such as parameter fluctuation amplitude, the occurrence of abnormal operating states, and the duration of continuous stable operation of components. The dual-objective performance of all parameter combinations is compared pairwise. If one parameter combination exhibits better energy efficiency improvement and operational stability than another, and no other parameter combination can improve operational stability without reducing its energy efficiency, or vice versa, then this parameter combination is at the Pareto front. Following this logic, all parameter combinations are compared and screened one by one, eliminating non-optimal combinations that are comprehensively surpassed by other combinations. All parameter combinations at the Pareto front are then compiled to form an optimized parameter solution set for a photothermal electrolysis hydrogen production system that simultaneously satisfies the dual objectives of energy efficiency improvement and operational stability. This solution set covers the optimal parameter selection under different energy efficiency and stability trade-offs.

[0124] Each parameter combination in the optimized parameter solution set is loaded one by one into the operating environment of the digital twin model. The continuous operation of the system under this parameter combination is simulated, and the stability indicators during the model operation are continuously monitored. These indicators include whether the operating temperature of each component remains within a safe range, whether the operating pressure remains stable within the rated range, whether the vibration frequency is below the critical threshold, whether the energy output fluctuation amplitude is controlled within the allowable limit, whether the component has fault alarm signals, and whether the system can maintain continuous and stable operation for the preset duration. The stability monitoring data of each parameter combination is systematically analyzed, and parameter combinations that exhibit unstable conditions such as parameter exceeding limits, excessive fluctuations, or component abnormalities during operation are eliminated. Parameter combinations that meet all monitoring indicators and have a stable operating state are retained. From the retained stable parameter combinations, their energy efficiency improvement performance is comprehensively evaluated, and combinations with more significant energy efficiency improvement are prioritized. If multiple combinations have similar energy efficiency performance, the combination with better operating stability indicators is selected. Finally, the parameter combination with stable operation and optimal energy efficiency is determined as the correction parameter set for the photothermal electrolysis hydrogen production system.

[0125] and These are artificially set weighting factors, used to measure the importance of energy efficiency improvement targets and operational stability targets in multi-objective optimization. It is the baseline energy loss value obtained by statistically analyzing the system's energy loss under baseline operating conditions using a digital twin model; It is the predicted energy loss value calculated by the digital twin model during simulation after parameter adaptation and adjustment; It is the monitoring result of the digital twin model on the stability of system operation after parameter adaptation and adjustment; This is the statistical result of the system's operational stability under baseline operating conditions, derived from the digital twin model. This formula is used to comprehensively evaluate the optimization effect of the photothermal electrolysis hydrogen production system under the dual objectives of energy efficiency improvement and operational stability. Among them, " "It partially quantifies the degree of energy efficiency improvement; the higher the value, the greater the reduction in energy loss and the more significant the energy efficiency improvement." The numerical value quantifies the degree of operational stability; a larger value indicates a smaller difference between the current operational stability and the baseline stability, signifying greater operational stability. Combining these two parts yields the multi-objective optimization function. This is used to determine the overall performance of the parameter combination in simultaneously meeting energy efficiency and stability targets. When the predicted energy loss value... When it decreases, The value of "" will increase; when the current operating stability Stability of operation relative to baseline When the difference decreases, " The value of "" will increase. Therefore, as the degree of energy efficiency improvement and operational stability increase, the multi-objective optimization function... The values ​​will show an increasing trend, indicating that the parameter combination performs better under the multi-objective optimization framework. The beneficial effects include: accurately extracting efficiency bottleneck types from the root cause analysis report; relying on a pre-set optimization strategy library structured by bottleneck type; precisely matching corresponding optimization strategy templates through keywords; combining energy loss points and related component information to form a targeted preliminary optimization plan, ensuring a high degree of alignment between the optimization direction and the bottleneck type; analyzing the adaptability of the initial reference parameters of the preliminary optimization plan with the current state based on real-time operating status data from the digital twin model and simulated environmental conditions and operating load levels; fine-tuning parameters through real-time monitoring of operating feedback to make the optimization plan more aligned with the actual operating needs of the system and improve the feasibility of the plan; and clarifying the system temperature... Stability constraints such as temperature, pressure, energy output fluctuations, and vibration frequency are considered. Focusing on multiple objectives—reducing energy loss, improving conversion efficiency, and extending component lifespan—the adjusted parameters are progressively refined and optimized, balancing multiple objectives with system stability, resulting in a scientifically sound set of corrected parameters. Through dedicated interfaces and corresponding data transmission paths, the integrity and format compatibility of the corrected parameter set are verified. Parameters are then accurately transmitted to each functional module of the model and updated, ensuring successful parameter loading and stable model operation with the new parameters. This achieves precise optimization of the photothermal electrolysis hydrogen production system, effectively addressing efficiency bottlenecks and improving system energy utilization efficiency and operational stability.

[0126] In this embodiment of the invention, S6, when the energy loss of the digital twin model reaches the minimum loss constraint, the correction parameters corresponding to the optimization strategy are output as the target optimization strategy of the photothermal electrolysis hydrogen production system.

[0127] To determine the minimum loss constraint of the photothermal electrolysis hydrogen production system, it is necessary to collect the historical best operating data of the system under ideal operating conditions, covering the minimum energy loss of each link, such as the minimum energy loss of photothermal conversion, electrolysis hydrogen production, and heat transfer. At the same time, combined with the energy consumption lower limit standard of the system design, the minimum constraint threshold of energy loss is comprehensively determined to ensure that the threshold meets the hardware performance limit of the system and is practically achievable.

[0128] After loading the modified parameter set into the digital twin model, the model is continuously simulated and run. Energy loss data of each module in the model, such as the photothermal acquisition module, the electrolytic hydrogen production module, and the pipeline transmission module, are collected in real time. These data are summarized in real time to calculate the overall energy loss value of the system. At the same time, the stability indicators such as the operating temperature, pressure, and vibration of each component are monitored to ensure that the system operation remains stable while monitoring energy loss.

[0129] The real-time calculated system energy loss value is continuously compared with the pre-determined minimum loss constraint threshold. If the energy loss value is equal to or less than the threshold and all system operation stability indicators meet the requirements, it is determined that the energy loss of the digital twin model has reached the minimum loss constraint. If the energy loss value is still higher than the threshold, the parameter set is adjusted and the process of loading, simulating, monitoring and comparing is repeated until the constraint conditions are met.

[0130] Once the energy loss of the digital twin model reaches the minimum loss constraint, the correction parameters loaded into the model at this time are extracted. These parameters cover the control, operation, and adjustment parameters of various modules such as photothermal harvesting, electrolytic hydrogen production, and energy conversion. The completeness and accuracy of these parameters are verified to ensure that the parameters are compatible with the various modules of the model. Subsequently, these parameters are compiled into a standardized parameter list, clarifying the module, function, and value corresponding to each parameter. This list serves as the target optimization strategy output for the photothermal electrolytic hydrogen production system, providing precise parameter guidance for the optimization and adjustment of the actual system.

[0131] The beneficial effects are as follows: by comprehensively analyzing historical best data and design standards, the minimum constraint threshold for system energy loss is clearly defined, providing a precise and practically achievable basis for determining optimization objectives; after loading the correction parameters into the digital twin model, the energy loss and stability indicators of each module are collected in real time, and the loss is continuously compared with the constraint threshold to ensure that the system operates stably while meeting the energy loss standard; when the model's energy loss meets the minimum constraint, the correction parameters are extracted and verified, and a standardized target optimization strategy is output, providing precise and reliable parameter guidance for the actual optimization and adjustment of the photothermal electrolysis hydrogen production system, helping the system achieve the dual goals of minimizing energy loss and ensuring operational stability.

[0132] like Figure 2 The diagram shown is a functional block diagram of an energy loss optimization system for a photothermal electrolysis hydrogen production system according to an embodiment of the present invention.

[0133] The energy loss optimization system 100 for a photothermal electrolysis hydrogen production system described in this invention can be installed in an electronic device. Depending on the functions implemented, the energy loss optimization system 100 may include a data acquisition and integration module 101, a digital twin modeling module 102, an energy flow simulation module 103, a loss point identification and root cause analysis module 104, an optimization strategy generation and parameter mapping module 105, and a target optimization strategy output module 106. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0134] In this embodiment, the functions of each module / unit are as follows:

[0135] The data acquisition and integration module 101 is used to collect real-time physical parameters and historical operating data of the photothermal electrolysis hydrogen production system through the Internet of Things platform, so as to obtain multi-source heterogeneous data of the photothermal electrolysis hydrogen production system.

[0136] The digital twin modeling module 102 is used to initialize the model of the photothermal electrolysis hydrogen production system based on the multi-source heterogeneous data, so as to construct a digital twin model of the photothermal electrolysis hydrogen production system.

[0137] The energy flow simulation module 103 is used to simulate the energy flow of the photothermal electrolysis hydrogen production system through the digital twin model, and obtain the energy flow distribution map of the photothermal electrolysis hydrogen production system;

[0138] The loss point identification and root cause analysis module 104 is used to identify energy loss points in the energy flow distribution diagram and analyze the root cause analysis report of the energy loss points based on the energy balance equation and efficiency optimization criteria of the photothermal electrolysis hydrogen production system.

[0139] The optimization strategy generation and parameter mapping module 105 is used to generate an optimization strategy for the photothermal electrolysis hydrogen production system based on the root cause analysis report, and to map the correction parameters of the optimization strategy to the digital twin model in real time.

[0140] The target optimization strategy output module 106 is used to output the correction parameters corresponding to the optimization strategy as the target optimization strategy for the photothermal electrolysis hydrogen production system when the energy loss of the digital twin model reaches the minimum loss constraint.

[0141] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0142] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0143] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0144] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0145] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0146] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for optimizing energy loss in a photothermal electrolysis hydrogen production system, characterized in that, The method includes: S1. Collect real-time physical parameters and historical operating data of the photothermal electrolysis hydrogen production system through the Internet of Things platform to obtain multi-source heterogeneous data of the photothermal electrolysis hydrogen production system. S2. Based on the multi-source heterogeneous data, the photothermal electrolysis hydrogen production system is initialized to construct a digital twin model of the photothermal electrolysis hydrogen production system. S3. Simulate the energy flow of the photothermal electrolysis hydrogen production system using the digital twin model to obtain the energy flow distribution map of the photothermal electrolysis hydrogen production system; S4. Identify the energy loss points in the energy flow distribution diagram, and analyze the root cause analysis report of the energy loss points based on the energy balance equation and efficiency optimization criteria of the photothermal electrolysis hydrogen production system; The identification of energy loss points in the energy flow distribution map includes: Based on the energy intensity gradient change, the energy flow distribution map is divided into multiple analysis regions to obtain the regional energy distribution feature set of the photothermal electrolysis hydrogen production system. Based on a preset loss mode feature library, suspicious regions that conform to abnormal energy loss characteristics in the energy distribution feature set of the region are identified, and a set of suspicious regions of the photothermal electrolysis hydrogen production system is obtained. Spatial correlation analysis was performed on the set of suspected areas to obtain the energy loss points of the photothermal electrolysis hydrogen production system; The root cause analysis report on the energy loss point based on the energy balance equation and efficiency optimization criteria of the photothermal electrolysis hydrogen production system includes: By traversing the spatial distribution information of the energy loss points and the topological relationship of the system components, a component association map of the energy loss points is constructed. Multi-dimensional rule matching is performed on the component association graph, and key components and interaction relationships that cause energy loss are identified through parallel rule reasoning to generate the component interaction network of the energy loss points. Based on the deviation between the real-time operating data and the benchmark parameters at the energy loss points, the efficiency bottleneck of energy transfer between components is analyzed to obtain a root cause analysis report of the energy loss points. S5. Based on the root cause analysis report, generate an optimization strategy for the photothermal electrolysis hydrogen production system, and map the correction parameters of the optimization strategy to the digital twin model in real time. The process of generating an optimization strategy for the photothermal electrolysis hydrogen production system based on the root cause analysis report, and mapping the correction parameters of the optimization strategy to the digital twin model in real time, includes: Based on the efficiency bottleneck type in the root cause analysis report, the corresponding optimization strategy template is retrieved from the preset optimization strategy library to obtain the preliminary optimization scheme of the photothermal electrolysis hydrogen production system. Based on the current operating status of the digital twin model, the parameters of the preliminary optimization scheme are adjusted for adaptation. Under stability constraints, the parameters for parameter adaptation and adjustment are optimized using multiple objectives to obtain the modified parameter set of the photothermal electrolysis hydrogen production system; The modified parameter set is dynamically loaded into the runtime environment of the digital twin model; The parameters for parameter adaptation and adjustment under stability constraints are subjected to multi-objective optimization to obtain a set of corrected parameters for the photothermal electrolysis hydrogen production system, including: The optimization framework is based on multiple objectives, including energy efficiency improvement and operational stability. In the multi-objective optimization framework, the parameter space is explored for the parameters that are adapted and adjusted under stability constraints. By identifying candidate parameter combinations that satisfy the multi-objective optimization framework through Pareto front analysis, the optimal parameter solution set of the photothermal electrolysis hydrogen production system is obtained. The stable set of optimized parameters is used as the set of corrected parameters for the photothermal electrolysis hydrogen production system. The calculation method for the multi-objective optimization is as follows: ; In the formula, For multi-objective optimization functions, The weighting factor corresponding to the energy efficiency improvement target. The weighting factor corresponding to the stable operation target. This is the baseline energy loss value for the digital twin model. The energy loss prediction value of the digital twin model after adapting and adjusting the parameters. The current system stability of the digital twin model after the parameters have been adapted and adjusted. S6. When the energy loss of the digital twin model reaches the minimum loss constraint, the correction parameters corresponding to the optimization strategy are output as the target optimization strategy of the photothermal electrolysis hydrogen production system.

2. The method for optimizing energy loss in a photothermal electrolysis hydrogen production system as described in claim 1, characterized in that, The process involves collecting real-time physical parameters and historical operating data of the photothermal electrolysis hydrogen production system through an IoT platform to obtain multi-source heterogeneous data of the system, including: Distributed data integration of multi-source heterogeneous data is performed to obtain a unified data platform for the photothermal electrolysis hydrogen production system. Through the unified data platform, the dynamic behavior characteristics of the photothermal electrolysis hydrogen production system that are highly related to energy loss are analyzed to obtain the feature description set of the photothermal electrolysis hydrogen production system. The feature description set is loaded into a predefined model template and initial parameters are configured to construct a digital twin model of the photothermal electrolysis hydrogen production system.

3. The method for optimizing energy loss in a photothermal electrolysis hydrogen production system as described in claim 1, characterized in that, The process of simulating the energy flow of the photothermal electrolysis hydrogen production system using the digital twin model to obtain the energy flow distribution map of the photothermal electrolysis hydrogen production system includes: After inputting the current system state parameters of the photothermal electrolysis hydrogen production system into the digital twin model, the energy transfer path in the digital twin model is dynamically tracked to obtain the energy flow trajectory data of the photothermal electrolysis hydrogen production system. By identifying the energy conversion nodes and loss nodes in the energy flow trajectory data, the node energy distribution dataset of the photothermal electrolysis hydrogen production system is obtained. Based on the energy flow direction and intensity gradient in the node energy distribution dataset, an energy flow distribution map of the photothermal electrolysis hydrogen production system is generated.

4. The method for optimizing energy loss in a photothermal electrolysis hydrogen production system as described in claim 3, characterized in that, The process of identifying energy conversion nodes and loss nodes in the energy flow trajectory data to obtain the node energy distribution dataset of the photothermal electrolysis hydrogen production system includes: Extract the energy flow rate change rate and conversion efficiency index from the energy flow trajectory data to obtain the node feature set of the photothermal electrolysis hydrogen production system; By comparing the input and output differences of node energy in the node feature set, the energy anomaly nodes of the photothermal electrolysis hydrogen production system are obtained. The node type and corresponding energy value of the energy anomaly nodes are labeled to generate a node energy distribution dataset for the photothermal electrolysis hydrogen production system.

5. An energy loss optimization system for a photothermal electrolysis hydrogen production system, used to implement the energy loss optimization method for a photothermal electrolysis hydrogen production system as described in claim 1, the system comprising: The data acquisition and integration module is used to collect real-time physical parameters and historical operating data of the photothermal electrolysis hydrogen production system through the Internet of Things platform, so as to obtain multi-source heterogeneous data of the photothermal electrolysis hydrogen production system. The digital twin modeling module is used to initialize the model of the photothermal electrolysis hydrogen production system based on the multi-source heterogeneous data, so as to construct a digital twin model of the photothermal electrolysis hydrogen production system. The energy flow simulation module is used to simulate the energy flow of the photothermal electrolysis hydrogen production system through the digital twin model, and obtain the energy flow distribution map of the photothermal electrolysis hydrogen production system; The energy loss point identification and root cause analysis module is used to identify energy loss points in the energy flow distribution diagram and analyze the root cause analysis report of the energy loss points based on the energy balance equation and efficiency optimization criteria of the photothermal electrolysis hydrogen production system. The identification of energy loss points in the energy flow distribution map includes: Based on the energy intensity gradient change, the energy flow distribution map is divided into multiple analysis regions to obtain the regional energy distribution feature set of the photothermal electrolysis hydrogen production system. Based on a preset loss mode feature library, suspicious regions that conform to abnormal energy loss characteristics in the energy distribution feature set of the region are identified, and a set of suspicious regions of the photothermal electrolysis hydrogen production system is obtained. Spatial correlation analysis was performed on the set of suspected areas to obtain the energy loss points of the photothermal electrolysis hydrogen production system; The root cause analysis report on the energy loss point based on the energy balance equation and efficiency optimization criteria of the photothermal electrolysis hydrogen production system includes: By traversing the spatial distribution information of the energy loss points and the topological relationship of the system components, a component association map of the energy loss points is constructed. Multi-dimensional rule matching is performed on the component association graph, and key components and interaction relationships that cause energy loss are identified through parallel rule reasoning to generate the component interaction network of the energy loss points. Based on the deviation between the real-time operating data and the benchmark parameters at the energy loss points, the efficiency bottleneck of energy transfer between components is analyzed to obtain a root cause analysis report of the energy loss points. The optimization strategy generation and parameter mapping module is used to generate an optimization strategy for the photothermal electrolysis hydrogen production system based on the root cause analysis report, and to map the correction parameters of the optimization strategy to the digital twin model in real time. The process of generating an optimization strategy for the photothermal electrolysis hydrogen production system based on the root cause analysis report, and mapping the correction parameters of the optimization strategy to the digital twin model in real time, includes: Based on the efficiency bottleneck type in the root cause analysis report, the corresponding optimization strategy template is retrieved from the preset optimization strategy library to obtain the preliminary optimization scheme of the photothermal electrolysis hydrogen production system. Based on the current operating status of the digital twin model, the parameters of the preliminary optimization scheme are adjusted for adaptation. Under stability constraints, the parameters for parameter adaptation and adjustment are optimized using multiple objectives to obtain the modified parameter set of the photothermal electrolysis hydrogen production system; The modified parameter set is dynamically loaded into the runtime environment of the digital twin model; The parameters for parameter adaptation and adjustment under stability constraints are subjected to multi-objective optimization to obtain a set of corrected parameters for the photothermal electrolysis hydrogen production system, including: The optimization framework is based on multiple objectives, including energy efficiency improvement and operational stability. In the multi-objective optimization framework, the parameter space is explored for the parameters that are adapted and adjusted under stability constraints. By identifying candidate parameter combinations that satisfy the multi-objective optimization framework through Pareto front analysis, the optimal parameter solution set of the photothermal electrolysis hydrogen production system is obtained. The stable set of optimized parameters is used as the set of corrected parameters for the photothermal electrolysis hydrogen production system. The calculation method for the multi-objective optimization is as follows: ; In the formula, For multi-objective optimization functions, The weighting factor corresponding to the energy efficiency improvement target. The weighting factor corresponding to the stable operation target. This is the baseline energy loss value for the digital twin model. The energy loss prediction value of the digital twin model after adapting and adjusting the parameters. The current system stability of the digital twin model after the parameters have been adapted and adjusted. The system baseline operational stability of the digital twin model; The target optimization strategy output module is used to output the correction parameters corresponding to the optimization strategy as the target optimization strategy of the photothermal electrolysis hydrogen production system when the energy loss of the digital twin model reaches the minimum loss constraint.

Images