Museum cultural relic environment monitoring method and system based on fusion technology

Abstract

The invention discloses a museum cultural relic environment monitoring method and system based on a fusion technology. The method comprises the following steps: monitoring a cultural relic circuit in real time by using a graph convolutional neural network, extracting node features to construct a logic circuit directed graph, determining fault nodes, and obtaining a fault diagnosis result; obtaining fixed reference objects and environment perception data in the museum to form environment information, inputting the environment information into a multi-reference object system analysis model, and analyzing the influence of environment factors on fault nodes to obtain an environment analysis result; combining the two results to position a cultural relic associated equipment power device which needs to be optimized emphatically; carrying out refined segmentation, simulation and optimization adjustment on the devices by utilizing a simulation technology, and generating a targeted power regulation and control strategy; and outputting the strategy so as to carry out environment adjustment and equipment maintenance treatment according to the strategy. By implementing the method provided by the invention, collaborative linkage of cultural relic state monitoring, environmental risk assessment and equipment power regulation and control can be realized, and comprehensive guarantee is provided for safety of cultural relics in a museum.

Description

Technical Field

[0001] This invention relates to the field of museum artifact monitoring technology, and more specifically to a method and system for environmental monitoring of museum artifacts based on fusion technology. Background Technology

[0002] The preservation environment of cultural relics in museums is complex and ever-changing, with inherent malfunctions of the relics themselves intertwined with risks from the surrounding environment. Traditional monitoring methods are often independent, failing to achieve synergistic analysis of malfunctions in relics and environmental factors, making it difficult to accurately and efficiently ensure the safety of relics. On the one hand, museum exhibition halls contain various environmental factors, such as temperature, humidity, light intensity, and air quality. These factors not only directly affect the preservation status of relics but may also indirectly affect the performance of their electrical systems. For example, increased humidity may lead to corrosion of circuit components and an increased risk of short circuits, while prolonged overload of light intensity may degrade the performance of photosensitive components in the circuit. However, most existing methods for diagnosing circuit malfunctions in relics ignore the coupling effect of these external environmental factors, analyzing only the electrical characteristics of the circuit itself. This limits the malfunction diagnosis models in complex real-world environments, making it difficult to comprehensively and accurately capture signs of malfunctions. On the other hand, the preciousness and non-renewable nature of museum relics determine the special importance of their monitoring. As important carriers of human history and culture, cultural relics possess extremely high historical, artistic, and scientific value. Once relics are damaged due to malfunctions or environmental factors, it will cause irreparable loss. Meanwhile, as people's awareness of cultural heritage protection continues to rise, public concern about the safety of museum artifacts is also increasing. The inadequacy of traditional monitoring methods not only fails to meet the high requirements of modern museums for artifact safety but may also affect the professional image of museums and public trust.

[0003] Therefore, it is necessary to design a new method to achieve coordinated linkage between cultural relic status monitoring, environmental risk assessment and equipment power regulation, so as to provide comprehensive protection for the safety of museum cultural relics. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and system for environmental monitoring of museum artifacts based on fusion technology.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a museum artifact environmental monitoring method based on fusion technology, comprising:

[0006] Graph convolutional neural network technology is used to monitor the circuit system involved in museum artifacts in real time, extract the features of the circuit nodes of the artifacts to construct a directed graph of logic circuits, and identify fault nodes to obtain fault diagnosis results.

[0007] Obtain fixed reference points and environmental perception data within the museum to obtain environmental information;

[0008] The environmental information is input into a multi-reference system analysis model to analyze the potential impact of environmental factors on the fault node, so as to obtain environmental analysis results.

[0009] Based on the fault diagnosis results and the environmental analysis results, the power devices of cultural relic-related equipment that need to be optimized should be identified.

[0010] By utilizing simulation technology in conjunction with equipment operating conditions and cultural relic protection needs, the power devices of the cultural relic-related equipment are finely segmented, simulated, and optimized to generate targeted power control strategies.

[0011] The power control strategy is output to facilitate environmental adjustments and equipment maintenance based on the power control strategy.

[0012] On the other hand, the present invention also provides a museum artifact environmental monitoring system based on fusion technology, comprising:

[0013] The fault diagnosis unit is used to monitor the circuit system involved in the museum artifacts in real time using graph convolutional neural network technology, in order to extract the features of the artifact circuit nodes to construct a directed graph of logic circuits and identify fault nodes to obtain fault diagnosis results.

[0014] The environmental information acquisition unit is used to acquire fixed reference objects and environmental perception data within the museum to obtain environmental information.

[0015] The environmental analysis unit is used to input the environmental information into the multi-reference system analysis model to analyze the potential impact of environmental factors on the fault node, so as to obtain the environmental analysis results.

[0016] The positioning unit is used to locate the power devices of cultural relic-related equipment that need to be optimized based on the fault diagnosis results and the environmental analysis results.

[0017] The simulation unit is used to perform fine segmentation, simulation and optimization adjustment of the power devices of the cultural relic-related equipment by combining simulation technology with the equipment operating conditions and cultural relic protection needs, and to generate targeted power control strategies.

[0018] The output unit is used to output the power control strategy so as to facilitate environmental adjustments and equipment maintenance based on the power control strategy.

[0019] The advantages of this invention compared to existing technologies are as follows: This invention uses graph convolutional neural network technology to monitor the circuit system of museum artifacts in real time, accurately extracts circuit node features to construct a directed graph of logic circuits, quickly identifies faulty nodes, and obtains fault diagnosis results. Simultaneously, it acquires fixed reference objects and environmental perception data within the museum to obtain comprehensive environmental information, which is then input into a multi-reference object system analysis model to deeply analyze the potential impact of environmental factors on faulty nodes, yielding environmental analysis results. Combining the fault diagnosis and environmental analysis results, it accurately locates the power devices of artifact-related equipment that require key optimization. Next, using simulation technology, based on equipment operating conditions and artifact protection needs, it performs refined segmentation, simulation, and optimization adjustments of the power devices, generating targeted power control strategies. Finally, it outputs the power control strategies, providing a clear basis for environmental adjustments and equipment maintenance, achieving synergistic linkage between artifact status monitoring, environmental risk assessment, and equipment power control, comprehensively ensuring the safety of museum artifacts.

[0020] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram illustrating an application scenario of the museum artifact environmental monitoring method based on fusion technology provided in an embodiment of the present invention.

[0023] Figure 2 A flowchart illustrating the museum artifact environmental monitoring method based on fusion technology provided in this embodiment of the invention;

[0024] Figure 3 A schematic block diagram of a museum artifact environmental monitoring system based on fusion technology provided in an embodiment of the present invention;

[0025] Figure 4 A schematic block diagram of a computer device provided for an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0028] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0029] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0030] Please see Figure 1 , Figure 1 This is a schematic diagram illustrating an application scenario of the museum artifact environmental monitoring method based on fusion technology provided in an embodiment of the present invention. Figure 2 This is a schematic flowchart illustrating a museum artifact environmental monitoring method based on fusion technology, provided in an embodiment of the present invention. This method is applied to a server. The server interacts with the terminal, achieving coordinated monitoring of artifact status, environmental risk assessment, and equipment power regulation through the fusion of graph convolutional neural networks, multi-reference system analysis models, and simulation technology, providing comprehensive protection for the safety of museum artifacts. First, graph convolutional neural networks are used to monitor artifact circuits in real time, constructing a directed graph of the logic circuits and identifying fault nodes to obtain fault diagnosis results. Simultaneously, fixed reference objects and environmental perception data within the museum are acquired to form environmental information, which is then input into the multi-reference system analysis model to analyze the impact of environmental factors on fault nodes, obtaining environmental analysis results. Next, combining the fault diagnosis and environmental analysis results, the power devices of artifact-related equipment requiring key optimization are located. Then, simulation technology is used to finely segment, simulate, and optimize these devices, generating targeted power regulation strategies. Finally, the power regulation strategies are output so that environmental adjustments and equipment maintenance can be carried out accordingly, ensuring the stability of the artifact preservation environment and the reliable operation of the equipment.

[0031] Figure 2 This is a flowchart illustrating the museum artifact environmental monitoring method based on fusion technology provided in an embodiment of the present invention. Figure 2 As shown, the method includes the following steps S110 to S160.

[0032] S110. Graph convolutional neural network technology is used to monitor the circuit system involved in museum artifacts in real time, so as to extract the features of the circuit nodes of the artifacts to construct a directed graph of logic circuits and identify fault nodes to obtain fault diagnosis results.

[0033] In this embodiment, the fault diagnosis result refers to the accurate identification of fault nodes and their severity through real-time monitoring of the circuit system, generating a diagnostic report that includes the fault location, type, and scope of impact, providing key basis for subsequent environmental analysis and equipment optimization.

[0034] In one embodiment, step S110 may include steps S111 to S115.

[0035] S111. Obtain the netlist file of the combinational logic artifact circuit to be diagnosed, construct a directed graph of the logic circuit, extract the feature vectors of each node containing electrical attributes, logical functions and connection relationship features, and integrate them to generate a node feature matrix.

[0036] In this embodiment, the node feature matrix refers to a matrix formed by integrating the feature vectors of each node (including electrical attributes, logical functions, and connection relationship features). This matrix comprehensively characterizes the features of each node in the circuit, providing data support for subsequent fault diagnosis model training.

[0037] Specifically, a netlist file is an important document describing the circuit structure and connections. It contains the logic gate types, input pin information, output pin information, and connections between logic gates in the circuit. By reading the netlist file, the basic information needed to construct the directed graph of the logic circuit can be obtained.

[0038] Based on the information in the netlist file, a directed graph of the logic circuit is constructed. Nodes in the graph represent logic gates or components in the circuit, and edges represent the connections between them. During the construction process, the netlist file is first read to identify each component and its connection relationships. Then, a data structure (such as an adjacency list) for storing nodes and directed edges is initialized. Next, each component in the netlist file is traversed, mapped to nodes in the data structure, and corresponding edges are added to represent the connection relationships.

[0039] For each node in a directed graph of a logic circuit, extract its feature vector, which contains electrical attributes, logical functions, and connectivity characteristics. Electrical attributes may include parameters such as the node's voltage and current; logical functions refer to the type of logic gate the node represents (such as AND gate, OR gate, etc.); connectivity characteristics describe how and where the node is connected to other nodes. For example, the node's degree (i.e., the number of edges connected to it), the types of adjacent nodes, and the node's level.

[0040] The extracted feature vectors from each node are integrated to form a node feature matrix. Each row of the matrix represents a node, and each column represents a specific feature. In this way, complex circuit structures and node characteristics can be transformed into a matrix form that is easy to analyze and process, providing a unified data format for subsequent fault diagnosis model training.

[0041] S112. Inject a random number of faults into random node positions of the combinational logic artifact circuit to be diagnosed, generate random test vectors, and use a circuit simulation model to obtain the failure response data of the circuit after fault injection.

[0042] In this embodiment, failure response data refers to data reflecting changes in circuit performance obtained through a simulation model after fault injection. This data is used to construct a fault element graph set, providing a foundation for subsequent fault training data generation.

[0043] In this embodiment, failure response data refers to data reflecting changes in circuit performance obtained through a simulation model after fault injection. This data is used to construct a fault element graph set, providing a foundation for subsequent fault training data generation.

[0044] To simulate potential fault conditions in real-world circuits, a random number of faults are injected into random node locations of the combinational logic artifact circuit to be diagnosed. Fault types can include short circuits, open circuits, parameter drift, etc., and the injection locations and numbers are determined using a specific algorithm or random number generator.

[0045] Based on the circuit's input characteristics, a series of random test vectors are generated. These test vectors serve as input signals to the circuit, stimulating its behavior and observing its response. The test vectors should be designed to cover as many operating states as possible to comprehensively assess the circuit's performance.

[0046] Using professional circuit simulation tools or custom simulation algorithms, a simulation model of the combinational logic artifact circuit to be diagnosed is constructed. This model can simulate the circuit's behavior under normal and fault conditions and output corresponding response data.

[0047] The generated random test vectors are input into the circuit simulation model, and the simulation is run. During the simulation, the circuit's output response after fault injection is recorded; this data constitutes the failure response data. By analyzing the failure response data, we can understand the performance changes of the circuit under different fault conditions, thus providing a basis for subsequent fault analysis and diagnosis.

[0048] S113. Construct a fault element graph set from the failure response data using the fault element graph method, calculate the fault score of each node and associate it with the corresponding fault, and generate fault training data containing fault features and association information.

[0049] In this embodiment, fault training data refers to a dataset containing fault features and related information. By preprocessing and extracting features from the failure response data, a fault element graph set is constructed, and the fault score of each node is calculated, ultimately forming a dataset used to train the fault diagnosis model.

[0050] Specifically, the construction process of the fault element graph set is as follows:

[0051] The failure response data is preprocessed to remove noise and interference, and key features that reflect the fault state are extracted using a feature extraction algorithm to form a fault feature set.

[0052] Based on the circuit design documents, draw a basic circuit diagram that includes nodes, connections, and logic functions;

[0053] The feature values ​​in the fault feature set are mapped to the corresponding nodes of the basic circuit diagram to generate a feature map that integrates fault information.

[0054] The feature map is analyzed to identify nodes with significant features in order to obtain potential fault points;

[0055] All potential fault points and their interrelationships are integrated into a fault element diagram to form a complete fault element diagram. Specifically, association rules between all potential fault points are established based on the circuit topology and fault propagation characteristics. An association matrix is ​​created to quantify the relationships between all potential fault points. A graphical tool is used to draw the fault element diagram, with potential fault points as nodes. Lines are drawn between relevant nodes according to the association matrix. Fault information is added to present the details of each point. The accuracy and practicality of the diagram are verified and optimized by comparing it with actual faults.

[0056] Specifically, by analyzing the feature map, nodes whose feature values ​​significantly deviate from the normal range are identified as potential fault points. For example, a sudden and significant increase in the voltage value or abnormal fluctuations in the current value of a node. Based on the circuit topology and fault propagation characteristics, association rules between potential fault points are established. For example, if there is a direct electrical connection between two potential fault points, and the feature value of the other node also shows abnormal changes when one fault point fails, then a correlation is considered to exist between them. This can be considered from the following aspects:

[0057] Physical connection relationship: If there is a direct physical connection between two nodes, such as a wire connection or a direct connection of components, their fault states may affect each other.

[0058] Signal transmission relationship: In digital circuits, signals are transmitted from one node to another. If the signal source node fails, it may cause abnormal signals at the receiving node. This causal relationship can also be used as a basis for correlation.

[0059] Fault propagation path: By analyzing the fault modes and effects of the circuit, the possible propagation paths of the fault are determined, and there may be correlations between the nodes along these paths.

[0060] Constructing the association matrix: Create an association matrix to represent the relationships between potential failure points. The rows and columns of the matrix represent the individual potential failure points, and the elements in the matrix represent the strength or type of association between two failure points. Binary representation (1 for association, 0 for no association) or more complex weighted representations (such as numerical values ​​for association strength) can be used.

[0061] Use graphical tools (such as professional drawing software, programming libraries, etc.) to draw a fault element diagram. Represent each potential fault point as a node in the diagram, and draw lines between related nodes according to the correlation matrix. The lines can use different colors, line styles, or thicknesses to represent the strength and type of the correlation. For example, strong correlations are represented by thick solid lines, and weak correlations by thin dashed lines, etc.

[0062] Add labels to each node in the fault element diagram to display detailed fault information for that node, such as fault type (short circuit, open circuit, parameter drift, etc.), fault severity (e.g., percentage of voltage deviation), and characteristic values. This helps to more intuitively understand the specific situation of each potential fault point.

[0063] The constructed fault element diagram is compared and verified with actual circuit fault conditions. Through real-world fault cases or fault injection experiments, the potential fault points and their relationships identified in the diagram are checked to ensure they match the actual situation. Based on the verification results, the fault element diagram is adjusted and optimized, such as correcting inaccurate relationships and adding missing potential fault points, to improve its accuracy and usability.

[0064] By following the steps above, all potential fault points and their interrelationships can be integrated into a fault element diagram, forming an intuitive and accurate fault visualization tool that provides strong support for fault diagnosis and repair.

[0065] The failure response data is filtered and denoised to remove outliers and interference, retaining only the valid information reflecting the circuit fault state. Appropriate methods (such as wavelet analysis and principal component analysis) are used to extract key data characterizing the fault features, forming a fault feature set. Based on the circuit design documents, a detailed basic circuit diagram including nodes, connections, and logical functions is drawn, clearly defining each component and its connection method. Feature values ​​from the fault feature set are mapped to corresponding nodes in the basic circuit diagram, generating a feature map that integrates fault information, visually presenting the distribution of faults in the circuit. The feature map is analyzed to identify nodes with significant features, i.e., potential fault points. Based on the circuit topology and fault propagation characteristics, association rules between potential fault points are established, and an association matrix is ​​created to quantify their relationships. A fault element diagram is drawn using graphical tools, with potential fault points as nodes. Connections are drawn between relevant nodes according to the association matrix, and fault information is added to present detailed information about each point. The accuracy and practicality of the diagram are verified and optimized by comparing it with actual faults, ultimately forming a complete set of fault element diagrams.

[0066] Based on the saliency of the features of each node in the fault element graph, a fault score is calculated for each node. For example, the fault score is determined based on factors such as the frequency of occurrence and the scope of influence of the node in the fault element graph; the fault score of each node is then associated with the corresponding fault type to clarify the fault mode corresponding to each fault score.

[0067] By integrating the fault scores and corresponding fault type information of each node, fault training data containing fault features and correlation information is generated. This data will be used to train the fault diagnosis model, enabling it to learn the mapping relationship between fault features and fault types.

[0068] S114. Input the fault training data into the preprocessed graph convolutional neural network model, set the model parameters and initialize the weights and bias parameters, and continuously optimize the model parameters through forward propagation and backward propagation until the maximum number of iterations is reached, so as to build an accurate fault diagnosis model.

[0069] In this embodiment, the fault diagnosis model refers to a trained graph convolutional neural network model that can accurately analyze real-time failure response data and identify fault nodes.

[0070] Specifically, the graph convolutional neural network model is preprocessed to ensure that it can receive fault training data and be trained effectively.

[0071] Determine the structure of the graph convolutional neural network model, including the number of layers, node feature dimensions, and hidden layer dimensions. Initialize the model's weights and bias parameters, typically using random initialization methods such as Xavier initialization or He initialization.

[0072] The fault training data is input into a graph convolutional neural network model for forward propagation. Graph convolution layers perform graph convolution operations on the node feature matrix and neighboring node information, aggregating neighboring node information and updating the node feature matrix. The updated node feature matrix is ​​then input into a fully connected layer to convert it into predicted fault scores.

[0073] A loss function is calculated based on the predicted fault score and the actual fault score. Commonly used loss functions include mean squared error (MSE) and cross-entropy loss.

[0074] The gradient of the loss function with respect to the model parameters is calculated using the backpropagation algorithm. Then, the model weights and bias parameters are updated based on the calculated gradient using gradient descent.

[0075] The process of forward propagation, loss function calculation, backpropagation, and parameter update is repeated until the preset maximum number of iterations is reached. Ultimately, a fault diagnosis model capable of accurately analyzing real-time failure response data and identifying fault nodes is constructed.

[0076] S115. Based on the constructed fault diagnosis model, analyze the real-time failure response data of the combinational logic circuit artifact to be diagnosed, determine the fault nodes in the circuit, and obtain the fault diagnosis results.

[0077] In this embodiment, real-time failure response data of the combinational logic circuit artifact to be diagnosed is collected during actual operation.

[0078] The acquired real-time failure response data is preprocessed, including noise removal and normalization, to make it conform to the input requirements of the fault diagnosis model.

[0079] The preprocessed real-time failure response data is input into the constructed fault diagnosis model. Based on the fault characteristics and correlation information learned during training, the model analyzes the real-time data to determine the fault nodes in the circuit.

[0080] Based on the model's output, fault diagnosis results are generated. These results include information such as the location, type, and severity of the faulty node, providing guidance for subsequent repair and maintenance work.

[0081] S120. Obtain fixed reference objects and environmental perception data within the museum to obtain environmental information.

[0082] In this embodiment, environmental information refers to various data related to the preservation environment of cultural relics within the museum. These data comprehensively reflect the environmental conditions within the museum, providing a basis for subsequent analysis of the potential impact of the environment on the circuit fault nodes of cultural relics.

[0083] Specifically, the fixed reference objects include display cases, building structures, etc., which have definite spatial locations and physical attributes within the museum and can be used to locate and calibrate environmental sensing data. The environmental sensing data includes real-time collected temperature, humidity, and illumination information. This data is collected through a sensor network deployed within the museum and reflects real-time changes in environmental conditions. The environmental information integrates the location attributes of the fixed reference objects and the real-time monitoring results of the environmental sensing data, providing detailed data support for subsequent analysis of the impact of environmental factors on fault nodes using a multi-reference system analysis model.

[0084] In the museum artifact monitoring method, step S120 involves acquiring fixed reference points and environmental sensing data within the museum to construct comprehensive environmental information. Environmental information is a comprehensive description of the museum's environmental state, integrating detailed data from various environmental parameters and fixed reference points. This provides a foundation for subsequent analysis of the potential impact of the environment on artifact circuit fault nodes. Specifically, fixed reference points mainly include display cases and building structures. These reference points have definite locations and physical attributes within the museum, providing spatial positioning and structural references for environmental analysis. Environmental sensing data encompasses key environmental parameters such as real-time temperature, humidity, and light intensity. This data is collected through various sensors deployed within the museum, reflecting real-time changes in the museum's environment. By integrating fixed reference point information and environmental sensing data, the resulting environmental information not only includes the physical layout of the museum but also reflects the current environmental state, providing detailed data support for subsequent analysis of the impact of environmental factors on fault nodes using a multi-reference point system analysis model.

[0085] Specifically, in the process of monitoring museum artifacts, environmental data, such as temperature, humidity, light intensity, and air quality, are first collected in real time by various sensors deployed in the exhibition hall. Simultaneously, the location and attribute information of fixed reference objects within the museum (such as display cases, building columns, and walls) are recorded. This information is then integrated to obtain comprehensive environmental data. Next, this environmental information is input into a multi-reference system analysis model. This model, combining the relative positions of the fixed reference objects and environmental changes, analyzes the potential impact of environmental factors (such as changes in temperature and humidity, and fluctuations in light intensity) on the artifact circuit fault nodes identified in the previous steps, thereby obtaining environmental analysis results. Specifically, the model dynamically adjusts the parameters in the artifact circuit fault feature database based on the spatial relationships (distance, orientation, etc.) between the reference objects and the artifact's circuit equipment, as well as environmental change trends (such as the rate of increase or decrease in temperature and humidity, and the magnitude of changes in light intensity), to reflect the influence of environmental factors on fault characteristics. Meanwhile, the model continuously receives real-time reference data and environmental parameters. When abnormal signs are detected in the artifact's circuit (such as abnormal fluctuations in current and voltage, sudden changes in node temperature, etc.) and the abnormal area is near potential environmental risk points (such as being close to high humidity areas or areas with strong light), the model will comprehensively analyze the blocking effect of the reference objects on environmental risks (such as the isolation effect of the display case's sealing degree on humidity, the blocking of light by the building structure, etc.) and environmental change trends, thereby determining the risk level of the fault and providing a basis for decision-making in subsequent fault handling and environmental control.

[0086] S130. Input the environmental information into the multi-reference system analysis model to analyze the potential impact of environmental factors on the fault node, so as to obtain the environmental analysis results.

[0087] In this embodiment, the environmental analysis results refer to the quantitative assessment of how environmental factors within the museum affect the fault nodes of the artifact circuits. It provides detailed information on the impact of the environment on the fault nodes, providing a basis for subsequent fault risk assessment and equipment optimization.

[0088] In one embodiment, step S130 described above may include steps S131 to S133.

[0089] S131. Construct a multi-reference system analysis model, initialize the model parameters, input the environmental information into the multi-reference system analysis model, and integrate environmental features and spatial location information of reference objects to construct a feature vector.

[0090] The construction of a multi-reference system analysis model is fundamental to analyzing the impact of environmental factors on failure nodes. This model, based on the museum's architectural structure, integrates fixed reference points and environmental perception data, enabling a comprehensive analysis of environmental factors within the museum.

[0091] Initializing model parameters is crucial for the accuracy and reliability of the model. By setting the initial parameters appropriately, the stability of the model during training and operation can be ensured.

[0092] The collected environmental information is input into the multi-reference system analysis model, enabling the model to obtain the real-time environmental status within the museum.

[0093] In the model, environmental features and spatial location information of reference objects are integrated to form a comprehensive feature vector. This feature vector can comprehensively reflect the environmental conditions within the museum and the positional relationships of fixed reference objects, providing data support for subsequent environmental factor analysis.

[0094] In this embodiment, step S131 may include steps S1311 to S1314.

[0095] S1311. Construct a three-dimensional spatial framework based on the museum building structure, import the position and attribute information of fixed reference objects, design the input, hidden and output layers of the model, and complete the construction of the multi-reference object system analysis model; among which, the hidden layer includes a feature extraction sub-layer that uses neurons to extract key features, a feature fusion sub-layer that integrates environmental and reference object features to form a comprehensive feature vector, and a nonlinear transformation sub-layer that enhances the model's expressive ability through activation functions.

[0096] S1312. Initialize the model parameters using historical data and expert experience, normalize the input data, and verify and adjust the model parameters using known data to complete the initialization of the model parameters.

[0097] S1313. The environmental information is preprocessed, and the data format is converted before being input into the multi-reference system analysis model;

[0098] S1314. Extract key environmental features from the environmental information and spatial location features from the fixed reference information, and fuse these features to construct a comprehensive feature vector.

[0099] Specifically, a three-dimensional spatial framework is established based on the museum's architectural structure. This framework can be created using the museum's architectural drawings, laser scan data, or Building Information Modeling (BIM), ensuring that the framework accurately reflects the museum's internal spatial layout.

[0100] The location and attribute information of fixed reference objects (such as display cases, building columns, walls, etc.) are imported into the three-dimensional spatial framework. This data can be obtained through manual surveying, BIM models, or the museum's existing facility management database, ensuring that the location and attributes of each reference object in three-dimensional space are accurately recorded.

[0101] The internal structure of the design model includes defining the input layer, hidden layer, and output layer. The input layer receives environmental perception data and fixed reference information, the hidden layer contains neurons or computational units for data processing and feature extraction, and the output layer outputs the model's analysis results, such as the degree of influence of environmental factors on faulty nodes.

[0102] The parameters in the model are initialized based on historical data, expert experience, or machine learning algorithms. For example, the blocking coefficient of a fixed reference object to different environmental factors is set, as well as the weight of the influence of environmental parameters on the circuit failure of the cultural relic. For neural network models, the connection weights and bias terms between neurons are initialized.

[0103] Normalizing the input data transforms environmentally perceived data and fixed reference information with different dimensions and numerical ranges into the same numerical interval (e.g., [0, 1] or [-1, 1]). This helps improve the training efficiency and stability of the model, ensuring that different types of input data contribute equally to the model.

[0104] The model was initially validated using some known environmental data and failure cases to evaluate its performance and accuracy. Based on the validation results, the model parameters were adjusted and optimized to ensure that the model can accurately reflect the relationship between environmental factors and circuit failures in cultural relics.

[0105] The collected environmental perception data and fixed reference object information undergo preprocessing, including data cleaning, noise removal, and missing value imputation. Outliers and erroneous records are checked and corrected to ensure the quality of the data input to the model. The preprocessed data is then converted into a format that the model can recognize and process. For example, environmental parameters such as temperature, humidity, and light intensity are converted into numerical vectors, and the location and attribute information of fixed reference objects are converted into coordinate and attribute feature vectors. The converted environmental information data is then input into the multi-reference object system analysis model. The order and format of the data input are ensured to match the model's input layer structure, enabling the model to correctly receive and process the data.

[0106] Key environmental features are extracted from the input environmental perception data, such as the rate of change of temperature and humidity, the range of light intensity fluctuations, and abrupt changes in air quality parameters. Simultaneously, spatial location features are extracted from fixed reference object information, such as the coordinates of the reference objects and their distance and orientation from the cultural relics' electrical equipment.

[0107] The extracted environmental and spatial features are fused to construct a comprehensive feature vector. Various feature fusion methods can be employed, such as concatenation, weighted summation, and principal component analysis (PCA), to combine different types of features into a single feature vector that comprehensively reflects the relationship between the environment and space.

[0108] Based on the model's requirements and characteristics, the fused feature vectors are further optimized and adjusted. For example, the most important features can be selectively retained, while redundant or less relevant features can be removed to improve the model's computational efficiency and performance. Simultaneously, the feature vectors are normalized or standardized to make them suitable for input into subsequent processing layers of the model for further analysis and computation.

[0109] Furthermore, the model's architecture is as follows:

[0110] Input layer design:

[0111] Environmental sensing data input nodes: Based on the scale and type of the sensor network, multiple input nodes are designed to receive environmental sensing data such as temperature and humidity, light intensity, and air quality. Each input node corresponds to one type of environmental data, ensuring that the data can be accurately input into the model. For example, one input node can be designed to receive data transmitted from the temperature and humidity sensor network, another input node can receive data from the light intensity sensor, and so on.

[0112] Fixed Reference Object Information Input Nodes: Design dedicated input nodes to receive attribute information such as the position, size, shape, and material of fixed reference objects. These nodes convert the geometric and physical properties of the reference objects into numerical forms that the model can process. For example, design input nodes for display cases to receive their position coordinates, length, width, height dimensions, material parameters, etc.

[0113] Hidden layer design:

[0114] Feature extraction sublayer: A feature extraction sublayer is designed within the hidden layers. Neurons or computational units perform preliminary processing on the input environmental and reference data to extract key features. Convolutional Neural Networks (CNNs) can be used to automatically extract spatial features, or Recurrent Neural Networks (RNNs) can be used to process time-series data.

[0115] Feature Fusion Sublayer: A feature fusion sublayer is designed to fuse extracted environmental features and spatial location features of reference objects. Fully connected layers (dense layers) or attention mechanisms can be used to integrate features from different sources, forming a comprehensive feature vector.

[0116] Nonlinear Transformation Sublayer: Introducing nonlinear activation functions (such as ReLU, Sigmoid, etc.) to perform nonlinear transformations on the fused features enhances the model's expressive and learning capabilities. Simultaneously, normalization layers (such as BatchNormalization) can be added to accelerate model convergence and improve stability.

[0117] In one embodiment, it can also be:

[0118] Feature extraction sublayer:

[0119] Environmental Feature Extraction: A convolutional neural network (CNN) structure is used to extract spatial features from the input environmental perception data. For example, a convolutional kernel is used to slide across the spatial distribution map of temperature and humidity data to extract local spatial features.

[0120] Reference Object Feature Extraction: A Multilayer Perceptron (MLP) is used to process the attribute information of a fixed reference object, extracting its key features. Simultaneously, a combination of MLP and self-attention mechanisms is employed. The MLP processes the attribute information of the fixed reference object by constructing input, hidden, and output layers, utilizing activation functions and training optimization to extract key features. Meanwhile, a self-attention mechanism is introduced into the Transformer architecture to encode environmental and reference object features, calculating query, key, and value vectors. Attention weights are obtained through scaled dot product attention and the Softmax function, capturing global dependencies of features.

[0121] Feature Fusion Sublayer: This sublayer fuses environmental and reference features to form a comprehensive feature vector. Fully connected layers are used to concatenate and fuse the extracted features, and an attention mechanism is employed to weight the fused features, highlighting important features and suppressing irrelevant ones. For example, temperature and humidity features are fused with the material characteristics of the display case to create a key feature combination that influences circuit malfunctions in cultural relics.

[0122] Nonlinear Transformation Sublayer: Nonlinear activation functions (such as ReLU, Sigmoid, etc.) are introduced to perform nonlinear transformations on the fused features, enhancing the model's expressive and learning capabilities. Simultaneously, a normalization layer (such as BatchNormalization) is added to normalize the features, accelerating model convergence and improving stability.

[0123] Output layer design:

[0124] Environmental Impact Assessment Nodes: These nodes are designed to assess the potential impact of environmental factors on the failure node. Each output node can correspond to a specific failure type or risk level, outputting the degree of influence of the environmental factor on that failure type.

[0125] Risk prediction node: Design output nodes to predict the level of failure risk. The softmax function can be used to convert the output into a probability distribution, representing the likelihood of different risk levels, thus providing an intuitive basis for decision-making.

[0126] Through the above detailed steps, the model can more effectively process input data, extract and fuse key features, and accurately output the potential impact of environmental factors on fault nodes and risk prediction results.

[0127] S132. Use spatial relationship analysis to determine the positional relationship between the fault node and the reference object, combine physical model and data-driven method to evaluate the impact of environmental factors on the fault node, and dynamically update the cultural relic circuit fault feature library.

[0128] In this embodiment, the cultural relic circuit fault feature database refers to a database or data structure used to store and manage the fault features of cultural relic circuits. It comprehensively records different fault types and their characteristic manifestations, providing a knowledge base for fault diagnosis.

[0129] Specifically, using the museum's three-dimensional spatial model and geometric algorithms, the relative positional relationship between the fault node and various fixed reference objects (such as display cases, building columns, etc.) is calculated. This includes calculating geometric parameters such as the distance, orientation, and spatial angle between the fault node and the reference objects.

[0130] On the one hand, physical models are applied to simulate the propagation and attenuation patterns of environmental factors (such as temperature, humidity, and light) within the museum space. On the other hand, data-driven methods are used to analyze the correlation between environmental factors and historical data on failure points. Machine learning algorithms are employed to uncover potential patterns between environmental factors and failure occurrence.

[0131] Based on the above assessment results, the feature vectors in the cultural relic circuit fault feature database are updated. The updated feature database is then fed back into the multi-reference system analysis model to optimize the model parameters and improve the model's prediction accuracy regarding the impact of environmental factors.

[0132] In one embodiment, step S132 may include steps S1321 to S1323.

[0133] S1321. Calculate the relative positional relationship between the fault node and each fixed reference object using the museum's three-dimensional spatial model and geometric algorithms.

[0134] In this embodiment, the location coordinates of the fault node and each reference object are determined based on the three-dimensional model of the museum building.

[0135] Geometric algorithms are used to calculate the relative positional relationships, such as distance and orientation, between the faulty node and reference objects. For example, the straight-line distance and azimuth angle from the faulty node to the nearest display case can be calculated.

[0136] S1322. Apply physical models to simulate the propagation and attenuation of environmental factors in space, and combine data-driven methods to analyze the correlation between environmental factors and historical data of fault nodes to obtain evaluation results.

[0137] In this embodiment, the evaluation result refers to a quantitative assessment of the degree of influence of environmental factors on the fault node, obtained through physical model simulation and data-driven analysis, including information such as the intensity, scope, and trend of the influence.

[0138] A physical propagation model of environmental factors is constructed, simulating how temperature and humidity propagate with distance and airflow, or how light intensity decreases with spatial location. Statistical analysis or machine learning methods are used to uncover correlation patterns between environmental factors and historical data of fault nodes. For example, regression analysis can be used to determine the quantitative relationship between humidity and fault occurrence. The simulation results of the physical model are combined with the analysis results of data-driven methods to form a comprehensive assessment of the impact of environmental factors.

[0139] S1323. Update the feature vectors in the cultural relic circuit fault feature library based on the evaluation results, and feed the updated feature library back to the multi-reference system analysis model to optimize the model parameters.

[0140] Specifically, in the process of monitoring museum artifacts, determining the positional relationship between fault nodes and reference objects using spatial relationship analysis, assessing the impact of environmental factors on fault nodes using physical models and data-driven methods, and dynamically updating the artifact circuit fault feature database is a complex and crucial task. The following is a detailed explanation:

[0141] The coordinate information of fault nodes and various fixed reference objects (such as display cases, building columns, walls, etc.) is obtained from the museum's 3D spatial model. This coordinate data is preprocessed, including noise removal and missing value filling, to ensure the accuracy and completeness of the data.

[0142] Geometric algorithms are used to calculate the spatial relationship between the faulty node and each reference point. This includes calculating Euclidean distances to determine straight-line distances and determining relative orientations (such as east, south, west, and north) through vector analysis. For example, the straight-line distance from the faulty node to the nearest display case is calculated to be 1.2 meters, located due east of the display case.

[0143] A spatial relationship matrix is ​​constructed to record the distance and orientation information between the fault node and all reference objects. This matrix will be used for subsequent environmental impact analysis to help the model understand the location and layout of the fault node within the museum space.

[0144] The appropriate physical model is selected based on the type of environmental factor. For example, for temperature and humidity changes, a heat transfer model is used to simulate the propagation and diffusion of temperature and humidity within the museum space. The actual temperature and humidity changes at the fault point are calculated, taking into account factors such as airflow and heat conduction from building materials. For the impact of lighting, an optical model is used to analyze the attenuation and distribution of light intensity in space, and to assess the thermal and photoelectric effects of lighting on the circuitry and equipment of the artifacts near the fault point.

[0145] Historical fault data and corresponding environmental information are collected, and machine learning algorithms (such as random forests, support vector machines, or deep learning models) are used to establish statistical relationships between environmental factors and fault occurrence. The model is then trained to identify the correlation between environmental change patterns and fault patterns. For example, analysis of historical data reveals that the failure rate of a certain type of artifact circuitry significantly increases when relative humidity exceeds 60% and temperature is above 25°C.

[0146] The calculation results of the physical model are integrated with the analysis results of the data-driven model. Through weight allocation or ensemble learning methods, the deterministic predictions of the physical model and the statistical regularities of the data-driven model are comprehensively considered to obtain a comprehensive assessment of the impact of environmental factors. For example, it is determined that under the current environmental conditions, the probability of the cultural relic's circuit equipment at the fault node being affected by temperature and humidity is 70%, and the probability of being affected by light is 30%.

[0147] The evaluation results are matched with feature vectors in a fault feature database for cultural relics circuits. Based on the pre-stored feature patterns in the database, the most likely fault type under the current environmental conditions is identified. For example, the matching reveals that the current environmental conditions are highly similar to the fault feature "insulation performance degradation due to high humidity" in the feature database.

[0148] The feature vectors in the fault feature library are updated based on the evaluation results. If new environmental factors are found to be associated with the fault or the strength of existing associations changes, the corresponding parameters in the feature vectors are adjusted. For example, if a new fault mode is detected under a specific combination of light and temperature and humidity, this combination is added as a new feature vector; if the influence weight of humidity on a certain fault is found to increase, the weight value is increased accordingly.

[0149] The updated fault feature library is fed back into the multi-reference system analysis model to optimize its parameters and structure. Simultaneously, the model is retrained using the new feature library to improve its ability to predict future faults. For example, the neural network model is fine-tuned using the new feature library, updating its connection weights and bias terms to more accurately predict the impact of environmental factors on faulty nodes.

[0150] Through the above steps, the positional relationship between the fault node and the reference object is accurately determined, the impact of environmental factors is comprehensively assessed, and the fault feature database is dynamically updated, thereby improving the accuracy and timeliness of fault diagnosis of museum artifact circuits.

[0151] Furthermore, the fault feature database of cultural relic circuits played a crucial role in the construction and analysis of the multi-reference system analysis model. Specifically, through various means such as historical fault records, expert experience, and laboratory simulations, characteristic data of cultural relic circuits under different fault types (such as short circuits, open circuits, component aging, etc.) were collected. These characteristic data include changes in voltage, current, and power of circuit nodes, as well as sensitivity changes related to environmental factors (such as temperature, humidity, and light).

[0152] The collected data is analyzed to extract key features that characterize the type and severity of faults. For example, the amplitude of sudden voltage changes at nodes, the frequency of current fluctuations, and the abnormal rate of power increase during a fault are extracted. Simultaneously, the influence of environmental factors on these features is considered, such as increased leakage current due to decreased insulation material performance caused by increased humidity.

[0153] The extracted features are organized into a structured database, namely the Cultural Relics Circuit Fault Feature Database. Each fault type corresponds to a set of feature vectors, which contain various abnormal behaviors of the circuit when the fault occurs. Each entry in the feature database is also associated with the environmental conditions in which the fault occurred (such as temperature and humidity range, light intensity, etc.) so as to correct and match fault features under different environments.

[0154] In multi-reference system analysis models, when extracting features from environmental perception data and fixed reference information, a fault feature database of cultural relic circuits is referenced. The model uses information from the feature database to identify fault characteristics that may occur under the current environmental conditions and compares these characteristics with real-time collected data. For example, if the feature database records leakage fault characteristics caused by decreased insulation performance in cultural relic circuits under high humidity environments, then under the current high humidity conditions, the model will focus on leakage-related features (such as minor changes in grounding current).

[0155] When analyzing the potential impact of environmental factors on fault nodes in the artifact's circuitry, the model invokes a fault feature library. Based on the correlation between fault features recorded in the library and environmental factors, the model can more accurately assess the types and severity of faults that current environmental changes (such as rising temperature or increased light intensity) may trigger. For example, the feature library might indicate that parameters of certain electronic components drift when temperatures rise, leading to abnormal circuit function. The model then analyzes the impact of current temperature changes on corresponding components in the artifact's circuitry and predicts potential fault nodes.

[0156] The multi-reference system analysis model dynamically updates the fault feature library of cultural relic circuits based on real-time monitoring data and environmental changes. When the model detects new fault features or new correlations between environmental factors and fault features, it feeds this information back into the feature library. For example, if a new fault mode is discovered during actual monitoring, or if the impact of a certain environmental factor on the fault is found to be greater or less than previously estimated, the model will update the feature vectors and correlations in the feature library accordingly. This allows the feature library to be continuously enriched and improved, enhancing the model's predictive ability and accuracy for future faults.

[0157] In assessing the risk level of a fault and providing decision support, the model relies on comprehensive information from a database of fault characteristics in cultural relics circuits. This database provides a mapping between fault characteristics and fault types and severity, enabling the model to accurately determine the fault risk level based on currently monitored characteristics and environmental conditions. Museum staff can quickly understand the nature of the fault and its potential impact based on the risk assessment results provided by the model and the fault characteristic descriptions in the database, thus taking targeted measures such as adjusting environmental control equipment and arranging for maintenance personnel to inspect and repair it.

[0158] S133. Continuously receive real-time data and optimize its own parameters to comprehensively assess the fault risk level in order to obtain environmental analysis results.

[0159] In this embodiment, a sensor network and data acquisition system deployed within the museum continuously collect environmental perception data (such as real-time temperature and humidity, light intensity, air quality, etc.) and equipment operation data (such as voltage and current of artifact circuits, etc.) to ensure the timeliness and accuracy of the data.

[0160] By utilizing machine learning algorithms, the parameters of a multi-reference system analysis model are automatically adjusted based on real-time data, improving the model's sensitivity to environmental changes and its prediction accuracy. For example, if real-time data shows that humidity has a significant impact on a particular fault node, the weight of humidity-related parameters can be appropriately increased.

[0161] By combining optimized model parameters and real-time data, a risk assessment algorithm is used to comprehensively consider various environmental factors and equipment status, quantitatively assess the failure risk of each fault node, determine its risk level (such as low risk, medium risk, high risk), and output environmental analysis results, providing a scientific basis for preventive maintenance and risk control.

[0162] S140. Based on the fault diagnosis results and the environmental analysis results, locate the power devices of cultural relic-related equipment that need to be optimized.

[0163] In one embodiment, step S140 described above may include steps S141 to S143.

[0164] S141. Based on the fault diagnosis results, identify the fault nodes and their severity in the circuit, and based on the environmental analysis results, determine the environmental risk factors and their levels that affect the fault nodes.

[0165] S142. Integrate the fault diagnosis results with the environmental analysis results, identify the fault nodes affected by environmental factors through correlation analysis, and comprehensively evaluate and determine the optimization priority to obtain the priority evaluation results.

[0166] In this embodiment, the priority evaluation result refers to

[0167] Integrating fault diagnosis results and environmental analysis results into a unified data structure facilitates subsequent processing. For example, a data table can be created that includes fault node information, fault type, fault severity, related environmental risk factors, and environmental risk levels. Correlation analysis can then be used to identify fault nodes affected by specific environmental risk factors. For instance, it might be discovered that a particular fault node is more prone to failure in high humidity environments.

[0168] By combining the severity of the failure with the level of environmental risk, a weighted scoring method is used to calculate the overall impact score, which is then used to assess the overall impact of each failure node.

[0169] The faulty nodes that need to be prioritized for optimization are determined based on their overall impact scores. For example, the top five nodes with the highest overall impact scores are selected as priority targets.

[0170] S143. Based on the priority evaluation results, identify the power devices of the equipment associated with the fault node, consider their load and operating status, and determine the key optimization devices to obtain the power devices of the cultural relics associated with the equipment that need to be optimized.

[0171] Specifically, a graph convolutional neural network model is used to diagnose faults in the circuits of cultural relics, identifying faulty nodes in the circuits. These faulty nodes may be due to abnormal voltage, abnormal current, or aging components.

[0172] Based on the output of the fault diagnosis model, the severity of each fault node is assessed. For example, if the voltage fluctuation of a node exceeds 20% of the normal range, it indicates that the node has a relatively serious fault.

[0173] By using a multi-reference system analysis model, environmental risk factors affecting the failure node can be identified. These factors include, for example, high humidity, strong sunlight, or temperature fluctuations.

[0174] Based on the output of the environmental analysis model, assess the level of each environmental risk factor. For example, humidity exceeding 60% and continuing to rise indicates a high humidity risk.

[0175] Integrate fault diagnosis results and environmental analysis results into a unified data structure. For example, create a data table that includes fault nodes, fault types, fault severity, related environmental risk factors, and environmental risk levels.

[0176] Correlation analysis can identify which faulty nodes are affected by specific environmental risk factors. For example, it may be found that the failure frequency of a certain faulty node increases significantly in a high humidity environment.

[0177] The overall impact of each failure node is comprehensively assessed by combining the severity of the failure and the level of environmental risk. For example, a weighted scoring method can be used, assigning different weights to the severity of the failure and the level of environmental risk to calculate a comprehensive impact score.

[0178] Based on the overall impact score, the faulty nodes that need to be prioritized for optimization are identified. For example, the top five nodes with the highest overall impact scores will be given priority consideration.

[0179] Based on the topology and functional design of the artifact's circuitry, identify the power devices associated with the faulty nodes. For example, a faulty node might be related to the power devices in the display case's lighting or temperature control system.

[0180] Assess the load and operating status of these power devices. For example, check whether the current, voltage, and temperature of the power devices are within their normal operating range.

[0181] Based on the comprehensive impact assessment and device load conditions, identify the power devices that require priority optimization. For example, prioritize optimizing power devices that have the greatest impact on the faulty node and have a high load.

[0182] Develop specific optimization plans, including replacing aging components, adjusting power distribution, or upgrading equipment. For example, plans may include replacing power components whose insulation performance deteriorates in high humidity environments and adding protective measures.

[0183] By combining the fault diagnosis results and environmental analysis results, we can effectively identify the power devices of the cultural relics-related equipment that need to be optimized, thereby improving the stability and reliability of the circuits of museum cultural relics.

[0184] More specifically, the core objective of this step is to identify the power devices in the equipment associated with cultural relics that require priority optimization. This is done based on the previously obtained fault diagnosis and environmental analysis results. By comprehensively utilizing these two sets of results, it is possible to accurately determine which power devices, due to fault conditions and environmental factors, require priority optimization.

[0185] First, identify the faulty nodes in the circuit based on the fault diagnosis results. This is similar to finding the location of a "lesion" in a complex circuit network. For example, in the circuit system of an artifact display device, by detecting abnormal changes in parameters such as current and voltage, it may be found that the insulation performance of a certain capacitor node has deteriorated, resulting in leakage. This capacitor node is the faulty node.

[0186] At the same time, it is also necessary to determine the severity of these fault nodes. This can be judged based on the degree to which the fault node affects the function of the entire circuit system. For example, if a fault node causes the entire cultural relic lighting system to fail immediately, then its severity is very high; but if it is just a minor node with a slight parameter deviation, and its impact on the overall function of the system is small, then its severity is relatively low.

[0187] Based on the environmental analysis results, the environmental risk factors and their levels affecting the failure node are then determined. Environmental factors include many aspects such as temperature, humidity, dust, and electromagnetic interference.

[0188] For example, in a high-humidity exhibition environment, humidity may be a major environmental risk factor affecting power devices in artifact-related equipment (such as power transistors on electronic control boards). Moreover, if the humidity remains high and unstable for a prolonged period, the environmental risk factor of humidity will be rated as high.

[0189] This process integrates fault diagnosis results and environmental analysis results. It's like putting two separate puzzle pieces together to form a more complete and accurate picture of the equipment's condition.

[0190] For example, fault diagnosis revealed that the diodes in the power conversion module of a certain artifact storage device showed signs of damage, while environmental analysis indicated that the storage environment suffered from severe electromagnetic interference. Integrating these two findings clarified that the electromagnetic interference was likely the environmental factor affecting the diodes in the power conversion module, leading to the malfunction.

[0191] Correlation analysis is used to identify failure nodes affected by environmental factors. This requires analyzing the degree of correlation between the failure nodes and environmental risk factors. For example, if the damage to a certain failure node is highly synchronous with the occurrence of an environmental risk factor, then it can be determined that the failure node is affected by that environmental factor.

[0192] Then, the optimization priority is determined by comprehensively considering the severity of the fault node and the level of environmental risk factors. For example, if the equipment corresponding to a fault node is critical to its function and also faces a high level of environmental risk, then the power devices associated with this fault node will be given a higher priority during optimization.

[0193] Based on the prior priority assessment results, identify the power devices associated with the faulty node. This step narrows down the specific range of power devices.

[0194] For example, based on the preceding analysis, several specific power transistors in a power regulation module were identified as being associated with high-priority fault nodes. These power transistors may have experienced performance degradation due to prolonged exposure to high-temperature environments (environmental risk factors), thereby affecting the normal operation of the entire module (fault nodes).

[0195] After identifying the relevant power devices, their load and operating status should also be considered to determine the key devices for optimization.

[0196] For example, regarding the power transistors mentioned above, if they are under high load conditions for a long period of time (i.e., carrying large currents or voltages) and their current operating state has shown signs of instability (such as intermittent short circuits), then these power transistors will be identified as key optimization devices. This is because they pose the greatest potential threat to the normal operation of cultural relic-related equipment under high load and unstable operating conditions, and therefore require priority optimization.

[0197] S150. Using simulation technology combined with equipment operating conditions and cultural relic protection needs, the power devices of the cultural relic-related equipment are finely segmented, simulated and optimized to generate targeted power control strategies.

[0198] In this embodiment, the power control strategy refers to a series of precise power control measures formulated through refined segmentation, simulation, and optimization of the power devices in the cultural relic-related equipment. These measures aim to rationally regulate the power supply, distribution, and use of the power devices based on the actual operating conditions of the equipment and the specific needs of cultural relic protection, thereby ensuring the safety of the cultural relic and improving the operating efficiency of the equipment. This strategy encompasses everything from the circuit design and parameter input of the power devices to the final power control scheme generated through simulation technology, which is then applied to the actual equipment while undergoing real-time monitoring and feedback adjustments.

[0199] Specifically, this step mainly utilizes simulation technology, combined with equipment operating conditions and cultural relic protection needs, to perform refined segmentation, simulation, and optimization adjustments of the power devices in the cultural relic-related equipment, ultimately generating targeted power control strategies. The aim is to ensure the safety of the cultural relic while simultaneously achieving efficient equipment operation through precise modeling and simulation of the power devices in the cultural relic-related equipment.

[0200] In one embodiment, step S150 described above may include steps S151 to S157.

[0201] S151. Input the circuit design and parameter information of the power device of the cultural relic-related equipment. According to the coordinate position and preset rules, divide the power device into multiple basic units. The basic units include driving, protection and output units.

[0202] In this embodiment, the circuit design and parameter information of the power devices associated with the cultural relics are input. This information forms the basis for subsequent segmentation and simulation. The circuit design includes the overall architecture of the power devices, while the parameter information includes specific details such as device dimensions and material properties.

[0203] Based on coordinate positions and preset rules, the power device is divided into multiple basic units. This division is similar to breaking down a complex device into several relatively simple parts for processing. The basic units include drive, protection, and output units. For example, the drive unit provides drive signals to the device, the protection unit prevents overload or damage from other abnormal conditions, and the output unit processes and transmits the device's output signals. This division method allows for more precise simulation and optimization tailored to the characteristics of each basic unit.

[0204] S152. Call the simulation code library to model each basic unit, set the operating parameters, generate simulation modules, and combine them into a preliminary structure according to their actual positions.

[0205] In this embodiment, a simulation code library is invoked to model each basic unit. The simulation code library contains various code modules for modeling, each corresponding to different physical phenomena and device characteristics. By invoking these code modules, each basic unit can be constructed in a virtual environment.

[0206] Operating parameters are set, simulating various conditions the equipment would experience during actual operation, such as temperature, voltage, and current. The generated simulation modules are then assembled into a preliminary structure according to their actual positions. This step is akin to building a virtual equipment model, combining the basic units according to their positions within the actual equipment to provide a complete preliminary model for subsequent simulations.

[0207] S153. Perform thermal and electromagnetic simulation on the back side of the preliminary structure to generate a back side module, and integrate it with the preliminary structure to form a complete device structure.

[0208] In this embodiment, thermal and electromagnetic simulations are performed on the back side of the preliminary structure to generate a back side module. The back side structure may be affected by thermal and electromagnetic factors in actual equipment; simulation can predict the potential impact of these factors on equipment performance. For example, thermal simulation of the back side can analyze heat dissipation during equipment operation, while electromagnetic simulation can analyze electromagnetic interference.

[0209] Integrating the rear module with the preliminary structure forms a complete device structure. This allows the virtual model to more closely resemble the actual device, including the various physical characteristics and interactions of the front and rear sides.

[0210] S154. Optimize the mesh density of each simulation module according to the preset adjustment rules, calculate the adjustment coefficients, and construct the simulation model.

[0211] In this embodiment, the mesh density of each simulation module is optimized according to preset adjustment rules. Mesh density refers to the density of the elements used to divide the simulation model into smaller units. Optimizing mesh density aims to improve simulation efficiency while maintaining simulation accuracy. For critical parts or modules that significantly impact simulation results, the mesh density can be appropriately increased, while for relatively less important parts, the mesh density can be appropriately decreased.

[0212] The adjustment coefficients are calculated, and these coefficients are determined comprehensively based on factors such as the characteristics of the equipment and the requirements of the simulation. By adjusting these coefficients, each simulation module is adjusted, thereby constructing a more reasonable and accurate simulation model.

[0213] Specifically, mesh density is a key factor in simulation models. Higher mesh density reflects physical phenomena more precisely but increases computational load and resource consumption; while lower mesh density improves simulation efficiency but may affect the accuracy of the results. Therefore, it is necessary to optimize the mesh density of each simulation module according to preset adjustment rules. For critical components such as key drive, protection, and output units in the power devices of artifact-related equipment, appropriately increasing the mesh density can more accurately capture their operating status and changes. For some minor components that have a relatively small impact on the simulation results, the mesh density can be appropriately reduced to minimize unnecessary computational resource consumption.

[0214] The adjustment coefficients are determined after comprehensively considering factors such as equipment characteristics and simulation requirements. Equipment characteristics include the type, size, and materials of power devices, while simulation requirements involve the required accuracy and computation time. Adjustment coefficients are calculated based on equipment characteristics and simulation requirements using specific calculation methods and formulas. Then, these adjustment coefficients are used to adjust the mesh density of each simulation module, reasonably increasing the mesh density of critical parts and appropriately decreasing the mesh density of less important parts. This results in a more reasonable and accurate simulation model, improving simulation efficiency while maintaining accuracy.

[0215] S155. Configure the contact electrodes according to the rules, run the simulation, and obtain the power loss and efficiency electrical performance parameters to obtain the simulation results.

[0216] In this embodiment, the simulation results refer to the set of electrical performance parameters such as power loss and efficiency obtained by running the simulation model, which can reflect the performance of power devices of cultural relic-related equipment during the simulated operation.

[0217] The contact electrodes are configured according to the rules. These electrodes are crucial for the connection between the equipment and external circuits, and their configuration affects the equipment's operating characteristics and simulation results. Then, the simulation is run to obtain electrical performance parameters such as power loss and efficiency. These parameters reflect the equipment's performance during simulated operation, and can be used to preliminarily determine whether the equipment meets the design requirements.

[0218] Specifically, contact electrodes are a crucial part of the connection between the device and external circuits, and their configuration directly affects the device's operating characteristics and simulation results. The position, shape, and number of contact electrodes should be configured appropriately according to preset rules. For example, in power devices, different types of contact electrodes (such as ohmic contact electrodes and Schottky contact electrodes) are suitable for different boundary conditions and operating conditions. Correct configuration ensures accurate simulation of current input, output, and signal transmission processes during the simulation process.

[0219] After configuring the contact electrodes, run the simulation program. Simulation yields electrical performance parameters such as power loss and efficiency. Power loss reflects the degree to which electrical energy consumed by the equipment is converted into heat or other forms of energy during operation, and is a crucial indicator for evaluating equipment energy efficiency. Efficiency, on the other hand, indicates the effective utilization of input electrical energy by the equipment, reflecting its performance. These parameters accurately reflect the equipment's performance during simulated operation. By comparing these parameters with design requirements or expected targets, a preliminary assessment can be made as to whether the equipment meets the design requirements.

[0220] S156. Compare the simulation results with the actual data to verify the simulation model, and optimize the simulation model based on the results.

[0221] In this embodiment, the simulation model is verified by comparing the simulation results with actual data. Actual data can come from past experience data, experimental data, etc. If the simulation results match the actual data well, the simulation model is reliable; if there are significant differences, the simulation model needs to be optimized based on the results. This step is to ensure that the simulation model can realistically reflect the actual operating conditions of the equipment, providing an accurate basis for subsequent power control strategies.

[0222] Specifically, to verify the reliability of the simulation model, the simulation results need to be compared with actual data. Actual data can be obtained from multiple sources, such as previous experimental data, operating data from similar equipment, and industry standard data. Through comparative analysis, the consistency between the simulation results and actual data in key electrical performance parameters such as power loss and efficiency is examined. If the two match well, it indicates that the simulation model can accurately reflect the actual operating conditions of the equipment and is reliable.

[0223] If the simulation results differ significantly from the actual data, the simulation model needs to be optimized based on the comparison results. This may involve adjusting equipment parameters in the model, correcting boundary conditions, and improving the simulation algorithm, among other things. Through continuous comparison and optimization, the simulation model gradually approaches the actual situation of the equipment, thereby providing an accurate basis for subsequent power control strategies.

[0224] S157. Based on the verified simulation model, formulate targeted power control strategies and apply the power control strategies to actual equipment to ensure the safety of cultural relics. At the same time, monitor the equipment operation status in real time and refine the simulation model based on feedback.

[0225] In this embodiment, a targeted power control strategy is developed based on a validated simulation model and applied to actual equipment. During practical application, it is also necessary to monitor the equipment's operating status in real time and refine the strategy based on feedback. In this way, the safety of cultural relics can be dynamically ensured while the equipment operates in optimal condition, achieving a balance between cultural relic protection and equipment operating efficiency.

[0226] Specifically, based on the verification and reliability of the simulation model, targeted power control strategies are formulated according to the equipment operating conditions and the needs of cultural relic protection. These strategies aim to ensure the safety of cultural relics while improving the operating efficiency and stability of the equipment by rationally controlling the power supply, distribution, and use of power devices. For example, the optimal operating voltage and current range of power devices under different operating conditions, as well as protective measures in case of abnormal situations, are determined based on simulation results.

[0227] The established power control strategy is applied to the actual equipment associated with cultural relics. During actual operation, the equipment may be affected by various external factors, such as environmental changes and load fluctuations, necessitating real-time monitoring of its operating status. By monitoring the equipment's electrical parameters, operating temperature, and other information, timely operational feedback is obtained. Based on this feedback, the power control strategy is refined and adjusted to adapt to the actual operating conditions of the equipment, achieving a dynamic balance between ensuring the safety of cultural relics and optimizing equipment operation, thus achieving a balance between cultural relic protection and equipment operating efficiency.

[0228] Here are some examples illustrating power regulation strategies:

[0229] The output voltage of the power devices is dynamically adjusted according to the actual load changes of the equipment associated with the cultural relics, so as to ensure the stable operation of the equipment under different load conditions and at the same time reduce energy consumption.

[0230] The load current of the equipment is monitored in real time by a current sensor; based on changes in the load current, a PID control algorithm is used to dynamically adjust the output voltage. For example:

[0231] When the load current increases, the output voltage should be increased appropriately to ensure the normal operation of the equipment.

[0232] When the load current decreases, the output voltage is reduced to reduce energy consumption.

[0233] Application scenarios: Suitable for scenarios where the load changes significantly in cultural relic protection equipment, such as cultural relic display lighting systems (when the brightness of the lights needs to be adjusted according to the ambient light).

[0234] Improve the power factor of equipment associated with cultural relics, reduce harmonic pollution, and improve power quality.

[0235] Power Factor Correction (PFC) Circuit: Adding a PFC circuit to the device input terminal, and adjusting the input current to a sine wave in phase with the input voltage through a Boost circuit.

[0236] Harmonic detection and suppression: The Fast Fourier Transform (FFT) algorithm is used to detect power grid harmonics in real time, and the harmonic current is dynamically compensated by an active power filter (APF).

[0237] Application scenarios: Suitable for scenarios in cultural relic protection equipment where high power quality is required, such as precision temperature and humidity control systems for cultural relic storage environments.

[0238] When multiple cultural relic-related devices are running simultaneously, overall energy consumption is optimized through collaborative control.

[0239] Based on the actual needs of each device, the load should be allocated reasonably to avoid overloading some devices while lightly loading others.

[0240] The central control system manages the operating status of all equipment and dynamically adjusts the power output of each device based on the overall load. For example:

[0241] When multiple devices are running simultaneously, reduce the power output of non-critical devices appropriately.

[0242] When some devices stop operating, the load is distributed to other devices to ensure the normal operation of critical equipment.

[0243] Application scenarios: Suitable for scenarios where multiple devices (such as multiple artifact display cases, environmental monitoring equipment, etc.) are operating simultaneously in cultural relic protection sites.

[0244] Adjust the operating parameters of the power devices according to changes in ambient temperature to ensure stable operation of the equipment under different environmental conditions.

[0245] Temperature sensors are installed in critical parts of the equipment (such as heat sinks for power devices) to monitor the ambient temperature in real time.

[0246] Based on feedback from the temperature sensor, the operating voltage and current of the power devices are dynamically adjusted. For example:

[0247] When the ambient temperature rises, reduce the operating current of the power devices appropriately to prevent overheating.

[0248] When the ambient temperature drops, appropriately increase the operating current of the power devices to ensure normal equipment operation.

[0249] Application scenarios: Suitable for temperature-sensitive scenarios in cultural relic protection equipment, such as constant temperature and humidity control systems for cultural relic storage environments.

[0250] Take timely measures in the early stages of equipment overload or failure to protect the equipment and cultural relics and reduce the risk of damage.

[0251] Overload detection: Real-time monitoring of the equipment's current and voltage; when an overload is detected, the protection mechanism is immediately triggered.

[0252] Fault early warning: By analyzing equipment operating data (such as temperature, current harmonics, etc.), potential faults can be predicted in advance. For example:

[0253] An overload warning is issued when the current exceeds 80% of the rated value.

[0254] When the temperature exceeds the set threshold, the output power of the power device is automatically reduced.

[0255] When a fault is detected, the power supply is automatically cut off and an alarm is issued.

[0256] Application scenarios: Suitable for scenarios in cultural relic protection equipment where the safety and reliability requirements of the equipment are high, such as cultural relic restoration equipment and cultural relic display lighting systems.

[0257] These power regulation strategies, by combining simulation technology to optimize equipment operation, not only ensure the safety of cultural relics but also improve the operating efficiency and reliability of the equipment, making them suitable for various cultural relic protection scenarios.

[0258] S160. Output the power control strategy to facilitate environmental adjustments and equipment maintenance based on the power control strategy.

[0259] In this embodiment, a detailed and clear power control strategy is output, which can provide clear guidance for subsequent actual operations, enabling operators to accurately adjust the environment and perform equipment maintenance in accordance with the strategy, thereby ensuring the effective implementation of the strategy.

[0260] Outputting power control strategies in an appropriate manner can transform complex technical information obtained from simulation analysis into specific instructions that are easy to understand and operate, and transmit them to relevant maintenance personnel, equipment managers, and environmental control personnel, ensuring that all parties accurately grasp the requirements for equipment operation and cultural relic protection.

[0261] By implementing power control strategies, timely adjustments to the environment and maintenance of equipment can effectively prevent the risk of damage to cultural relics caused by power problems, while ensuring that the equipment is always in good operating condition and reducing the impact of equipment failures on cultural relic protection.

[0262] The power control strategy should be presented in the form of a written report, which should include the basis for the strategy, specific operational steps, expected results, and precautions. For example, it should detail the circumstances under which the equipment's output voltage needs to be adjusted, how to adjust it, and the adjusted voltage range.

[0263] Visual aids such as charts and graphs are used to display key information about power control strategies, including voltage and current variation curves of equipment under different load conditions, and corresponding control measures. This allows operators to more intuitively understand the core elements of the strategy, facilitating quick and accurate execution.

[0264] Key parameters in the power control strategy are output as data files for direct reading and application by the equipment's control system. Examples include target values ​​for output power factor correction and harmonic suppression thresholds, enabling the automatic control system to adjust the equipment's operating status in real time based on these parameters.

[0265] According to the power control strategy, the temperature and humidity of the environment where the cultural relic protection equipment is located should be adjusted. For example, if the strategy indicates that the heat generated during the operation of the equipment will raise the ambient temperature and thus affect the protection of cultural relics, then measures such as increasing the cooling capacity of the air conditioner or adjusting the air supply volume can be taken to lower the ambient temperature and maintain a suitable environment for the protection of cultural relics.

[0266] For artifact display equipment, the light intensity and duration are adjusted according to the power control strategy. If the strategy indicates that the current light power is too high and may cause light damage to the artifacts, then the brightness of the lights is reduced or the daily lighting time is shortened accordingly to meet the lighting requirements for artifact preservation.

[0267] Based on equipment operating status assessment information in power control strategies, preventative maintenance plans are developed. For example, if the strategy indicates that certain power devices will age faster after prolonged high-load operation, then these power devices can be inspected, replaced, or repaired in advance to prevent sudden equipment failures from damaging cultural relics.

[0268] When equipment experiences a power failure, maintenance personnel can quickly locate the cause of the fault by referring to the output power control strategy and carry out repairs according to the guidance in the strategy. For example, the strategy clearly defines the fault points and repair methods corresponding to different fault phenomena, enabling maintenance personnel to quickly take effective repair measures, reduce equipment downtime, and minimize the adverse impact on cultural relic protection work.

[0269] The aforementioned museum artifact environmental monitoring method based on fusion technology uses graph convolutional neural network technology to monitor the circuit system of museum artifacts in real time, accurately extracting circuit node features to construct a directed graph of logic circuits, quickly identifying fault nodes and obtaining fault diagnosis results. Simultaneously, it acquires data from fixed reference objects and environmental perception within the museum to obtain comprehensive environmental information, which is then input into a multi-reference system analysis model to deeply analyze the potential impact of environmental factors on fault nodes, yielding environmental analysis results. Combining the fault diagnosis and environmental analysis results, it accurately locates the power devices of artifact-related equipment requiring key optimization. Next, using simulation technology, based on equipment operating conditions and artifact protection needs, it performs refined segmentation, simulation, and optimization adjustments of the power devices, generating targeted power control strategies. Finally, it outputs the power control strategies, providing a clear basis for environmental adjustments and equipment maintenance, achieving synergistic linkage between artifact status monitoring, environmental risk assessment, and equipment power control, comprehensively ensuring the safety of museum artifacts.

[0270] Figure 3 This is a schematic block diagram of a museum artifact environmental monitoring system 300 based on fusion technology provided in an embodiment of the present invention. Figure 3 As shown, corresponding to the above-described museum artifact environmental monitoring method based on fusion technology, the present invention also provides a museum artifact environmental monitoring system 300 based on fusion technology. This museum artifact environmental monitoring system 300 based on fusion technology includes a unit for executing the above-described museum artifact environmental monitoring method based on fusion technology, and the system can be configured in a server. Specifically, please refer to... Figure 3 The museum artifact environmental monitoring system 300 based on fusion technology includes a fault diagnosis unit 301, an environmental information acquisition unit 302, an environmental analysis unit 303, a positioning unit 304, a simulation unit 305, and an output unit 306.

[0271] The fault diagnosis unit 301 is used to monitor the circuit system of the museum artifacts in real time using graph convolutional neural network technology to extract the features of the artifact circuit nodes, construct a directed graph of the logic circuit, and identify the fault nodes to obtain fault diagnosis results. The environmental information acquisition unit 302 is used to acquire fixed reference objects and environmental perception data within the museum to obtain environmental information. The environmental analysis unit 303 is used to input the environmental information into a multi-reference object system analysis model to analyze the potential impact of environmental factors on the fault nodes to obtain environmental analysis results. The positioning unit 304 is used to locate the power devices of the artifact-related equipment that need to be optimized based on the fault diagnosis results and the environmental analysis results. The simulation unit 305 is used to use simulation technology combined with the equipment operating conditions and artifact protection needs to perform fine segmentation, simulation, and optimization adjustment of the power devices of the artifact-related equipment to generate targeted power control strategies. The output unit 306 is used to output the power control strategies to facilitate environmental adjustments and equipment maintenance based on the power control strategies.

[0272] In one embodiment, the fault diagnosis unit 301 is used for:

[0273] The netlist file of the combinational logic artifact circuit to be diagnosed is obtained, a directed graph of the logic circuit is constructed, and feature vectors containing electrical attributes, logical functions, and connection relationships of each node are extracted and integrated to generate a node feature matrix. A random number of faults are injected into random node positions of the combinational logic artifact circuit to be diagnosed, generating random test vectors. Failure response data of the circuit after fault injection is obtained using a circuit simulation model. The failure response data is used to construct a fault element graph set using the fault element graph method, and the fault score of each node is calculated and associated with the corresponding fault to generate fault training data containing fault features and association information. The fault training data is input into a pre-processed graph convolutional neural network model, and model parameters are set and weights and bias parameters are initialized. The model parameters are continuously optimized through forward and backward propagation until the maximum number of iterations is reached, constructing an accurate fault diagnosis model. Based on the constructed fault diagnosis model, the real-time failure response data of the combinational logic artifact circuit to be diagnosed is analyzed to determine the fault nodes in the circuit and obtain the fault diagnosis results.

[0274] In one embodiment, the fault diagnosis unit 301 is used to preprocess the failure response data to remove noise and interference, and extract key features that reflect the fault state through a feature extraction algorithm to form a fault feature set; draw a basic circuit diagram containing nodes, connection relationships, and logical functions according to the circuit design document; map the feature values ​​in the fault feature set to the corresponding nodes of the basic circuit diagram to generate a feature map that integrates fault information; analyze the feature map to identify nodes with significant features to obtain potential fault points; integrate all potential fault points and their interrelationships into a fault element diagram to form a complete fault element diagram; wherein, based on the circuit topology and fault propagation characteristics, association rules between all potential fault points are established, an association matrix is ​​created to quantify the relationship between all potential fault points, and a graphical tool is used to draw the fault element diagram, with potential fault points as nodes, lines are drawn between relevant nodes according to the association matrix, fault information is added to present the details of each point, and the accuracy and practicality of the diagram are verified and optimized by comparison with actual faults.

[0275] In one embodiment, the environmental analysis unit 303 is used for:

[0276] A multi-reference system analysis model is constructed, model parameters are initialized, and the environmental information is input into the multi-reference system analysis model. Feature vectors are constructed by integrating environmental features and spatial location information of reference objects. Spatial relationship analysis is used to determine the positional relationship between the fault node and the reference objects. The impact of environmental factors on the fault node is evaluated by combining physical models and data-driven methods. The fault feature database of cultural relics circuits is dynamically updated. Real-time data is continuously received and its own parameters are optimized. The fault risk level is comprehensively evaluated to obtain environmental analysis results.

[0277] In one embodiment, the environmental analysis unit 303 is further configured to:

[0278] A three-dimensional spatial framework is constructed based on the museum's architectural structure. The location and attribute information of fixed reference objects are imported, and the input, hidden, and output layers of the model are designed to complete the construction of a multi-reference object system analysis model. The hidden layer includes a feature extraction sublayer that uses neurons to extract key features, a feature fusion sublayer that integrates environmental and reference object features to form a comprehensive feature vector, and a nonlinear transformation sublayer that enhances the model's expressive power through activation functions. Historical data and expert experience are used to initialize the model parameters, and the input data is normalized. The model parameters are then verified and adjusted using known data to complete the initialization. The environmental information is preprocessed, and the data format is converted before being input into the multi-reference object system analysis model. Key environmental features from the environmental information and spatial location features from the fixed reference object information are extracted, and these features are fused to construct a comprehensive feature vector.

[0279] In one embodiment, the environmental analysis unit 303 is further configured to:

[0280] The relative positional relationship between the fault node and each fixed reference object is calculated using the museum's three-dimensional spatial model and geometric algorithms. A physical model is applied to simulate the propagation and attenuation of environmental factors in space, and a data-driven approach is used to analyze the historical data correlation between environmental factors and fault nodes to obtain evaluation results. Based on the evaluation results, the feature vectors in the cultural relic circuit fault feature library are updated, and the updated feature library is fed back to the multi-reference system analysis model to optimize the model parameters.

[0281] In one embodiment, the positioning unit 304 is further configured to:

[0282] Based on the fault diagnosis results, fault nodes and their severity in the circuit are identified. Based on the environmental analysis results, environmental risk factors and their levels affecting the fault nodes are determined. The fault diagnosis results and environmental analysis results are integrated, and fault nodes affected by environmental factors are identified through correlation analysis. The optimization priorities are determined through comprehensive evaluation to obtain priority evaluation results. Based on the priority evaluation results, power devices associated with the fault nodes are identified. Considering their load and operating status, key optimization devices are determined to obtain the power devices of cultural relics-related equipment that need to be optimized.

[0283] In one embodiment, the simulation unit 305 is used for:

[0284] The circuit design and parameter information of the power devices of the cultural relic-related equipment are input. According to the coordinate position and preset rules, the power devices are divided into multiple basic units, including driving, protection and output units 306. The simulation code library is called to model each basic unit, set the operating condition parameters, generate simulation modules and combine them into a preliminary structure according to the actual position. Thermal and electromagnetic simulations are performed on the back of the preliminary structure to generate a back module, which is then integrated with the preliminary structure to form a complete equipment structure. According to preset adjustment rules, the mesh density of each simulation module is optimized, the adjustment coefficient is calculated, and a simulation model is constructed. Contact electrodes are configured according to the rules, the simulation is run, and power loss, efficiency and electrical performance parameters are obtained to obtain simulation results. The simulation results are compared with actual data to verify the simulation model, and the simulation model is optimized based on the results. Based on the verified simulation model, a targeted power control strategy is formulated and applied to the actual equipment to ensure the safety of the cultural relic. At the same time, the operating status of the equipment is monitored in real time, and the simulation model is refined based on the feedback.

[0285] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the museum artifact environmental monitoring system 300 based on fusion technology and its various units can be found in the corresponding descriptions in the aforementioned method embodiments. For the sake of convenience and brevity, these details will not be repeated here.

[0286] The aforementioned museum artifact environmental monitoring system 300 based on fusion technology can be implemented as a computer program, which can be used in, for example... Figure 4 It runs on the computer device shown.

[0287] Please see Figure 4 , Figure 4 This is a schematic block diagram of a computer device provided in an embodiment of this application. The computer device 500 can be a server, wherein the server can be a standalone server or a server cluster composed of multiple servers.

[0288] See Figure 4 The computer device 500 includes a processor 502, a memory, and a network interface 505 connected via a system bus 501. The memory may include a non-volatile storage medium 503 and internal memory 504.

[0289] The non-volatile storage medium 503 may store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions that, when executed, cause the processor 502 to perform a museum artifact environmental monitoring method based on fusion technology.

[0290] The processor 502 provides computing and control capabilities to support the operation of the entire computer device 500.

[0291] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute a museum artifact environmental monitoring method based on fusion technology.

[0292] This network interface 505 is used for network communication with other devices. Those skilled in the art will understand that... Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device 500 to which the present application is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0293] The processor 502 is used to run a computer program 5032 stored in the memory to implement all steps of the museum artifact environmental monitoring method based on fusion technology.

[0294] It should be understood that in the embodiments of this application, the processor 502 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0295] It will be understood by those skilled in the art that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program includes program instructions and can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiments of the above methods.

[0296] Therefore, the present invention also provides a storage medium. This storage medium can be a computer-readable storage medium. The storage medium stores a computer program, wherein when executed by a processor, the computer program causes the processor to perform all steps of the museum artifact environmental monitoring method based on fusion technology.

[0297] The storage medium can be any computer-readable storage medium capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), magnetic disk, or optical disk.

[0298] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0299] In the embodiments provided by this invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of each unit is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0300] The steps in the method of this invention can be adjusted, merged, or reduced in order according to actual needs. The units in the system of this invention can be merged, divided, or reduced according to actual needs. Furthermore, the functional units in the various embodiments of this 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.

[0301] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0302] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for monitoring the environment of museum artifacts based on fusion technology, characterized in that, include: Graph convolutional neural network technology is used to monitor the circuit system involved in museum artifacts in real time, extract the features of the circuit nodes of the artifacts to construct a directed graph of logic circuits, and identify fault nodes to obtain fault diagnosis results. Obtain fixed reference points and environmental perception data within the museum to obtain environmental information; The environmental information is input into a multi-reference system analysis model to analyze the potential impact of environmental factors on the fault node, so as to obtain environmental analysis results. Based on the fault diagnosis results and the environmental analysis results, the power devices of cultural relic-related equipment that need to be optimized should be identified. By utilizing simulation technology in conjunction with equipment operating conditions and cultural relic protection needs, the power devices of the cultural relic-related equipment are finely segmented, simulated, and optimized to generate targeted power control strategies. The power control strategy is output to facilitate environmental adjustments and equipment maintenance based on the power control strategy.

2. The museum artifact environment monitoring method based on fusion technology according to claim 1, characterized in that, The method employs graph convolutional neural network technology to monitor the circuit systems involved in museum artifacts in real time, extracting the features of circuit nodes to construct a directed graph of logic circuits, and identifying faulty nodes to obtain fault diagnosis results, including: Obtain the netlist file of the combinational logic artifact circuit to be diagnosed, construct a directed graph of the logic circuit, extract feature vectors of each node containing electrical attributes, logical functions and connection relationships, and integrate them to generate a node feature matrix; A random number of faults are injected into random node positions of the combinational logic artifact circuit to be diagnosed, a random test vector is generated, and the circuit failure response data after fault injection is obtained using a circuit simulation model. The failure response data is used to construct a set of failure element graphs using the failure element graph method. The failure score of each node is calculated and associated with the corresponding failure to generate failure training data containing failure features and association information. The fault training data is input into the preprocessed graph convolutional neural network model, the model parameters are set and the weights and bias parameters are initialized. The model parameters are continuously optimized through forward propagation and backward propagation until the maximum number of iterations is reached, and an accurate fault diagnosis model is constructed. Based on the established fault diagnosis model, the real-time failure response data of the combinational logic circuit artifact to be diagnosed is analyzed to identify the fault nodes in the circuit and obtain the fault diagnosis results.

3. The museum artifact environmental monitoring method based on fusion technology according to claim 2, characterized in that, The process of constructing the fault element graph set is as follows: The failure response data is preprocessed to remove noise and interference, and key features that reflect the fault state are extracted using a feature extraction algorithm to form a fault feature set. Based on the circuit design documents, draw a basic circuit diagram that includes nodes, connections, and logic functions; The feature values ​​in the fault feature set are mapped to the corresponding nodes of the basic circuit diagram to generate a feature map that integrates fault information. The feature map is analyzed to identify nodes with significant features in order to obtain potential fault points; All potential fault points and their interrelationships are integrated into a fault element diagram to form a complete fault element diagram. Specifically, association rules between all potential fault points are established based on the circuit topology and fault propagation characteristics. An association matrix is ​​created to quantify the relationships between all potential fault points. A graphical tool is used to draw the fault element diagram, with potential fault points as nodes. Lines are drawn between relevant nodes according to the association matrix. Fault information is added to present the details of each point. The accuracy and practicality of the diagram are verified and optimized by comparing it with actual faults.

4. The museum artifact environmental monitoring method based on fusion technology according to claim 1, characterized in that, The fixed reference objects include display cases and building structures; the environmental sensing data includes real-time collected temperature, humidity, and light information.

5. The museum artifact environmental monitoring method based on fusion technology according to claim 4, characterized in that, The step of inputting the environmental information into a multi-reference system analysis model to analyze the potential impact of environmental factors on the fault node and obtain environmental analysis results includes: A multi-reference system analysis model is constructed, the model parameters are initialized, the environmental information is input into the multi-reference system analysis model, and the environmental features and spatial location information of the reference objects are fused to construct a feature vector; Spatial relationship analysis is used to determine the positional relationship between the fault node and the reference object. The influence of environmental factors on the fault node is evaluated by combining physical models and data-driven methods. The fault feature library of cultural relics circuits is dynamically updated. It continuously receives real-time data and optimizes its own parameters to comprehensively assess the level of fault risk in order to obtain environmental analysis results.

6. The museum artifact environmental monitoring method based on fusion technology according to claim 5, characterized in that, The construction of the multi-reference system analysis model, initialization of model parameters, input of environmental information into the multi-reference system analysis model, and fusion of environmental features and spatial location information of reference objects to construct feature vectors include: Using the museum's architectural structure as a base, a three-dimensional spatial framework is constructed. The position and attribute information of fixed reference objects are imported, and the input, hidden, and output layers of the model are designed to complete the construction of a multi-reference object system analysis model. Among them, the hidden layer includes a feature extraction sublayer that uses neurons to extract key features, a feature fusion sublayer that integrates environmental and reference object features to form a comprehensive feature vector, and a nonlinear transformation sublayer that enhances the model's expressive ability through activation functions. The model parameters are initialized using historical data and expert experience, the input data is normalized, and the model parameters are verified and adjusted using known data to complete the initialization of the model parameters. The environmental information is preprocessed, the data format is converted, and then it is input into the multi-reference system analysis model; Key environmental features are extracted from the environmental information and spatial location features from the fixed reference information. These features are then fused to construct a comprehensive feature vector.

7. The museum artifact environmental monitoring method based on fusion technology according to claim 6, characterized in that, The process of determining the positional relationship between the fault node and the reference object using spatial relationship analysis, evaluating the impact of environmental factors on the fault node using physical models and data-driven methods, and dynamically updating the cultural relic circuit fault feature database includes: The relative positional relationship between the fault node and each fixed reference object is calculated using the museum's three-dimensional spatial model and geometric algorithms; Physical models are used to simulate the propagation and attenuation of environmental factors in space, and data-driven methods are combined to analyze the correlation between environmental factors and historical data of fault nodes in order to obtain evaluation results. The feature vectors in the cultural relic circuit fault feature library are updated based on the evaluation results, and the updated feature library is fed back to the multi-reference system analysis model to optimize the model parameters.

8. The museum artifact environmental monitoring method based on fusion technology according to claim 1, characterized in that, The step of locating the power devices of cultural relic-related equipment that require key optimization based on the fault diagnosis results and the environmental analysis results includes: Based on the fault diagnosis results, the fault nodes and their severity in the circuit are identified, and based on the environmental analysis results, the environmental risk factors and their levels affecting the fault nodes are determined. The fault diagnosis results are integrated with the environmental analysis results. Correlation analysis is used to identify fault nodes affected by environmental factors. The optimization priorities are determined through comprehensive evaluation to obtain priority evaluation results. Based on the priority assessment results, the power devices associated with the fault nodes are identified. Considering their load and operating status, the key optimization devices are determined to obtain the power devices of the cultural relics associated with them that need to be optimized.

9. The museum artifact environmental monitoring method based on fusion technology according to claim 8, characterized in that, The method utilizes simulation technology combined with equipment operating conditions and cultural relic protection needs to perform refined segmentation, simulation, and optimization adjustment of the power devices of the cultural relic-related equipment, generating targeted power control strategies, including: Input the circuit design and parameter information of the power device of the cultural relic-related equipment, and divide the power device into multiple basic units according to the coordinate position and preset rules. The basic unit includes driving, protection and output units. The simulation code library is called to model each basic unit, set the operating parameters, generate simulation modules, and combine them into a preliminary structure according to their actual positions. Thermal and electromagnetic simulations were performed on the back side of the preliminary structure to generate a back side module, which was then integrated with the preliminary structure to form a complete device structure. Based on the preset adjustment rules, the mesh density of each simulation module is optimized, the adjustment coefficients are calculated, and the simulation model is constructed. Configure the contact electrodes according to the rules, run the simulation, and obtain the power loss and efficiency electrical performance parameters to obtain the simulation results; Compare the simulation results with actual data to verify the simulation model, and optimize the simulation model based on the results. Based on the validated simulation model, targeted power control strategies are developed and applied to actual equipment to ensure the safety of cultural relics. At the same time, the operating status of the equipment is monitored in real time, and the simulation model is refined and adjusted based on the feedback.

10. A museum artifact environmental monitoring system based on fusion technology, characterized in that, include: The fault diagnosis unit is used to monitor the circuit system involved in the museum artifacts in real time using graph convolutional neural network technology, in order to extract the features of the artifact circuit nodes to construct a directed graph of logic circuits and identify fault nodes to obtain fault diagnosis results. The environmental information acquisition unit is used to acquire fixed reference objects and environmental perception data within the museum to obtain environmental information. The environmental analysis unit is used to input the environmental information into the multi-reference system analysis model to analyze the potential impact of environmental factors on the fault node, so as to obtain the environmental analysis results. The positioning unit is used to locate the power devices of cultural relic-related equipment that need to be optimized based on the fault diagnosis results and the environmental analysis results. The simulation unit is used to perform fine segmentation, simulation and optimization adjustment of the power devices of the cultural relic-related equipment by combining simulation technology with the equipment operating conditions and cultural relic protection needs, and to generate targeted power control strategies. The output unit is used to output the power control strategy so as to facilitate environmental adjustments and equipment maintenance based on the power control strategy.

Images