Information management method, system, electronic device and storage medium

Abstract

This invention relates to the fields of data processing and industrial internet technology, and discloses information management methods, systems, electronic devices, and storage media. The methods include receiving heterogeneous data from multiple sources such as carbon emissions and energy consumption, mapping them into holographic data packets containing a unified spatiotemporal reference; dynamically constructing a spatiotemporal causal correlation spectrum based on the correlation weights between data nodes; calculating the topological potential energy value of a physical region, and generating hardware control commands when the potential energy value exceeds a warning threshold to adaptively adjust the sampling frequency of the front-end device; monitoring the correlation entropy of data nodes, and when a phase transition condition is met, serializing and migrating a stable subgraph to a second storage layer and generating a hash index; retrieving data based on the index and performing integrity verification. This invention achieves dynamic adjustment of the sampling frequency through topological potential energy feedback and uses correlation entropy to drive storage evolution, optimizing hardware resource configuration while ensuring the spatiotemporal correlation and reliable traceability of data.

Description

Technical Field

[0001] This invention relates to the fields of data processing and industrial internet technology, and in particular to information management methods, systems, electronic devices and storage media. Background Technology

[0002] In the process of information management of modern industrial parks or large facilities, heterogeneous data from various sources, such as carbon emission management, energy consumption monitoring, plant data, and operation and maintenance records, are typically involved. However, existing technologies often lack a unified spatiotemporal benchmark when processing this data, resulting in data from different business systems being isolated from each other. This makes it impossible to establish effective quantitative correlations at the temporal, spatial, and semantic levels, and it is difficult to form a topological structure that reflects the inherent logic between the data, thus limiting the ability to comprehensively analyze complex scenarios.

[0003] In terms of data acquisition, front-end devices typically operate at a preset fixed sampling frequency. This rigid acquisition mode cannot perceive the urgency or frequency of events occurring within a physical area. When events are frequent in the area, a low sampling rate can lead to the loss of critical details; while during periods of stability, continuous high-frequency sampling can generate a large amount of redundant data, resulting in wasted hardware power consumption and ineffective use of transmission bandwidth.

[0004] Furthermore, in data storage and lifecycle management, existing cold and hot data tiered storage strategies are mostly based on simple time periods, lacking dynamic assessment of the stability or disorder of the data structure itself. This leads to high-performance storage resources being occupied by relatively stable data for extended periods. Simultaneously, during data archiving and migration, there is often a lack of endogenous verification mechanisms based on content characteristics, making it difficult to verify the integrity of historical data after long-term storage and failing to meet high standards of reliable traceability and tamper-proof requirements. Summary of the Invention

[0005] The purpose of this invention is to provide information management methods, systems, electronic devices, and storage media, which solve the problems of existing information management technologies in processing multi-source heterogeneous data, such as the lack of a unified spatiotemporal benchmark, the inability to adaptively adjust the data acquisition frequency according to the urgency of events, and the lack of a dynamic evolution mechanism for cold and hot data storage strategies, resulting in weak data correlation, low hardware resource utilization, and difficulty in tracing key information.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The first aspect of this invention provides an information management method, comprising the following steps:

[0008] S1. Initialization and Spatiotemporal Mapping of Multi-Source Heterogeneous Data: Receive multi-source heterogeneous data from the front-end acquisition device. This multi-source heterogeneous data includes at least one of carbon emission management data, smart energy consumption data, smart plant data, and smart operation and maintenance data. Map the received data into a standardized holographic data packet containing a unified spatiotemporal reference.

[0009] S2. Constructing a spatiotemporal causal correlation spectrum: Instantiate the holographic data packet into data nodes, calculate the correlation weight between data nodes, and construct and dynamically update a spatiotemporal causal correlation spectrum with a topological structure based on the correlation weight.

[0010] S3. Adaptive frequency-varying acquisition feedback based on topological potential energy: Calculate the topological potential energy value of the physical region in the spatiotemporal causal correlation spectrum map, and use the topological potential energy value as the basis for characterizing the density of data correlation within the region; when the topological potential energy value exceeds the warning threshold, generate hardware control instructions to adjust the data sampling frequency of the front-end acquisition device.

[0011] S4. Storage state evolution based on association entropy: Monitor the association entropy value of data nodes. When the association entropy value meets the phase transition condition, it is determined that the data structure of the data node tends to be stable. Then, the corresponding stable subgraph is serialized into a binary holographic data block, cross-layer migration from the first storage layer to the second storage layer is performed, and a hash index value is generated.

[0012] S5. Trusted Traceability Interaction: Retrieves binary holographic data blocks from the second storage layer based on hash index values, performs integrity verification, and completes information traceability management.

[0013] Preferably, in step S1, the holographic data packet is configured as a four-dimensional vector structure, which includes a content carrier field, a spatial coordinate vector field, a timestamp field, and a device attribute vector field. In step S2, the specific operations for constructing and dynamically updating a spatiotemporal causal correlation spectrum with a topological structure based on the correlation weights between data nodes include: calculating the temporal dimension difference, spatial dimension distance, and content semantic similarity between any two data nodes; calculating the correlation weight based on the weighted sum of the negative exponential function term of the temporal dimension difference, the negative exponential function term of the spatial dimension distance, and the content semantic similarity term; setting a connection threshold, and establishing a connection edge between two data nodes only when the correlation weight exceeds this threshold.

[0014] Preferably, in step S1, the generation method of the spatial coordinate vector field in the holographic data packet includes: for fixed-type front-end acquisition devices, retrieving the corresponding spatial coordinates in a pre-stored static device location table based on the hardware identifier; for mobile-type front-end acquisition devices, obtaining real-time location information by reading the dynamic positioning interface and converting it into spatial coordinates.

[0015] Preferably, in step S3, the topological potential value is calculated by weighted summation based on the sum of the connectivity degrees of all data nodes within the physical region and the sum of the weights of all connected edges. The hardware control instruction includes a target sampling frequency parameter, which is a value for configuring the data sampling frequency. The target sampling frequency parameter is calculated based on the product of the base sampling frequency and the logarithmic gain term of the topological potential value relative to the warning threshold, and the calculated target sampling frequency parameter is limited to the maximum sampling frequency range supported by the front-end acquisition device.

[0016] Preferably, in step S3, the hardware control instructions also include a modal activation instruction; when the topological potential energy value exceeds the warning threshold, the modal activation instruction drives the auxiliary sensor in the front-end acquisition device to switch from a dormant state to an active state. Furthermore, this step also includes a hysteresis recovery mechanism, i.e., setting a recovery threshold lower than the warning threshold, and generating a reset instruction only when the topological potential energy value remains below the recovery threshold within a stable time window to restore the basic sampling frequency and cut off the power supply to the auxiliary sensor.

[0017] Preferably, in step S4, the association entropy value is used to quantify the degree of disorder in the causal connection structure of the data nodes in the current spatiotemporal context. Step S4 also performs a tombstone replacement operation, that is, after migrating the stable subgraph to the second storage layer, a tombstone data node is created at the original location in the first storage layer, and the tombstone data node contains a pointer to the physical storage address in the second storage layer.

[0018] Preferably, in step S4, generating the hash index value specifically involves calculating the combination of the metadata header, attribute vector sequence, spatial coordinate vector sequence, and binary weight sequence of the binary holographic data block using a one-way cryptographic hash function. In step S5, performing the integrity verification specifically involves calculating the real-time hash value of the binary holographic data block retrieved from the second storage layer and comparing this real-time hash value with the hash index value generated in step S4. If they match, the integrity verification is deemed successful.

[0019] A second aspect of the present invention provides an information management system, comprising:

[0020] The multimodal sensing access module is configured to receive multi-source heterogeneous data from the front-end acquisition device and map it into a standardized holographic data packet containing a unified spatiotemporal reference. The multi-source heterogeneous data covers at least one of carbon emission management data, smart energy consumption data, smart factory data, and smart operation and maintenance data.

[0021] The spectrum dynamic construction module is configured to instantiate holographic data packets into data nodes, and construct and dynamically update a spatiotemporal causal correlation spectrum with topological structure based on the correlation weights between data nodes.

[0022] The topology feedback control module is configured to calculate the topological potential energy value of the physical region in the spatiotemporal causal correlation spectrum, and generate hardware control commands to adjust the data sampling frequency of the front-end acquisition device when the topological potential energy value exceeds the warning threshold.

[0023] The storage evolution management module is configured to monitor the correlation entropy value of data nodes, and when the correlation entropy value meets the phase transition condition, it serializes the corresponding stable subgraph into a binary holographic data block, performs cross-layer migration from the first storage layer to the second storage layer, and generates a hash index value.

[0024] The trusted traceability interaction module is configured to retrieve binary holographic data blocks from the second storage layer based on hash index values ​​and perform integrity verification to complete information traceability management.

[0025] A third aspect of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the information management method of any one of the first aspects described above.

[0026] A fourth aspect of the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the information management method of any one of the first aspects described above.

[0027] This invention standardizes heterogeneous data through a unified spatiotemporal mapping and establishes a topological structure between data using a spatiotemporal causal correlation spectrum. A feedback mechanism based on topological potential energy automatically adjusts the hardware sampling frequency according to the density of data correlations within a region, increasing the sampling rate to capture details during periods of high event density and decreasing the sampling rate to conserve resources during periods of low event density. Simultaneously, a storage evolution mechanism based on correlation entropy migrates stable data structures to secondary storage and generates hash indexes, optimizing storage space utilization while ensuring data tamper resistance and traceability through hash verification.

[0028] In summary, the present invention has at least one of the following beneficial technical effects:

[0029] 1. This invention maps multi-source heterogeneous data into holographic data packets with a unified spatiotemporal reference, and constructs a spatiotemporal causal correlation spectrum based on multi-dimensional weights. This enables the standardized fusion and structured correlation of data from different business systems, quantifies the causal relationships of different data nodes at the temporal, spatial, and semantic levels, and solves the problem of traditional multi-source data being isolated from each other and lacking internal logical correlation. It provides a unified topological data foundation for comprehensive data analysis in complex scenarios.

[0030] 2. This invention characterizes the density of data association by calculating the topological potential energy value of a physical region, and uses this as a feedback signal to adaptively adjust the sampling frequency of the front-end acquisition device. In areas with active events and high topological potential energy, the sampling frequency is automatically increased to capture key details, while in areas with stable conditions, the frequency is reduced to reduce redundant data. Thus, while ensuring the integrity of data acquisition, hardware power consumption and data transmission bandwidth usage are effectively reduced.

[0031] 3. This invention determines the evolution state of the data structure by monitoring the association entropy value of data nodes, serializes and migrates the data subgraphs that tend to be stable to the second storage layer and generates a hash index, realizing dynamic hierarchical storage based on data state, releasing high-performance storage resources, and at the same time using the hash index generated by cross-layer migration to perform integrity verification on archived data, ensuring the anti-tampering capability of historical data and the reliable traceability throughout the entire life cycle. Attached Figure Description

[0032] Figure 1 This is a flowchart illustrating the overall process of an adaptive acquisition and storage method based on a spatiotemporal causal correlation spectrum according to an embodiment of the present invention.

[0033] Figure 2 This is a schematic diagram illustrating the principle of dynamic construction of spatiotemporal causal correlation spectrum according to an embodiment of the present invention;

[0034] Figure 3 This is a schematic diagram of an adaptive feedback control logic based on topological potential energy according to an embodiment of the present invention;

[0035] Figure 4 This is a schematic diagram of the functional module architecture of an information management system according to an embodiment of the present invention. Detailed Implementation

[0036] The following is in conjunction with the appendix Figure 1 -Appendix Figure 4 The present invention will be further described in detail below.

[0037] This invention provides an information management method, system, electronic device, and storage medium.

[0038] See attached document Figure 1 This invention provides an information management method applicable to a control platform that includes multiple functional subsystems such as carbon emission management, smart energy consumption, smart factory operations, smart park, power distribution monitoring, and smart operation and maintenance.

[0039] The information management method relies on an information management system, which in its physical architecture includes: multiple front-end acquisition devices distributed in a predetermined physical space, computing devices connected to the front-end acquisition devices through an industrial communication network, and data storage devices connected to the computing devices.

[0040] The front-end acquisition device is equipped with a programmable control interface, supporting modification of the sampling frequency and operating mode via external commands. The computing device runs a spatiotemporal mapping module, a correlation analysis module, a feedback control module, and a storage management module. The data storage device includes random access memory for performing high-frequency graph data operations and large-capacity non-volatile memory for persistent storage.

[0041] This information management method includes the following steps:

[0042] Step S1: Initialization and spatiotemporal mapping of multi-source heterogeneous data.

[0043] The information acquisition module receives source data from different functional subsystems via an industrial bus or API interface. The source data includes structured data (such as sensor readings) and unstructured data (such as monitoring video frames and maintenance work order text). The system has a pre-installed spatial mapping database, which stores a three-dimensional digital model of the physical space and the installation coordinates of each front-end acquisition device.

[0044] Upon receiving the source data, the spatiotemporal mapping module performs data cleaning and standardization operations. For each piece of source data, the system retrieves the corresponding three-dimensional spatial coordinates from the spatial mapping database based on the unique identifier of its source device. Simultaneously, the system uses a time source calibrated with a network time protocol or a precision time protocol to mark the source data with nanosecond-level timestamps. The system encapsulates the cleaned content data, retrieved 3D spatial coordinates, timestamps, and device attribute information into a standardized holographic data packet. The data structure of the holographic data packet includes: a content carrier field, a spatial coordinate vector field, a timestamp field, and a device attribute vector field.

[0045] Step S2: Construct a spatiotemporal causal relationship spectrum.

[0046] The system maintains a dynamic graph data structure in random access memory, namely a spatiotemporal causal correlation spectrum. The spatiotemporal causal correlation spectrum consists of a set of data nodes and a set of edges. Each spatiotemporal data packet generated in step S1 is instantiated as a data node in the spectrum. For newly added data nodes and existing data nodes in the spectrum that are within the current sliding time window, the correlation analysis module calculates the correlation weight between them using a preset correlation strength algorithm.

[0047] The association weight is a numerical value calculated based on a weighted function of temporal differences, spatial distances, and semantic similarity. Specifically, the calculation logic is configured as follows: the association weight is negatively exponentially correlated with the absolute value of the time difference between the two data nodes, negatively exponentially correlated with the Euclidean distance between the two data nodes, and positively correlated with the semantic or causal similarity of the content of the two data nodes. The time decay factor and spatial decay factor are configurable parameters used to adjust the system's sensitivity to spatiotemporal proximity. Semantic similarity is determined for text data based on cosine similarity using a vector space model, and for numerical data based on a pre-stored causal rule table.

[0048] The system compares the calculated association weights with a preset connection threshold. If the association weight is greater than the connection threshold, a directed or undirected edge connection is established between the two data nodes, and the calculated association weight is assigned to the edge, thereby organizing the discrete data nodes into an association network with a topological structure.

[0049] Step S3: Adaptive frequency-varying acquisition feedback based on topological potential energy.

[0050] The system periodically scans the spatiotemporal causal correlation spectrum to identify data node clusters corresponding to specific physical regions. For each physical region, the feedback control module calculates the current topological potential value. The topological potential value is used to quantify the density of information associations within the region. The calculation of the topological potential value is based on the sum of the connectivity degrees of all data nodes within the region and the sum of the weights of all edges within the region, and is weighted and accumulated according to the importance coefficients of the data nodes.

[0051] After calculation, the system compares the topological potential energy value with a preset warning threshold. When the topological potential energy value exceeds the warning threshold, the system determines that the area is in a period of active data association and then generates hardware control commands. The system queries a preset device address mapping table to find the network addresses (IP addresses or bus addresses) of all front-end acquisition devices in the physical area and sends the hardware control commands to the corresponding devices.

[0052] Hardware control instructions include frequency modulation instructions and mode activation instructions.

[0053] The frequency modulation command includes a target sampling frequency parameter, which is positively correlated with the current topological potential energy value. The front-end acquisition device responds to the frequency modulation command by adjusting the data sampling frequency to the target value by modifying the interrupt period of its internal timer.

[0054] The modal activation command contains code to enable the functional module. In response to the modal activation command, the front-end acquisition device supplies power to the internal auxiliary sensors (such as infrared modules or high-frequency spectrum analysis modules), switching them from sleep mode to working mode.

[0055] Step S4: Dynamically related data cluster generation and index construction.

[0056] The system executes graph partitioning algorithms (such as the Louvain algorithm or label propagation algorithm) on the spatiotemporal causal correlation spectrum. Based on the edge weight distribution, the system divides the set of closely connected data nodes into independent subgraphs, each of which is defined as a correlated data cluster. The correlated data cluster logically aggregates a group of heterogeneous data that are spatiotemporally proximate and causally related in content.

[0057] The index building module extracts feature vectors for each related data cluster and constructs a multi-level index. The multi-level index structure includes an inverted index in the time dimension, a grid index in the spatial dimension, and a vector index in the content dimension. When the query parser receives a query request, it locates the relevant related data cluster by matching the multi-level index and returns all data nodes contained within that cluster.

[0058] Step S5: Storage state evolution based on association entropy.

[0059] The system monitors the state of the generated associated data clusters and calculates the structural entropy value of the associated data clusters. The structural entropy value is used to quantify the stability of the internal topology of the associated data clusters. The calculation of the structural entropy value is based on the probability entropy of the connection distribution of data nodes within the cluster and the time-varying derivative of the average edge weight within the cluster.

[0060] The system determines the data storage state based on the structure entropy value:

[0061] Dynamic graph storage state: When the structure entropy value is higher than the preset stability threshold, it indicates that the relationships within the data cluster are still changing drastically. At this time, the associated data cluster is maintained in a graph database format in random access memory, allowing real-time data node insertion and edge weight update operations to be performed.

[0062] Static serialization storage state: When the structure entropy value drops below a preset stability threshold and remains stable within a preset observation window, the system performs a storage state transition operation. The storage state transition operation includes: serializing all data node data and their topological relationships within the associated data cluster into a contiguous binary data block; writing the binary data block to a large-capacity non-volatile memory for archiving; generating a hash digest of the data block; and sending the hash digest to the business databases of each functional subsystem as an index anchor.

[0063] Through the above steps, the present invention realizes an adaptive information management method that can automatically adjust the front-end acquisition parameters according to the correlation density of information content and automatically switch the storage medium according to the stability of the data structure.

[0064] See attached document Figure 1In this embodiment, step S1 is to convert raw data from different physical interfaces, following different communication protocols and with different formats into standardized holographic data packets that can be processed uniformly by the system.

[0065] Step S1 is performed by the data preprocessing subsystem in the computing device. The data preprocessing subsystem is logically divided into a protocol parsing unit, a spatiotemporal alignment unit, and a feature extraction unit.

[0066] When the raw data stream enters the system, the protocol parsing unit first performs multi-protocol identification and unpacking operations. Since the front-end acquisition devices cover various industrial and network protocols such as Modbus RTU, BACnet, OPCUA, RTSP, and HTTP / MQTT, the protocol parsing unit is equipped with a protocol signature library. The protocol parsing unit identifies the communication protocol type of the data source by matching the signature codes in the data frame header, and strips the headers of each communication protocol layer to extract the payload data. For raw voltage or current signals transmitted by analog signal sensors, the system converts them into digital quantities through an analog-to-digital converter interface.

[0067] After acquiring the payload data, the spatiotemporal alignment unit performs three-dimensional spatial coordinate mapping. The system pre-builds a spatial mapping database in the storage device. This database, based on Building Information Modeling (BIM) or Geographic Information System (GIS), stores a digitized three-dimensional mesh of the physical environment. The spatial mapping database maintains a static device location table, which records the unique hardware identifier and three-dimensional spatial coordinates of each fixed acquisition device. The correspondence.

[0068] For fixed data acquisition devices (such as wall temperature sensors and fixed surveillance cameras), the spatiotemporal alignment unit reads the source device ID of the data and retrieves the corresponding spatial coordinate vector from the static device location table using a hash lookup algorithm. .

[0069] For mobile data acquisition devices (such as inspection robots and maintenance personnel wearing smart bracelets), the system is equipped with a dynamic positioning interface. When the data source is identified as a mobile device ID, the spatiotemporal alignment unit queries the real-time location information of the device at the moment the data was generated (e.g., obtained through UWB ultra-wideband positioning or LiDAR SLAM system) through the dynamic positioning interface, and converts the real-time location into a spatial coordinate vector in a unified coordinate system. .

[0070] Simultaneously, the spatiotemporal alignment unit performs a time base unification operation. Although each acquisition device has its own local clock, considering clock drift between devices and network transmission latency, the system uses a unified arrival timestamp or a generated timestamp corrected by the network time protocol. The system assigns a unified timestamp with nanosecond-level precision to each unpacked data segment. .

[0071] Subsequently, the feature extraction unit standardizes and vectorizes the content of the payload data to generate a content carrier. To address the different types of heterogeneous data, the system employs a traffic splitting strategy:

[0072] For scalar data (such as temperature, voltage, and air pressure), the system performs normalization processing, converting it into floating-point values ​​under preset dimensions, and directly using it as the content carrier. The numerical components.

[0073] For text-based data (such as fault descriptions submitted by maintenance personnel and system alarm logs), the system calls a pre-trained natural language processing model (such as a lightweight version of BERT or Word2Vec) to convert the text string into a fixed-dimensional semantic feature vector, which serves as the content carrier. .

[0074] For image or video frame data, the system calls a convolutional neural network feature extractor to extract key features of the image (such as texture, edge histogram, or object category probability) and generate a high-dimensional feature vector as the content carrier. .

[0075] After completing the above processing, the data encapsulation unit assembles the components into standardized holographic data packets. The mathematical expression for a holographic data packet is defined as follows:

[0076] ;

[0077] in, Represents standardized holographic data packets; The content carrier is represented by a scalar value or a vectorized feature vector, which is used to calculate semantic or causal similarity in subsequent calculations. Representing a spatial coordinate vector, specifically in the form of: ,in , , , respectively, represent the x-coordinate, y-coordinate, and vertical coordinate values ​​in a unified three-dimensional Cartesian coordinate system; It represents a unified timestamp, which is a 64-bit long integer value representing the number of nanoseconds since a specific epoch; The attribute vector contains the source device type code (such as sensor, actuator, camera), data priority marker, and subsystem affiliation code (such as belonging to HVAC system or security system).

[0078] Through step S1, the system eliminates the format differences of the original data, uniformly projecting information from different locations, times, and forms in the physical world into a four-dimensional mathematical space. This provides a standardized input basis for constructing the spatiotemporal causal correlation spectrum and calculating the correlation weights between data nodes in the subsequent step S2. Step S1 not only includes information about the data itself but also solidifies the spatiotemporal environmental characteristics at the time of data generation, ensuring that subsequent feedback analysis based on topological potential energy can accurately correspond to physical entities.

[0079] See attached document Figure 1 and attached Figure 2 Step S2 utilizes the correlation analysis module in the computing device to organize discrete standardized holographic data packets in the random access memory using a dynamic graph structure.

[0080] Output holographic data packets in step S1 Subsequently, the association analysis module instantiates it as a data node in the graph database. To ensure that computing resources are focused on currently active relationships, the system maintains a time-based sliding window. Only when a new data node is generated... timestamp With existing data nodes in the graph database timestamp The system only applies adjustments to that data node when the absolute value of the difference between them is less than a preset correlation time threshold. Start the association weight calculation program.

[0081] For each pair of data nodes that meet the time window condition The system uses a built-in association strength calculation algorithm to calculate the association weight between the two. This weight value quantifies the degree of coupling between two independent data points within the same spatiotemporal and logical context. (Association weight) The calculation formula is as follows:

[0082] ;

[0083] in, Represents data nodes With data nodes The correlation weight between them; The base of the natural logarithm; Represents data nodes Uniform timestamp With data nodes Uniform timestamp The absolute value of the difference between them represents the degree of closeness in the time dimension; This represents the time decay coefficient, a constant greater than zero, used to control the rate at which the correlation strength decays as the time difference increases; Represents data nodes spatial coordinate vector With data nodes spatial coordinate vector The Euclidean distance between them represents the degree of proximity in spatial dimensions; This represents the spatial attenuation coefficient, a constant greater than zero, used to control the rate at which the correlation strength decreases with increasing spatial distance; This represents the content similarity and causal probability function used to calculate data nodes. Content carrier With data nodes Content carrier The degree of logical connection between them; Indicates the weighting coefficient for the time term; Represents the weighting coefficients for the spatial terms; This represents the weighting coefficient for the content items. , , All are non-negative real numbers, and satisfy the following conditions: The system is based on data nodes. and attribute vector and The data type label in the table is used to find and determine the specific values ​​of the three coefficients in the preset weight configuration table.

[0084] Regarding the content similarity and causal probability function in the formula The association analysis module is based on attribute vectors and Execute branch selection logic:

[0085] Vector space similarity calculation branch: when attribute vectors and When the corresponding data content is identified as unstructured data features (such as text vectors or image feature vectors), the system will... Calculate the cosine similarity between two vectors:

[0086] ;

[0087] in, Represents the dot product of vectors. and These represent the magnitudes of the vectors.

[0088] Causal rule mapping branch: when the attribute vector and When the data content corresponding to the identifier is a structured status code or device operating parameters, the system constructs a query key pair. The system uses a query key to search a pre-defined causal rule matrix. This matrix stores fault propagation probability values ​​between different types of device states in key-value pairs. The system then directly assigns the retrieved probability values ​​to the matrix. .

[0089] Calculate the association weight Then, the system compares it with a preset connection threshold. If If the connection threshold is exceeded, the system creates a connection data node in the graph database. and The edge, and It is stored as the weight attribute of that edge.

[0090] In addition, the system is configured with a periodic graph maintenance task. This task scans the existing edges in the graph database at preset intervals and updates the time difference term in the above formula using the current system time. The system then recalculates the weights. When the weight of an edge decays to below a preset removal threshold, the system removes that edge from the graph structure to maintain the sparsity and real-time performance of the spatiotemporal causal correlation spectrum.

[0091] See attached document Figure 1 and attached Figure 3 In step S3 of this embodiment, the feedback control module in the computing device calculates the potential energy of the physical region based on the topological characteristics of the spatiotemporal causal correlation spectrum constructed in step S2, and controls the operating parameters of the front-end acquisition device in reverse.

[0092] While the spectral structure is continuously updated in step S2, the feedback control module performs periodic region scanning tasks. The system has a pre-set spatial region mapping table, which divides the physical space to be managed into several logical regions. .

[0093] For each logical region The feedback control module retrieves the spatial coordinate vector from the data node attributes. Identify all data nodes that fall within the spatial range of the current region, forming a regional data node set. .

[0094] Subsequently, the feedback control module calculates this logical region. topological potential value Topological potential energy quantifies the information aggregation density and event correlation complexity of a physical region within the current time window. A higher value indicates a greater number of data nodes within the region, and stronger spatiotemporal or causal relationships between these nodes. The calculation formula is as follows:

[0095] ;

[0096] in, Represents logical region The topological potential energy value; Indicates that it is located in the logical region A set of regional data nodes encompassing all data nodes within a spatial range; Represents data nodes Importance weighting coefficients, the system from data nodes Corresponding holographic data packets attribute vector Extract the priority field and determine it according to the preset mapping table. Numerical values, such as alarm data The value is greater than that of regular monitoring data. value; Represents data nodes The degree, that is, the relationship between the data nodes in the spatiotemporal causal correlation spectrum. The total number of directly connected edges; Represents data nodes The set of all neighboring data nodes; Represents data nodes With data nodes The correlation weight between them.

[0097] After the calculation is completed, the feedback control module will obtain the result. Compared with the system's preset potential energy warning threshold Compare. If System determination area Under normal monitoring conditions, maintain the default acquisition parameters of the front-end acquisition devices in this area. If System determination area When in a high-density association state, hardware control commands are generated for devices within that region. The system first queries the spatial region mapping table to obtain the logical region. The system includes device identifiers for all front-end data acquisition devices and further queries the device address table to obtain the network communication address (such as IP address, port number, or fieldbus slave address) of each device. Subsequently, the system issues control commands at two levels:

[0098] First level: Nonlinear modulation of sampling frequency

[0099] The system is based on the current topological potential value Calculate the target sampling frequency To balance data accuracy and hardware load, the system employs a logarithmic gain adjustment strategy and sets a hardware frequency upper limit truncation. The calculation formula is as follows:

[0100] ;

[0101] in, This indicates the target sampling frequency sent to the front-end acquisition device, in Hertz; This indicates the maximum supported sampling frequency limited by the hardware specifications of the front-end acquisition device; This represents a function that takes the minimum value, ensuring that the calculated frequency does not exceed the physical limits of the hardware. ; This indicates the preset basic sampling frequency of the front-end data acquisition device; This represents the frequency adjustment gain coefficient, used to set the response sensitivity of the frequency as the potential energy changes; Represents the natural logarithm operation; This indicates the potential energy warning threshold; Represents logical region The topological potential value.

[0102] The feedback control module will calculate Encapsulated as a frequency write command. For sensors supporting the Modbus protocol, the system constructs a message with function code 0x06 (write a single register) or 0x10 (write multiple registers), which will... Writes to the device-specific sampling period configuration register. The instruction is sent to the target device via an industrial communication network, driving it to adjust the sampling period of its internal analog-to-digital converter.

[0103] Second level: Multimodal sleep / wake-up

[0104] System traverses logical regions Check all devices within the system for any auxiliary data acquisition devices that are in low-power sleep mode. The check should be based on the operating mode field in the device status table and the multimodal capability identifier in the device attribute table. If... The feedback control module sends modal activation commands to these auxiliary acquisition devices (e.g., sending a Wake-on-LAN data packet or a specific GPIO high-level trigger signal). The modal activation commands drive high-energy-consuming devices such as infrared thermal imagers and high-frequency vibration analyzers to switch from sleep mode to operating mode in order to acquire supplementary physical information.

[0105] After receiving the frequency modulation command, the front-end acquisition device's internal microcontroller (MCU) performs hardware-level parameter changes. The communication processor then parses the target sampling frequency. Then, the reload value of the hardware timer is recalculated. Reload value The calculation formula is as follows:

[0106] ;

[0107] in, This indicates the value written to the timer auto-reload register, which determines the time interval between two analog-to-digital conversion interrupts; This indicates the floor function; This indicates the system clock frequency of the microcontroller in the front-end data acquisition device; This represents the prescaler coefficient of the timer; This indicates the target sampling frequency sent to the front-end acquisition device, in Hertz; This is a correction factor used to compensate for one clock cycle when the counter returns to zero.

[0108] microcontroller will Writing to the register and forcibly updating the event immediately changes the time resolution of the signal acquisition. Simultaneously, the awakened auxiliary acquisition device, after a warm-up and stabilization period, begins outputting multidimensional data. This newly added high-frequency, multimodal data then enters step S1, where it is mapped to new data nodes, thereby dynamically updating the spectral structure in the next cycle, forming a closed-loop feedback.

[0109] To prevent the topological potential value At the potential energy warning threshold Minor fluctuations in the surrounding environment cause the front-end data acquisition device to frequently switch between high-frequency and low-frequency modes, resulting in control jitter. The system introduces a hysteresis response mechanism. The system sets a recovery threshold. (in The system state determination logic is defined as follows:

[0110] ;

[0111] in, Indicates the target control state of the system at the next moment; Indicates the active acquisition status of high-frequency multimodal data; This indicates the normal acquisition state of the fundamental frequency single-mode; Indicates the recovery threshold; This represents the stable time window, which requires the low potential energy state to last for this duration before the system can issue a reset command.

[0112] When the system execution state is changed Switch to At that time, the feedback control module generates a reset command. The reset command contains the value... The frequency parameters and modal sleep control codes are recorded. After receiving the reset command, the front-end acquisition device restores the timer reload value to the corresponding basic sampling frequency. The system reads the value and cuts off the power supply to the auxiliary sensing module, returning the device to a low-power normal monitoring mode. Through this hysteresis response mechanism, the system achieves its acquisition strategy by minimizing unnecessary energy consumption and storage resource usage while ensuring the capture of critical events.

[0113] See attached document Figure 1 In step S4 of this embodiment, the storage evolution module in the computing device uses the concept of thermodynamic entropy to quantitatively evaluate the structural stability of data nodes and perform cross-media storage migration and lightweight index construction.

[0114] In step S3, a high-frequency, multimodal enhanced data stream is generated through feedback control. After being updated to the spectral structure in step S2, the number of data nodes in the graph database grows exponentially. To resolve the conflict between storage resources and retrieval efficiency, the storage evolution module introduces association entropy as a core indicator to measure the activity and stability of data nodes.

[0115] Correlation Entropy In this embodiment, data nodes are characterized. In the current spatiotemporal context, the degree of disorder and information dispersion of its causal connection structure are considered. The higher the correlation entropy of a data node, the more complex and evenly distributed the strong correlation between that data node and a large number of surrounding data nodes are, indicating that the data node is in the core region of the event's evolution and its state is extremely unstable (i.e., the information is in a high-energy state or liquid state). Conversely, if the correlation entropy is low, it means that the data node is maintained by only a few definite causal chains, or that the correlation with surrounding data nodes has become solidified, indicating that the event represented by the data node has subsided and is suitable for archiving (i.e., the information is in a low-energy state or solid state).

[0116] The storage evolution module iterates through the active data nodes in the graph database according to a preset evolution cycle (e.g., every 10 minutes) and calculates their association entropy. The calculation formula is as follows:

[0117] ;

[0118] To calculate the probability distribution term in the above formula The system first needs to calculate the data nodes. Local total weight The calculation logic is as follows:

[0119] ;

[0120] in, Represents data nodes The correlation entropy value, in nanots; Represents data nodes The set of all neighboring data nodes; Represents a set Any neighboring data node in the denominator is used for summation operations; Represents data nodes With data nodes The correlation weight between them; Represents data nodes data nodes with neighbors The correlation weight between them; Represents data nodes The sum of the weights of all associated edges is used for normalization. Represents data nodes relative data nodes The normalized association probability satisfies and . Represents the natural logarithm operation.

[0121] Based on the calculated correlation entropy, the storage evolution module divides the data lifecycle into two physical states: a primary storage layer (liquid active phase) and a secondary storage layer (solid archive phase). The primary storage layer consists of an in-memory database, supporting high-frequency read and write operations; the secondary storage layer consists of large-capacity object storage or hard disk arrays, which are cost-effective.

[0122] The system has a preset phase transition threshold. When data nodes During generation, it resides in the first-level storage layer by default. The system periodically scans data nodes to determine whether the phase transition triggering conditions are met. The determination logic is as follows:

[0123] ;

[0124] in, Represents data nodes The phase transition trigger state is triggered when the value is True; Represents data nodes The correlation entropy value, in nanots; This indicates the system's preset phase transition threshold; This represents the logical AND operation; Indicates the current system timestamp; Represents data nodes The generated timestamp; This represents the minimum aging time window, used to prevent misjudgment due to the lack of established connections in the early stages of data node generation.

[0125] when When true, the system initiates the solidification and encapsulation process.

[0126] The first step, subgraph extraction: based on data nodes Extract stable subgraph structure centered on the core. .

[0127] The second step is serialization compression: [The text abruptly ends here, likely due to an incomplete sentence or a formatting error Attribute vectors in Spatial coordinates and edge weight Serialized into binary holographic blocks .

[0128] The third step is physical migration: transferring the binary holographic block. Write to the secondary storage layer to obtain the physical storage address .

[0129] Step 4, Tombstone Replacement: The original graph structure is deleted in the first-level storage layer, retaining only one tombstone data node. This tombstone data node stores... With data summaries, when the system attempts to access the data node again, it can automatically trigger a readback loading operation from the secondary storage layer, achieving transparent hierarchical data access.

[0130] In generating binary holographic blocks Subsequently, to support efficient indexing and integrity verification across systems, the storage evolution module immediately executes the digital fingerprint generation procedure. The system calculates the hash index value of the holographic block. The calculation formula is as follows:

[0131] ;

[0132] in, Represents a binary holographic block The hash index value; This represents a one-way cryptographic hash function; The metadata header of the holographic block (including timestamp and version number); Represents the binary XOR operation; The ordered concatenation operator refers to sequentially concatenating the sets according to their data node IDs. The information of all data nodes in the process is concatenated into a binary stream; The information of all data nodes in the process is concatenated into a binary stream; The attribute vector contains the source device type code (such as sensor, actuator, camera), data priority marker, and subsystem affiliation code (such as belonging to HVAC system or security system). This represents the bit-level concatenation operator; Representing a spatial coordinate vector, specifically in the form of: ,in , , , respectively, represent the x-coordinate, y-coordinate, and vertical coordinate values ​​in a unified three-dimensional Cartesian coordinate system; Indicates the association weight The binary representation is converted to a fixed-point number format to avoid inconsistent hash calculations caused by differences in floating-point precision.

[0133] After the calculation is completed, the system constructs a lightweight global index table with the following entry structure: When an external system (such as an audit terminal) requests a query for historical data, the computing device only returns index metadata. The external system then... Download data stream directly from the secondary storage layer And calculate the consistency coefficient locally. :

[0134] ;

[0135] in, The value indicates the verification result: 1 indicates completeness, and 0 indicates tampering. This represents the Kronecker function, which outputs 1 when the two parameters are equal, and 0 otherwise. This represents the downloaded holographic block data stream; This represents a one-way cryptographic hash function.

[0136] Through this mechanism, step S4 realizes the evolution of data from high-cost liquid to low-cost solid, and provides underlying protection based on hash fingerprint for trusted traceability in step S5.

[0137] See attached document Figure 4 The present invention also provides an information management system, which includes a multimodal sensing access module, a spectrum dynamic construction module, a topology feedback control module, a storage evolution management module, and a trusted traceability interaction module. These modules are implemented at the hardware level by one or more processors of a computing device executing computer program instructions stored in memory. At the logical level, they achieve high-throughput data interaction through a message middleware based on a publish-subscribe model, jointly realizing the complete technical process from step S1 to step S5.

[0138] The multimodal sensing access module is equipped with a heterogeneous protocol parsing unit and a spatiotemporal alignment unit. The multimodal sensing access module connects to the front-end physical sensor network via an industrial fieldbus (such as RS485 or CAN) or an Ethernet interface.

[0139] To address compatibility issues with heterogeneous devices, the heterogeneous protocol parsing unit loads a configurable address mapping file (such as an XML or JSON descriptor). This address mapping file defines the correspondence between register addresses of devices from different vendors and standard internal system attributes. Based on this, the parsing unit automatically unpacks the raw binary packets and maps them to observations, eliminating the need to rewrite code for each new device.

[0140] Subsequently, the spatiotemporal alignment unit appends a uniform system timestamp to each independent observation. and spatial coordinate vector And it maps unstructured observation data into standardized data node attribute vectors. The multimodal sensing access module, as the system's input, publishes standardized data packets to the system's internal message bus topic for subsequent modules to subscribe to and consume.

[0141] The dynamic spectrum construction module is the core processing unit of the system, integrating an in-memory graph database and a causal computation unit. This module subscribes to data streams from the access modules and maintains the real-time topology of the spatiotemporal causal relation spectrum.

[0142] The causal computation unit is responsible for executing the core algorithm in step S2, for each pair of data nodes. Calculate association weights The dynamic graph construction module uses copy-on-write technology to dynamically update the adjacency matrix of the graph database. When the weight exceeds a preset threshold, a connection edge is created; when the weight decays below the threshold, the connection is disconnected. This module provides real-time snapshots of the graph topology to the topology feedback control module and the storage evolution management module, including the set of regional data nodes. Neighbor data node set and weight matrix .

[0143] The topology feedback control module is used to realize the closed-loop adaptive adjustment of the system, and it includes a potential energy analysis unit and a command generation unit.

[0144] Potential energy analysis unit periodically scans logic region Based on the topological potential energy calculation logic, the information aggregation density within the region is quantified. When the calculated topological potential energy value... Exceeding the warning threshold At that time, the potential energy analysis unit triggers an alarm and sends a signal to the instruction generation unit.

[0145] The instruction generation unit calculates the target sampling frequency based on the potential energy deviation using a logarithmic gain strategy. It also generates a reverse control message containing device addressing information.

[0146] The topology feedback control module sends commands to the front-end physical devices through the network interface, dynamically adjusts the data acquisition density in step S1, and realizes closed-loop control of sensing, computing, and feedback.

[0147] The storage evolution management module is responsible for managing the lifecycle and hierarchical storage of data, and includes an entropy monitoring unit, a phase change execution unit, and a hash index unit.

[0148] The entropy monitoring unit monitors the structural stability of active data nodes in real time based on the correlation entropy calculation logic. When a data node is detected... correlation entropy value When the phase change threshold is below the threshold and the aging time condition is met, the phase change execution unit is triggered.

[0149] The phase-change execution unit performs the physical migration of data. It extracts a stable subgraph from the first-level storage layer (in-memory database). Serialize it into a binary holographic block The data is then written to the secondary storage layer (cold storage medium). Simultaneously, the phase-change execution unit creates a tombstone data node at the original location in the primary storage layer. This tombstone data node contains a pointer to the physical address of the secondary storage layer, used for transparent data redirection in subsequent queries.

[0150] The hash index unit generates a unique digital fingerprint of the holographic block simultaneously with the holographic block generation, based on hash calculation logic. Furthermore, a lightweight global index table was constructed to address the issue of index bloat caused by massive historical data.

[0151] The Trusted Traceability Interaction Module provides a human-computer interaction interface and API interface, and includes a query parsing unit and an integrity verification unit.

[0152] Upon receiving an external tracing request, the query parsing unit retrieves the corresponding hash index and physical address from the lightweight index table. The system then retrieves the holographic data block stored in the secondary storage layer based on the address. Before data display, the integrity verification unit calculates the real-time hash value of the data stream and compares it with the original fingerprint according to consistency verification logic. Only when the verification passes is the reconstructed spatiotemporal causal correlation spectrum rendered to the visualization terminal, thus providing users with tamper-proof evidence of event evolution.

[0153] This invention also provides an electronic device designed to function as a computing device, hosting and running an information management system. This device is not limited to a general-purpose computer, but can also be an edge server optimized for high-concurrency graph computing, an industrial control workstation, or a cloud computing data node. The electronic device includes a processor, memory, a communication interface, and a system bus.

[0154] The processor (heterogeneous computing unit) is the computing and control center of an electronic device. Given that step S2 involves a large number of data node attribute vectors... The cosine similarity between them is calculated, and the association entropy of the data nodes in the whole graph is calculated in step S4. For logarithmic probability operations, the processor preferably uses a central processing unit that supports the AVX-512 or SVE (Scalable Vector Extension) instruction set. This instruction set architecture allows the processor to perform multiple floating-point multiplication and addition operations in parallel within a single clock cycle, thereby reducing computational cost. and The time complexity.

[0155] Furthermore, in order to address the one-way encrypted hash function in step S5 For computation and integrity verification of holographic blocks, the processor can also integrate or connect external hardware acceleration units (such as FPGAs or ASIC encryption cards). These hardware acceleration units are specifically designed to offload the hash function computation load from the main CPU, ensuring that the system maintains millisecond-level response latency even when backtracking massive amounts of historical data.

[0156] The memory (hierarchical phase-change memory architecture) achieves physical isolation and coordination between the first-level and second-level storage layers in step S4 at the physical hardware level.

[0157] High-speed volatile memory area: Corresponding to the first-level memory layer, it consists of DDR5 SDRAM or HBM (High Bandwidth Memory). This area is used to reside in the memory database, storing the spatiotemporal causal correlation spectrum in a liquid active phase. Its nanosecond-level random access characteristics support the real-time dynamic update of the graph topology in step S2 (data node insertion and edge weight adjustment).

[0158] High-capacity non-volatile storage area: Corresponding to the secondary storage layer, it consists of NVMe SSD arrays or 3DXPoint storage media. This area is used for persistent storage of solid-state holographic data blocks generated by serialization in step S4. This area has high sequential write throughput and can handle historical data streams migrated from the high-speed storage area.

[0159] The memory stores a computer program that, when executed by a processor, causes the processor to perform the method steps described in the foregoing embodiments. Specifically, the memory also maintains a global hash index table, which establishes the hash index values ​​of the holographic data blocks. With physical storage address The mapping relationship between them.

[0160] The communication interface (full-duplex feedback channel) is used to enable interaction between electronic devices and the physical world and external systems. The specific configuration is as follows:

[0161] Industrial fieldbus interface: including RS485, CAN, or EtherCAT interfaces. In step S1, this interface acts as an input terminal, receiving raw observation data from the front-end acquisition device; in step S3, this interface acts as an output terminal, transmitting data including the target sampling frequency. The hardware control commands for the modal activation code are sent to the front-end device, forming a physical closed loop.

[0162] Broadband network interface: including Gigabit / 10 Gigabit Ethernet or fiber optic interface. This interface is used to respond to the HTTP / HTTPS tracing request initiated by the external audit terminal in step S5 and to transmit the data stream that has undergone integrity verification.

[0163] The system bus connects the above components. In order to meet the high-speed direct memory access of the holographic data block from memory to disk in step S4, the system bus adopts the PCIe 5.0 or higher bus standard to eliminate the bandwidth bottleneck when data is migrated between hierarchical storage media.

[0164] Furthermore, one embodiment of the present invention also provides a computer-readable storage medium. This storage medium can be any non-transitory medium capable of storing program code (such as an optical disc, hard disk, flash drive, or cloud storage object). A computer program is stored on the storage medium, which, when run on an electronic device, causes the electronic device to perform the following operations:

[0165] Execution step S1: Map the original data to include a unified timestamp through heterogeneous protocol parsing. and spatial coordinate vector Standardized holographic data packets;

[0166] Execution step S2: Based on content similarity and causal probability function The spatiotemporal causal relationship spectrum is dynamically constructed and updated in memory.

[0167] Execute step S3: Calculate the topological potential energy of the region. When the potential energy exceeds the threshold, it generates a value used to adjust the sampling frequency of the front-end device. Feedback instructions;

[0168] Execution step S4: Monitor data node association entropy Serialize the low-entropy stable subgraph into a binary holographic block. The data nodes will be migrated to cold storage while retaining the tombstone data nodes in primary storage.

[0169] Execute step S5: Utilize hash index The retrieved historical data is verified against tampering.

[0170] The data structure stored in this storage medium includes not only executable instructions but also a holographic data file format generated according to the method of this invention. The holographic data file format consists of a metadata header, a compressed sequence of data node attributes, and a tail hash signature for integrity verification.

Claims

1. An information management method, characterized in that, Includes the following steps: S1. Initialization and spatiotemporal mapping of multi-source heterogeneous data: Receive multi-source heterogeneous data from the front-end acquisition device and map it into a standardized holographic data packet containing a unified spatiotemporal reference. The multi-source heterogeneous data includes at least one of carbon emission management data, smart energy consumption data, smart plant data, and smart operation and maintenance data. S2. Construct a spatiotemporal causal correlation spectrum: Instantiate the holographic data packet into data nodes, and construct and dynamically update a spatiotemporal causal correlation spectrum with a topological structure based on the correlation weights between data nodes. S3. Adaptive frequency-varying acquisition feedback based on topological potential energy: Calculate the topological potential energy value of the physical region in the spatiotemporal causal correlation spectrum, and when the topological potential energy value exceeds the warning threshold, generate hardware control instructions to adjust the data sampling frequency of the front-end acquisition device; S4. Storage state evolution based on correlation entropy: Monitor the correlation entropy value of the data node, and when the correlation entropy value meets the phase transition condition, serialize the corresponding stable subgraph into a binary holographic data block, perform cross-layer migration from the first storage layer to the second storage layer, and generate a hash index value. S5. Trusted traceability interaction: Retrieve the binary holographic data block from the second storage layer based on the hash index value, and perform integrity verification to complete information traceability management.

2. The information management method according to claim 1, characterized in that, In step S1, the holographic data packet is configured as a four-dimensional vector structure, which includes a content carrier field, a spatial coordinate vector field, a timestamp field, and a device attribute vector field. In step S2, the construction and dynamic updating of a spatiotemporal causal correlation spectrum with topological structure based on the correlation weights between data nodes specifically includes: Calculate the temporal dimension difference, spatial dimension distance, and semantic content similarity between any two data nodes; The association weight is calculated by weighting the negative exponential function term of the time dimension difference, the negative exponential function term of the spatial dimension distance, and the content semantic similarity term. Set a connection threshold, and establish connection edges between the data nodes when the association weight exceeds the connection threshold.

3. The information management method according to claim 1, characterized in that, In step S3, the topological potential energy value is calculated by weighted summation based on the sum of the connectivity degree of all data nodes in the physical region and the sum of the weights of all connected edges. The hardware control command includes a target sampling frequency parameter, which is a value that configures the data sampling frequency. The target sampling frequency parameter is calculated based on the product of the base sampling frequency and the logarithmic gain term of the topological potential value relative to the warning threshold, and the target sampling frequency parameter is limited to the maximum sampling frequency range supported by the front-end acquisition device.

4. The information management method according to claim 3, characterized in that, In step S3, the hardware control instructions further include modal activation instructions, which specifically include: When the topological potential energy value exceeds the warning threshold, the modal activation command drives the auxiliary sensor in the front-end acquisition device to switch from a dormant state to a working state. The method also includes a hysteresis recovery mechanism, which sets a recovery threshold less than the warning threshold. Only when the topological potential energy value remains below the recovery threshold within a stable time window, a reset command is generated to restore the basic sampling frequency and cut off the power supply to the auxiliary sensor.

5. The information management method according to claim 1, characterized in that, In step S4, the correlation entropy value is used to quantify the degree of disorder in the causal connection structure of the data node in the current spatiotemporal context; Step S4 also includes a tombstone replacement operation, which involves creating a tombstone data node at the original location of the first storage layer after migrating the stable subgraph to the second storage layer. The tombstone data node contains a pointer to a physical storage address in the second storage layer.

6. The information management method according to claim 1, characterized in that, In step S4, generating the hash index value specifically involves calculating the hash index value by using a one-way cryptographic hash function to combine the metadata header, attribute vector sequence, spatial coordinate vector sequence, and binary weight sequence of the binary holographic data block. In step S5, the integrity verification specifically involves calculating the real-time hash value of the binary holographic data block retrieved from the second storage layer, comparing the real-time hash value with the hash index value generated in step S4, and determining that the integrity verification passes when the real-time hash value matches the hash index value.

7. The information management method according to claim 2, characterized in that, In step S1, the method for generating the spatial coordinate vector field in the holographic data packet includes: For the fixed type of front-end acquisition device, the corresponding spatial coordinates are retrieved from the pre-stored static device location table based on the hardware identifier; For mobile-type front-end acquisition devices, real-time location information is obtained by reading the dynamic positioning interface and converted into spatial coordinates.

8. An information management system, characterized in that, The information management method according to any one of claims 1-7 includes: A multimodal sensing access module is configured to receive multi-source heterogeneous data from front-end acquisition devices and map it into a standardized holographic data packet containing a unified spatiotemporal reference. The multi-source heterogeneous data includes at least one of carbon emission management data, smart energy consumption data, smart factory data, and smart operation and maintenance data. The spectrum dynamic construction module is configured to instantiate the holographic data packet into data nodes, and construct and dynamically update a spatiotemporal causal correlation spectrum with topological structure based on the correlation weights between data nodes. The topology feedback control module is configured to calculate the topological potential energy value of the physical region in the spatiotemporal causal correlation spectrum, and generate hardware control instructions to adjust the data sampling frequency of the front-end acquisition device when the topological potential energy value exceeds a warning threshold. The storage evolution management module is configured to monitor the correlation entropy value of the data node, and when the correlation entropy value meets the phase transition condition, serialize the corresponding stable subgraph into a binary holographic data block, perform cross-layer migration from the first storage layer to the second storage layer, and generate a hash index value. The trusted traceability interaction module is configured to retrieve the binary holographic data block from the second storage layer based on the hash index value and perform integrity verification to complete information traceability management.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the information management method as described in any one of claims 1-7.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the information management method as described in any one of claims 1-7.

Images