Cloud-computing-based natural resource engineering monitoring method and system

Abstract

The application discloses a natural resource engineering monitoring method and system based on cloud computing, relates to the technical field of engineering monitoring, constructs a cloud access layer, forms a standardized monitoring data set, establishes a monitoring quality anomaly label set, generates an engineering state vector set based on a space-time aligned monitoring data set, outputs a risk event set and generates a risk evidence chain, receives an execution receipt and establishes a corresponding relationship, and realizes the standardization and unified management of multi-source monitoring data through cloud access, guarantees data accuracy in combination with a time synchronization, space mapping and quality label mechanism, constructs an engineering state profile and generates a consistent snapshot, supports historical comparison and tracing, generates a risk event set and a risk evidence chain through a causal graph atlas inference, realizes risk full-chain tracing, further generates a scheduling instruction and forms a closed-loop management in combination with an execution receipt, supports differential evaluation and dynamic adjustment, so as to realize the linkage and continuous optimization of monitoring, risk identification and engineering scheduling.

Description

Technical Field

[0001] This invention relates to the field of engineering monitoring technology, and in particular to a cloud computing-based method and system for monitoring natural resource engineering. Background Technology

[0002] With the advancement of large-scale water conservancy projects, mining development, and transportation infrastructure construction, the demand for safety monitoring in natural resource engineering is increasing. Existing monitoring technologies largely rely on single-type sensors or localized monitoring methods, making it difficult to achieve the fusion and processing of multi-source heterogeneous data across time and space. Furthermore, traditional monitoring systems mostly focus on data acquisition and alarms, lacking unified spatiotemporal alignment mechanisms and quality control measures, resulting in incomparability and low reliability of monitoring data. In addition, the linkage between monitoring results and project scheduling is insufficient, failing to achieve real-time decision-making and execution closed-loop based on monitoring data.

[0003] Currently, Chinese invention patent CN117011110A discloses a monitoring method for natural resource ecological protection and restoration projects. This method includes the following steps: constructing a digital ecological space to form a data resource system for the integrated protection and restoration of mountains, rivers, forests, fields, lakes, grasslands, and deserts; constructing an IoT sensing module, a monitoring and early warning module, and a comprehensive evaluation module; and constructing a project closed-loop management module to form an integrated collaborative supervision application system for the entire lifecycle of engineering projects. This monitoring method for natural resource ecological protection and restoration projects, through digital means, achieves real-time, dynamic, visualized, and traceable comprehensive monitoring and supervision of ecological protection and restoration projects, supporting the overall intelligent governance of all elements, the entire lifecycle, and the entire process of ecological protection and restoration projects for mountains, rivers, forests, fields, lakes, grasslands, and deserts.

[0004] The aforementioned technologies are insufficient to achieve unified access, spatiotemporal alignment, and quality control of multi-source monitoring data, and to form a closed-loop linkage between monitoring results and scheduling execution. Summary of the Invention

[0005] The technical problem solved by this invention is that existing technologies are unable to achieve unified access, spatiotemporal alignment and quality control of multi-source monitoring data, and to form a closed-loop linkage between monitoring results and scheduling execution.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] A cloud-based method for monitoring natural resource engineering includes the following steps:

[0008] Step S1: Construct a cloud access layer, collect raw monitoring data uploaded by edge acquisition nodes, and generate source identifier, device identifier, and time identifier to form a standardized monitoring dataset;

[0009] Step S2 involves performing time synchronization, spatial mapping, and quality label generation on the standardized monitoring dataset to obtain a spatiotemporally aligned monitoring dataset and establishing a set of monitoring quality anomaly labels.

[0010] Step S3: Generate a set of engineering status vectors based on the spatiotemporal aligned monitoring dataset, construct a unified monitoring profile, and store it in versioned storage to generate a verifiable consistency snapshot.

[0011] Step S4: Run causal graph inference on the set of engineering state vectors, output a set of risk events and generate a risk evidence chain, record and store it in the audit log. The risk evidence chain includes the monitoring data source corresponding to the risk event, the relationship of monitoring parameters involved in the inference path, the threshold basis and the trigger node.

[0012] Step S5: Generate a set of scheduling instructions based on the monitoring profile and risk evidence chain and send them to the edge execution nodes, receive execution receipts and establish corresponding relationships;

[0013] Step S6: Generate a differential evaluation report based on the execution receipt and the new spatiotemporal alignment monitoring dataset, update the threshold inheritance graph and causal graph versions, and write the change records to the audit log;

[0014] Step S3 includes the following sub-steps:

[0015] Step S301: Generate an engineering state vector set based on the spatiotemporal aligned monitoring dataset. The engineering state vector set includes displacement, seepage, pore water pressure, and environmental conditions, and is aggregated to form a multidimensional vector set.

[0016] Step S302: Construct a unified monitoring profile based on the engineering state vector set, combine the engineering state vectors corresponding to each monitoring point in a unified spatial coordinate system to generate a visual profile, and maintain a one-to-one correspondence with the engineering state vector set.

[0017] Step S303: Store the engineering status vector set and monitoring profile into versioned storage, generate version number and verification information for each write, and form a verifiable consistency snapshot.

[0018] Step S5 includes the following sub-steps:

[0019] Step S501: Generate a set of scheduling instructions based on the monitoring profile and the risk evidence chain. The set of scheduling instructions includes the execution action, priority and constraints. During the generation of the set of scheduling instructions, the instructions are bound to the corresponding risk evidence chain.

[0020] Step S502: The set of scheduling instructions is sent to the edge execution node and the task is executed. The edge execution node parses and executes the received instructions. The content of the execution includes equipment start-up and shutdown operations, operating parameter adjustment and emergency handling operations.

[0021] Step S503: Receive the execution receipts returned by the edge execution nodes and establish a corresponding relationship. The execution receipts include the execution results, completion time, device status and abnormal information. The cloud stores the execution receipts and scheduling instruction sets accordingly.

[0022] Step S6 includes the following sub-steps:

[0023] Step S601: Generate a differential evaluation report based on the execution receipt and the new spatiotemporal alignment monitoring dataset. The differential evaluation report includes the deviation between the scheduling instructions and the actual response, data trend changes, monitoring profile comparison results, and anomaly statistics.

[0024] Step S602: Update the threshold inheritance graph and causal graph versions. During the update process, incorporate the new round of monitoring data and differential evaluation results into the calculation, correct the historical threshold, output the inheritance threshold, and adjust the node relationships and edge weights in the causal graph.

[0025] Step S603: Write the threshold inheritance graph update and causal graph change records into the audit log, and attach a timestamp, version number, reason for modification and corresponding monitoring data source to the log.

[0026] The set of scheduling instructions and execution receipts form a closed-loop management system. The closed-loop management includes a one-to-one correspondence between scheduling instructions and execution receipts, recording of execution result deviations, classification of abnormal feedback, and automatic adjustment of scheduling strategies. The closed-loop management supports differential evaluation and auditing.

[0027] Preferably, step S1 includes the following sub-steps:

[0028] Step S101: Collect slope displacement monitoring data, seepage pressure monitoring data, pore water pressure monitoring data, rainfall monitoring data, GNSS base station monitoring data, and construction equipment operation logs, and upload the slope displacement monitoring data, seepage pressure monitoring data, pore water pressure monitoring data, rainfall monitoring data, GNSS base station monitoring data, and construction equipment operation logs to the cloud to form an original monitoring data set;

[0029] Step S102 involves generating source identifiers, device identifiers, and time identifiers for each piece of data in the original monitoring data set, and then appending them to the original monitoring data set to form a standardized monitoring dataset.

[0030] Preferably, step S2 includes the following sub-steps:

[0031] Step S201: Perform time synchronization processing on the standardized monitoring dataset;

[0032] Step S202: Perform spatial mapping processing on the standardized monitoring dataset;

[0033] Step S203: Generate monitoring quality labels for the standardized monitoring dataset and establish a set of monitoring quality anomaly labels. Label slope displacement monitoring data, seepage pressure monitoring data, pore water pressure monitoring data, rainfall monitoring data, GNSS base station monitoring data and construction equipment operation logs that have outliers, missing segments and signal noise anomalies with anomaly labels, and output the spatiotemporally aligned monitoring dataset.

[0034] Preferably, step S4 includes the following sub-steps:

[0035] Step S401: Run causal graph inference on the engineering state vector set to generate a risk event set. The causal graph inference identifies links that may cause engineering anomalies by analyzing the causal relationship and time series changes between monitoring parameters and outputting a risk event set, which includes displacement mutation, seepage anomaly, pore water pressure increase and rainfall overload.

[0036] Step S402: Generate a risk evidence chain based on the risk event set;

[0037] Step S403: Record the risk evidence chain and store it in the audit log.

[0038] Preferably, the risk evidence chain enables full-chain tracing of the formation process of a risk event, specifically including:

[0039] Indicate the source of the original monitoring data corresponding to the triggering event, the inference node in the causal graph, the threshold conditions on which the judgment is based, and the monitoring unit involved.

[0040] The cloud-based natural resource engineering monitoring system includes a data unification module, a spatiotemporal alignment module, a monitoring construction module, a map inference module, an instruction execution module, and a closed-loop evaluation module.

[0041] The data unification module is used to build a cloud access layer, collect raw monitoring data uploaded by edge acquisition nodes, and generate source identifiers, device identifiers, and time identifiers to form a standardized monitoring dataset.

[0042] The spatiotemporal alignment module is used to perform time synchronization, spatial mapping, and quality label generation on the standardized monitoring dataset to obtain a spatiotemporal aligned monitoring dataset and to establish a set of monitoring quality anomaly labels.

[0043] The monitoring construction module is used to generate a set of engineering status vectors based on the spatiotemporally aligned monitoring dataset, construct a unified monitoring profile, and store it in versioned storage to generate verifiable consistency snapshots.

[0044] The graph inference module is used to run causal graph inference on the set of engineering state vectors, output a set of risk events and generate a risk evidence chain, and record and store it in the audit log.

[0045] The instruction execution module is used to generate a set of scheduling instructions based on the monitoring profile and the risk evidence chain and send them to the edge execution nodes, receive execution receipts and establish corresponding relationships;

[0046] The closed-loop evaluation module is used to generate a differential evaluation report based on the execution receipt and the new spatiotemporal alignment monitoring dataset, update the threshold inheritance graph and causal graph versions, and write the change records to the audit log.

[0047] The beneficial effects of this invention are as follows: This invention achieves standardized and unified management of multi-source monitoring data through cloud access, and ensures data accuracy by combining time synchronization, spatial mapping and quality labeling mechanisms. The method constructs an engineering status profile and generates a consistent snapshot, supports historical comparison and traceability, and generates a set of risk events and a risk evidence chain through causal graph inference, realizing full-chain traceability of risks. Furthermore, it generates scheduling instructions and forms a closed-loop management system by combining execution receipts, supporting differential evaluation and dynamic adjustment, thereby realizing the linkage and continuous optimization of monitoring, risk identification and engineering scheduling. Attached Figure Description

[0048] Figure 1 A flowchart illustrating the steps of a cloud computing-based natural resource engineering monitoring method according to an embodiment of the present invention;

[0049] Figure 2 This is a basic flowchart of a cloud-based natural resource engineering monitoring system provided as an embodiment of the present invention. Detailed Implementation

[0050] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0051] Example 1, referring to Figure 1 It provides a cloud-based method for monitoring natural resource engineering, including the following steps:

[0052] Step S1: Construct a cloud access layer, collect raw monitoring data uploaded by edge acquisition nodes, and generate source identifier, device identifier, and time identifier to form a standardized monitoring dataset.

[0053] Step S2 involves performing time synchronization, spatial mapping, and quality label generation on the standardized monitoring dataset to obtain a spatiotemporally aligned monitoring dataset and establishing a set of monitoring quality anomaly labels.

[0054] Step S3: Generate a set of engineering status vectors based on the spatiotemporal aligned monitoring dataset, construct a unified monitoring profile, and store it in versioned storage to generate a verifiable consistency snapshot.

[0055] Step S4: Run causal graph inference on the set of engineering state vectors, output a set of risk events and generate a risk evidence chain, record and store it in the audit log. The risk evidence chain includes the monitoring data source corresponding to the risk event, the relationship of monitoring parameters involved in the inference path, the threshold basis and the trigger node.

[0056] Step S5: Generate a set of scheduling instructions based on the monitoring profile and risk evidence chain and send them to the edge execution nodes, receive execution receipts and establish corresponding relationships.

[0057] Step S6: Generate a differential evaluation report based on the execution receipt and the new spatiotemporal alignment monitoring dataset, update the threshold inheritance graph and causal graph versions, and write the change record to the audit log.

[0058] This invention achieves standardized and unified management of multi-source monitoring data through cloud access, and ensures data accuracy by combining time synchronization, spatial mapping and quality labeling mechanisms. The method constructs an engineering status profile and generates a consistent snapshot, supports historical comparison and traceability, and generates a set of risk events and a risk evidence chain through causal graph inference, realizing full-chain traceability of risks. Furthermore, it generates scheduling instructions and forms a closed-loop management system with execution receipts, supports differential evaluation and dynamic adjustment, thereby realizing the linkage and continuous optimization of monitoring, risk identification and engineering scheduling.

[0059] Step S1 includes the following sub-steps:

[0060] Step S101: Collect slope displacement monitoring data, seepage pressure monitoring data, pore water pressure monitoring data, rainfall monitoring data, GNSS base station monitoring data, and construction equipment operation logs, and upload the slope displacement monitoring data, seepage pressure monitoring data, pore water pressure monitoring data, rainfall monitoring data, GNSS base station monitoring data, and construction equipment operation logs to the cloud to form an original monitoring data set.

[0061] Step S101 collects various types of monitoring data, such as slope displacement, seepage pressure, pore water pressure, rainfall, GNSS base station and construction equipment operation logs, which can comprehensively cover the structural safety, environmental hydrology and construction disturbance of natural resource projects, forming a complete set of original monitoring data.

[0062] Step S102 involves generating source identifiers, device identifiers, and time identifiers for each piece of data in the original monitoring data set, and then appending them to the original monitoring data set to form a standardized monitoring dataset.

[0063] Step S102 generates source identifier, device identifier and time identifier for each piece of original monitoring data and attaches them to the data to form a standardized monitoring dataset. This enables consistent data management and traceability, avoiding analytical bias caused by unknown sources or format differences.

[0064] Step S1 enables unified access and standardized management of multi-source monitoring data, ensuring that monitoring data from different sources and in different formats can form a consistent and traceable data foundation in the cloud, providing a reliable data source for subsequent spatiotemporal alignment, state vector generation, and causal inference.

[0065] Step S2 includes the following sub-steps:

[0066] Step S201: Perform time synchronization processing on the standardized monitoring dataset.

[0067] Step S201 uses time synchronization processing to unify the data uploaded by different sensors and acquisition nodes to the same time reference, eliminating errors caused by acquisition delay and clock drift, and ensuring the consistency of various monitoring data in the time dimension.

[0068] Step S202: Perform spatial mapping processing on the standardized monitoring dataset.

[0069] Step S202 uses spatial mapping to map the data from each monitoring point to a unified coordinate system, correcting differences in the collection locations and enabling data from different sources to be superimposed and compared in space, thus ensuring the spatial consistency of the analysis results.

[0070] Step S203: Generate monitoring quality labels for the standardized monitoring dataset and establish a set of monitoring quality anomaly labels. Label slope displacement monitoring data, seepage pressure monitoring data, pore water pressure monitoring data, rainfall monitoring data, GNSS base station monitoring data and construction equipment operation logs that have outliers, missing segments and signal noise anomalies with anomaly labels, and output the spatiotemporally aligned monitoring dataset.

[0071] Step S203 generates monitoring quality labels and establishes a set of monitoring quality anomaly labels to identify abnormal values, missing segments, or noise data in slope displacement, seepage pressure, pore water pressure, rainfall, GNSS base station and construction equipment operation logs, forming traceable quality anomaly records, and outputs a cleaned and labeled spatiotemporally aligned monitoring dataset.

[0072] Step S2 unifies the multi-source monitoring data in both time and space dimensions, and combines a quality control mechanism to screen out abnormal samples, forming a spatiotemporally aligned monitoring dataset, thereby ensuring the accuracy and reliability of the subsequent generation of engineering state vectors and monitoring profiles.

[0073] Step S3 includes the following sub-steps:

[0074] Step S301: Generate an engineering state vector set based on the spatiotemporal aligned monitoring dataset. The engineering state vector set includes displacement, seepage, pore water pressure, and environmental conditions, and is aggregated to form a multidimensional vector set.

[0075] Step S301 generates an engineering state vector set based on a spatiotemporally aligned monitoring dataset, which aggregates multi-dimensional monitoring parameters such as displacement, seepage, pore water pressure, and environmental conditions, enabling a systematic characterization of the operation status of natural resource engineering in each monitoring unit and forming a comprehensive numerical representation.

[0076] Step S302: Construct a unified monitoring profile based on the engineering state vector set. Combine the engineering state vectors corresponding to each monitoring point in a unified spatial coordinate system to generate a visual profile and maintain a one-to-one correspondence with the engineering state vector set.

[0077] Step S302 constructs a unified monitoring profile based on the set of engineering state vectors, combines the state vectors corresponding to each monitoring point in a unified spatial coordinate system, generates an intuitive visualization profile, realizes a one-to-one correspondence between monitoring data and spatial location, and facilitates an intuitive analysis of the overall safety status of the dam body, bedrock and construction area.

[0078] Step S303: Store the engineering status vector set and monitoring profile into versioned storage, and generate version number and verification information for each write to form a verifiable consistency snapshot.

[0079] Step S303 stores the engineering status vector set and monitoring profile into versioned storage, and generates a version number and verification information each time it is written, which can form a verifiable consistency snapshot, realize the historical traceability, version comparison and long-term consistency verification of monitoring data and profile.

[0080] Step S3 realizes the structured expression and spatial visualization of multi-source monitoring data, forming a multi-dimensional vector set that can reflect the overall engineering status and intuitively display the monitoring profile of key parts. It also establishes verifiable and consistent snapshots through versioned storage, providing a reliable basis for historical comparison and tracing.

[0081] Step S4 includes the following sub-steps:

[0082] Step S401: Run causal graph inference on the engineering state vector set to generate a risk event set. Causal graph inference identifies links that may cause engineering anomalies by analyzing the causal relationship and temporal changes between monitoring parameters. The output risk event set includes displacement mutation, seepage anomaly, pore water pressure increase and rainfall overload.

[0083] Step S401, by running causal graph inference on the set of engineering state vectors, can analyze the causal relationships and temporal changes between monitoring parameters, identify causal links that may lead to engineering anomalies, and output a set of risk events. This set covers key risk types such as displacement mutation, seepage anomaly, pore water pressure increase and rainfall overload, and comprehensively reflects the potential safety hazards of the project.

[0084] Step S402: Generate a risk evidence chain based on the risk event set.

[0085] Step S402 generates a risk evidence chain based on the risk event set, connecting the monitoring data sources, causal inference paths, threshold criteria, and triggering nodes corresponding to the risk events to form a complete evidence chain, ensuring that each risk conclusion has a clear formation process and interpretability, which facilitates subsequent responsibility determination and review.

[0086] Step S403: Record the risk evidence chain and store it in the audit log.

[0087] Step S403, by recording and storing the risk evidence chain in the audit log, can solidify the entire process of risk identification and inference, ensure the traceability of risk sources, and provide reliable data for regulatory inspections and historical audits.

[0088] Step S4 implements causal inference analysis based on the set of engineering state vectors, which can identify potential risk events and form a risk evidence chain, thereby providing interpretable and traceable risk evidence for engineering safety monitoring, and solidifying the risk analysis process through audit logs.

[0089] Step S5 includes the following sub-steps:

[0090] Step S501: Generate a set of scheduling instructions based on the monitoring profile and the risk evidence chain. The set of scheduling instructions includes the execution action, priority and constraints. During the generation of the set of scheduling instructions, the instructions are bound to the corresponding risk evidence chain.

[0091] Step S501 generates a set of scheduling instructions based on the monitoring profile and the risk evidence chain, which can transform the abstract risk identification results into specific execution actions, set priorities and constraints, and bind the corresponding risk evidence chain to achieve transparency and traceability of scheduling decisions.

[0092] Step S502: The set of scheduling instructions is sent to the edge execution node and the task is executed. The edge execution node parses and executes the received instructions. The execution includes equipment start-up and shutdown operations, operating parameter adjustments and emergency response operations.

[0093] Step S502 sends the set of scheduling instructions to the edge execution node for parsing and execution, enabling on-site equipment start-up and shutdown, operating parameter adjustment and emergency response operations to be completed directly, so that the scheduling strategy generated in the cloud can be implemented in a timely manner, ensuring the real-time and effectiveness of risk handling.

[0094] Step S503: Receive the execution receipts returned by the edge execution nodes and establish a corresponding relationship. The execution receipts include the execution results, completion time, device status and abnormal information. The cloud stores the execution receipts and scheduling instruction sets accordingly.

[0095] Step S503, by receiving the execution receipts returned by the edge execution nodes and establishing a corresponding relationship, can store the execution results, completion time, device status, and abnormal information in a one-to-one correspondence with the original scheduling instructions, forming a closed-loop record of instructions and execution, providing a complete basis for subsequent differential evaluation, strategy optimization, and accountability.

[0096] Step S5 realizes a closed loop of scheduling execution based on monitoring profile and risk evidence chain, transforms risk identification results into operable scheduling instructions and implements them at edge nodes, and then stores them in conjunction with execution receipts, thereby opening up the entire process link.

[0097] Step S6 includes the following sub-steps:

[0098] Step S601: Generate a differential evaluation report based on the execution receipt and the new spatiotemporal alignment monitoring dataset. The differential evaluation report includes the deviation between the scheduling instructions and the actual response, data trend changes, monitoring profile comparison results, and anomaly statistics.

[0099] Step S601 generates a differential evaluation report based on the execution receipt and the new spatiotemporal alignment monitoring dataset. This report can quantify the difference between scheduling instructions and actual responses, track data trend changes, compare the differences between the old and new monitoring profiles, and statistically analyze anomalies, thereby providing a comprehensive evaluation of scheduling effectiveness and risk management effectiveness.

[0100] Step S602: Update the threshold inheritance graph and causal graph versions. During the update process, incorporate the new round of monitoring data and differential evaluation results into the calculation, correct the historical threshold, output the inheritance threshold, and adjust the node relationships and edge weights in the causal graph.

[0101] Step S602 updates the threshold inheritance graph and causal graph versions, incorporates the new round of monitoring data and differential evaluation results into the calculation, corrects historical thresholds, outputs inheritance thresholds that are more consistent with the current working conditions, and adjusts the node relationships and edge weights of the causal graph, thereby maintaining the synchronization and accuracy of the model with the actual operating state of the project.

[0102] Step S603: Write the threshold inheritance graph update and causal graph change records into the audit log, and attach a timestamp, version number, reason for modification and corresponding monitoring data source to the log.

[0103] Step S603 writes the threshold inheritance graph update and causal graph change records into the audit log, and adds timestamps, version numbers, reasons for modification and monitoring data sources. This can retain complete evidence for each update, ensure that the modification process is traceable and transparent, and provide long-term reliable record basis for supervision and operation and maintenance.

[0104] Step S6 achieves closed-loop evaluation and model update of monitoring data and scheduling execution. Through differential evaluation reports, execution deviations and new risk characteristics are discovered, and thresholds and causal graphs are corrected accordingly. Finally, the update process is solidified in the form of logs to build a monitoring and control system that can be continuously iterated and optimized.

[0105] The risk evidence chain enables full-chain tracing of the formation process of a risk event, specifically including:

[0106] Indicate the source of the original monitoring data corresponding to the triggering event, the inference node in the causal graph, the threshold conditions on which the judgment is based, and the monitoring unit involved.

[0107] The set of scheduling instructions and execution receipts form a closed-loop management system. The closed-loop management includes a one-to-one correspondence between scheduling instructions and execution receipts, recording of execution result deviations, classification of abnormal feedback, and automatic adjustment of scheduling strategies. The closed-loop management supports differential evaluation and auditing.

[0108] Example 2, refer to Figure 2 It provides a cloud-based natural resource engineering monitoring system, including a data unification module, a spatiotemporal alignment module, a monitoring construction module, a map inference module, an instruction execution module, and a closed-loop evaluation module.

[0109] The data unification module is used to build a cloud access layer, collect raw monitoring data uploaded by edge acquisition nodes, and generate source identifiers, device identifiers, and time identifiers to form a standardized monitoring dataset.

[0110] The spatiotemporal alignment module is used to perform time synchronization, spatial mapping, and quality label generation on the standardized monitoring dataset to obtain a spatiotemporally aligned monitoring dataset and to establish a set of monitoring quality anomaly labels.

[0111] The monitoring building module is used to generate a set of engineering status vectors based on the spatiotemporally aligned monitoring dataset, construct a unified monitoring profile, and store it in versioned storage to generate verifiable consistency snapshots.

[0112] The graph inference module is used to run causal graph inference on the set of engineering state vectors, output a set of risk events and generate a risk evidence chain, which is then recorded and stored in the audit log.

[0113] The instruction execution module is used to generate a set of scheduling instructions based on the monitoring profile and risk evidence chain and send them to the edge execution nodes, receive execution receipts and establish corresponding relationships.

[0114] The closed-loop evaluation module is used to generate differential evaluation reports based on execution receipts and new spatiotemporal alignment monitoring datasets, update the threshold inheritance graph and causal graph versions, and write change records to the audit log.

[0115] This invention achieves standardized processing and unified identification of multi-source monitoring data, such as slope displacement, seepage pressure, pore water pressure, rainfall, GNSS base station and construction equipment operation logs, through a cloud access layer. This improves data traceability and consistency. A unified spatiotemporal framework is established through time synchronization and spatial mapping, and an abnormal monitoring data is identified by a quality labeling mechanism, ensuring the accuracy and reliability of monitoring results. A unified monitoring profile is constructed and verifiable consistency snapshots are generated, giving the monitoring data versioning and long-term traceability, providing support for historical comparison and trend analysis. A set of risk events is generated based on causal graph inference, forming a risk evidence chain, realizing full-chain traceability of the risk event formation process, with interpretability and responsibility definition capabilities. A set of scheduling instructions is generated and issued to edge execution nodes, and a closed-loop management mechanism is established with execution receipts, supporting differential evaluation and strategy adjustment, realizing real-time linkage between monitoring and engineering scheduling. The differential evaluation report drives the dynamic updating of threshold inheritance graph and causal graph, forming an iteratively optimized monitoring and control system that adapts to changes in the engineering environment.

[0116] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0117] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A cloud computing-based method for monitoring natural resource engineering, characterized in that, Includes the following steps: Step S1: Construct a cloud access layer, collect raw monitoring data uploaded by edge acquisition nodes, and generate source identifier, device identifier, and time identifier to form a standardized monitoring dataset; Step S2 involves performing time synchronization, spatial mapping, and quality label generation on the standardized monitoring dataset to obtain a spatiotemporally aligned monitoring dataset and establishing a set of monitoring quality anomaly labels. Step S3: Generate a set of engineering status vectors based on the spatiotemporal aligned monitoring dataset, construct a unified monitoring profile, and store it in versioned storage to generate a verifiable consistency snapshot. Step S4: Run causal graph inference on the set of engineering state vectors, output a set of risk events and generate a risk evidence chain, record and store it in the audit log. The risk evidence chain includes the monitoring data source corresponding to the risk event, the relationship of monitoring parameters involved in the inference path, the threshold basis and the trigger node. Step S5: Generate a set of scheduling instructions based on the monitoring profile and risk evidence chain and send them to the edge execution nodes, receive execution receipts and establish corresponding relationships; Step S6: Generate a differential evaluation report based on the execution receipt and the new spatiotemporal alignment monitoring dataset, update the threshold inheritance graph and causal graph versions, and write the change record to the audit log; Step S3 includes the following sub-steps: Step S301: Generate an engineering state vector set based on the spatiotemporal aligned monitoring dataset. The engineering state vector set includes displacement, seepage, pore water pressure, and environmental conditions, and is aggregated to form a multidimensional vector set. Step S302: Construct a unified monitoring profile based on the engineering state vector set, combine the engineering state vectors corresponding to each monitoring point in a unified spatial coordinate system to generate a visual profile, and maintain a one-to-one correspondence with the engineering state vector set. Step S303: Store the engineering status vector set and monitoring profile into versioned storage, generate version number and verification information for each write, and form a verifiable consistency snapshot. Step S5 includes the following sub-steps: Step S501: Generate a set of scheduling instructions based on the monitoring profile and the risk evidence chain. The set of scheduling instructions includes the execution action, priority and constraints. During the generation of the set of scheduling instructions, the instructions are bound to the corresponding risk evidence chain. Step S502: The set of scheduling instructions is sent to the edge execution node and the task is executed. The edge execution node parses and executes the received instructions. The content of the execution includes equipment start-up and shutdown operations, operating parameter adjustment and emergency handling operations. Step S503: Receive the execution receipts returned by the edge execution nodes and establish a corresponding relationship. The execution receipts include the execution results, completion time, device status and abnormal information. The cloud stores the execution receipts and scheduling instruction sets accordingly. Step S6 includes the following sub-steps: Step S601: Generate a differential evaluation report based on the execution receipt and the new spatiotemporal alignment monitoring dataset. The differential evaluation report includes the deviation between the scheduling instructions and the actual response, data trend changes, monitoring profile comparison results, and anomaly statistics. Step S602: Update the threshold inheritance graph and causal graph versions. During the update process, incorporate the new round of monitoring data and differential evaluation results into the calculation, correct the historical threshold, output the inheritance threshold, and adjust the node relationships and edge weights in the causal graph. Step S603: Write the threshold inheritance graph update and causal graph change records into the audit log, and attach the timestamp, version number, reason for modification and corresponding monitoring data source to the log; The set of scheduling instructions and execution receipts form a closed-loop management system. The closed-loop management includes a one-to-one correspondence between scheduling instructions and execution receipts, recording of execution result deviations, classification of abnormal feedback, and automatic adjustment of scheduling strategies. The closed-loop management supports differential evaluation and auditing.

2. The cloud computing-based natural resource engineering monitoring method as described in claim 1, characterized in that, Step S1 includes the following sub-steps: Step S101: Collect slope displacement monitoring data, seepage pressure monitoring data, pore water pressure monitoring data, rainfall monitoring data, GNSS base station monitoring data, and construction equipment operation logs, and upload the slope displacement monitoring data, seepage pressure monitoring data, pore water pressure monitoring data, rainfall monitoring data, GNSS base station monitoring data, and construction equipment operation logs to the cloud to form an original monitoring data set; Step S102 involves generating source identifiers, device identifiers, and time identifiers for each piece of data in the original monitoring data set, and attaching them to the original monitoring data set to form a standardized monitoring dataset.

3. The cloud computing-based natural resource engineering monitoring method as described in claim 2, characterized in that, Step S2 includes the following sub-steps: Step S201: Perform time synchronization processing on the standardized monitoring dataset; Step S202: Perform spatial mapping processing on the standardized monitoring dataset; Step S203: Generate monitoring quality labels for the standardized monitoring dataset and establish a set of monitoring quality anomaly labels. Label slope displacement monitoring data, seepage pressure monitoring data, pore water pressure monitoring data, rainfall monitoring data, GNSS base station monitoring data and construction equipment operation logs that have outliers, missing segments and signal noise anomalies with anomaly labels, and output the spatiotemporally aligned monitoring dataset.

4. The cloud computing-based natural resource engineering monitoring method as described in claim 3, characterized in that, Step S4 includes the following sub-steps: Step S401: Run causal graph inference on the engineering state vector set to generate a risk event set. The causal graph inference identifies links that may cause engineering anomalies by analyzing the causal relationship and time series changes between monitoring parameters and outputting a risk event set, which includes displacement mutation, seepage anomaly, pore water pressure increase and rainfall overload. Step S402: Generate a risk evidence chain based on the risk event set; Step S403: Record the risk evidence chain and store it in the audit log.

5. The cloud computing-based natural resource engineering monitoring method as described in claim 4, characterized in that, The risk evidence chain enables full-chain tracing of the formation process of a risk event, specifically including: Indicate the source of the original monitoring data corresponding to the triggering event, the inference node in the causal graph, the threshold conditions on which the judgment is based, and the monitoring unit involved.

6. A cloud computing-based natural resource engineering monitoring system, applied in any one of the cloud computing-based natural resource engineering monitoring methods as described in claims 1-5, characterized in that, It includes a data unification module, a spatiotemporal alignment module, a monitoring and construction module, a graph inference module, an instruction execution module, and a closed-loop evaluation module; The data unification module is used to build a cloud access layer, collect raw monitoring data uploaded by edge acquisition nodes, and generate source identifiers, device identifiers, and time identifiers to form a standardized monitoring dataset. The spatiotemporal alignment module is used to perform time synchronization, spatial mapping, and quality label generation on the standardized monitoring dataset to obtain a spatiotemporal aligned monitoring dataset and to establish a set of monitoring quality anomaly labels. The monitoring construction module is used to generate a set of engineering status vectors based on the spatiotemporally aligned monitoring dataset, construct a unified monitoring profile, and store it in versioned storage to generate verifiable consistency snapshots. The graph inference module is used to run causal graph inference on the set of engineering state vectors, output a set of risk events and generate a risk evidence chain, and record and store it in the audit log. The instruction execution module is used to generate a set of scheduling instructions based on the monitoring profile and risk evidence chain and send them to the edge execution nodes, receive execution receipts and establish corresponding relationships; The closed-loop evaluation module is used to generate a differential evaluation report based on the execution receipt and the new spatiotemporal alignment monitoring dataset, update the threshold inheritance graph and causal graph versions, and write the change records to the audit log.

Images