An in-situ multi-parameter cooperative monitoring method for high-parameter equipment

Abstract

The application discloses an in-situ multi-parameter cooperative monitoring method for high-parameter equipment, and relates to the technical field of special equipment monitoring. The method comprises the following steps: accurately determining the key monitoring area of the high-parameter equipment based on failure mode analysis, thereby improving the pertinence and effectiveness of monitoring; synchronously collecting multi-source sensing data in the key monitoring area through a unified clock in a non-invasive manner; sequentially performing space-time alignment and multi-level fusion on the multi-source sensing data, so as to generate a comprehensive index representing the health state of the equipment, deeply mine the complementary and related information among the multi-parameters, and significantly improve the extraction ability of early weak fault features; performing graded early warning based on the comprehensive index and dynamic criterion, and iteratively updating the fusion weight or the dynamic criterion according to historical feedback data, so that the system is self-adaptive to equipment aging and working condition changes, the early warning threshold and the fusion model are continuously optimized, the false alarm and missed alarm rates are effectively reduced, and the closed-loop intelligent safety monitoring in the whole life cycle is realized.

Description

Technical Field

[0001] This application relates to the field of special equipment monitoring technology, and in particular to an in-situ multi-parameter collaborative monitoring method for high-parameter equipment. Background Technology

[0002] High-parameter core equipment, such as nuclear power pressure vessels, coal chemical reactors, hydrogen energy storage equipment, supercritical units, and long-distance high-pressure pipelines, are subjected to harsh conditions such as high temperature, high pressure, strong corrosion, and alternating loads for a long time. They are prone to early hidden damage such as stress concentration, wall thickness reduction, crack initiation, sealing failure, local overheating, and media erosion.

[0003] Existing monitoring technologies for these high-parameter core devices have the following drawbacks: they primarily rely on single-parameter or dual-parameter monitoring, making them susceptible to fluctuations in operating conditions and resulting in high false alarm and false negative rates; sensor installation often requires drilling, welding, or disassembly, compromising the integrity and sealing safety of the pressure-bearing structure; conventional electronic sensors lack sufficient temperature resistance, corrosion resistance, and electromagnetic interference resistance, making long-term stable operation difficult; multi-source data lacks a unified spatiotemporal synchronization mechanism, hindering deep collaborative fusion; and early warning systems rely on fixed thresholds, failing to adapt to equipment aging and changes in operating conditions, making it difficult to identify early, subtle fault characteristics.

[0004] Therefore, existing technologies cannot meet the collaborative monitoring needs of high-parameter equipment for long-term, in-situ, online, high-precision, and interference-resistant operation. Summary of the Invention

[0005] The purpose of this application is to provide an in-situ multi-parameter collaborative monitoring method for high-parameter equipment, which can achieve non-invasive, highly reliable and accurate early warning of early defects in high-parameter equipment under extreme working conditions.

[0006] To achieve the above objectives, this application provides the following solution: An in-situ multi-parameter collaborative monitoring method for high-parameter equipment includes the following steps: The measurement point determination step, based on failure mode analysis, determines the key monitoring area of ​​the high-parameter equipment.

[0007] The data acquisition step involves non-invasively collecting multi-source sensor data synchronously in the key monitoring area using a unified clock.

[0008] The status assessment step involves sequentially performing spatiotemporal alignment and multi-level fusion on the multi-source sensor data to generate a comprehensive index that can characterize the health status of the device.

[0009] The early warning and optimization steps involve performing tiered early warnings based on the comprehensive index and dynamic criteria, and iteratively updating the weights of the multi-level fusion or the dynamic criteria based on historical feedback data.

[0010] Optionally, the step of determining the measuring point specifically includes the following steps: Based on failure mode analysis, the potential failure modes of the high-parameter equipment are analyzed.

[0011] Assess the probability of occurrence, severity, and detectability of each potential failure mode, and determine the risk priority of each potential failure mode.

[0012] Select one or more locations with the highest risk priority as the key monitoring areas.

[0013] Optionally, in the data acquisition step, the non-invasive method includes deploying the data on the outer wall of the high-parameter equipment using high-temperature resistant adhesives or detachable clamps; the multi-source sensing data is acquired by a passive composite sensing array; the passive composite sensing array includes at least two of the following: fiber optic grating sensing units, high-temperature acoustic emission sensors, electromagnetic ultrasonic thickness probes, and distributed temperature sensors.

[0014] Optionally, the data acquisition step further includes: The sampling frequency is automatically switched according to the steady-state or transient operating conditions of the high-parameter equipment; wherein, when the fluctuation amplitude and change rate of each physical parameter of the high-parameter equipment are both lower than the preset threshold, it is determined to be a steady-state operating condition, otherwise it is determined to be a transient operating condition.

[0015] Optionally, the state assessment step involves spatiotemporal alignment of the multi-source sensing data, specifically including the following steps: The multi-source sensor data is subjected to noise reduction, temperature drift compensation, and outlier removal.

[0016] Assign a unified timestamp and spatial coordinates to the processed multi-source sensor data.

[0017] Optionally, the state assessment step involves multi-level fusion of the spatiotemporally aligned multi-source sensor data, specifically including the following steps: Data layer fusion: Based on the signal-to-noise ratio of each data source, weighting coefficients are dynamically allocated, and spatiotemporally aligned multi-source sensor data are fused.

[0018] Feature layer fusion: Extracting time-frequency domain features of the fused data and coupling correlation features that characterize the linkage relationship between multiple parameters.

[0019] Decision-level fusion: Based on the time-frequency domain features and the coupling correlation features, the comprehensive index is calculated by combining the confidence weight of multi-evidence fusion through the evidence reasoning model.

[0020] Optionally, the formula for calculating the signal-to-noise ratio is: in, For signal-to-noise ratio, The signal amplitude, The noise amplitude is used; the dynamic allocation of weighting coefficients is as follows: the data source with a higher signal-to-noise ratio is assigned a larger weight.

[0021] Optionally, the evidence reasoning model used in the decision-making layer fusion is either the DS evidence theory model or a deep learning model.

[0022] Optionally, the warning levels in the warning and optimization steps include four levels: "Normal," "Attention," "Alarm," and "Emergency"; the method further includes the following steps: Based on the warning level, different operation and maintenance strategies are linked; among them, the "normal" level is matched with a continuous monitoring strategy, the "attention" level is matched with an encrypted monitoring strategy, the "alarm" level is matched with a special detection strategy, and the "emergency" level is matched with a partial maintenance and shutdown handling strategy.

[0023] Optionally, the early warning and optimization steps involve iterative updates based on historical feedback data, specifically including the following steps: Based on historical data throughout the entire lifecycle and operation and maintenance feedback, the weights in the multi-level fusion are updated with the goal of improving the accuracy of early warnings.

[0024] The threshold of the dynamic criterion is adaptively adjusted based on the original design specifications of the equipment and its aging status.

[0025] The identification model used in the early warning and optimization steps is incrementally trained using the newly added data, and the model parameters are updated.

[0026] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides an in-situ multi-parameter collaborative monitoring method for high-parameter equipment. First, based on failure mode analysis, the method accurately identifies key monitoring areas of the equipment, thereby avoiding blind deployment and improving the targeting and effectiveness of monitoring. Then, a non-invasive method is used to synchronously collect multi-source sensor data in the key monitoring areas using a unified clock. This allows for the acquisition of microsecond-level synchronized raw data of multiple physical quantities without downtime or damage to the equipment structure, ensuring equipment integrity and data consistency while reducing subsequent fusion errors. Next, the multi-source sensor data is sequentially spatiotemporally aligned and fused at multiple levels to generate a comprehensive index characterizing the equipment's health status. This deeply mines the complementary and correlated information between multiple parameters, significantly improving the ability to extract early, subtle fault characteristics and the accuracy of assessment. Finally, based on the comprehensive index and dynamic criteria, hierarchical early warning is executed, and the fusion weights or dynamic criteria are iteratively updated according to historical feedback data. This enables the monitoring system to adapt to equipment aging and changes in operating conditions, continuously optimizing the early warning threshold and fusion model, effectively reducing false alarms and missed alarms, and achieving closed-loop intelligent safety monitoring throughout the entire lifecycle. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a flowchart of an in-situ multi-parameter collaborative monitoring method for high-parameter equipment, provided as an embodiment of this application.

[0029] Figure 2 This is a flowchart of step A1 of an in-situ multi-parameter collaborative monitoring method for high-parameter equipment provided in an embodiment of this application.

[0030] Figure 3 This is a schematic diagram of multi-level fusion in an in-situ multi-parameter collaborative monitoring method for high-parameter equipment, provided in an embodiment of this application.

[0031] Figure 4 This is a schematic diagram of a hierarchical early warning and operation and maintenance strategy linkage mechanism in an in-situ multi-parameter collaborative monitoring method for high-parameter equipment, provided in an embodiment of this application.

[0032] Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0033] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0034] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0035] This application provides an in-situ multi-parameter collaborative monitoring method for high-parameter equipment. In one exemplary embodiment, such as... Figure 1 As shown, the method includes the following steps: A1. Measurement Point Determination Steps: Based on Failure Mode and Effects Analysis (FMEA), determine the key monitoring areas of the high-parameter equipment. This step aims to identify the parts of the equipment most prone to failure, providing a basis for the accurate deployment of subsequent sensors.

[0036] In a specific embodiment, such as Figure 2 As shown, the step of determining the measuring point further includes: Step A11: Based on failure mode analysis, analyze the potential failure modes of the high-parameter equipment. Analyze the failure modes that the equipment may experience under high temperature, high pressure and other conditions, such as weld cracking, stress corrosion in the transition zone of the end cap, leakage of the sealing surface and erosion thinning, etc.

[0037] Step A12: Assess the probability of occurrence, severity, and detectability of each potential failure mode, and determine the risk priority of each potential failure mode. For each identified failure mode, assess its probability of occurrence (O), severity (S), and detectability (D), and calculate the Risk Priority Number (RPN) using the formula RPN = O × S × D. The higher the RPN value, the greater the risk of that failure mode.

[0038] Step A13: Select the areas with the highest risk priority as the key monitoring areas. Based on the RPN value assessment results, select one or more areas with the highest risk priority and determine them as the key monitoring areas. The FMEA process will also output information such as the failure mechanism of the area and the recommended parameter types for monitoring, providing a technical basis for subsequent deployment plans.

[0039] A2. Data Acquisition Steps: Multi-source sensor data is synchronously acquired in key monitoring areas using a unified clock via a non-invasive method. Specifically, a passive composite sensor array is deployed on the outer wall of the equipment using non-invasive methods such as high-temperature resistant bonding or detachable clamps. This passive composite sensor array includes at least two of the following: fiber optic grating sensing units, high-temperature acoustic emission sensors, electromagnetic ultrasonic thickness probes, and distributed temperature sensors, forming a multi-physical quantity sensing network resistant to electromagnetic interference. Microsecond-level spatiotemporal synchronous acquisition is achieved through a unified hardware clock, obtaining multi-dimensional raw data such as shell temperature, surface strain, internal pressure, acoustic emission signals, and electromagnetic ultrasonic wall thickness / defect echo signals.

[0040] In one specific embodiment, the data acquisition step further includes adaptive discrimination of operating conditions to achieve automatic switching of the sampling frequency: Step A21: Calculate the fluctuation amplitude and rate of change of each physical parameter in real time. Specifically, based on the synchronously acquired shell temperature, surface strain, and internal pressure signals, calculate their amplitude fluctuations (such as the difference between the maximum and minimum values) and the rate of change per unit time in real time.

[0041] Step A22: Determine whether the steady-state operating conditions are met. When the fluctuation amplitude and rate of change of each physical parameter are both below the preset threshold and there is no obvious step change, it is determined to be a steady-state operating condition; otherwise, when any parameter of pressure, temperature, or strain experiences rapid rise or fall, step drift, or sudden distortion or jump in acoustic emission or electromagnetic ultrasonic echo signals, it is automatically identified as a transient operating condition.

[0042] Step A23: Switch the sampling frequency according to the operating condition. If the condition is determined to be steady state, use the low-frequency sampling mode; if the condition is determined to be transient, immediately switch to the high-frequency sampling mode to fully preserve the abnormal feature information.

[0043] In an exemplary embodiment, taking a high-temperature and high-pressure hydrogenation reactor as an example, fiber optic temperature / strain sensors, high-temperature acoustic emission sensors, and electromagnetic ultrasonic thickness probes are deployed in key areas such as the circumferential weld of the cylinder, the transition zone of the end cap, and the flange sealing surface using a high-temperature resistant bonding method to form a passive composite sensing network. Subsequently, microsecond-level synchronous acquisition is achieved through a unified clock to obtain multi-dimensional data on temperature, strain, pressure, acoustic emission, and wall thickness.

[0044] A3. Status assessment steps: Spatiotemporal alignment and multi-level fusion of multi-source sensor data are performed sequentially to generate a comprehensive index HI that can characterize the health status of the equipment.

[0045] First, the multi-source sensor data is spatiotemporally aligned. This includes denoising the multi-source sensor data (e.g., adaptive wavelet denoising), temperature drift compensation (baseline temperature drift compensation algorithm), and outlier removal (e.g., outlier removal based on statistical distribution). Then, the processed data is assigned a unified timestamp and spatial coordinates to construct a standardized multi-parameter time-series fusion dataset.

[0046] Secondly, multi-level fusion is performed on the spatiotemporally aligned multi-source sensor data. In a specific embodiment, such as... Figure 3 As shown, multi-level fusion includes data layer fusion, feature layer fusion, and decision layer fusion.

[0047] Data layer fusion: Fusion is performed by dynamically assigning weighting coefficients based on the signal-to-noise ratio (SNR) of each data source. Taking electromagnetic ultrasonic defect echo signals and acoustic emission waveform signals as examples, their SNRs are calculated separately using the following formulas: in, This represents the signal-to-noise ratio, measured in decibels (dB). Represents the signal amplitude. This represents the noise amplitude. Shell temperature, surface strain, internal pressure, acoustic emission characteristic parameters, electromagnetic ultrasonic wall thickness, and defect characteristic parameters are used as the fusion input data sources. Weights are dynamically assigned based on the real-time signal-to-noise ratio (SNR) of the two signals; the data source with a higher SNR receives a larger weight. Specifically, weighted summation or weighted averaging can be used for fusion to obtain preliminary fused data.

[0048] Feature layer fusion: Extracting time-frequency domain features and coupling correlation features representing the interrelationships between multiple parameters from the fused data. Time-frequency domain features include time-domain features (mean, variance, peak value, kurtosis, rate of change, fluctuation amplitude) and frequency-domain features (dominant frequency, spectral amplitude, band energy, power spectral density). Coupling correlation features focus on the linkage changes, correlations, temporal lags, and cooperative mutation patterns among five types of monitoring parameters: shell temperature, surface strain, internal pressure, acoustic emission, and electromagnetic ultrasound.

[0049] Decision-level fusion: Based on the aforementioned time-frequency domain features and coupled correlation features, the comprehensive index HI is calculated using an evidence reasoning model. This evidence reasoning model can employ either the DS evidence theory model or a deep learning model. The DS evidence theory model performs uncertainty fusion on multi-source evidence through a basic probability allocation function; the deep learning model (such as a deep neural network) learns the mapping relationship between defect types and health status based on historical feature data. The final output is a value between 0 and 1 as the comprehensive health index HI, which intuitively reflects the current overall health status of the device.

[0050] Taking the high-temperature and high-pressure hydrogenation reactor as an example, the collected signals are denoised, temperature drift compensated, and spatiotemporally aligned to construct a standardized multi-parameter time-series dataset. The comprehensive health index HI is calculated through three-level collaborative fusion. When HI continues to decline and strain and acoustic emission characteristics are synchronously abnormal, it is determined to be the initiation of microcracks, triggering a level-two warning.

[0051] A4. Early warning and optimization steps: Implement hierarchical early warning based on comprehensive index and dynamic criteria, and iteratively update the weights or dynamic criteria of multi-level fusion based on historical feedback data.

[0052] This embodiment implements a tiered early warning system. Based on equipment design specifications and industry standard values ​​as the fundamental threshold, and combined with historical data from long-term steady-state operation and real-time signal-to-noise ratio, dynamic criteria are adaptively adjusted to form a judgment. The degree of feature deviation is quantified by calculating the spatial distance, similarity, and correlation residuals between real-time features and the standard benchmark feature library. Combining the Health Index (HI) with the degree of feature deviation, four early warning levels—"Normal," "Attention," "Alarm," and "Emergency"—are established, simultaneously identifying defect types (corrosion, cracks, stress concentration, leaks, etc.) and their corresponding areas.

[0053] like Figure 4As shown, this embodiment links tiered early warning with operation and maintenance strategies: when the early warning level is "normal," a "continuous monitoring" strategy is matched, and monitoring is performed at the normal frequency; when the early warning level is "attention," a "encrypted monitoring" strategy is matched, increasing the data collection frequency and closely tracking the status trend; when the early warning level is "alarm," a "specialized inspection" strategy is matched, initiating a specialized inspection process (such as ultrasonic re-inspection) to verify the defect status; when the early warning level is "urgent," a "partial maintenance" and "shutdown" strategy is matched, prioritizing partial maintenance, and immediately shutting down the system if the risk continues to escalate. This forms a closed-loop process of "monitoring-diagnosis-early warning-response."

[0054] In a specific embodiment, the iterative update in step A4 specifically includes the following steps: Update fusion weights: Based on historical data throughout the entire lifecycle and operation and maintenance feedback, and with the early warning accuracy rate as the optimization objective, update the weights in the multi-level fusion. Specifically, based on historical monitoring data throughout the entire lifecycle, defect diagnosis results, and operation and maintenance handling feedback data, and with the early warning accuracy rate, defect identification accuracy, and false alarm / missed alarm rate as optimization objectives, recalculate the dynamic weighting coefficients of the data layer and the evidence confidence weights of the decision layer through an iterative optimization algorithm.

[0055] Update dynamic criterion thresholds: Based on the original design specifications of the equipment and its aging status, the thresholds of the dynamic criteria are adaptively adjusted. Specifically, by integrating previous early warning results and operation and maintenance verification feedback, and combining the equipment's operating time, aging status, and media corrosion and erosion patterns, the upper and lower limits of the dynamic criterion thresholds are adaptively adjusted by region and operating condition, eliminating abnormal threshold deviations caused by operating condition interference and noise interference.

[0056] Update the identification model: Incrementally train the identification model used in the early warning and optimization steps using the newly added data, and update the model parameters. In this embodiment, for the DS evidence theory model, the evidence trust function and basic probability allocation function are reallocated, and the evidence fusion rules are corrected; for the deep learning model, newly added historical monitoring feature data, defect annotation data, and operation and maintenance result data are included in the training dataset for incremental training and fine-tuning, continuously improving the accuracy of health index calculation and defect location identification.

[0057] Taking the high-temperature and high-pressure hydrogenation reactor as an example, the system automatically generates ultrasonic re-inspection work orders and updates the model weights and early warning thresholds simultaneously to achieve adaptive optimization.

[0058] The above embodiments of this application take a high-temperature and high-pressure hydrogenation reactor as an example. However, those skilled in the art will understand that the above methods of this application can also be widely extended to various high-parameter pressure-bearing equipment such as nuclear power, hydrogen energy, coal chemical industry, power, and long-distance pipelines to achieve intelligent, routine, and long-term safety monitoring.

[0059] By implementing steps A1 to A4 above, the method provided in this application uses FMEA (Failure Mode and Effects Analysis) to determine key measurement points, and deploys a passive composite sensor network in a non-invasive manner to achieve microsecond-level synchronous acquisition of multiple parameters such as temperature, strain, pressure, acoustic emission, and electromagnetic ultrasound. After signal preprocessing and spatiotemporal alignment, a comprehensive health index HI is constructed through a three-level collaborative fusion of "data layer fusion—feature layer fusion—decision layer fusion," achieving multi-parameter coupled intelligent diagnosis and four-level graded early warning. It also links with operation and maintenance strategies to form a closed loop, continuously iterating and optimizing the model using historical data. This application can achieve accurate early warning of equipment defects under extreme operating conditions without shutting down or disassembling the equipment. It has advantages such as strong anti-interference, high reliability, and wide adaptability, and is suitable for the full life-cycle safety monitoring of various high-parameter pressure-bearing equipment.

[0060] Based on the same inventive concept, this application also provides a system for implementing the above-described in-situ multi-parameter collaborative monitoring method for high-parameter equipment. The solution provided by this system is similar to the implementation described in the above method; therefore, the specific limitations in one or more system embodiments provided below can be found in the above-described limitations of the in-situ multi-parameter collaborative monitoring method for high-parameter equipment, and will not be repeated here.

[0061] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 5 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it can implement the in-situ multi-parameter collaborative monitoring method for high-parameter devices provided in the previous embodiment.

[0062] Figure 5The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0063] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0064] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0065] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0066] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with relevant regulations and be authorized by the owner of the corresponding device.

[0067] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0068] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0069] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0070] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for in-situ multi-parameter collaborative monitoring of high-parameter equipment, characterized in that, include: The measurement point determination step, based on failure mode analysis, determines the key monitoring area of ​​the high-parameter equipment; The data acquisition step involves using a non-invasive method to synchronously collect multi-source sensor data in the key monitoring area via a unified clock. The status assessment step involves sequentially performing spatiotemporal alignment and multi-level fusion on the multi-source sensor data to generate a comprehensive index that can characterize the health status of the device. The early warning and optimization steps involve performing tiered early warnings based on the comprehensive index and dynamic criteria, and iteratively updating the weights of the multi-level fusion or the dynamic criteria based on historical feedback data.

2. The in-situ multi-parameter collaborative monitoring method for high-parameter equipment according to claim 1, characterized in that, The steps for determining the measuring points specifically include: Based on failure mode analysis, the potential failure modes of the high-parameter equipment are analyzed; Assess the probability of occurrence, severity, and detectability of each potential failure mode, and determine the risk priority of each potential failure mode; Select one or more locations with the highest risk priority as the key monitoring areas.

3. The in-situ multi-parameter collaborative monitoring method for high-parameter equipment according to claim 1, characterized in that, include: In the data acquisition step, the non-invasive method includes deploying the data on the outer wall of the high-parameter equipment using high-temperature resistant adhesives or detachable clamps; the multi-source sensing data is acquired by a passive composite sensing array; the passive composite sensing array includes at least two of the following: fiber optic grating sensing units, high-temperature acoustic emission sensors, electromagnetic ultrasonic thickness probes, and distributed temperature sensors.

4. The in-situ multi-parameter collaborative monitoring method for high-parameter equipment according to claim 1, characterized in that, The data acquisition steps also include: The sampling frequency is automatically switched according to the steady-state or transient operating conditions of the high-parameter equipment; wherein, when the fluctuation amplitude and change rate of each physical parameter of the high-parameter equipment are both lower than the preset threshold, it is determined to be a steady-state operating condition, otherwise it is determined to be a transient operating condition.

5. The in-situ multi-parameter collaborative monitoring method for high-parameter equipment according to claim 1, characterized in that, The state assessment step involves spatiotemporal alignment of the multi-source sensor data, specifically including: The multi-source sensor data is subjected to noise reduction, temperature drift compensation, and outlier removal. Assign a unified timestamp and spatial coordinates to the processed multi-source sensor data.

6. The in-situ multi-parameter collaborative monitoring method for high-parameter equipment according to claim 1, characterized in that, The state assessment step involves multi-level fusion of spatiotemporally aligned multi-source sensor data, specifically including: Data layer fusion: Based on the signal-to-noise ratio of each data source, weighting coefficients are dynamically allocated, and spatiotemporally aligned multi-source sensor data are fused. Feature layer fusion: Extracting time-frequency domain features and coupling correlation features representing the linkage relationship between multiple parameters from the fused data; Decision-level fusion: Based on the time-frequency domain features and the coupling correlation features, the comprehensive index is calculated by combining the confidence weight of multi-evidence fusion through the evidence reasoning model.

7. The in-situ multi-parameter collaborative monitoring method for high-parameter equipment according to claim 6, characterized in that, The formula for calculating the signal-to-noise ratio is: in, For signal-to-noise ratio, The signal amplitude, The noise amplitude is used; the dynamic allocation of weighting coefficients is as follows: the data source with a higher signal-to-noise ratio is assigned a larger weight.

8. The in-situ multi-parameter collaborative monitoring method for high-parameter equipment according to claim 6, characterized in that, The evidence reasoning model used in the decision-making layer fusion is either the DS evidence theory model or a deep learning model.

9. The in-situ multi-parameter collaborative monitoring method for high-parameter equipment according to claim 1, characterized in that, The warning and optimization steps include four warning levels: "Normal," "Attention," "Alarm," and "Emergency." The method also includes: Based on the warning level, different operation and maintenance strategies are linked; among them, the "normal" level is matched with a continuous monitoring strategy, the "attention" level is matched with an encrypted monitoring strategy, the "alarm" level is matched with a special detection strategy, and the "emergency" level is matched with a partial maintenance and shutdown handling strategy.

10. The in-situ multi-parameter collaborative monitoring method for high-parameter equipment according to claim 1, characterized in that, The early warning and optimization steps involve iterative updates based on historical feedback data, specifically including: Based on historical data throughout the entire lifecycle and operation and maintenance feedback, with the goal of improving early warning accuracy, the weights in the multi-level fusion are updated. The threshold of the dynamic criterion is adaptively adjusted based on the original design specifications of the equipment and the degree of aging during operation. The identification model used in the early warning and optimization steps is incrementally trained using the newly added data, and the model parameters are updated.

Images