Disease diagnosis method for big wood component of ancient building and pseudo-classic building

Abstract

The invention relates to the technical field of building structure health monitoring, and discloses an ancient building and pseudo-classic building big wood component disease diagnosis method comprising the following steps: carrying out multi-modal baseline representation on a big wood component to calibrate a disease high-risk area; executing a preset environmental stress test protocol for the region; environment stress data and metabolic gas concentration serving as metabolic feedback data are synchronously collected; extracting instantaneous dynamic characteristics based on the collected data, and constructing a stress-feedback fingerprint representing a response mode of the instantaneous dynamic characteristics; training a disease development prediction model based on the fingerprints, and constructing a response surface model to invert an environmental critical point triggering disease activity; and finally, in combination with a prediction result and the critical point, generating a comprehensive diagnosis report and defining a safe operation envelope for guiding preventive maintenance. According to the invention, through active stress testing and metabolic fingerprint analysis, early quantitative diagnosis of diseases in the incubation period is realized, and an accurate quantitative basis is provided for preventive microenvironment regulation and control.

Description

Technical Field

[0001] This invention relates to the field of building structural health monitoring technology, and in particular to a method for diagnosing diseases of large timber components in ancient and imitation ancient buildings. Background Technology

[0002] The structural health of large wooden components in ancient and imitation ancient buildings, as core load-bearing parts, is of paramount importance. These wooden components are exposed to the natural environment for a long time and are highly susceptible to biological diseases (including fungal decay and termite infestation).

[0003] Existing diagnostic methods largely rely on periodic inspections, employing techniques primarily including visual inspection, drilling, and conventional non-destructive testing (NDT) using stress wave or micro-drilling resistance gauges. These methods have significant limitations in application. They primarily depend on identifying existing physical damage, including internal voids, decay, or a significant reduction in wood density, thus resulting in a severe lag in diagnostic results. For bioactivity in its early stages, latent phases, or dormant states, existing technologies struggle to effectively detect and assess it.

[0004] Furthermore, traditional testing results are mostly qualitative descriptions of damage or local physical parameters, lacking precise quantitative assessment methods for the level of disease biological activity, and even less capable of predicting the future development trend of diseases. This results in managers being unable to identify the key environmental thresholds for activating latent diseases in preventative maintenance practices, causing environmental protection measures to remain largely passive and experience-based, making it difficult to achieve proactive suppression based on precise data, and thus failing to effectively prevent early diseases from transforming into irreversible physical damage. Summary of the Invention

[0005] The purpose of this invention is to provide a method for diagnosing diseases of large timber components in ancient and imitation ancient buildings, which solves the problem that existing technologies lack the ability to make early quantitative diagnoses of latent diseases in large timber components and cannot define the critical point of their activation environment, leading to blind prevention and maintenance.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for diagnosing diseases of large timber components in ancient and imitation ancient buildings, characterized by comprising the following steps;

[0007] S1. Perform multimodal baseline characterization on large timber components placed in a controlled environment to obtain their physical state data, and identify at least one high-risk area for disease based on the physical state data.

[0008] S2. Design and execute a preset environmental stress test protocol for the at least one high-risk area of ​​the disease;

[0009] S3. During the execution of the environmental stress test protocol, environmental parameters are collected synchronously as environmental stress data, and the concentration of metabolic gas in at least one high-risk disease area is monitored in real time as metabolic feedback data.

[0010] S4. Based on the environmental stress data and the metabolic feedback data, extract instantaneous dynamic features and construct a stress-feedback fingerprint that characterizes the response pattern of the high-risk area under specific stress.

[0011] S5. Based on the physical state data and the stress-feedback fingerprint, train the disease development prediction model and construct a response surface model to fit the relationship between environmental parameters and biological activity, so as to invert the environmental critical point that triggers disease activity.

[0012] S6. Combining the disease development prediction model and the environmental critical point, generate a comprehensive diagnostic report and define a safe operating envelope to guide preventive maintenance.

[0013] Preferably, step S1 includes:

[0014] Non-destructive testing techniques were used to obtain the internal physical state data of the large timber components.

[0015] The internal physical state data is fused with the external geometric morphology data to construct a baseline digital twin model;

[0016] The voxels in the baseline digital twin model are quantified using a disease risk assessment function to identify high-risk areas for the disease.

[0017] Preferably, the step of marking the at least one high-risk area for disease in step S1 specifically includes:

[0018] The risk of voxels within the large timber components is quantified using a disease risk assessment function.

[0019] The disease risk assessment function is constructed based on wood density, local moisture content, and physical damage indicator functions, combined with their respective preset weighting coefficients, and is used to assess the disease risk value of the voxels.

[0020] Furthermore, the internal physical state data and external geometric morphology data are spatially registered and fused to construct a baseline digital twin model. This model comprises a set of three-dimensional voxel meshes, each voxel... Associated physical properties, including at least wood density. and local moisture content

[0021] Through a disease risk assessment function For the voxel Quantify the risks:

[0022] ;

[0023] In the formula, Represents a voxel in a three-dimensional voxel grid. voxels Disease risk assessment value, voxels wood density, voxels Local moisture content, This is a density risk function used to characterize the negative correlation between wood density and disease risk, i.e. The lower, The higher the function value, This is a moisture content risk function used to characterize the positive correlation between local moisture content and disease risk, i.e. The higher, The higher the function value, This is a physical damage indicator function used to mark voxels with existing cavities or decay areas obtained from S104. If v is located in the damaged area, The value is 1 otherwise the value is 0. The preset weighting coefficients for the three risk factors mentioned above are determined based on expert experience or historical data.

[0024] Preferably, the environmental stress testing protocol in step S2 includes a humidity surge protocol or a temperature and humidity cycle protocol;

[0025] The humidity surge protocol is defined as increasing the relative humidity from a preset baseline value to a preset peak value, characterized by the rate of humidity change.

[0026] The temperature and humidity cycle protocol is defined as setting up a cycle unit that includes a high temperature and low humidity stage and a low temperature and high humidity stage, and repeatedly executing the cycle unit.

[0027] Furthermore, the environmental stress testing protocol includes a humidity surge protocol (…). The protocol Defined as within the time interval Inside, relative humidity From a preset baseline value Increase to a preset peak value The protocol A key parameter is the rate of change of humidity. ;

[0028] ;

[0029] In the formula, Indicates the rate of change of humidity. Indicates peak humidity. Indicates the humidity reference value. This indicates the point in time when humidity reaches its peak. This indicates the time point corresponding to the humidity baseline value.

[0030] Preferably, the metabolic feedback data in step S3 includes at least carbon dioxide concentration; the synchronous acquisition is achieved by assigning a unified timestamp to the environmental stress data and the metabolic feedback data acquired within the same sampling period.

[0031] Preferably, the instantaneous dynamic features in step S4 include at least:

[0032] Response delay is the time difference between the moment when the environmental stress data begins to change significantly and the moment when the metabolic feedback data begins to show an observable response; peak concentration is the maximum concentration value reached by the metabolic feedback data within the execution time interval of the environmental stress test protocol.

[0033] Peak response rate is the maximum rate at which the concentration of the metabolic feedback data changes over time; total metabolic release is the cumulative integral of the metabolic feedback data relative to the baseline concentration.

[0034] Furthermore, the instantaneous dynamic features are used to quantify the high-risk areas of the disease. Internal biological activity on the environmental stress testing protocol Parameters of the response mode. The instantaneous dynamic characteristics include at least:

[0035] Response delay From the environmental stress data The point at which significant changes begin to occur, up to the point in the metabolic feedback data. The time difference between the moments when an observable response begins to occur;

[0036] Peak concentration The metabolic feedback data In the protocol The maximum concentration value reached within the execution time interval;

[0037] Peak response rate Metabolic feedback data Concentration over time The maximum rate of change;

[0038] Total metabolic release Metabolic feedback data (by (For example) in the protocol Execution time interval (from arrive Within, relative to a baseline concentration The cumulative integral value;

[0039] :

[0040] In the formula, For the first High-risk areas in the agreement Total metabolic release. For this region in time of concentration, Baseline Concentration (pre-stress steady-state concentration). For the agreement The start and end times, For area code, This is the test protocol number.

[0041] Preferably, in step S4, the stress-feedback fingerprint is a feature vector composed of the instantaneous dynamic features, which include response delay, peak concentration, peak response rate, and total metabolic release.

[0042] Preferably, the step of training the disease development prediction model in step S5 includes:

[0043] Construct a fusion input feature vector, which includes static features derived from the physical state data, dynamic features derived from the stress-feedback fingerprint, and environmental features derived from the environmental stress test protocol;

[0044] The disease development prediction model is trained using a machine learning model to establish the relationship between the fused input feature vector and the future development status of the high-risk area of ​​the disease.

[0045] Preferably, the step of reversing the environmental critical point that triggers disease activity in step S5 includes:

[0046] The response surface model is constructed to fit the relationship between the environmental parameters and a key bioactivity indicator.

[0047] The inflection point of the response curve is determined by mathematical analysis of the derivative of the response surface model, and the inflection point is defined as the environmental critical point.

[0048] Preferably, in step S6, the safe operating envelope is defined as a set of environmental parameters, wherein the environmental parameters in the set cause the output value of the response surface model to be no greater than a preset upper limit of biological activity safety.

[0049] In summary, the present invention has at least one of the following beneficial technical effects:

[0050] 1. This invention, by applying a preset environmental stress testing protocol and monitoring the concentration of metabolic gases in real time as metabolic feedback data, can actively stimulate and quantify the instantaneous responses of biological pathogens in a dormant or low-activity state. This overcomes the limitation of traditional static detection methods, which can only identify physical damage that has already occurred, and achieves early warning before structural damage occurs.

[0051] 2. This invention constructs a stress-feedback fingerprint by extracting instantaneous dynamic features such as response delay, peak concentration, and total metabolic release, transforming biological activity from a qualitative judgment into a quantifiable feature vector. Furthermore, this fingerprint is used as a key dynamic feature to train a disease development prediction model, which can predict the future development status of high-risk disease areas and provide data support for graded maintenance.

[0052] 3. This invention constructs a response surface model to fit the relationship between environmental parameters and biological activity, and reversely calculates the environmental critical point that triggers disease activity. Based on the safe operating envelope defined by this critical point, it provides precise parameter boundaries for the microenvironmental regulation of controllable environments, enabling preventive maintenance to transform from a passive response to a threshold-based active inhibition, effectively preventing disease activation. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0054] Figure 2 This is a schematic diagram illustrating the baseline characterization and high-risk disease area marking of the present invention;

[0055] Figure 3 This is a schematic diagram of the stress-feedback fingerprint extraction process of the present invention;

[0056] Figure 4 This is a schematic diagram illustrating the safe operation envelope definition of the present invention. Detailed Implementation

[0057] The following is in conjunction with the appendix Figure 1 -Appendix Figure 4 The present invention will be further described in detail below.

[0058] This invention provides a method for diagnosing diseases of large timber components in ancient and imitation ancient buildings, comprising the following steps:

[0059] S1, perform multimodal baseline characterization on large timber components placed in a controlled environment to obtain their physical state data, and identify at least one high-risk area for disease based on the physical state data;

[0060] Specifically, in step S1 of this embodiment, multimodal baseline characterization is performed on the large wooden components placed in a controlled environment to obtain their physical state data, and at least one high-risk area for disease is identified based on the physical state data.

[0061] In one specific implementation, S1 may include the following sub-steps;

[0062] S101, The large timber component to be tested is placed in a controlled environment. Specifically, the controlled environment is a customizable climate chamber, which has the function of regulating its internal environmental parameters, including at least the ambient temperature. and relative humidity .

[0063] S102, Perform multimodal baseline characterization on the large timber components to obtain their physical state data. Multimodal baseline characterization is a non-destructive data acquisition process, and the physical state data includes the external geometric morphology data and internal physical state data of the components.

[0064] S103, Acquire external geometric data. Specifically, 3D laser scanning technology is used to scan the surface of the large timber component, generating a high-precision external point cloud model characterizing its external shape, dimensional parameters, and surface texture, denoted as... .

[0065] S104, Obtain Internal Physical State Data. Internal physical state data describes the physical properties of the component's internal structure. This step uses non-destructive testing techniques to obtain this data. In one embodiment, the non-destructive testing technology includes X-ray computed tomography (X-CT) and ultrasonic testing. X-CT is used to obtain information on the density distribution, voids, cracks, and decayed areas inside the component; ultrasonic testing is used to obtain information on the elastic modulus distribution or sound velocity attenuation of the component, and to assist in calibrating the local moisture content.

[0066] S105, Construct the baseline digital twin model of the large timber components. This step uses the external point cloud model obtained in S103. Internal physical state data acquired with S104 Spatial registration and data fusion are performed. Internal physical state data. It is processed into a three-dimensional voxel mesh, where each voxel in the three-dimensional voxel mesh... Each is associated with a unique set of spatial coordinates and a set of physical properties, including at least wood density. and local moisture content The fused model is the baseline digital twin model. .

[0067] S106, Based on physical condition data, at least one high-risk area for disease is identified. This step utilizes a baseline digital twin model. The physical attributes stored in the data are assessed through a disease risk assessment function. For each voxel within the large timber component Risk quantification. Disease risk assessment function. The construction principle is that areas with low wood density, high local moisture content, or existing physical damage are at higher risk of developing biological diseases (such as fungal decay or termite infestation).

[0068] In a preferred embodiment, the disease risk assessment function Defined as:

[0069] :

[0070] In the formula, Represents a voxel in a three-dimensional voxel grid. voxels Disease risk assessment value, voxels wood density, voxels Local moisture content, This is a density risk function used to characterize the negative correlation between wood density and disease risk, i.e. The lower, The higher the function value, This is a moisture content risk function used to characterize the positive correlation between local moisture content and disease risk, i.e. The higher, The higher the function value, This is a physical damage indicator function used to mark voxels with existing cavities or decay areas obtained from S104. If v is located in the damaged area, The value is 1 otherwise the value is 0. The preset weighting coefficients for the three risk factors mentioned above are determined based on expert experience or historical data.

[0071] S107, identifies high-risk areas for disease. A preset risk threshold is set. Traversing the baseline digital twin model All voxels in the set of conditions The voxels were selected. A spatially connected set of voxels that meets the above conditions is defined as an independent high-risk disease zone (HRZ) and uniquely numbered as follows: (Total number of high-risk areas identified). High-risk areas for disease. This will serve as the target area for in situ metabolic monitoring in subsequent steps.

[0072] S2, for at least one high-risk area for disease, design and execute a pre-defined environmental stress test protocol;

[0073] Specifically, step S2 in this embodiment may include the following sub-steps:

[0074] S201, Design of Environmental Stress Testing Protocol. The design principle of the environmental stress testing protocol is to stimulate high-risk areas of disease by actively applying non-steady-state changes in environmental parameters. The protocol simulates the metabolic activity of potentially dormant or inactive pathogenic organisms (such as fungi and termites) within the timber structure. It also aims to simulate high-risk weather events that large timber components might encounter during actual service.

[0075] S202, the design basis of the environmental stress testing protocol includes historical meteorological data of the location of large timber components, especially environmental patterns that are strongly correlated with the occurrence of wood decay or insect infestation, such as persistent high humidity during the rainy season, a sharp rise in humidity after heavy rain, or a day-night cycle of high temperature and high humidity.

[0076] S203 defines the environmental stress test protocol as one or more preset environmental parameter change trajectories. ,in The protocol number ( , (Total number of protocols). Environmental parameter change trajectory. It is a time series vector, which includes at least the ambient temperature. and relative humidity Over time The path of change.

[0077] S204 provides a specific embodiment of an environmental stress testing protocol, namely a humidity surge protocol. Humidity surge agreement This is used to simulate scenarios where components encounter sudden high humidity (such as heavy rain or leaks). Protocol Specifically defined as: within the time interval Inside, the relative humidity in the controlled environment From a preset baseline value To boost linearly or non-linearly to a preset peak value .

[0078] S205, Humidity Sudden Increase Protocol A key parameter is the rate of change of environmental parameters, specifically the rate of change of humidity. For a linear increase, the rate of humidity change... The calculation method is as follows:

[0079] ;

[0080] In the formula, Indicates the rate of change of humidity. Indicates peak humidity. Indicates the humidity reference value. This indicates the point in time when humidity reaches its peak. This indicates the time point corresponding to the humidity baseline value.

[0081] Humidity change rate It is itself a key stress factor used to test the sensitivity of diseased organisms to the rate of environmental change.

[0082] S206 provides a second specific embodiment of the environmental stress testing protocol, namely the temperature and humidity cycling protocol. Temperature and humidity cycle protocol Used to simulate environments with day-night temperature differences or alternating wet and dry conditions. Protocol Specifically, it is defined as: setting up a loop unit that includes a high temperature and low humidity stage and a low temperature and high humidity stage, and repeatedly executing the loop unit. Second-rate( .protocol Used to test the excitation effect of temperature and humidity coupling on biological activity and the cumulative effect under cyclic stress.

[0083] S207 provides a third specific embodiment of the environmental stress testing protocol, namely the gradient constant protocol. Gradient constancy protocol Environmental parameter thresholds used to detect the activation of biological activity.

[0084] protocol Specifically defined as: taking environmental parameters (such as relative humidity) Increase to the first setting value And maintain a constant duration Then, it is raised to the second setting value. And maintain the same constant duration Repeat this process. There are several gradients.

[0085] S208, Execute the environmental stress test protocol. In one embodiment, the data processing and analysis center of the diagnostic system, according to the selected protocol... and the corresponding environmental parameter change trajectory The system sends commands to the control unit of the controlled environment (i.e., the customizable climate chamber in S101). The control unit automatically adjusts its internal temperature and humidity regulation components to ensure that the actual environmental parameters within the controlled environment strictly follow the preset environmental parameter change trajectory. run.

[0086] S3, during the execution of the environmental stress test protocol, simultaneously collects environmental parameters as environmental stress data, and monitors the metabolic gas concentration in at least one high-risk disease area in real time as metabolic feedback data.

[0087] Specifically, step S3 in this embodiment includes the following sub-steps;

[0088] S301, the environmental stress testing protocol defined in S200 The entire execution time interval (from) arrive Within this system, the data processing and analysis center of the diagnostic system operates on a fixed sampling cycle. Continuously collect data.

[0089] S302, Collecting environmental stress data Environmental stress data It refers to real-time measurements of actual environmental parameters within a controlled environment. The data originates from environmental sensors deployed within the controlled environment and includes environmental stress data. At least ambient temperature and relative humidity .

[0090] S303, collects metabolic feedback data. Metabolic feedback data This refers to the high-risk areas for diseases identified by S107. Real-time measurements of internal metabolic gas concentrations, data sourced from areas connected to high-risk disease zones. High-sensitivity gas analyzer with in-situ gas monitoring probe, metabolic feedback data At least carbon dioxide concentration .

[0091] S304, achieving simultaneous acquisition of environmental stress data and metabolic feedback data; the specific implementation method of simultaneous acquisition is: the data processing and analysis center performs data acquisition within the same sampling period. All data collected internally (including environmental stress data) and all Metabolic feedback data of high-risk areas for individual diseases Assign a uniform timestamp ,in , This is the sampling sequence number.

[0092] S305, Construct a synchronized dataset ;exist At any given moment, environmental stress data With all Metabolic feedback data of high-risk areas for individual diseases Combined into a single data record point And store in the synchronized dataset. Middle; data recording point Defined as;

[0093] :

[0094] In the formula, For the first The name or identifier of a synchronized dataset. For the first A synchronized dataset at time [time] A single data record point, For time indexing, the first Each sampling time or observation time point In time Collected environmental stress data, They represent time. Corresponding to the To the Metabolic feedback data measured in high-risk areas of disease. This represents the total number of high-risk areas for disease. Index the dataset. Time sampling index.

[0095] S306, fixed sampling period The setup is designed to achieve high-frequency data acquisition, the purpose of which is to capture the instantaneous dynamic response of pathogenic biological activity to environmental stress; a fixed sampling period. The value is less than the characteristic time of the metabolic response of the disease organism, which is the shortest expected time from the start of changes in environmental parameters to the start of significant changes in metabolic feedback data.

[0096] S4, based on environmental stress data and metabolic feedback data, extracts instantaneous dynamic features and constructs a stress-feedback fingerprint that characterizes the response pattern of high-risk areas under specific stress.

[0097] Specifically, step S4 in this embodiment includes the following sub-steps:

[0098] S401, regarding the synchronized dataset obtained in S300 The analysis will be conducted. The core of the analysis is to establish environmental stress data. (As input) and metabolic feedback data The temporal response relationship between (as outputs) is quantified by extracting a set of instantaneous dynamic features.

[0099] S403, Extract instantaneous dynamic features, which include at least: response delay. Response delay This refers to data from environmental stress. (e.g., relative humidity) From the moment when significant changes begin to occur (i.e., the onset of stress) to the point when metabolic feedback data... (e.g., carbon dioxide concentration) The time difference between the moment when an observable response begins to occur.

[0100] S404, Extract instantaneous dynamic features, which include at least: peak concentration. Peak concentration This refers to the environmental stress testing protocol. Metabolic feedback data within the execution time interval The maximum concentration value reached.

[0101] S405, Extract instantaneous dynamic features, which include at least: peak response rate. Peak response rate This refers to the environmental stress testing protocol. Metabolic feedback data within the execution time interval Concentration over time The maximum rate of change (i.e., the maximum slope of the time series curve).

[0102] S406, Extract transient dynamic features, which include at least: total metabolic release. Total metabolic release This refers to the environmental stress testing protocol. Execution time interval (from arrive Internal metabolic feedback data (by (For example) relative to a baseline concentration The cumulative integral value. Baseline concentration. To execute the agreement Previous steady-state gas concentration. Total metabolic release. The calculation formula is:

[0103] :

[0104] In the formula, For the first High-risk areas in the agreement Total metabolic release. For this region in time of concentration, Baseline Concentration (pre-stress steady-state concentration). For the agreement The start and end times, For area code, This is the test protocol number.

[0105] S407, constructing a stress-feedback fingerprint. A stress-feedback fingerprint is a unique method for characterizing high-risk areas for disease. Environmental stress testing protocol Feature vectors of the response mode eigenvectors It is composed of at least one instantaneous dynamic feature extracted from S403 to S406. In a preferred embodiment, the feature vector... Defined as;

[0106] ;

[0107] In the formula, For a unique representation region In the agreement Feature vectors of the response mode To respond to latency; from Apply ( The time difference between when the metabolic / signal activity in that region first reaches a preset threshold or significantly deviates from the baseline. Peak concentration / amplitude The peak rate of change, This represents the cumulative release amount divided by the total response amount.

[0108] S408, Constructing a stress-feedback fingerprint database Stress-Feedback Fingerprint Database It is a structured database used to store feature vectors constructed in S407.

[0109] S409, in the stress-feedback fingerprint database In the middle, each stored feature vector Each is associated with a set of metadata. The metadata includes at least: high-risk areas for disease. The identifier and its physical state data acquired in S100 (such as...) ), and environmental stress testing protocols The identifier and its key parameters (such as) Stress-Feedback Fingerprint Database It serves as the data foundation for model training and critical point inversion in S500.

[0110] S5, based on physical state data and stress-feedback fingerprints, trains a disease development prediction model and constructs a response surface model to fit the relationship between environmental parameters and biological activity, so as to invert the environmental critical point that triggers disease activity.

[0111] Specifically, step S5 in this embodiment may include the following sub-steps:

[0112] S501 utilizes the stress-feedback fingerprint database built in S400. From the stress-feedback fingerprint database Extract data from the dataset and construct a fusion input feature vector for model training.

[0113] S502, Fusion of input feature vectors At least including:

[0114] Static features The static features are derived from the physical state data obtained by S1, specifically the high-risk areas for disease. Wood density at the location and local moisture content ;

[0115] Dynamic features The dynamic features are derived from the stress-feedback fingerprint constructed by S4, specifically as feature vectors. (including response delay) Peak concentration and total metabolic release );

[0116] Environmental characteristics The environmental characteristics are derived from the S2 environmental stress testing protocol. The key parameter is the peak humidity. and rate of change of humidity .

[0117] S503, Training a Disease Development Prediction Model Disease development prediction model It is a machine learning model, which in a non-limiting embodiment is a gradient boosting tree model or a long short-term memory network model.

[0118] S504, Disease Development Prediction Model The fused input feature vector constructed in S502 As model input, the model input integrates the static physical properties of the components with the dynamic bioactivity response characteristics obtained by S4.

[0119] S505, Disease Development Prediction Model Model output High-risk areas for disease The future development status. The future development status includes the disease type, disease activity level (e.g., dormant, low activity, high activity level), or structural damage rate.

[0120] S506, Constructing a response surface model Response surface model It is an independent model, constructed to invert high-risk areas that trigger diseases. Environmental critical point of biological activity Response surface model It is specifically designed to fit nonlinear causal relationships between environmental parameters (inputs) and bioactivity indicators (outputs).

[0121] S507, Response Surface Model The model input consists of environmental parameters defined in S200, which include at least the ambient temperature. and relative humidity Response surface model The model output is a key bioactivity indicator, which in a preferred embodiment is the total metabolic release as defined in S406. .

[0122] S508, Response Surface Model One specific implementation method is as follows:

[0123] ;

[0124] In the formula, For the first Total metabolic release in high-risk areas for disease. For the region The specific function model obtained through training, For ambient humidity, For ambient temperature, The rate of change of humidity.

[0125] in, It utilizes a stress-feedback fingerprint database The data in the middle targets high-risk areas for diseases. A specific function model obtained through training.

[0126] S509, Inversion Environment Critical Point Environmental critical point Defined as making the response surface model Output (i.e., total metabolic release) ) Parameter points where the rate of change of the input (i.e., environmental parameters) changes significantly.

[0127] S510, the significantly changed parameter point, physically corresponds to the environmental threshold for the transition of disease biological activity from a dormant state to an activated state.

[0128] S511, Environmental Critical Point One specific solution method is to use mathematical analysis of the response surface model. The derivative is used to determine this. For example, the critical humidity threshold. (Environmental critical point) One component can be solved by solving the model. right The point where the absolute value of the second derivative is the largest is used to determine the inflection point of the response curve.

[0129] ;

[0130] In the formula, This is the critical humidity threshold. For environmental humidity variables, For the region defined in the preceding formula S508 Response surface model, For the model The second partial derivative with respect to humidity, This indicates the value of the independent variable that makes the expression within the parentheses reach its maximum value.

[0131] S6 combines disease development prediction models and environmental critical points to generate a comprehensive diagnostic report and defines a safe operating envelope to guide preventive maintenance.

[0132] Specifically, step S6 in this embodiment may include the following sub-steps;

[0133] S601, Generate a comprehensive diagnostic report. The comprehensive diagnostic report is a comprehensive assessment and prediction of the disease status of large timber components.

[0134] S602, the comprehensive diagnostic report includes: a disease status assessment. This assessment is based on the physical condition data obtained in S1 (used to identify high-risk areas for disease). (physical location and spatial extent), and combined with the stress-feedback fingerprint constructed by S4. (Used to quantify high-risk areas for disease) (Current level of bioactivity).

[0135] S603, the comprehensive diagnostic report includes: disease development trend prediction. This disease development trend prediction is based on the disease development prediction model trained in S5. The output, the output High-risk areas for disease were predicted. The future types of disease or the rate of structural damage.

[0136] S604, the comprehensive diagnostic report includes: identification of key environmental triggers. The identification of key environmental triggers is based on the environmental critical points obtained from the inversion in S5. (Including critical humidity thresholds) and critical temperature threshold ), environmental critical point It clearly points out the need to activate high-risk areas for disease. Environmental parameter thresholds for biological activity.

[0137] S605 defines the Safe Operating Envelope (SOE). The Safe Operating Envelope is the environmental critical point derived from S500. In conjunction with a preset upper limit for bioactivity safety. To be defined together.

[0138] S605 defines the Safe Operating Envelope (SOE). The Safe Operating Envelope is the environmental critical point derived from S500. In conjunction with a preset upper limit for bioactivity safety. To be defined together.

[0139] S606, the safe operating envelope is a set of environmental parameters (including ambient temperature). and relative humidity This constitutes a multidimensional parameter space region. Any combination of environmental parameters falling within the safe operating envelope will not affect the response surface model constructed in S500. Output (i.e., total metabolic release) Exceeding the preset safety limit for bioactivity .

[0140] S607, the Safe Operating Envelope (SOE) is mathematically defined as an environmental parameter that satisfies the following conditions. The set of:

[0141] ;

[0142] In the formula, Indicates the safe operating envelope. Represents an environment parameter vector. Indicates that for the first High-risk areas for disease The trained response surface model, Indicates the upper limit of biological activity safety. This condition must apply to all Several high-risk disease areas were established simultaneously. Indicates that for the first High-risk areas for disease The response surface model obtained through training.

[0143] In S608, the Safe Operating Envelope (SOE) is ultimately output as a quantitative basis for guiding preventive maintenance of large timber components. Specifically, the SOE is used to guide the microenvironmental regulation of the controllable environment (customizable climate chamber) in S101, ensuring that the actual service environment parameters of the large timber components are always kept within the SOE, thereby achieving proactive inhibition of disease biological activity and proactive prevention of disease development.

Claims

1. A method for diagnosing diseases of large timber components in ancient and imitation ancient buildings, characterized in that, Includes the following steps; S1. Perform multimodal baseline characterization on large timber components placed in a controlled environment to obtain their physical state data, and identify at least one high-risk area for disease based on the physical state data. S2. Design and execute a preset environmental stress test protocol for the at least one high-risk area of ​​the disease; S3. During the execution of the environmental stress test protocol, environmental parameters are collected synchronously as environmental stress data, and the concentration of metabolic gas in at least one high-risk disease area is monitored in real time as metabolic feedback data. S4. Based on the environmental stress data and the metabolic feedback data, extract instantaneous dynamic features and construct a stress-feedback fingerprint that characterizes the response pattern of the high-risk area under specific stress. S5. Based on the physical state data and the stress-feedback fingerprint, train the disease development prediction model and construct a response surface model to fit the relationship between environmental parameters and biological activity, so as to invert the environmental critical point that triggers disease activity. S6. Combining the disease development prediction model and the environmental critical point, generate a comprehensive diagnostic report and define a safe operating envelope to guide preventive maintenance.

2. The method for diagnosing diseases of large timber components in ancient and imitation ancient buildings according to claim 1, characterized in that, Step S1 includes: Non-destructive testing techniques were used to obtain the internal physical state data of the large timber components. The internal physical state data is fused with the external geometric morphology data to construct a baseline digital twin model; The voxels in the baseline digital twin model are quantified using a disease risk assessment function to identify high-risk areas for the disease.

3. The method for diagnosing diseases of large timber components in ancient and imitation ancient buildings according to claim 2, characterized in that, The step of marking the at least one high-risk area for disease in step S1 specifically includes: The risk of voxels within the large timber components is quantified using a disease risk assessment function. The disease risk assessment function is constructed based on wood density, local moisture content, and physical damage indicator functions, combined with their respective preset weighting coefficients, and is used to assess the disease risk value of the voxels.

4. The method for diagnosing diseases of large timber components in ancient and imitation ancient buildings according to claim 1, characterized in that, The environmental stress testing protocol in step S2 includes a humidity surge protocol or a temperature and humidity cycle protocol. The humidity surge protocol is defined as increasing the relative humidity from a preset baseline value to a preset peak value, characterized by the rate of humidity change. The temperature and humidity cycle protocol is defined as setting up a cycle unit that includes a high temperature and low humidity stage and a low temperature and high humidity stage, and repeatedly executing the cycle unit.

5. The method for diagnosing diseases of large timber components in ancient and imitation ancient buildings according to claim 1, characterized in that, The metabolic feedback data in step S3 includes at least carbon dioxide concentration; the synchronous acquisition is achieved by assigning a unified timestamp to the environmental stress data and the metabolic feedback data collected within the same sampling period.

6. The method for diagnosing diseases of large timber components in ancient and imitation ancient buildings according to claim 1, characterized in that, The instantaneous dynamic features in step S4 include at least the following: Response delay is the time difference between the moment when the environmental stress data begins to change significantly and the moment when the metabolic feedback data begins to show an observable response; peak concentration is the maximum concentration value reached by the metabolic feedback data within the execution time interval of the environmental stress test protocol. Peak response rate is the maximum rate at which the concentration of the metabolic feedback data changes over time; total metabolic release is the cumulative integral of the metabolic feedback data relative to the baseline concentration.

7. The method for diagnosing diseases of large timber components in ancient and imitation ancient buildings according to claim 1, characterized in that, In step S4, the stress-feedback fingerprint is a feature vector composed of the instantaneous dynamic features, which include response delay, peak concentration, peak response rate, and total metabolic release.

8. The method for diagnosing diseases of large timber components in ancient and imitation ancient buildings according to claim 1, characterized in that, The step of training the disease development prediction model in step S5 includes: Construct a fusion input feature vector, which includes static features derived from the physical state data, dynamic features derived from the stress-feedback fingerprint, and environmental features derived from the environmental stress test protocol; The disease development prediction model is trained using a machine learning model to establish the relationship between the fused input feature vector and the future development status of the high-risk area of ​​the disease.

9. A method for diagnosing diseases of large timber components in ancient and imitation ancient buildings according to claim 1, characterized in that, The step S5, which involves reversing the environmental critical point that triggers disease activity, includes: The response surface model is constructed to fit the relationship between the environmental parameters and a key bioactivity indicator. The inflection point of the response curve is determined by mathematical analysis of the derivative of the response surface model, and the inflection point is defined as the environmental critical point.

10. A method for diagnosing diseases of large timber components in ancient and imitation ancient buildings according to claim 1, characterized in that, In step S6, the safe operating envelope is defined as a set of environmental parameters, wherein the environmental parameters in the set ensure that the output value of the response surface model is not greater than a preset biological activity safety limit.

Images