Timber building fire hazard and structure instability comprehensive linkage alarm system

Abstract

The application discloses a comprehensive linkage alarm system for fire hazards and structure instability of timber buildings, and relates to the technical field of timber safety monitoring, which comprises the following modules: a multi-dimensional heterogeneous signal acquisition module for collecting data and constructing real-time state monitoring data sets; a structure spectrum benchmark modeling module for constructing a thermal-rigidity attenuation theoretical model; a thermal-frequency correlation analysis module for generating a thermal-frequency deviation degree index; a structure state determination module for identifying timber pyrolysis states and structure instability risks; a cascade response module for executing corresponding fixed-point spraying control, evacuation path planning and alarm output; and a self-adaptive feedback correction module for warehousing data and correcting thermal aging parameters. The application establishes a correlation model of thermal and rigidity by fusing thermal imaging and vibration data, effectively eliminates environmental wind load noise, realizes accurate identification of internal hazards of timber buildings, and improves the accuracy of building monitoring and the efficiency of emergency response.

Description

Technical Field

[0001] This invention relates to the field of timber structure safety monitoring technology, specifically to a comprehensive alarm system for fire hazards and structural instability in timber buildings. Background Technology

[0002] As a traditional architectural form, timber-framed buildings occupy an important position in the protection of ancient buildings and modern green buildings. Timber-framed buildings are mainly composed of wooden components such as beams and columns, which are connected by mortise and tenon joints. Their load-bearing capacity and fire resistance are directly related to the overall lifespan and safety of the building. In order to ensure the safety and durability of the building, it is usually necessary to carry out fire prevention and daily inspection and monitoring of the structural health of timber-framed buildings during daily use and maintenance.

[0003] To ensure safety, it is usually necessary to install several sensors around key nodes, beams and columns of wooden buildings. These sensors monitor the condition of the wooden building in real time. The purpose of the monitoring is usually to ensure the safety of the main structure. Under the influence of the external environment, wooden buildings are prone to carbonization of components, degradation of material properties, and reduction of structural load-bearing capacity. Existing technologies include acquiring monitoring data of wooden buildings and classifying safety based on the monitoring data.

[0004] However, in reality, the test data contains significant noise and some data that has little impact on the structural condition. For example, wind loads or occasional vibrations in the external environment can cause vibration signals in the structure, and simple surface temperature monitoring is unlikely to detect slow-moving hidden dangers inside. If safety classification is directly based on the current test data, the safety classification may be inaccurate, leading to significant safety hazards. Summary of the Invention

[0005] The purpose of this invention is to provide a comprehensive alarm system for fire hazards and structural instability in wooden buildings, thus solving the problems existing in the background technology.

[0006] To address the aforementioned technical problems, this invention provides a comprehensive alarm system for fire hazards and structural instability in wooden buildings, comprising: a multi-dimensional heterogeneous signal acquisition module configured to acquire surface thermal imaging temperature field data, structural vibration response data, and environmental parameters of the target wooden building, and align timestamps to construct a real-time status monitoring dataset;

[0007] Structural Spectrum Benchmark Modeling Module: Configured to establish a structural spectrum benchmark based on modal scan data, and combine wood pyrolysis kinetic parameters to construct a thermal-stiffness decay theoretical model describing the nonlinear mapping relationship between temperature variables and elastic modulus;

[0008] Thermo-frequency correlation analysis module: configured to calculate the correlation characteristics between the rate of change of temperature over time and the rate of change of natural frequency, and generate a thermo-frequency deviation index;

[0009] Structural state determination module: configured to identify the pyrolysis state of wood and structural instability risk based on the residual analysis results of the thermal-frequency deviation index and the thermal-stiffness decay theoretical model;

[0010] The graded linkage response module is configured to execute corresponding fixed-point sprinkler control, evacuation route planning, and alarm output based on the identified status level.

[0011] Adaptive feedback correction module: configured to store temperature change data and frequency change data during the event process into a database to correct the thermal aging parameters in the thermal-stiffness decay theoretical model.

[0012] Preferably, the modules are implemented using the following method:

[0013] S1. Collect surface thermal imaging temperature field data, structural vibration response data and environmental parameters of the target wooden building to construct a real-time condition monitoring dataset;

[0014] S2. Perform modal scanning on the target wooden structure, establish a structural spectrum baseline at room temperature, and pre-define the thermal-stiffness attenuation theoretical model to define the functional relationship between temperature change and elastic modulus attenuation.

[0015] S3. Real-time monitoring of surface thermal imaging temperature field data and structural vibration response data. When the local temperature exceeds the preset safety value or the natural frequency shows non-periodic decay, the high-frequency sampling mode is activated.

[0016] S4. In high-frequency sampling mode, extract temperature gradient features and frequency change acceleration features, calculate thermal-frequency deviation, and input the thermal-frequency deviation into the thermal-stiffness attenuation theoretical model for residual analysis to generate confidence evaluation results of the impact of fire source on structure.

[0017] S5. Based on the confidence evaluation results of the impact of the fire source on the structure and the attenuation of the current structural frequency relative to the baseline, determine the fire hazard level and the structural instability time window, and output graded alarm commands and control signals.

[0018] S6. In response to the graded alarm command, execute the corresponding physical intervention measures, and feed back the heterogeneous data of this monitoring cycle to the structural spectrum benchmark modeling module to update the thermal aging parameters.

[0019] Preferably, S1 specifically includes:

[0020] Infrared thermal imaging sensing units and piezoelectric vibration sensing units are deployed at key component nodes of the target wooden structure to obtain spatial location information of each node.

[0021] During the sampling period, the surface temperature distribution matrix is ​​acquired through the infrared thermal imaging sensing unit, and the time-series data of structural micro-vibration acceleration are acquired through the piezoelectric vibration sensing unit.

[0022] Obtain ambient wind speed and humidity data as a reference for ambient noise;

[0023] The surface temperature distribution matrix, structural micro-vibration acceleration time series data, environmental wind speed data, and environmental humidity data are time-stamped and denoised, and then combined to generate a real-time condition monitoring dataset.

[0024] Preferably, S2 specifically includes:

[0025] S21. During the system initialization phase, apply excitation to the target wooden structure or utilize environmental excitation, collect structural response signals, extract multiple natural frequencies and damping ratios through fast Fourier transform, and construct a structural spectrum baseline.

[0026] S22. Based on the properties of wood materials, a pyrolysis kinetic equation is introduced to establish a nonlinear functional relationship between temperature variables and elastic modulus, and a thermal-stiffness decay theoretical curve is generated.

[0027] S23. Calculate the first derivative of the thermal-stiffness attenuation theoretical curve, determine the theoretical frequency attenuation threshold corresponding to a unit temperature increase, and use it as the benchmark for thermal-stiffness attenuation rate.

[0028] Preferably, S4 specifically includes:

[0029] S41. Based on the data in the high-frequency sampling mode, calculate the temperature time change rate of the surface temperature distribution matrix, and calculate the natural frequency change rate and frequency change acceleration corresponding to the time series data of structural micro-vibration acceleration.

[0030] S42. Use frequency change acceleration to filter environmental wind load interference, remove oscillating frequency change components, and retain monotonically decaying frequency change components.

[0031] S43. Normalize the retained monotonic decay-type frequency change component and the temperature-time change rate, and calculate the cross-correlation coefficient between the two on the time axis.

[0032] S44. Substitute the normalized data into the thermal-stiffness attenuation theoretical model and calculate the residual value between the observed frequency attenuation rate and the attenuation rate predicted by the theoretical model.

[0033] S45. Based on the cross-correlation coefficient and residual value, generate thermal-frequency deviation. If the cross-correlation coefficient is greater than the preset correlation threshold and the residual value is less than the preset model tolerance, it is judged as a high-confidence structural pyrolysis, and the confidence level of the fire source's impact on the structure, which characterizes the possibility of structural damage, is output.

[0034] Preferably, S5 specifically includes:

[0035] S51. Set the surface interference judgment logic, mechanical damage judgment logic and structural pyrolysis judgment logic;

[0036] S52. If the rate of change of temperature over time is higher than the preset temperature rise threshold and the rate of change of natural frequency is lower than the preset frequency change threshold, it is determined to be non-structural surface fire source interference.

[0037] S53. If the rate of change of the natural frequency is higher than the preset frequency change threshold and the rate of change of temperature over time is lower than the preset temperature rise threshold, then it is determined to be non-fire mechanical damage.

[0038] S54. If the confidence level of the fire source's impact on the structure meets the high-confidence structural pyrolysis condition, it is determined to be a structural fire hazard, and the total attenuation ratio of the current natural frequency relative to the structural spectrum baseline is further calculated.

[0039] S55. Based on the total attenuation ratio, predict the remaining load-bearing life time window of the structure and generate a comprehensive state assessment result that includes the type of hidden danger, confidence level and remaining life.

[0040] Preferably, S6 specifically includes:

[0041] S61. Receive the comprehensive status assessment results and implement a graded strategy based on the risk level;

[0042] S62. If it is determined to be a structural fire hazard and has not reached the instability threshold, execute the first level response: locate the coordinates of the pyrolysis component, start the fixed-point spray device for the coordinates of the component, and issue a smoldering warning signal.

[0043] S63. If it is determined that the structure is about to become unstable or the total attenuation ratio exceeds the preset collapse threshold, execute the second level response: based on the predicted value of the remaining life of the structure, calculate the structural stability weight of each evacuation path node, plan the optimal evacuation path that avoids high-risk beam and column areas, and control the evacuation indicator light to dynamically display the path.

[0044] S64. After the event ends, extract the temperature-frequency evolution data of the entire process of this event, use the data to calibrate the attenuation coefficient in the thermal-stiffness attenuation theoretical model, and update the thermal aging parameters in the database.

[0045] Preferably, the specific steps for filtering environmental wind load interference in S42 are as follows:

[0046] Monitor the time-domain variation characteristics of the natural frequency, calculate the difference between the natural frequency at the current moment and the natural frequency at the previous moment, and obtain the frequency change. If the frequency change shows a periodic fluctuation characteristic around the center frequency, it is marked as elastic deformation caused by wind load or environmental vibration, and the frequency change of this part is set to zero.

[0047] If the frequency change shows an irreversible monotonically decreasing trend, and the direction of the frequency change acceleration remains constant, it is marked as plastic damage or pyrolysis softening caused by material stiffness degradation, and this part of the frequency change is retained for subsequent analysis.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] By constructing a multidimensional dataset containing surface thermal imaging temperature field data and structural vibration response data, a correlation model between temperature variables and elastic modulus was established. This model can analyze the correlation between temperature change trends and natural frequency change trends by combining signal characteristics from different dimensions. Through this joint analysis of multidimensional data, the stiffness reduction caused by heating of the structure can be distinguished from conventional environmental disturbances from a physical mechanism perspective. This enables a comprehensive assessment of the internal state of the wood and the overall stability of the structure, improves the ability to identify hidden hazards, and ensures the reliability of monitoring results.

[0050] By utilizing the acceleration characteristics of frequency changes, the monitored signals are classified and processed. By monitoring the time-domain variation characteristics of the frequency, signals exhibiting periodic fluctuations can be identified and classified as elastic deformation caused by wind load or environmental vibration, and then filtered out. For signals exhibiting irreversible monotonic changes with constant direction, they are identified as manifestations of material stiffness degradation or pyrolytic softening, effectively eliminating environmental noise interference, preserving true structural damage information, and ensuring the accuracy of the alarm system.

[0051] It has the functions of hierarchical linkage response and life prediction. Based on the calculated confidence evaluation results and structural frequency attenuation, it can estimate the remaining load-bearing life window of the structure and generate a comprehensive assessment result including the type of hidden danger and the remaining life. Based on the assessment result, it can automatically plan evacuation routes to avoid high-risk areas and control the corresponding physical intervention facilities for targeted treatment, providing a quantitative basis for emergency decision-making. It can implement targeted protective measures while ensuring the safe evacuation of personnel, thereby reducing disaster losses. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is a logic block diagram of the system of the present invention. Detailed Implementation

[0054] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0055] Please see Figure 1 This invention provides a comprehensive alarm system for fire hazards and structural instability in wooden buildings, including: a multi-dimensional heterogeneous signal acquisition module configured to acquire surface thermal imaging temperature field data, structural vibration response data and environmental parameters of the target wooden building, and align timestamps to construct a real-time status monitoring dataset;

[0056] Structural Spectrum Benchmark Modeling Module: Configured to establish a structural spectrum benchmark based on modal scan data, and combine wood pyrolysis kinetic parameters to construct a thermal-stiffness decay theoretical model describing the nonlinear mapping relationship between temperature variables and elastic modulus;

[0057] Thermo-frequency correlation analysis module: configured to calculate the correlation characteristics between the rate of change of temperature over time and the rate of change of natural frequency, and generate a thermo-frequency deviation index;

[0058] Structural state determination module: configured to identify the pyrolysis state of wood and structural instability risk based on the residual analysis results of the thermal-frequency deviation index and the thermal-stiffness decay theoretical model;

[0059] The graded linkage response module is configured to execute corresponding fixed-point sprinkler control, evacuation route planning, and alarm output based on the identified status level.

[0060] Adaptive feedback correction module: configured to store temperature change data and frequency change data during the event process into a database to correct the thermal aging parameters in the thermal-stiffness decay theoretical model.

[0061] This embodiment details the architecture logic and physical implementation of the integrated alarm system, aiming to solve the problem of missed smoldering alarms caused by data silos in traditional monitoring. The multi-dimensional heterogeneous signal acquisition module is implemented through a hardware interface layer, integrating a FLIRLepton infrared thermal imaging module and a MEMS piezoelectric accelerometer to acquire sampling rates of [missing information]. Temperature field data With sampling rate Vibration data This module has a built-in NTP time synchronization unit that aligns data packets from different sensors with millisecond-level precision, constructing a unified time index. Real-time status monitoring dataset:

[0062]

[0063] For environmental parameters;

[0064] The structural spectral reference modeling module runs during the system initialization phase, extracting the structural reference frequency using the environmental excitation method. It also preloads a thermal-stiffness attenuation theory model based on the Arrhenius equation, which is ultimately mapped to a frequency prediction function:

[0065]

[0066] It is a parameter vector containing physical constants such as activation energy;

[0067] The thermal-frequency correlation analysis module is configured to perform sliding window cross-correlation calculations to calculate the rate of temperature change. With the rate of change of frequency The Pearson correlation coefficient is used to generate a thermal-frequency deviation index. Based on this, the structural state determination module inputs the observed data into the theoretical model to calculate the residuals. If the residual is less than the preset tolerance If the condition is met, it is determined to meet the characteristics of pyrolysis; then, the graded linkage response module outputs GPIO control signals according to the risk level to drive the spray solenoid valve or LED evacuation indicator; finally, the adaptive feedback correction module uses the least squares method to adjust the model parameters using event data. Perform iterative updates:

[0068]

[0069] The adaptive learning rate parameter is used to enable lifelong learning of the system.

[0070] The modules are connected in the following way:

[0071] S1. Collect surface thermal imaging temperature field data, structural vibration response data and environmental parameters of the target wooden building to construct a real-time condition monitoring dataset;

[0072] S2. Perform modal scanning on the target wooden structure, establish a structural spectrum baseline at room temperature, and pre-define the thermal-stiffness attenuation theoretical model to define the functional relationship between temperature change and elastic modulus attenuation.

[0073] S3. Real-time monitoring of surface thermal imaging temperature field data and structural vibration response data. When the local temperature exceeds the preset safety value or the natural frequency shows non-periodic decay, the high-frequency sampling mode is activated.

[0074] S4. In high-frequency sampling mode, extract temperature gradient features and frequency change acceleration features, calculate thermal-frequency deviation, and input the thermal-frequency deviation into the thermal-stiffness attenuation theoretical model for residual analysis to generate confidence evaluation results of the impact of fire source on structure.

[0075] S5. Based on the confidence evaluation results of the impact of the fire source on the structure and the attenuation of the current structural frequency relative to the baseline, determine the fire hazard level and the structural instability time window, and output graded alarm commands and control signals.

[0076] S6. In response to the graded alarm command, execute the corresponding physical intervention measures, and feed back the heterogeneous data of this monitoring cycle to the structural spectrum benchmark modeling module to update the thermal aging parameters.

[0077] This embodiment defines the timing and logic of the specific working method of the above system, realizing frequency conversion monitoring and closed-loop control.

[0078] Steps S1 and S2 are the initialization and normalization phases, during which the system operates in low-power mode with the sampling rate set to [value missing]. It must meet the minimum requirements of the Nyquist sampling theorem for the fundamental frequency;

[0079] Step S3 introduces a frequency conversion triggering mechanism: setting a temperature threshold. With frequency attenuation threshold When detected or If the system does not exhibit periodic response characteristics, an interrupt is triggered, switching the system to a high-frequency sampling mode and increasing the sampling rate to [value missing]. ,For example ;

[0080] Step S4 is the core calculation step, which calculates the temperature time gradient in high-frequency mode. With frequency second derivative Calculate the confidence level using a theoretical model. The confidence level is a probability value between 0 and 1, representing the likelihood that the current frequency decrease is caused by pyrolysis;

[0081] Step S5 is based on confidence level Compared with stiffness attenuation ratio A two-factor lookup table is used to determine the risk level, such as Level I, Level II, or Level III, and to predict the instability time. ;

[0082] Step S6 executes physical actions and closes the data loop, marking the complete time series data segment that caused the alarm as a "positive sample" for subsequent model parameter calibration.

[0083] S1 specifically includes:

[0084] Infrared thermal imaging sensing units and piezoelectric vibration sensing units are deployed at key component nodes of the target wooden structure to obtain spatial location information of each node.

[0085] During the sampling period, the surface temperature distribution matrix is ​​acquired through the infrared thermal imaging sensing unit, and the time-series data of structural micro-vibration acceleration are acquired through the piezoelectric vibration sensing unit.

[0086] Obtain ambient wind speed and humidity data as a reference for ambient noise;

[0087] The surface temperature distribution matrix, structural micro-vibration acceleration time series data, environmental wind speed data, and environmental humidity data are time-stamped and denoised, and then combined to generate a real-time condition monitoring dataset.

[0088] This embodiment details the specific algorithm implementation for data preprocessing; spatial location information is obtained through BIM model coordinates. Calibration is performed; during the data acquisition phase, the infrared thermal imaging unit outputs... Temperature matrix The output length of the piezoelectric vibration unit is acceleration vector To address the issue of inconsistent sampling rates among multiple data sources, such as thermal imaging (10Hz), vibration (1000Hz), and environmental data (1Hz), this embodiment employs a high-frequency alignment strategy with a low-frequency frequency or a unified resampling strategy; specifically, a unified time axis is selected. For low-frequency data, namely environmental and temperature data, cubic spline interpolation is used for upsampling to align it with the timestamps of the vibration data. In the denoising process, a wavelet thresholding algorithm is used for the acceleration data: a 5-level decomposition is performed using the Daubechiesdb4 wavelet basis, and a soft thresholding function is applied to the high-frequency coefficients.

[0089]

[0090] General threshold ;

[0091] The noise standard deviation is usually estimated by the absolute median (MAD) of the detail coefficients of the first-level wavelet decomposition. Finally, the signal is reconstructed to remove high-frequency environmental noise.

[0092] For the temperature matrix, median filtering is used to remove salt-and-pepper noise; the final generated real-time condition monitoring dataset It is a multidimensional tensor that contains synchronized temperature, vibration, and environmental parameters.

[0093] S2 specifically includes:

[0094] S21. During the system initialization phase, apply excitation to the target wooden structure or utilize environmental excitation, collect structural response signals, extract multiple natural frequencies and damping ratios through fast Fourier transform, and construct a structural spectrum baseline.

[0095] S22. Based on the properties of wood materials, a pyrolysis kinetic equation is introduced to establish a nonlinear functional relationship between temperature variables and elastic modulus, and a thermal-stiffness decay theoretical curve is generated.

[0096] S23. Calculate the first derivative of the thermal-stiffness attenuation theoretical curve, determine the theoretical frequency attenuation threshold corresponding to a unit temperature increase, and use it as the benchmark for thermal-stiffness attenuation rate.

[0097] This embodiment defines in detail the mathematical construction process of the physical model, ensuring the physical interpretability of the monitoring benchmark;

[0098] In step S21, using the acceleration response signal under environmental excitation, the front of the structure is identified through power spectral density (PSD) analysis and peak picking method. First natural frequency , serving as the structural spectrum baseline;

[0099] In step S22, based on the properties of wood material, a pyrolysis kinetic equation is introduced to establish a nonlinear functional relationship between temperature variables and elastic modulus. Specifically, to address the mismatch between the dimensionality of the surface temperature distribution matrix and the theoretical model input variables, this embodiment defines the model input variables. The temperature is the most unfavorable point, and the unit of temperature is Kelvin (K). First, an Arrhenius dynamic model is constructed to show the decay of the elastic modulus with temperature:

[0100]

[0101] The maximum value in the surface temperature distribution matrix at the current moment, i.e. It is used to characterize the stiffness decay in the region of the structure most severely heated;

[0102] The elastic modulus at room temperature;

[0103] Ambient temperature;

[0104] The characteristic pyrolysis temperature of wood;

[0105] This is the stiffness attenuation coefficient;

[0106] This refers to the nonlinear index of material pyrolysis.

[0107] Furthermore, based on the principles of structural dynamics, the natural frequencies of the structure And elastic stiffness, i.e., elastic modulus There exists The physical relationship; in order to accurately characterize this physical mechanism, this embodiment substitutes the above elastic modulus attenuation model into the frequency-stiffness relationship to derive the theoretical curve formula for frequency changing with temperature:

[0108]

[0109] By introducing an index This corrects the nonlinear difference between the frequency decay rate and the modulus decay rate, ensuring that the theoretical baseline conforms to the actual physical evolution law.

[0110] In step S23, the above-mentioned thermal-stiffness attenuation theoretical curve is calculated. First derivative with respect to temperature The theoretical frequency attenuation threshold corresponding to a unit temperature increase was determined and used as the benchmark for thermal-stiffness attenuation rate. This is used for residual analysis during subsequent monitoring.

[0111] S4 specifically includes:

[0112] S41. Based on the data in the high-frequency sampling mode, calculate the temperature time change rate of the surface temperature distribution matrix, and calculate the natural frequency change rate and frequency change acceleration corresponding to the time series data of structural micro-vibration acceleration.

[0113] S42. Use frequency change acceleration to filter environmental wind load interference, remove oscillating frequency change components, and retain monotonically decaying frequency change components.

[0114] S43. Normalize the retained monotonic decay-type frequency change component and the temperature-time change rate, and calculate the cross-correlation coefficient between the two on the time axis.

[0115] S44. Substitute the normalized data into the thermal-stiffness attenuation theoretical model and calculate the residual value between the observed frequency attenuation rate and the attenuation rate predicted by the theoretical model.

[0116] S45. Based on the cross-correlation coefficient and residual value, generate thermal-frequency deviation. If the cross-correlation coefficient is greater than the preset correlation threshold and the residual value is less than the preset model tolerance, it is judged as a high-confidence structural pyrolysis, and the confidence level of the fire source's impact on the structure, which characterizes the possibility of structural damage, is output.

[0117] This embodiment provides a specific mathematical definition of the core algorithm and corrects the variable signs to distinguish between the original function and the derivative;

[0118] Step S41: Calculate the rate of change: Define the rate of temperature change. natural frequency change rate and frequency change acceleration A five-point difference scheme is used for numerical differentiation to reduce noise amplification;

[0119] In step S43, the normalized cross-correlation coefficient (NCC) is calculated; the sliding window width is set to... At any moment Calculate the rate of temperature change Inverse phase value of the rate of change of frequency Local correlation:

[0120]

[0121] and Sliding windows The arithmetic mean of the rate of change of internal temperature and the rate of change of frequency;

[0122] In step S44, the residual value is calculated. ; Measured temperature Substituting the theoretical derivative model determined in step S2, the theoretical attenuation rate is obtained. The residual is then defined as:

[0123]

[0124] In step S45, the confidence level of the fire source's impact on the structure is defined. Constructing fusion evaluation metrics using the Sigmoid function:

[0125]

[0126] and These are the weighting coefficients, and they satisfy the normalization constraint. Ensure the confidence output range is ;

[0127] It is an indicator function;

[0128] The relevant threshold;

[0129] The sensitivity coefficient has the following dimensions: Used to balance residuals Dimensions;

[0130] like If so, it is determined to be a high-confidence structural pyrolysis.

[0131] S5 specifically includes:

[0132] S51. Set the surface interference judgment logic, mechanical damage judgment logic and structural pyrolysis judgment logic;

[0133] S52. If the rate of change of temperature over time is higher than the preset temperature rise threshold and the rate of change of natural frequency is lower than the preset frequency change threshold, it is determined to be non-structural surface fire source interference.

[0134] S53. If the rate of change of the natural frequency is higher than the preset frequency change threshold and the rate of change of temperature over time is lower than the preset temperature rise threshold, then it is determined to be non-fire mechanical damage.

[0135] S54. If the confidence level of the fire source's impact on the structure meets the high-confidence structural pyrolysis condition, it is determined to be a structural fire hazard, and the total attenuation ratio of the current natural frequency relative to the structural spectrum baseline is further calculated.

[0136] S55. Based on the total attenuation ratio, predict the remaining load-bearing life time window of the structure and generate a comprehensive state assessment result that includes the type of hidden danger, confidence level and remaining life.

[0137] This embodiment constructs a classification decision tree based on multi-parameter logical thresholds;

[0138] Steps S51-S53 define the following logical criteria:

[0139] Surface interference: IF AND If the fire is classified as a surface fire, such as a trash can burning, it does not affect structural safety.

[0140] Mechanical damage: IF AND The cause was determined to be mechanical impact or loose joints;

[0141] Structural pyrolysis: IF It was determined to be a concealed structure pyrolysis (smoldering).

[0142] Step S54: Calculate the total attenuation ratio. ;

[0143] In step S55, the remaining load-bearing life is predicted. Based on the current stiffness decay rate To prevent during periods of gradual stiffness change, i.e. A division-by-zero error can cause computational overflow. This embodiment introduces a minimal regularization parameter. First-order Taylor expansion was used to predict when the collapse threshold would be reached. Time required:

[0144]

[0145] The system's final output vector: ;pass The function ensures that the denominator is always greater than 0, thus guaranteeing the numerical stability of the algorithm during the structurally stable period.

[0146] S6 specifically includes:

[0147] S61. Receive the comprehensive status assessment results and implement a graded strategy based on the risk level;

[0148] S62. If it is determined to be a structural fire hazard and has not reached the instability threshold, execute the first level response: locate the coordinates of the pyrolysis component, start the fixed-point spray device for the coordinates of the component, and issue a smoldering warning signal.

[0149] S63. If it is determined that the structure is about to become unstable or the total attenuation ratio exceeds the preset collapse threshold, execute the second level response: based on the predicted value of the remaining life of the structure, calculate the structural stability weight of each evacuation path node, plan the optimal evacuation path that avoids high-risk beam and column areas, and control the evacuation indicator light to dynamically display the path.

[0150] S64. After the event ends, extract the temperature-frequency evolution data of the entire process of this event, use the data to calibrate the attenuation coefficient in the thermal-stiffness attenuation theoretical model, and update the thermal aging parameters in the database.

[0151] This embodiment details the execution logic of the hierarchical response and feedback mechanism;

[0152] Step S62, i.e., Level 1 response: When When the system determines that the structure is undergoing pyrolysis, it reads the coordinates from the thermal imager. Mapped to the spray matrix index The output control signal opens the corresponding solenoid valve, and only the damaged point is micro-sprayed to prevent the water loss from expanding;

[0153] Step S63, i.e., second-level response: when or At that time, perform dynamic path planning; construct a building topology map. Define node weights Remaining life of the component associated with this node Inversely proportional:

[0154]

[0155] This is the saturation time threshold;

[0156] To prevent the elimination of zero factors;

[0157] The Dijkstra algorithm is used to search for the minimum weight path from the starting point to the safe exit, which is the safest path, and the RGB light strips along the way are used to indicate the direction.

[0158] Step S64, i.e., model calibration: After the event ends, the recorded data is compared with... Store in the database; solve the optimization problem using the nonlinear least squares method, ensuring that the mathematical structure of the calibration formula is completely consistent with the aforementioned physical model:

[0159]

[0160] Update model parameters This is to correct the heat-sensitive properties of wood after it has aged.

[0161] The specific steps for filtering environmental wind-borne interference in S42 are as follows:

[0162] Monitor the time-domain variation characteristics of the natural frequency, calculate the difference between the natural frequency at the current moment and the natural frequency at the previous moment, and obtain the frequency change. If the frequency change shows a periodic fluctuation characteristic around the center frequency, it is marked as elastic deformation caused by wind load or environmental vibration, and the frequency change of this part is set to zero.

[0163] If the frequency change shows an irreversible monotonically decreasing trend, and the direction of the frequency change acceleration remains constant, it is marked as plastic damage or pyrolysis softening caused by material stiffness degradation, and this part of the frequency change is retained for subsequent analysis.

[0164] This embodiment specifically illustrates the design of a signal classification filter based on second-order derivative features; the system maintains a length of... Frequency observation window ;

[0165] Criterion 1, i.e., wind load identification: due to inherent frequency Since the value is always positive, directly calculating its sign flip cannot characterize the fluctuation feature; therefore, this step first calculates the centered sequence of the frequency data within the window. :

[0166]

[0167] This is the arithmetic mean of the frequencies within the current window;

[0168] calculate sign function The number of times the window is flipped; if the number of flips exceeds a preset threshold. and the frequency mean If the deformation is determined to be elastic deformation caused by wind load or environmental vibration, then the effective change amount is determined. ;

[0169] Criterion 2, namely damage identification: if within the window It does not exhibit high-frequency reversal characteristics, and the original frequency... It exhibits an irreversible monotonically decreasing trend (i.e., first-order difference). (Continuous), while frequency changes acceleration If the direction remains constant and there is no reverse restoring force, it is determined to be plastic damage or pyrolysis softening caused by material stiffness degradation, and this part of the frequency change data is retained for subsequent analysis.

[0170] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A comprehensive alarm system for fire hazards and structural instability in wooden buildings, characterized in that, include: Multidimensional heterogeneous signal acquisition module: configured to acquire surface thermal imaging temperature field data, structural vibration response data and environmental parameters of the target wooden building, and align timestamps to build a real-time condition monitoring dataset; Structural Spectrum Benchmark Modeling Module: Configured to establish a structural spectrum benchmark based on modal scan data, and combine wood pyrolysis kinetic parameters to construct a thermal-stiffness decay theoretical model describing the nonlinear mapping relationship between temperature variables and elastic modulus; Thermo-frequency correlation analysis module: configured to calculate the correlation characteristics between the rate of change of temperature over time and the rate of change of natural frequency, and generate a thermo-frequency deviation index; Structural state determination module: configured to identify the pyrolysis state of wood and structural instability risk based on the residual analysis results of the thermal-frequency deviation index and the thermal-stiffness decay theoretical model; The graded linkage response module is configured to execute corresponding fixed-point sprinkler control, evacuation route planning, and alarm output based on the identified status level. Adaptive feedback correction module: configured to store temperature change data and frequency change data during the event process into a database to correct the thermal aging parameters in the thermal-stiffness decay theoretical model; The modules are connected in the following way: S1. Collect surface thermal imaging temperature field data, structural vibration response data and environmental parameters of the target wooden building to construct a real-time condition monitoring dataset; S2. Perform modal scanning on the target wooden structure, establish a structural spectrum baseline at room temperature, and pre-define the thermal-stiffness attenuation theoretical model to define the functional relationship between temperature change and elastic modulus attenuation. S3. Real-time monitoring of surface thermal imaging temperature field data and structural vibration response data. When the local temperature exceeds the preset safety value or the natural frequency shows non-periodic decay, the high-frequency sampling mode is activated. S4. In high-frequency sampling mode, extract temperature gradient features and frequency change acceleration features, calculate thermal-frequency deviation, and input the thermal-frequency deviation into the thermal-stiffness attenuation theoretical model for residual analysis to generate confidence evaluation results of the impact of fire source on structure. S5. Based on the confidence evaluation results of the impact of the fire source on the structure and the attenuation of the current structural frequency relative to the baseline, determine the fire hazard level and the structural instability time window, and output graded alarm commands and control signals. S6. In response to the graded alarm command, execute the corresponding physical intervention measures and feed back the heterogeneous data of this monitoring cycle to the structural spectrum benchmark modeling module to update the thermal aging parameters; S2 specifically includes: S21. During the system initialization phase, apply excitation to the target wooden structure or utilize environmental excitation, collect structural response signals, extract multiple natural frequencies and damping ratios through fast Fourier transform, and construct a structural spectrum baseline. S22. Based on the properties of wood materials, a pyrolysis kinetic equation is introduced to establish a nonlinear functional relationship between temperature variables and elastic modulus, and a thermal-stiffness decay theoretical curve is generated. S23. Calculate the first derivative of the thermal-stiffness attenuation theoretical curve, determine the theoretical frequency attenuation threshold corresponding to a unit temperature increase, and use it as the benchmark for thermal-stiffness attenuation rate.

2. The integrated alarm system for fire hazards and structural instability in wooden buildings according to claim 1, characterized in that, S1 specifically includes: Infrared thermal imaging sensing units and piezoelectric vibration sensing units are deployed at key component nodes of the target wooden structure to obtain spatial location information of each node. During the sampling period, the surface temperature distribution matrix is ​​acquired through the infrared thermal imaging sensing unit, and the time-series data of structural micro-vibration acceleration are acquired through the piezoelectric vibration sensing unit. Obtain ambient wind speed and humidity data as a reference for ambient noise; The surface temperature distribution matrix, structural micro-vibration acceleration time series data, environmental wind speed data, and environmental humidity data are time-stamped and denoised, and then combined to generate a real-time condition monitoring dataset.

3. The integrated alarm system for fire hazards and structural instability in wooden buildings according to claim 2, characterized in that, S4 specifically includes: S41. Based on the data in the high-frequency sampling mode, calculate the temperature time change rate of the surface temperature distribution matrix, and calculate the natural frequency change rate and frequency change acceleration corresponding to the time series data of structural micro-vibration acceleration. S42. Use frequency change acceleration to filter environmental wind load interference, remove oscillating frequency change components, and retain monotonically decaying frequency change components. S43. Normalize the retained monotonic decay-type frequency change component and the temperature-time change rate, and calculate the cross-correlation coefficient between the two on the time axis. S44. Substitute the normalized data into the thermal-stiffness attenuation theoretical model and calculate the residual value between the observed frequency attenuation rate and the attenuation rate predicted by the theoretical model. S45. Based on the cross-correlation coefficient and residual value, generate thermal-frequency deviation. If the cross-correlation coefficient is greater than the preset correlation threshold and the residual value is less than the preset model tolerance, it is judged as a high-confidence structural pyrolysis, and the confidence level of the fire source's impact on the structure, which characterizes the possibility of structural damage, is output.

4. The integrated alarm system for fire hazards and structural instability in wooden buildings according to claim 3, characterized in that, S5 specifically includes: S51. Set the surface interference judgment logic, mechanical damage judgment logic and structural pyrolysis judgment logic; S52. If the rate of change of temperature over time is higher than the preset temperature rise threshold and the rate of change of natural frequency is lower than the preset frequency change threshold, it is determined to be non-structural surface fire source interference. S53. If the rate of change of the natural frequency is higher than the preset frequency change threshold and the rate of change of temperature over time is lower than the preset temperature rise threshold, then it is determined to be non-fire mechanical damage. S54. If the confidence level of the fire source's impact on the structure meets the high-confidence structural pyrolysis condition, it is determined to be a structural fire hazard, and the total attenuation ratio of the current natural frequency relative to the structural spectrum baseline is further calculated. S55. Based on the total attenuation ratio, predict the remaining load-bearing life time window of the structure and generate a comprehensive state assessment result that includes the type of hidden danger, confidence level and remaining life.

5. The integrated alarm system for fire hazards and structural instability in wooden buildings according to claim 4, characterized in that, S6 specifically includes: S61. Receive the comprehensive status assessment results and implement a graded strategy based on the risk level; S62. If it is determined to be a structural fire hazard and has not reached the instability threshold, execute the first level response: locate the coordinates of the pyrolysis component, start the fixed-point spray device for the coordinates of the component, and issue a smoldering warning signal. S63. If it is determined that the structure is about to become unstable or the total attenuation ratio exceeds the preset collapse threshold, execute the second level response: based on the predicted value of the remaining life of the structure, calculate the structural stability weight of each evacuation path node, plan the optimal evacuation path that avoids high-risk beam and column areas, and control the evacuation indicator light to dynamically display the path. S64. After the event ends, extract the temperature-frequency evolution data of the entire process of this event, use the data to calibrate the attenuation coefficient in the thermal-stiffness attenuation theoretical model, and update the thermal aging parameters in the database.

6. The integrated alarm system for fire hazards and structural instability in wooden buildings according to claim 5, characterized in that, The specific steps for filtering environmental wind-borne interference in S42 are as follows: Monitor the time-domain variation characteristics of the natural frequency, calculate the difference between the natural frequency at the current moment and the natural frequency at the previous moment, and obtain the frequency change. If the frequency change exhibits a periodic fluctuation characteristic around the center frequency, it is marked as elastic deformation caused by wind load or environmental vibration, and the frequency change is set to zero. If the frequency change shows an irreversible monotonically decreasing trend and the direction of the frequency change acceleration remains constant, it is marked as plastic damage or pyrolysis softening caused by material stiffness degradation, and the frequency change is retained for subsequent analysis.

Images