Floor temperature and humidity detection method and system

Abstract

The invention is suitable for the technical field of building environment monitoring and diagnosis, and provides a floor temperature and humidity detection method and system, and the method comprises the steps: continuously collecting original data obtained through the detection of a temperature and humidity sensor preset in a target floor region, and obtaining a multi-dimensional feature; obtaining a physical structure change signal according to the multi-dimensional features; comparing the physical structure change signal with a preset physical degradation characteristic pattern library, if a comparison result is highly matched, judging that a potential physical problem exists, and generating diagnosis information with a clear physical root orientation; and according to the diagnosis information, an early warning signal is triggered, a detailed diagnosis report is generated, and the diagnosis report comprises a problem area, a diagnosed specific physical root and suggested preventive maintenance measures. According to the invention, the temperature and humidity information of the floor can be accurately detected.

Description

Technical Field

[0001] The present application relates to the technical field of building environment monitoring and diagnosis, and in particular to a floor temperature and humidity detection method and system. Background Art

[0002] Inside buildings, minute inaccuracies in floor temperature and humidity information can cause the building's temperature and humidity control system to operate based on these deviations. This can cause localized areas of the floor to experience suboptimal temperature and humidity for extended periods, accelerating the degradation of the floor material. This can lead to deformation of wood flooring, hollowing of tiles, or the growth of microcracks in concrete. When these issues accumulate to a certain extent, traditional detection methods can only report abnormal conditions but cannot trace the underlying process of the problem. They also cannot provide accurate, non-invasive diagnosis to determine the specific area and physical mechanism causing the problem, resulting in increased maintenance costs and even a negative impact on the long-term service life of the building. Summary of the Invention

[0003] The present application discloses a floor temperature and humidity detection method and system to solve the above-mentioned technical problems in the prior art.

[0004] In a first aspect, the present application discloses a method for detecting floor temperature and humidity, comprising the following steps: Continuously collect raw data from temperature and humidity sensors preset in the target floor area, and perform cleaning, denoising, and time synchronization on the raw data to obtain pre-processed temperature and humidity time series data; Perform multi-dimensional feature analysis on the pre-processed temperature and humidity time series data. The multi-dimensional feature analysis includes spectrum analysis, amplitude distribution analysis, data coherence analysis, and cross-correlation analysis with related physical quantities to extract the fluctuation patterns and correlation characteristics of the temperature and humidity data in different dimensions and obtain multi-dimensional features. Based on the multi-dimensional features, the source of the temperature and humidity measurement deviation is identified, and the temperature and humidity fluctuations that do not conform to the characteristics of environmental changes or sensor drift are identified as potential signals caused by changes in the physical structure of the floor, thereby obtaining a physical structure change signal; Obtaining multi-dimensional features of the physical structure change signal based on the physical structure change signal, and comparing the multi-dimensional features of the physical structure change signal with a preset physical degradation feature pattern library, which stores multi-dimensional feature patterns of temperature and humidity data corresponding to different types of floor materials and underlying structures during specific physical degradation processes; if the comparison results are highly matched, it is determined that a potential physical problem exists, and diagnostic information with a clear physical root cause is generated; and Based on this diagnostic information, early warning signals are triggered and a detailed diagnostic report is generated that contains the problem area, the specific physical root cause diagnosed, and recommended preventive maintenance actions.

[0005] Clearly, this solution ensures data quality and synchronization by continuously collecting raw temperature and humidity data from the floor area and performing meticulous preprocessing. Furthermore, multi-dimensional feature analysis is performed on the preprocessed temperature and humidity time series data, including spectrum, amplitude distribution, data coherence, and cross-correlation analysis with related physical quantities. This allows for comprehensive and in-depth analysis of the fluctuation patterns and correlation characteristics of the temperature and humidity data across different dimensions. Furthermore, based on these multi-dimensional features, the system intelligently identifies the source of temperature and humidity measurement deviations, accurately identifying temperature and humidity fluctuations that do not conform to normal environmental variations or sensor drift as potential signals caused by changes in the floor's physical structure. These multi-dimensional features are then compared against a pre-defined library of physical degradation signature patterns, containing temperature and humidity data characteristic patterns associated with specific physical degradation processes for different types of flooring materials and substructures. A high-quality match indicates the presence of a potential physical problem, generating diagnostic information that clearly identifies the physical root cause. Ultimately, this diagnostic information triggers early warning signals and generates a detailed diagnostic report that identifies the problem area, the specific physical root cause, and recommended preventive maintenance measures.

[0006] It can be seen that the present application can effectively distinguish between slight temperature and humidity measurement deviations caused by changes in the physical structure of the floor and environmental changes or sensor drift itself, thereby realizing early, accurate and physical root-directed diagnosis of potential physical problems of the floor, overcoming the shortcomings of existing technologies that cannot trace the hidden process of the problem and provide accurate diagnosis, significantly reducing maintenance costs and extending the service life of the building.

[0007] In some preferred embodiments, the multi-dimensional features of the physical structure change signal are compared with a preset physical degradation feature pattern library, which stores multi-dimensional feature patterns of temperature and humidity data corresponding to different types of floor materials and underlying structures during specific physical degradation processes. If the comparison results are highly matched, it is determined that there is a potential physical problem, and diagnostic information with a clear physical root cause is generated, specifically including the following steps: applying active acoustic excitation to the floor and collecting acoustic response data of the floor; constructing a local acoustic impedance spectrum of the floor based on the acoustic response data; performing correlation analysis on the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal; Based on the correlation analysis results, performing pattern comparison in the physical degradation characteristic pattern library, wherein the physical degradation characteristic pattern library stores acoustic impedance spectrum patterns and temperature and humidity response patterns corresponding to different types of floor materials and underlying structures during specific physical degradation processes; and If the comparison result is highly matched, it is determined that there is a potential physical problem, and diagnostic information with a clear physical root cause is generated.

[0008] Obviously, by introducing active acoustic excitation and acoustic impedance spectrum analysis, this solution can verify and supplement the structural status of the floor from another physical dimension, making the pattern comparison of physical degradation characteristics more comprehensive and accurate, significantly improving the reliability and accuracy of diagnosis, and reducing the possibility of misjudgment.

[0009] In some preferred embodiments, correlation analysis is performed on the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal, specifically comprising the following steps: Performing a time-lag correlation analysis on the abnormal feature of the local acoustic impedance spectrum and the multi-dimensional feature of the physical structure change signal to identify a time-lag relationship between the abnormal feature and the multi-dimensional feature; Calculating the dynamic response characteristics between the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal according to the time lag relationship; Conduct a contextual assessment of the dynamic response characteristics, taking into account current environmental parameters and historical degradation trends; and According to the context evaluation result, the physical directionality of the correlation relationship between the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal is determined.

[0010] Obviously, this scheme can deeply explore the intrinsic physical connection between acoustic response and temperature and humidity signals through time-delay correlation analysis, dynamic response characteristic calculation and contextual evaluation, so as to more accurately determine the physical directionality of the correlation between the two and provide a more solid physical basis for diagnosis.

[0011] In some preferred embodiments, a time-lag correlation analysis is performed on the abnormal feature of the local acoustic impedance spectrum and the multi-dimensional feature of the physical structure change signal to identify the time-lag relationship between the abnormal feature and the multi-dimensional feature, specifically comprising the following steps: Dynamically segmenting the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal according to current environmental parameters and the identified degradation stage; Performing a time-varying coherence analysis on the abnormal feature time series and the multi-dimensional feature time series within each dynamic segment to identify a dynamic phase difference between the abnormal feature and the multi-dimensional feature at different frequencies; Based on the dynamic phase difference, a time lag distribution between the abnormal feature and the multi-dimensional feature is constructed; and According to the shape of the time lag distribution and the context information of the dynamic segmentation, a time lag relationship between the abnormal feature and the multi-dimensional feature is determined.

[0012] Obviously, this scheme can more accurately capture the dynamic correlation between acoustic anomalies and temperature and humidity signals in different time periods and frequencies through dynamic segmentation and time-varying coherence analysis, thereby more accurately identifying the time lag relationship and improving the accuracy and robustness of correlation analysis.

[0013] In some preferred embodiments, the abnormal feature time series of the local acoustic impedance spectrum and the multi-dimensional feature time series of the physical structure change signal are dynamically segmented according to the current environmental parameters and the identified deterioration stage, specifically including the following steps: Real-time acquisition of local environmental parameters and representation of local environmental gradient; According to the local environmental parameters, the thermal and moisture diffusion characteristics of the floor material, and the sensor response time, the time series of the local environmental parameters, the abnormal feature time series of the local acoustic impedance spectrum, and the multi-dimensional feature time series of the physical structure change signal are time-aligned and lag-compensated; Analysis of the correlation between the time-aligned local environmental gradient and the abnormal feature time series of the local acoustic impedance spectrum and the multi-dimensional feature time series of the physical structure change signal; According to the correlation analysis result and the identified deterioration stage, a structural change point caused by environmental gradient change or deterioration stage transition in the time series is identified, and the abnormal feature time series of the local acoustic impedance spectrum and the multi-dimensional feature time series of the physical structure change signal are dynamically segmented accordingly.

[0014] Obviously, this scheme can achieve more intelligent and adaptive dynamic segmentation by considering local environmental parameters, material characteristics, and sensor response time for time alignment and lag compensation, and identifying change points caused by environment or deterioration stage, ensuring the accuracy of the context of the analysis, and avoiding misjudgment of environmental impact as structural deterioration.

[0015] In some preferred embodiments, the correlation between the time-aligned local environmental gradient and the abnormal feature time series of the local acoustic impedance spectrum and the multi-dimensional feature time series of the physical structure change signal is analyzed, specifically including the following steps: Multivariate time-frequency analysis is performed on the time-aligned local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multidimensional characteristic time series of the physical structure change signal to identify the synchronization or causal relationship between the local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multidimensional characteristic time series of the physical structure change signal at different frequencies and time scales.

[0016] Obviously, this scheme uses multivariate time-frequency analysis, which can reveal the complex synchronization or causal relationship between local environmental gradients and acoustic, temperature and humidity signals at different frequencies and time scales, thereby more comprehensively understanding the impact of environmental factors on floor degradation signals and laying the foundation for subsequent causal relationship identification.

[0017] In some preferred embodiments, identifying the synchronization or causal relationship between the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multidimensional characteristic time series of the physical structure change signal at different frequencies and time scales specifically includes the steps of: Performing multivariate time-frequency analysis on the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal to obtain a multivariate time-frequency analysis result; By analyzing the directional information flow in the multivariate time-frequency analysis results, the synchronization or causal relationship between the local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multidimensional characteristic time series of the physical structure change signal at different frequencies and time scales is identified.

[0018] Obviously, by analyzing the directional information flow in the multivariate time-frequency analysis results, this scheme can more accurately identify the causal relationship between environmental gradients and floor response signals, rather than just synchrony, thereby more effectively removing environmental interference and focusing on the real signal caused by the physical degradation of the floor.

[0019] In some preferred embodiments, analyzing the directional information flow in the multivariate time-frequency analysis results specifically includes the steps of: Dynamically adjust the time window or frequency range of the information flow analysis according to current environmental parameters and identified degradation stages; quantifying the amount and direction of information transfer between the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal within the dynamically adjusted time window or frequency range; Evaluate whether the information flow pattern formed by the quantified information transfer amount and direction conforms to the physical coupling characteristics under the degradation stage and environmental conditions; and According to the evaluation result, the physical directionality of the causal relationship between the local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal is determined.

[0020] Obviously, this scheme dynamically adjusts the analysis parameters and evaluates the conformity of the information flow pattern with the physical coupling characteristics, which can make the causal analysis more adaptable to the actual environment and degradation stage, thereby more accurately determining the causal relationship between environmental factors and floor degradation signals, and further improving the accuracy of diagnosis.

[0021] In some preferred embodiments, quantifying the amount and direction of information transfer between the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal specifically includes the steps of: Adjusting calculation parameters of information flow indicators according to the current environmental parameters, the identified degradation stage, and sensor response characteristics to adapt to different physical coupling conditions; Under the adjusted parameters, calculating information flow indicators between the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal to obtain initial information transfer amount and direction; analyzing the amount and direction of the initial information transfer for known information flow patterns related to the local environmental gradient or sensor characteristics; Filtering out the known information flow pattern related to the local environmental gradient or the sensor's own characteristics from the initial information transfer amount and direction to obtain the information flow representing the deterioration of the floor material; and The information flow caused by the deterioration of floor materials is quantified to obtain the final information transmission amount and direction.

[0022] Obviously, by adjusting the calculation parameters and filtering out the known information flow patterns caused by the environment or the sensor's own characteristics, this scheme can more purely extract the information flow caused by the deterioration of the floor material, thereby more accurately quantifying its transmission amount and direction, greatly improving the specificity and accuracy of the diagnosis.

[0023] In a second aspect, the present application also discloses a floor temperature and humidity detection system, which includes: The acquisition and processing module is used to continuously collect the raw data detected by the temperature and humidity sensors preset in the target floor area, and clean, denoise and time synchronize the raw data to obtain pre-processed temperature and humidity time series data; A feature analysis module is used to perform multi-dimensional feature analysis on the pre-processed temperature and humidity time series data. The multi-dimensional feature analysis includes spectrum analysis, amplitude distribution analysis, data coherence analysis, and cross-correlation analysis with related physical quantities to extract the fluctuation patterns and correlation characteristics of the temperature and humidity data in different dimensions to obtain multi-dimensional features; a deviation identification module, configured to identify the source of the temperature and humidity measurement deviation based on the multi-dimensional features, and identify temperature and humidity fluctuations that do not conform to environmental changes or sensor drift characteristics as potential signals caused by changes in the physical structure of the floor, thereby obtaining a physical structure change signal; a pattern comparison diagnostic module, configured to obtain multi-dimensional features of the physical structure change signal based on the physical structure change signal, and compare the multi-dimensional features of the physical structure change signal with a preset physical degradation feature pattern library, which stores multi-dimensional feature patterns of temperature and humidity data corresponding to different types of floor materials and underlying structures during specific physical degradation processes; if the comparison results are highly matched, it is determined that a potential physical problem exists, and diagnostic information with a clear physical root cause is generated; and The early warning reporting module is used to trigger an early warning signal based on the diagnostic information and generate a detailed diagnostic report, which includes the problem area, the diagnosed specific physical root cause and the recommended preventive maintenance measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flow chart of a floor temperature and humidity detection method provided in this application.

[0025] Figure 2 This is a structural diagram of a floor temperature and humidity detection system provided in this application. DETAILED DESCRIPTION

[0026] The technical solutions in this application will be described clearly and completely below with reference to the accompanying drawings in this application.

[0027] This application proposes a floor temperature and humidity detection method, such as Figure 1 As shown, the following steps are included: Continuously collect raw data from temperature and humidity sensors preset in the target floor area, and perform cleaning, denoising, and time synchronization on the raw data to obtain pre-processed temperature and humidity time series data; Multi-dimensional feature analysis is performed on the pre-processed temperature and humidity time series data. The multi-dimensional feature analysis includes spectrum analysis, amplitude distribution analysis, data coherence analysis, and cross-correlation analysis with related physical quantities to extract the fluctuation patterns and correlation characteristics of temperature and humidity data in different dimensions and obtain multi-dimensional features. Based on multi-dimensional features, the source of temperature and humidity measurement deviation is identified, and temperature and humidity fluctuations that do not conform to environmental changes or sensor drift characteristics are identified as potential signals caused by changes in the physical structure of the floor, thereby obtaining a physical structure change signal; Obtaining multi-dimensional features of the physical structure change signal based on the physical structure change signal, and comparing the multi-dimensional features of the physical structure change signal with a preset physical degradation feature pattern library. The physical degradation feature pattern library stores multi-dimensional feature patterns of temperature and humidity data corresponding to different types of floor materials and underlying structures during specific physical degradation processes. If the comparison results are highly matched, it is determined that a potential physical problem exists, and diagnostic information with a clear physical root cause is generated; and Based on the diagnostic information, early warning signals are triggered and a detailed diagnostic report is generated, which contains the problem area, the specific physical root cause diagnosed, and recommended preventive maintenance measures.

[0028] The present application continuously collects temperature and humidity data and pre-processes it, and then performs multi-dimensional feature analysis on the pre-processed data, which can comprehensively extract the fluctuation patterns and correlation characteristics of temperature and humidity data in different dimensions. Therefore, based on these multi-dimensional features, the source of temperature and humidity measurement deviations can be accurately identified, and potential signals caused by changes in the physical structure of the floor can be distinguished from environmental changes or sensor drift. Furthermore, by comparing the multi-dimensional features of the identified physical structure change signals with a preset physical degradation feature pattern library, the present application can determine whether there are potential physical problems and generate diagnostic information with a clear physical root cause. Ultimately, an early warning signal is triggered based on the diagnostic information and a detailed diagnostic report is generated, thereby achieving early, accurate, and non-invasive diagnosis of changes in the physical structure of the floor, effectively solving the problem of the inability to trace hidden degradation processes and provide accurate diagnosis in the prior art, significantly reducing maintenance costs and extending the service life of buildings.

[0029] Furthermore, by deeply analyzing temperature and humidity data across the floor area, potential structural issues can be identified and diagnosed early. This method is commonly applied to floor structures within various types of buildings, such as residences, commercial buildings, and industrial plants, and is particularly suitable for scenarios with high requirements for floor structural stability, material durability, and indoor environmental comfort.

[0030] Specifically, raw data refers to the unprocessed electrical or digital signals directly output by temperature and humidity sensors. Cleaning, denoising, and time synchronization of this raw data are key steps in data preprocessing. Cleaning removes outliers or erroneous records; denoising eliminates random fluctuations caused by sensor noise and environmental interference; and time synchronization ensures that data from different sensors or at different time points are accurately aligned, forming consistent temperature and humidity time series data, laying the foundation for subsequent analysis.

[0031] Multidimensional feature analysis is a comprehensive data analysis method designed to uncover deeper patterns related to changes in the floor's physical structure from temperature and humidity time series data. Spectral analysis reveals the energy distribution of temperature and humidity data at different frequencies, such as periodic fluctuations or abnormal oscillations at specific frequencies. Amplitude distribution analysis examines the amplitude characteristics of temperature and humidity data fluctuations, such as the presence of abnormal peaks or valleys. Data coherence analysis assesses the degree of correlation between data from different sensors or from the same sensor at different time points to identify synchronization or lag. Cross-correlation analysis with related physical quantities may involve comparing temperature and humidity data with other physical quantities, such as ambient temperature, humidity, air pressure, and floor stress, to distinguish external influences from internal structural changes. These analyses yield multidimensional features that represent the comprehensive representation of temperature and humidity data across multiple dimensions, including time, frequency, and space.

[0032] The physical degradation feature pattern library is a pre-established database that stores multi-dimensional characteristic patterns of temperature and humidity data corresponding to a large number of different types of floor materials (such as wood flooring, ceramic tile, concrete, and composite materials) and their underlying structures (such as leveling layers, moisture barriers, insulation layers, and structural layers) as they undergo specific physical degradation processes (such as hollowing, cracking, moisture, deformation, delamination, and corrosion). These patterns are typically obtained through experiments, simulations, or historical data accumulation and are key to diagnosing physical floor problems.

[0033] Diagnostic information, generated by the system based on the comparison results, is a report or prompt that clearly points to the physical root cause. It not only indicates the existence of a problem but, more importantly, its nature, possible causes, and affected areas. Ultimately, the generation of early warning signals and detailed diagnostic reports aims to promptly notify users or maintenance personnel and provide specific maintenance recommendations, enabling preventive maintenance and preventing further problems.

[0034] This application aims to solve the problem that the existing floor temperature and humidity detection system cannot effectively distinguish the slight measurement deviation caused by the physical structure change of the floor and cannot provide accurate and non-invasive diagnosis. The solution is as follows: Introducing a multi-dimensional feature analysis and physical degradation feature pattern library comparison mechanism. Unlike existing technologies that only perform simple data calibration and threshold alarms, this application performs spectrum analysis, amplitude distribution analysis, data coherence analysis, and cross-correlation analysis with related physical quantities on pre-processed temperature and humidity time series data. This allows the system to extract the fluctuation patterns and correlation characteristics of temperature and humidity data in different dimensions from a deeper and more subtle level. This multi-dimensional analysis enables the system to more accurately capture abnormal temperature and humidity signals caused by subtle changes in the physical structure of the floor, and distinguish them from environmental changes or sensor drift.

[0035] Furthermore, this application compares the multi-dimensional characteristics of identified physical structural change signals with a pre-defined library of physical degradation feature patterns, enabling a clear diagnosis of the physical root cause of flooring problems. Existing technologies often only provide a general indication of "temperature and humidity anomalies" when an anomaly is detected, failing to specify whether it is "hollowing," "cracking," or "moisture."

[0036] In some of the aforementioned embodiments of this application, multi-dimensional feature analysis of temperature and humidity data and comparison with a pre-set pattern library enables preliminary identification of changes in the floor's physical structure. However, relying solely on passive monitoring with temperature and humidity data may not provide sufficiently early or accurate localization of physical degradation in certain early stages of degradation or when temperature and humidity changes are not significant, resulting in limitations in diagnostic specificity and sensitivity. To address this, this application further proposes a diagnostic method that incorporates active acoustic excitation to enhance the ability to identify physical structural issues in the floor.

[0037] The above process compares the multi-dimensional features of the physical structure change signal with a preset physical degradation feature pattern library, which stores multi-dimensional feature patterns of temperature and humidity data corresponding to different types of floor materials and underlying structures during specific physical degradation processes. If the comparison results are highly matched, it is determined that a potential physical problem exists, and diagnostic information with a clear physical root cause is generated. The specific steps include the following: applying active acoustic excitation to the floor and collecting acoustic response data of the floor; constructing a local acoustic impedance spectrum of the floor based on the acoustic response data; performing correlation analysis on the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal; performing pattern comparison in the physical degradation characteristic pattern library based on the correlation analysis results, wherein the physical degradation characteristic pattern library stores acoustic impedance spectrum patterns and temperature and humidity response patterns corresponding to different types of floor materials and underlying structures during specific physical degradation processes; and If the comparison results are highly matched, it is determined that there is a potential physical problem, and diagnostic information with a clear physical root cause is generated.

[0038] Specifically, active acoustic excitation of the floor involves applying controlled acoustic or vibrational energy to a target area of ​​the floor using an external excitation source (e.g., a piezoelectric actuator, electromagnetic actuator, or small impact hammer). This excitation can be broadband or frequency-specific, with the goal of stimulating the natural vibration modes or acoustic response within the floor structure. Simultaneously, collecting acoustic response data from the floor involves using highly sensitive sensors (e.g., accelerometers, microphones, or laser vibrometers) to detect the vibration or acoustic signals generated by the floor after being excited. This response data contains rich information about the floor's current physical state, such as its stiffness, damping, and internal defects.

[0039] Constructing a floor's local acoustic impedance spectrum based on acoustic response data can be understood as performing signal processing and analysis on the collected acoustic response data, such as Fourier transforms and wavelet analysis, to determine the floor's acoustic impedance characteristics at different frequencies. Acoustic impedance is a physical quantity that describes a material's ability to resist sound wave propagation. Its changes can directly reflect abnormalities in the floor's internal structure (such as delamination, hollowing, cracks, or moisture intrusion). Local acoustic impedance spectroscopy focuses on the impedance characteristics of specific areas, helping to precisely locate problem areas.

[0040] Furthermore, based on the correlation analysis results, a pattern comparison is performed within the physical degradation characteristic pattern library. This pattern library not only stores multi-dimensional characteristic patterns of temperature and humidity data corresponding to specific physical degradation processes for different types of floor materials and substructures, but also includes corresponding acoustic impedance spectroscopy patterns. This expands the pattern library into a multimodal feature library, capable of simultaneously comparing temperature and humidity response patterns and acoustic impedance spectroscopy patterns. This multimodal comparison allows for more precise identification of the degree of match between the current floor condition and known degradation patterns.

[0041] If the comparison results are highly matched, a potential physical problem is determined, and diagnostic information with a clear physical root cause is generated. This high match indicates that the current floor's temperature, humidity, and acoustic response characteristics are highly consistent with the characteristic patterns of a specific physical deterioration in the pattern library (for example, floor delamination, hollowing of the base layer, or wood swelling or decay caused by moisture intrusion). This provides more specific and reliable diagnostic results than relying solely on temperature and humidity data.

[0042] Through the above technical solutions, this application can significantly improve the accuracy of diagnosis and early warning capabilities of physical problems of floors. Compared with the method that relies solely on temperature and humidity data, the introduction of active acoustic excitation and acoustic impedance spectrum analysis enables the system to more sensitively capture changes in the internal structure of the floor, and even small, early-stage deterioration can be effectively identified. This multimodal data fusion and comparison mechanism not only improves the reliability of diagnostic results and reduces false positives and missed reports, but also can provide diagnostic information with clear physical root causes, such as whether it is a structural problem caused by delamination, hollowing or moisture, thereby providing more accurate guidance for subsequent preventive maintenance measures and effectively avoiding potential structural damage and safety hazards.

[0043] However, in practice, the physical degradation of flooring is often a dynamic and complex process influenced by multiple factors. Simple correlation analysis may not fully reveal the temporal dependencies and causal relationships between different physical quantities, as well as the impact of environmental factors on these relationships, resulting in insufficient diagnostic accuracy and robustness.

[0044] In this regard, the present application further proposes that the steps of correlating the abnormal characteristics of the above-mentioned local acoustic impedance spectrum with the multi-dimensional characteristics of the physical structure change signal include: performing a time-lag correlation analysis on the abnormal features of the local acoustic impedance spectrum and the multi-dimensional features of the physical structure change signal to identify a time-lag relationship between the abnormal features and the multi-dimensional features; Calculating the dynamic response characteristics between the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal according to the time lag relationship; Conducting a contextual assessment of the dynamic response characteristics in conjunction with current environmental parameters and historical degradation trends; and According to the context evaluation result, the physical directionality of the correlation relationship between the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal is determined.

[0045] Specifically, time-lag correlation analysis involves calculating the correlation between two time series at different time offsets to identify whether there is a temporal lead or lag relationship between them. For example, methods such as cross-correlation functions, Granger causality tests, or dynamic time warping can be used for this analysis. Its goal is to accurately capture the temporal evolution of the relationship between the acoustic response caused by changes in the floor's physical structure and changes in temperature and humidity, which is crucial for understanding degradation mechanisms.

[0046] Dynamic response characteristics can be understood as the time-dependent response of one physical quantity to changes in another. For example, hollowing or delamination within a floor might immediately reveal abnormalities in its acoustic impedance spectrum, whereas abnormalities in the temperature and humidity signals might only manifest after a while due to thermal inertia or the lag of moisture diffusion. By calculating these dynamic response characteristics, the coupling strength and response speed between different physical phenomena can be quantified.

[0047] In practical applications, contextual assessment combines current environmental parameters (such as temperature, humidity, and air pressure) with historical degradation trends (e.g., the floor area's previous degradation history, repair history, and material aging models) to comprehensively assess dynamic response characteristics. For example, in high-temperature and high-humidity environments, certain degradation processes may accelerate, resulting in a shortened lag time or increased response amplitude in acoustic and temperature and humidity responses. This assessment eliminates environmental interference with measurement results and more accurately determines degradation status.

[0048] Therefore, determining the physical directionality of the correlation means clarifying the physical mechanism underlying the relationship between acoustic anomalies and temperature and humidity anomalies, taking into account time lag, dynamic response, and contextual information. For example, does acoustic change cause temperature and humidity changes, vice versa, or do both stem from a common physical degradation source? This helps us fundamentally understand the degradation process of flooring.

[0049] The solution of this application solves the possible limitations of correlation analysis in the existing technology by introducing time-lag correlation analysis, dynamic response characteristic calculation and contextual evaluation. Specifically, time-lag correlation analysis can reveal the temporal sequence and lag relationship between the abnormal characteristics of the acoustic impedance spectrum and the multi-dimensional characteristics of temperature and humidity, which is crucial for understanding the dynamic evolution of the physical degradation process. For example, certain structural defects (such as cracking and delamination) may first cause rapid changes in acoustic characteristics, while their impact on the temperature and humidity field may be delayed due to the thermal and moisture diffusion characteristics of the material. By identifying this time-lag relationship, it is possible to more accurately determine which signal is the "leading indicator", thereby achieving earlier degradation warning.

[0050] Furthermore, by calculating dynamic response characteristics based on the identified time lag relationships, it is possible to quantify the strength and speed of the mutual influence between different physical quantities. This allows the system to not only identify correlations but also gain a deeper understanding of the "dynamics" of these correlations. For example, the dynamic characteristics of acoustic and thermo-hygrometric responses can differ significantly for different types of degradation (such as expansion due to moisture and contraction due to drying).

[0051] Furthermore, contextual assessment, combining current environmental parameters with historical degradation trends, effectively eliminates interference from environmental factors (such as temperature and humidity fluctuations) on measurement results and calibrates dynamic response characteristics. For example, when the ambient temperature fluctuates dramatically, temperature and humidity sensor data may fluctuate. However, contextual assessment can distinguish whether these fluctuations are normal environmental phenomena or abnormal responses caused by changes in the floor's physical structure. Furthermore, the inclusion of historical degradation trends enables the system to learn and adapt to the typical behavior patterns of specific floor materials at different stages of degradation, thereby improving diagnostic accuracy and robustness.

[0052] Through this multi-level analysis, the physical directionality of the correlation between the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal can be determined. This means that the system not only understands the correlation between the two, but also understands how they are related and the specific physical mechanism behind this correlation, providing a more solid physical basis for the subsequent generation of diagnostic information.

[0053] Through the above technical solutions, this application can significantly improve the accuracy and reliability of the diagnosis of physical problems in the floor. By deeply analyzing the time lag relationship and dynamic response characteristics between the acoustic response and the temperature and humidity changes, the system can more accurately capture the early signals of changes in the internal physical structure of the floor, and effectively distinguish interference caused by environmental factors or sensor drift. In addition, the contextual evaluation based on environmental parameters and historical degradation trends further enhances the robustness of the diagnosis and reduces the risk of false positives and missed reports. Ultimately, the physical directionality of the correlation relationship is clarified, making the generated diagnostic information more interpretable and instructive, and can provide users with diagnostic results with clear physical root causes, thereby achieving more accurate preventive maintenance and intervention, effectively extending the service life of the floor and reducing maintenance costs.

[0054] Specifically, the above-mentioned time-lag correlation analysis of the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal to identify the time-lag relationship between the abnormal characteristics and the multi-dimensional characteristics may include the following steps: Dynamically segmenting the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal according to current environmental parameters and the identified degradation stage; Performing a time-varying coherence analysis on the abnormal feature time series and the multi-dimensional feature time series in each of the dynamic segments to identify the dynamic phase difference between the abnormal feature and the multi-dimensional feature at different frequencies; constructing a time lag distribution between the abnormal feature and the multi-dimensional feature based on the dynamic phase difference; and A time lag relationship between the abnormal feature and the multi-dimensional feature is determined according to the morphology of the time lag distribution and the context information of the dynamic segmentation.

[0055] Specifically, dynamic segmentation involves intelligently segmenting the continuously collected time series of abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal based on real-time environmental parameters (such as temperature, humidity, and air pressure) and the current degradation stage of the floor material (such as initial cracks, delamination, and hollowing). The purpose of this segmentation is to ensure that when analyzing data from different time periods, the impact of environmental changes on data characteristics and the potential changes in response caused by the degradation process itself are fully considered. For example, when the ambient temperature fluctuates drastically, the physical response of the floor may be different from when the temperature is stable. Dynamic segmentation can distinguish and process data under these different environmental conditions.

[0056] Furthermore, time-varying coherence analysis is an advanced signal processing technique performed on the data within each dynamic segment. This analysis aims to identify the dynamic phase differences at different frequencies between the anomalous characteristic time series of the local acoustic impedance spectrum and the multidimensional characteristic time series of the physical structure change signal. Coherence analysis can reveal the correlation between the two signals in the frequency domain, while time-varying coherence allows this correlation to dynamically adjust over time, thereby more accurately capturing the transient or nonlinear response caused by changes in the floor structure. By analyzing the dynamic phase difference, the temporal lead or lag relationship between the acoustic response and the temperature and humidity response can be inferred, which is crucial for understanding the energy transfer and information coupling during physical degradation processes.

[0057] Based on the identified dynamic phase differences, a time lag distribution between the anomaly signature and the multidimensional signature can be constructed. This distribution can be a probability distribution, a histogram, or a trend plot, quantifying the time delay between acoustic anomalies and temperature and humidity anomalies at different frequencies and time points. For example, if a tiny crack appears within the floor, the acoustic anomaly may appear before or after the temperature and humidity anomaly, and this time lag may vary as the crack propagates. Constructing a time lag distribution helps fully understand this dynamic relationship.

[0058] Ultimately, based on the shape of the constructed time-lagged distribution (e.g., whether it is concentrated at a specific time point or exhibits a broad distribution) and the contextual information of the dynamic segmentation (e.g., the environmental conditions and degradation stage corresponding to the segment), the time-lagged relationship between the anomaly signature and the multidimensional features can be determined. This determination is not simply a simple time difference, but rather a deeper correlation judgment that incorporates the physical context and degradation mechanism. For example, it can be determined whether the acoustic anomaly causes the temperature and humidity anomaly, or vice versa, or whether both are triggered by a potential physical event.

[0059] The solution of the present application can overcome the limitations of traditional time-lag analysis methods in dealing with non-stationary and nonlinear data by introducing dynamic segmentation, time-varying coherence analysis and time lag distribution construction. Specifically, dynamic segmentation enables the analysis to adapt to changing environmental conditions and floor degradation stages, ensuring the validity and comparability of data under different working conditions. Time-varying coherence analysis can capture the dynamic phase relationship between acoustic response and temperature and humidity response at different frequencies, which is crucial for revealing the energy transfer and information coupling mechanism in complex physical degradation processes. By constructing a time lag distribution, the dynamic time correlation between acoustic anomalies and temperature and humidity anomalies can be quantified and understood more comprehensively and finely, thereby providing a more accurate and reliable basis for subsequent physical root cause directional judgment. This multi-level, dynamic analysis method enables the system to more accurately identify potential signals caused by changes in the physical structure of the floor, and distinguish them from fluctuations caused by environmental changes or sensor drift.

[0060] Through the above technical solutions, the present application can significantly improve the accuracy and robustness of time-lag correlation analysis. Traditional time-lag analysis often assumes that the signal is stationary and the environmental conditions are unchanged, which is difficult to meet in actual floor degradation monitoring. The present application uses dynamic segmentation to enable the analysis to adapt to non-stationary data and dynamic environments, thereby avoiding misjudgments caused by environmental changes or degradation stage transitions. Time-varying coherence analysis can reveal the dynamic phase relationship between signals at different frequencies, which is crucial for understanding complex physical coupling mechanisms, especially when the microstructure of floor materials changes, it can more sensitively capture early signs of degradation. As a result, the determined time lag relationship is more physically directional and can more accurately reflect the true dynamic response of changes in the physical structure of the floor, thereby providing more reliable input for subsequent physical degradation characteristic pattern comparisons, ultimately improving the accuracy of floor physical problem diagnosis and early warning capabilities.

[0061] However, in practical applications, fluctuations in environmental parameters such as temperature, humidity, and air pressure, as well as the evolution of floor material degradation stages, can significantly impact temperature and humidity data and acoustic response data. Failure to fully consider these external factors and performing dynamic segmentation solely based on data fluctuations can lead to inaccurate segmentation boundaries, thereby affecting the accuracy of subsequent time-lag correlation analysis and even misinterpreting environmental fluctuations as structural change signals.

[0062] In this regard, this application further proposes a more accurate and robust dynamic segmentation method by introducing environmental parameters and degradation stage information to more accurately identify structural change points in time series.

[0063] In this regard, the present application further proposes the steps of dynamically segmenting the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal according to the current environmental parameters and the identified degradation stage, including: Acquire local environmental parameters in real time and characterize local environmental gradients; Based on local environmental parameters, the thermal and moisture diffusion characteristics of floor materials, and sensor response time, time alignment and lag compensation are performed on the local environmental parameter time series, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal; Analyze the correlation between the time-aligned local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal; and Based on the correlation analysis results and the identified degradation stages, the structural change points in the time series caused by environmental gradient changes or degradation stage transitions are identified, and accordingly the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal are dynamically segmented.

[0064] Specifically, real-time acquisition of local environmental parameters involves continuously monitoring and recording various physical parameters of the current environment through auxiliary sensors deployed in or near the target floor area, such as temperature, humidity, and air pressure sensors. These parameters are used to characterize the local environmental gradient—the rate of change of environmental parameters in space or time, such as temperature or humidity gradients—in order to capture dynamic environmental changes that may affect the response of floor materials.

[0065] Based on local environmental parameters, the thermal and moisture diffusion characteristics of the flooring material, and sensor response time, the time series of local environmental parameters, the time series of anomaly characteristics of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal are time-aligned and lag-compensated. This can be understood as precise time synchronization of time series data from different sources. Because changes in environmental parameters, the flooring material's response to environmental changes (thermal and moisture diffusion characteristics), and the sensor's own response speed all have certain time lags, these time series require precise time calibration using algorithms (such as cross-correlation analysis and dynamic time warping) to eliminate errors caused by time lag and ensure the accuracy of subsequent correlation analysis.

[0066] In practical applications, the correlation between the abnormal feature time series of the local environment gradient and the local acoustic impedance spectrum after time alignment and the multi-dimensional feature time series of the physical structure change signal is analyzed. Specifically, statistical methods or machine learning algorithms, such as Granger causality test, mutual information, or deep learning-based time series correlation model, are used to quantify and identify the mutual influence relationship between the environmental gradient change and the floor temperature and humidity and acoustic response abnormal features. The purpose is to distinguish between signal fluctuations caused by environmental changes and real signals caused by floor physical structure changes.

[0067] Further, according to the correlation analysis results and the identified degradation stages, the structural change points caused by environmental gradient changes or degradation stage transitions in the time series are identified, and the abnormal feature time series of the local acoustic impedance spectrum and the multi-dimensional feature time series of the physical structure change signal are dynamically segmented accordingly. This means that when performing time series segmentation, not only the statistical properties of the data itself are considered, but also the influence of environmental factors and the current degradation state of the floor. For example, when the environmental temperature changes dramatically, the temperature and humidity data may fluctuate accordingly, in which case the fluctuations should be attributed to environmental factors rather than floor structure changes. At the same time, when the floor enters another degradation stage (e.g., delamination intensification) from a degradation stage (e.g., initial cracking), its temperature and humidity and acoustic response patterns may change significantly, and these change points should also be identified as important segmentation boundaries. In this way, dynamic segmentation can more accurately reflect the real time points of floor physical structure changes, avoiding misjudgment of environmental noise or normal degradation process as abnormal signals.

[0068] The scheme of the present application introduces real-time acquisition of local environmental parameters and representation of environmental gradient, enabling the system to perceive and quantify the influence of external environment on floor temperature and humidity and acoustic response. By accurately time-aligning and lag-compensating the time series of environmental parameters and the time series of temperature and humidity and acoustic features, the time bias caused by different physical processes and sensor response speeds is eliminated, ensuring the synchronization and comparability of the data. Thus, when analyzing the correlation between the local environmental gradient after time alignment and the floor feature data, it can effectively distinguish between fluctuations caused by environmental changes and real signals caused by floor physical structure changes. Finally, combined with the correlation analysis results and the identified degradation stages, the system can intelligently identify the key time points in the time series that are truly caused by structural changes or degradation stage transitions, thereby achieving accurate dynamic segmentation of the abnormal feature time series of the local acoustic impedance spectrum and the multi-dimensional feature time series of the physical structure change signal. This method avoids the defect of traditional methods that may misjudge environmental noise or normal environmental response as structural abnormalities, significantly improving the accuracy and reliability of dynamic segmentation and providing a more pure and accurate data basis for subsequent time-lag correlation analysis.

[0069] In some of the above-mentioned embodiments of the present application, a multivariate time-frequency analysis method can be used to analyze the correlation between the time-aligned local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multidimensional characteristic time series of the physical structure change signal.

[0070] Specifically, the above-mentioned analysis of the association between the time-aligned local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multidimensional characteristic time series of the physical structure change signal specifically includes the steps of: performing multivariate time-frequency analysis on the time-aligned local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum and the multidimensional characteristic time series of the physical structure change signal to identify the synchronization or causal relationship between the local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multidimensional characteristic time series of the physical structure change signal at different frequencies and time scales.

[0071] Multivariate time-frequency analysis is a technique that can simultaneously analyze the relationships between multiple time series in both the time and frequency domains. This analysis method can reveal synchronized changes or lagged relationships between different variables at specific frequency components, as well as how these relationships evolve over time. For example, methods such as wavelet coherence analysis, cross-spectral density analysis, or Granger causality analysis can be used to achieve this. This type of analysis can identify how local environmental gradients (such as temperature and humidity changes) affect the acoustic response of the floor and the signals of changes in the physical structure of temperature and humidity, as well as the frequency ranges and timescales at which these effects occur. Synchronicity refers to the simultaneous changes of multiple signals at specific frequencies or time points, while causality further reveals whether changes in one signal cause changes in another.

[0072] The solution of this application, by introducing multivariate time-frequency analysis, can more finely analyze the complex interactions between local environmental gradients, abnormal characteristic time series of local acoustic impedance spectra, and multi-dimensional characteristic time series of physical structural change signals. Traditional time domain or frequency domain analysis may not be able to fully capture the dynamic correlation between these variables at different frequencies and time scales. Through time-frequency analysis, it is possible to identify how environmental changes cause abnormalities in the floor structure response within a specific frequency range, or how structural degradation causes specific fluctuation patterns in temperature and humidity signals. For example, acoustic impedance anomalies at certain frequencies may be synchronized with ambient temperature fluctuations at specific frequencies, which may indicate structural stress caused by material expansion or contraction caused by temperature changes. In addition, by analyzing information flow or Granger causality, it is possible to further determine whether environmental changes cause structural responses, or whether structural changes in turn affect local environmental perception, thereby providing more accurate physical directionality for subsequent identification and diagnosis of degradation stages.

[0073] Specifically, the above-mentioned identification of the synchronization or causal relationship between the abnormal characteristic time series of the local environmental gradient and the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal at different frequencies and time scales can be further achieved through the following steps.

[0074] Performing multivariate time-frequency analysis on the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal to obtain a multivariate time-frequency analysis result; By analyzing the directional information flow in the multivariate time-frequency analysis results, the synchronization or causal relationship between the local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multidimensional characteristic time series of the physical structure change signal at different frequencies and time scales is identified.

[0075] The results of multivariate time-frequency analysis refer to the output obtained by jointly analyzing multiple time series data in the time and frequency domains, such as the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multidimensional characteristic time series of physical structure change signals. This analysis aims to reveal the interrelationships between these variables at different frequency components and time points, such as their energy distribution, phase relationship, and correlation. Specifically, methods such as wavelet coherence analysis, cross-spectral density analysis, or Granger causality analysis can be used to generate such results.

[0076] Furthermore, directional information flow refers to the direction and intensity of the influence of one variable on another in a multivariable system. In the context of time-frequency analysis, directional information flow analysis can reveal how information is transmitted from one signal to another at specific frequencies and time scales. For example, this directional information flow can be quantified and identified by calculating metrics such as transfer entropy, conditional Granger causality, or partial coherence. The goal is to distinguish whether changes in floor temperature, humidity, and acoustic response are caused by environmental changes or by physical deterioration of the floor itself, thereby more accurately attributing the signal's source.

[0077] The solution of the present application can deeply reveal the intrinsic connection between the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal by analyzing the directional information flow in the multivariate time-frequency analysis results. Traditional methods may only be able to identify the correlation or synchronization between variables, but it is difficult to clarify the causal relationship or the direction of information transmission. By introducing the concept of directional information flow, the path and intensity of information flow from one variable to another can be quantified and identified. For example, it is possible to determine whether the temperature change of the ambient temperature causes the temperature and humidity response of the floor, or whether the structural deterioration inside the floor causes the abnormal temperature, humidity and acoustic response. This analysis helps to eliminate the interference of environmental factors and more accurately locate the signals caused by changes in the physical structure of the floor, thereby improving the accuracy of diagnosis.

[0078] However, in practical applications, the dynamic changes of environmental parameters and the evolution of floor degradation stages may lead to the complexity of information flow patterns, making it insufficient to simply identify synchronicity or causal relationships to accurately determine their physical directivity, which may affect the accuracy and reliability of diagnosis.

[0079] In this regard, the present application further proposes steps for analyzing the directional information flow in the multivariate time-frequency analysis results, which specifically include the following steps: Dynamically adjust the time window or frequency range of the information flow analysis according to current environmental parameters and the identified degradation stage; quantifying the amount and direction of information transfer between the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal within the dynamically adjusted time window or frequency range; evaluating whether an information flow pattern formed by the quantified information transfer amount and direction conforms to the physical coupling characteristics under the degradation stage and the environmental conditions; and According to the evaluation result, the physical directionality of the causal relationship between the local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal is determined.

[0080] Specifically, dynamically adjusting the time window or frequency range for information flow analysis means that, given that floor degradation is a dynamic process and its response to environmental changes may vary depending on the degradation stage and environmental conditions, the parameters of information flow analysis should not be fixed. For example, in the early stages of degradation, a wider frequency range may be required to capture weak structural change signals; in the later stages of degradation, a narrower frequency range may be required to focus on specific resonant frequency shifts. Dynamic adjustment of the time window can adapt to rapid or slow changes in environmental parameters, ensuring that the most relevant temporal information is captured during analysis.

[0081] To quantify the amount and direction of information transfer between local environmental gradients, the time series of anomaly characteristics of the local acoustic impedance spectrum, and the multidimensional time series of physical structure change signals, various information theory or statistical methods can be employed. For example, metrics such as Granger causality, transfer entropy, or mutual information can be used to quantify the information flow between different time series. These metrics can reveal the extent and direction of the impact of changes in one series on future changes in another, thereby determining the existence and strength of a causal relationship between them.

[0082] Assessing whether the information flow pattern formed by the quantified amount and direction of information transfer is consistent with the physical coupling characteristics of the degradation stage and environmental conditions is a key step in ensuring that the diagnostic results are physically meaningful. This means that the identified information flow pattern must be consistent with the known or expected physical response mechanism of the floor material during the specific degradation process. For example, if the diagnosis shows that humidity changes cause the floor to expand, the direction of information flow should be from humidity to structural response, and its strength should match the material's hygroscopic expansion coefficient. This assessment can be based on pre-established physical models, experimental data, or expert knowledge bases.

[0083] The assessment results thus confirm the physical directivity of the causal relationship between the local environmental gradient, the anomalous characteristic time series of the local acoustic impedance spectrum, and the multidimensional characteristic time series of the physical structural change signal. This means not only identifying correlations or synchronicities, but also clearly indicating which environmental factor or structural change caused the other phenomenon, thus providing a clear physical root cause for subsequent diagnosis and maintenance.

[0084] The solution of this application dynamically adjusts the parameters of information flow analysis, allowing the analysis process to better adapt to complex and changing environmental conditions and the stage of floor degradation. By quantifying the amount and direction of information transfer, it is possible to more accurately reveal the causal relationship between different physical quantities, rather than just correlation. Furthermore, by comparing and evaluating the quantified information flow patterns with known physical coupling characteristics, it ensures that the identified causal relationship has a solid physical basis, avoiding misdiagnosis or missed diagnosis. This method enables more in-depth and accurate diagnosis of changes in the floor's physical structure, capable of extracting signals truly caused by structural degradation from complex temperature and humidity data and clarifying their physical root causes.

[0085] In some embodiments of the present application, the information flow between the local environmental gradient, the abnormal feature time series of the local acoustic impedance spectrum, and the multi-dimensional feature time series of the physical structure change signal is quantified to identify the causal relationship between them. However, in practical applications, the direct quantification of information transmission amount and direction may be disturbed by various factors, such as rapid fluctuations of environmental parameters or response characteristics of the sensor itself, which may introduce noise or pseudo-correlation unrelated to the deterioration of the floor physical structure, thereby affecting the accuracy and reliability of the diagnosis.

[0086] To this end, the present application further proposes a step of quantifying the information transmission amount and direction between the local environmental gradient, the abnormal feature time series of the local acoustic impedance spectrum, and the multi-dimensional feature time series of the physical structure change signal, comprising: adjusting the calculation parameters of the information flow index according to the current environmental parameters, the identified deterioration stage, and the sensor response characteristics, to adapt to different physical coupling conditions; under the adjusted parameters, calculating the information flow index between the local environmental gradient, the abnormal feature time series of the local acoustic impedance spectrum, and the multi-dimensional feature time series of the physical structure change signal to obtain the initial information transmission amount and direction; analyzing the known information flow patterns related to the local environmental gradient or the sensor itself characteristics in the initial information transmission amount and direction; filtering out the known information flow patterns related to the local environmental gradient or the sensor itself characteristics from the initial information transmission amount and direction to obtain information flow characteristic of floor material deterioration; and quantifying the information flow characteristic of floor material deterioration to obtain the final information transmission amount and direction.

[0087] Specifically, in quantifying the information transmission amount and direction between the local environmental gradient, the abnormal feature time series of the local acoustic impedance spectrum, and the multi-dimensional feature time series of the physical structure change signal, first, the calculation parameters of the information flow index are adjusted according to the current environmental parameters, the identified deterioration stage, and the sensor response characteristics. The current environmental parameters can include temperature, humidity, air pressure, etc., the identified deterioration stage can refer to different deterioration states of the floor such as initial crack, delamination, and wet expansion, and the sensor response characteristics refer to the response speed and accuracy of the sensor to environmental changes. The adjustment of these parameters aims to ensure that the calculation of information flow can adapt to different physical coupling conditions, for example, in a humid environment, the thermal-hygro coupling characteristics of the material will change, and the corresponding calculation weight or model parameters need to be adjusted.

[0088] Next, under the adjusted parameters, information flow metrics are calculated between the local environmental gradient, the anomaly characteristic time series of the local acoustic impedance spectrum, and the multidimensional characteristic time series of the physical structure change signal to obtain the initial information transfer quantity and direction. Information flow metrics can be quantified using a variety of established methods, such as transfer entropy, Granger causality, or mutual information. These metrics quantify the ability of one time series to predict the future state of another, thereby revealing the information transfer relationship between them. The initial information transfer quantity and direction refer to the information flow results calculated directly from the raw data before any filtering or correction.

[0089] Furthermore, the initial information flow quantity and direction are analyzed for known information flow patterns related to local environmental gradients or sensor characteristics. For example, when the ambient humidity rises sharply, the temperature and humidity sensor may respond immediately and produce corresponding signal fluctuations. These fluctuations are directly caused by the environmental change, not by structural degradation of the floor. Similarly, sensors may have inherent drift or noise patterns, which should also be identified as known information flow patterns.

[0090] Then, known information flow patterns related to local environmental gradients or sensor characteristics are filtered out from the initial information flow volume and direction, resulting in information flows that represent floor material degradation. This filtering process can be implemented using a variety of signal processing techniques, such as adaptive filtering, pattern recognition, or machine learning classification, to remove information flow components that are not caused by structural degradation, thereby highlighting information flows that are truly caused by floor material degradation.

[0091] Finally, the information flow characterizing floor material degradation is quantified to determine the final information transfer quantity and direction. This final quantification is purer and more accurate, more directly reflecting the information transfer pattern caused by floor physical structural degradation, and providing a reliable basis for subsequent diagnosis.

[0092] The solution of the present application effectively solves the problem of interference from environmental factors and sensor characteristics in traditional information flow analysis by introducing dynamic adjustment of information flow indicator calculation parameters, calculation of initial information flow, identification and filtering of known interference patterns, and re-quantification of the final information flow. Specifically, by adjusting the calculation parameters according to the current environmental parameters, the identified degradation stage, and the sensor response characteristics, it ensures that the information flow model can adapt to the actual physical coupling conditions and improves the accuracy of the calculation. The calculation of the initial information transfer amount and direction provides the basis for subsequent refinement. The key is that by analyzing and filtering out known information flow patterns related to local environmental gradients or the characteristics of the sensor itself, the information flow components caused by non-structural degradation can be effectively stripped off, thereby avoiding the misjudgment of environmental fluctuations or sensor drift as structural degradation signals. As a result, the final quantified information flow can more accurately and purely characterize the information transfer caused by floor material deterioration, greatly improving the physical directionality and reliability of the diagnostic results.

[0093] The specific embodiment of the present application also discloses a floor temperature and humidity detection system, such as Figure 2 As shown, the system includes: Acquisition and processing module 1 is used to continuously collect raw data detected by temperature and humidity sensors preset in the target floor area, and clean, denoise and time synchronize the raw data to obtain pre-processed temperature and humidity time series data; Feature analysis module 2, configured to perform multi-dimensional feature analysis on the pre-processed temperature and humidity time series data, wherein the multi-dimensional feature analysis includes spectrum analysis, amplitude distribution analysis, data coherence analysis, and cross-correlation analysis with related physical quantities, so as to extract the fluctuation patterns and correlation characteristics of the temperature and humidity data in different dimensions and obtain multi-dimensional features; Deviation identification module 3, for identifying the source of the temperature and humidity measurement deviation based on the multi-dimensional features, identifying temperature and humidity fluctuations that do not conform to environmental changes or sensor drift characteristics as potential signals caused by changes in the physical structure of the floor, and obtaining a physical structure change signal; a pattern comparison and diagnosis module 4 for obtaining multi-dimensional features of the physical structure change signal based on the physical structure change signal, and comparing the multi-dimensional features of the physical structure change signal with a preset physical degradation feature pattern library, wherein the physical degradation feature pattern library stores multi-dimensional feature patterns of temperature and humidity data corresponding to different types of floor materials and underlying structures during specific physical degradation processes; if the comparison results are highly matched, it is determined that a potential physical problem exists, and diagnostic information with a clear physical root cause is generated; and The early warning reporting module 5 is used to trigger an early warning signal based on the diagnostic information and generate a detailed diagnostic report, wherein the diagnostic report includes the problem area, the diagnosed specific physical root cause and the recommended preventive maintenance measures.

[0094] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application.

Claims

1. A floor temperature and humidity detection method, characterized in that: The following steps are involved: Continuously collect raw data detected by temperature and humidity sensors preset in the target floor area, and perform cleaning, denoising, and time synchronization processing on the raw data to obtain pre-processed temperature and humidity time series data; Performing a multi-dimensional feature analysis on the pre-processed temperature and humidity time series data, wherein the multi-dimensional feature analysis includes spectrum analysis, amplitude distribution analysis, data coherence analysis, and cross-correlation analysis with related physical quantities, so as to extract the fluctuation patterns and correlation characteristics of the temperature and humidity data in different dimensions and obtain multi-dimensional features; Based on the multi-dimensional features, the source of the temperature and humidity measurement deviation is identified, and the temperature and humidity fluctuations that do not conform to the characteristics of environmental changes or sensor drift are identified as potential signals caused by changes in the physical structure of the floor, thereby obtaining a physical structure change signal; Obtaining multi-dimensional features of the physical structure change signal based on the physical structure change signal, and comparing the multi-dimensional features of the physical structure change signal with a preset physical degradation feature pattern library, wherein the physical degradation feature pattern library stores multi-dimensional feature patterns of temperature and humidity data corresponding to different types of floor materials and underlying structures during specific physical degradation processes; if the comparison results are highly matched, it is determined that a potential physical problem exists, and diagnostic information with a clear physical root cause is generated; and Based on the diagnostic information, early warning signals are triggered and a detailed diagnostic report is generated, which includes the problem area, the specific physical root cause diagnosed, and recommended preventive maintenance measures.

2. A floor temperature and humidity detection method according to claim 1, characterized in that: The method compares the multi-dimensional features of the physical structure change signal with a preset physical degradation feature pattern library, which stores multi-dimensional feature patterns of temperature and humidity data corresponding to different types of floor materials and underlying structures during specific physical degradation processes. If the comparison results are highly matched, it is determined that a potential physical problem exists, and diagnostic information with a clear physical root cause is generated. The method specifically includes the following steps: applying active acoustic excitation to the floor and collecting acoustic response data of the floor; constructing a local acoustic impedance spectrum of the floor based on the acoustic response data; performing correlation analysis on the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal; Based on the correlation analysis results, pattern comparison is performed in the physical degradation characteristic pattern library, wherein the physical degradation characteristic pattern library stores acoustic impedance spectrum patterns and temperature and humidity response patterns corresponding to different types of floor materials and underlying structures during specific physical degradation processes; as well as If the comparison results are highly matched, it is determined that there is a potential physical problem, and diagnostic information with a clear physical root cause is generated.

3. A floor temperature and humidity detection method according to claim 2, characterized in that: The correlating analysis of the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal specifically includes the following steps: performing a time-lag correlation analysis on the abnormal features of the local acoustic impedance spectrum and the multi-dimensional features of the physical structure change signal to identify a time-lag relationship between the abnormal features and the multi-dimensional features; calculating, based on the time lag relationship, a dynamic response characteristic between the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal; Conducting a contextual assessment of the dynamic response characteristics in conjunction with current environmental parameters and historical degradation trends; as well as According to the context evaluation result, the physical directionality of the correlation relationship between the abnormal characteristics of the local acoustic impedance spectrum and the multi-dimensional characteristics of the physical structure change signal is determined.

4. A floor temperature and humidity detection method according to claim 3, characterized in that: The performing of a time-lag correlation analysis on the abnormal features of the local acoustic impedance spectrum and the multi-dimensional features of the physical structure change signal to identify the time-lag relationship between the abnormal features and the multi-dimensional features specifically includes the following steps: Dynamically segmenting the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal according to current environmental parameters and the identified degradation stage; Performing a time-varying coherence analysis on the abnormal feature time series and the multi-dimensional feature time series in each of the dynamic segments to identify the dynamic phase difference between the abnormal feature and the multi-dimensional feature at different frequencies; constructing a time lag distribution between the abnormal feature and the multi-dimensional feature based on the dynamic phase difference; as well as A time lag relationship between the abnormal feature and the multi-dimensional feature is determined according to the morphology of the time lag distribution and the context information of the dynamic segmentation.

5. A floor temperature and humidity detection method according to claim 4, characterized in that: The method of dynamically segmenting the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal according to the current environmental parameters and the identified degradation stage specifically includes the following steps: Acquire local environmental parameters in real time and characterize local environmental gradients; Based on the local environmental parameters, the heat and moisture diffusion characteristics of the floor material, and the sensor response time, time alignment and lag compensation are performed on the local environmental parameter time series, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal; Analyzing the correlation between the time-aligned local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal; Based on the correlation analysis results and the identified degradation stages, structural change points in the time series caused by environmental gradient changes or degradation stage transitions are identified, and accordingly, the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal are dynamically segmented.

6. A floor temperature and humidity detection method according to claim 5, characterized in that: The analysis of the correlation between the time-aligned local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal specifically includes the steps of: Multivariate time-frequency analysis is performed on the time-aligned local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multidimensional characteristic time series of the physical structure change signal to identify the synchronization or causal relationship between the local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multidimensional characteristic time series of the physical structure change signal at different frequencies and time scales.

7. A floor temperature and humidity detection method according to claim 6, characterized in that: The identifying of the synchronization or causal relationship between the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal at different frequencies and time scales specifically includes the steps of: Performing multivariate time-frequency analysis on the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal to obtain a multivariate time-frequency analysis result; By analyzing the directional information flow in the multivariate time-frequency analysis results, the synchronization or causal relationship between the local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multidimensional characteristic time series of the physical structure change signal at different frequencies and time scales is identified.

8. A floor temperature and humidity detection method according to claim 7, characterized in that: The analyzing of the directional information flow in the multivariate time-frequency analysis result specifically comprises the steps of: Dynamically adjust the time window or frequency range of the information flow analysis according to current environmental parameters and the identified degradation stage; quantifying the amount and direction of information transfer between the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal within the dynamically adjusted time window or frequency range; Evaluate whether the information flow pattern formed by the quantified amount and direction of information transfer is consistent with the physical coupling characteristics under the described degradation stage and environmental conditions; and According to the evaluation result, the physical directionality of the causal relationship between the local environmental gradient and the abnormal characteristic time series of the local acoustic impedance spectrum and the multi-dimensional characteristic time series of the physical structure change signal is determined.

9. A floor temperature and humidity detection method according to claim 8, characterized in that: The quantification of the amount and direction of information transmission between the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal specifically includes the steps of: Adjusting calculation parameters of information flow indicators according to the current environmental parameters, the identified degradation stage, and sensor response characteristics to adapt to different physical coupling conditions; Under the adjusted parameters, calculating information flow indicators between the local environmental gradient, the abnormal characteristic time series of the local acoustic impedance spectrum, and the multi-dimensional characteristic time series of the physical structure change signal to obtain initial information transfer amount and direction; analyzing the amount and direction of the initial information transfer for known information flow patterns related to the local environmental gradient or sensor characteristics; filtering out the known information flow patterns related to the local environmental gradient or the sensor's own characteristics from the initial information transfer amount and direction, and obtaining the information flow characterizing the deterioration of the floor material; as well as The information flow caused by the deterioration of the floor material is quantified to obtain the final information transmission amount and direction.

10. A floor temperature and humidity detection system, characterized in that: The system includes: An acquisition and processing module is used to continuously acquire raw data detected by temperature and humidity sensors preset in the target floor area, and to clean, denoise, and time-synchronize the raw data to obtain pre-processed temperature and humidity time series data; a feature analysis module for performing multi-dimensional feature analysis on the pre-processed temperature and humidity time series data, wherein the multi-dimensional feature analysis includes spectrum analysis, amplitude distribution analysis, data coherence analysis, and cross-correlation analysis with related physical quantities, so as to extract the fluctuation patterns and correlation characteristics of the temperature and humidity data in different dimensions and obtain multi-dimensional features; a deviation identification module, configured to identify the source of the temperature and humidity measurement deviation based on the multi-dimensional features, and identify temperature and humidity fluctuations that do not conform to the characteristics of environmental changes or sensor drift as potential signals caused by changes in the physical structure of the floor, thereby obtaining a physical structure change signal; a pattern comparison diagnostic module, configured to obtain multi-dimensional features of the physical structure change signal based on the physical structure change signal, and compare the multi-dimensional features of the physical structure change signal with a preset physical degradation feature pattern library, wherein the physical degradation feature pattern library stores multi-dimensional feature patterns of temperature and humidity data corresponding to different types of floor materials and underlying structures during specific physical degradation processes; if the comparison results are highly matched, it is determined that a potential physical problem exists, and diagnostic information with a clear physical root cause is generated; and The early warning reporting module is used to trigger an early warning signal based on the diagnostic information and generate a detailed diagnostic report, wherein the diagnostic report includes the problem area, the diagnosed specific physical root cause, and the recommended preventive maintenance measures.

Images