Multi-sensor fusion building structure detection and evaluation system and method

Abstract

The invention belongs to the technical field of building structure detection and evaluation, and provides a multi-sensor fusion building structure detection and evaluation system and method.The system comprises a humidity detection module, an infrared thermal imaging module, a sound wave detection module, a data fusion module and an evaluation module, humidity abnormity caused by plant growth is identified; the infrared thermal imaging module captures the temperature distribution of the humidity abnormal area, and further locates a plant growth area; the sound wave detection module emits sound waves in humidity and temperature abnormal areas, and analyzes sound wave propagation characteristics to confirm growth activities of plant roots; the data fusion module integrates the multi-modal data to judge the growth position and state of the plant and the damage influence on the building structure; and the evaluation module evaluates the safety and residual life of the building accordingly. The method has the advantages that the internal structure of the building can be comprehensively and accurately detected, so that the service life of the building can be comprehensively and effectively evaluated.

Description

Technical Field

[0001] The present invention relates to the field of building detection and evaluation, and in particular to a building structure detection and evaluation system and method that integrates multiple sensors. Background Art

[0002] Plants growing in buildings are often found on the surface of building structures, walls, and internal locations such as wall joints and pipes. Plant growth is usually carried out through natural seed or spore propagation, and the pores and gaps of certain building materials such as masonry and concrete, as well as the microclimate environment around the building, often provide conditions for plant growth.

[0003] However, the growth of plants in buildings may cause a certain degree of physical and chemical damage to the building itself. Specifically, the growth of plant roots will destroy the integrity of building materials, and the additional weight may cause deformation and collapse of walls or structures. The acidic substances secreted by plants may also cause corrosion of building materials, and the activity of microorganisms may cooperate with plant roots to accelerate the damage of building materials.

[0004] Existing technologies mainly rely on manual inspection or a single technical means to detect plant growth inside buildings. Common technical methods include visual inspection and regular structural inspection, but these methods have limitations, especially when plants grow hidden inside buildings and cannot be seen directly. Traditional technologies cannot effectively detect hidden plants and cannot fully and effectively assess the damage caused by plants growing inside buildings. Summary of the invention

[0005] The first object of the present invention is to provide a multi-sensor fusion building structure detection and evaluation system, which has the advantage of being able to comprehensively and accurately detect the internal structure of a building, thereby being able to comprehensively and effectively evaluate the life of the building.

[0006] The above technical objectives of the present invention are achieved through the following technical solutions:

[0007] A multi-sensor fusion building structure detection and evaluation system, comprising:

[0008] Humidity detection modules, distributed in multiple locations of the building structure, are used to monitor humidity changes inside the building in real time and identify humidity anomalies caused by plant growth;

[0009] An infrared thermal imaging module, working in conjunction with the humidity detection module, is used to capture the temperature distribution on the surface and inside of the building, and further locate the plant growth area by analyzing the heat changes in the humidity abnormal area;

[0010] An acoustic wave detection module, linked with the humidity detection module and the infrared thermal imaging module, is used to transmit acoustic waves in the humidity abnormality area and the heat abnormality area and receive reflected signals, and analyze the propagation characteristics of the acoustic waves to confirm the growth activities of the plant roots;

[0011] A data fusion module is used to receive the detection data of the humidity detection module, the infrared thermal imaging module and the sound wave detection module, and comprehensively judge the growth position and state of the plant and its physical and chemical damage effects on the building structure through fusion analysis;

[0012] An assessment module, connected to the data fusion module, is used to assess the remaining life of the building based on the physical and chemical damage effects of plant growth on the building structure.

[0013] Further configuration: the humidity detection module includes:

[0014] A humidity sensor network, including humidity sensors arranged at multiple locations of the building structure, for collecting humidity data in real time;

[0015] A data acquisition unit is used to obtain the humidity change rate of each sensor and mark the area where the humidity change rate exceeds a threshold as an abnormal humidity area;

[0016] The regional analysis unit is used to calculate the mean and standard deviation of humidity within the monitoring unit and analyze the humidity distribution characteristics;

[0017] The time series analysis unit is used to perform sliding window calculations on humidity data to determine the persistence of humidity anomaly areas;

[0018] The data transmission unit is used to upload the location information and humidity characteristics of the humidity abnormality area to the data fusion module.

[0019] By adopting the above technical solutions, the humidity detection module can achieve rapid, dynamic and accurate positioning of hidden plant growth areas inside buildings through real-time data collection, abnormal area identification, time series analysis and distribution characteristic calculation;

[0020] By monitoring the humidity data inside the building structure in real time through the humidity sensor network and calculating the humidity change rate, it is possible to quickly and accurately identify areas with abnormal humidity, laying the foundation for the subsequent accurate detection of infrared thermal imaging and acoustic wave detection modules;

[0021] By using time series analysis, we can detect the persistence and change trend of humidity anomaly areas, realize dynamic monitoring of humidity detection, capture long-term humidity changes caused by plant growth, and avoid misjudgment caused by short-term fluctuations;

[0022] By calculating the mean and standard deviation of humidity within the monitoring unit, the spatial distribution pattern of humidity can be intuitively reflected, the spatial characteristics of the plant growth area (distribution range and concentration) can be provided, and the errors that may be caused by local data anomalies can be reduced, thereby improving the accuracy of the overall analysis.

[0023] Further configuration: the infrared thermal imaging module includes:

[0024] A heat map capture unit, used to scan the abnormal area marked by the humidity detection module and generate a heat map of the temperature distribution in the area;

[0025] The temperature difference analysis unit is used to calculate the temperature difference of each point in the thermal map and mark the temperature abnormal points according to the set temperature difference threshold;

[0026] The regional clustering unit is used to cluster adjacent temperature anomaly points to form potential plant growth areas;

[0027] The heat-humidity joint analysis unit is used to cross-validate the temperature anomaly area and the humidity anomaly area, and extract the overlapping part as the plant growth area.

[0028] By adopting the above technical solution and coordinating with the humidity detection module, the detection area can be accurately locked, redundant scanning can be reduced, and the efficiency of thermal imaging can be improved. The accuracy of positioning the plant growth area can be enhanced by combining humidity and heat characteristics. Infrared imaging technology can be used to achieve rapid area positioning, and the plant growth range can be refined through clustering algorithms.

[0029] Further configuration: the acoustic wave detection module includes:

[0030] An acoustic wave transmitting unit, used for transmitting acoustic waves in an area with abnormal humidity and temperature;

[0031] An acoustic wave receiving unit, used to receive the reflected signal and record the propagation time and amplitude of the signal;

[0032] A reflection analysis unit, used to calculate the spatial position and material properties of the reflection point according to the propagation time formula and the attenuation formula;

[0033] A feature extraction unit is used to extract the time delay, amplitude change and waveform distortion of the reflected signal to identify the growth activity of the plant root system;

[0034] The joint analysis unit is used to cross-validate the reflection point with the humidity anomaly area and the temperature anomaly area to determine the actual location of the plant roots.

[0035] By adopting the above technical solutions, the growth activities of plant roots can be accurately confirmed through sound wave propagation path, attenuation and reflection analysis; combined with humidity and thermal imaging data, the detection range can be greatly reduced and the positioning accuracy can be improved; material property analysis: the acoustic characteristics of plant roots are extracted based on the reflection signal to provide a basis for subsequent evaluation modules.

[0036] Further configuration: the joint analysis unit is configured with a plant root distribution model generation strategy, and the plant root distribution model generation strategy includes:

[0037] Data fusion step: receiving the data of humidity detection, infrared thermal imaging and acoustic wave detection to form a joint point set;

[0038] Three-dimensional point cloud generation steps: converting the joint point set into a three-dimensional point cloud model of the plant root system through an interpolation algorithm;

[0039] Three-dimensional surface fitting steps: spatial fitting of point cloud data based on radial basis function;

[0040] Voxelization processing step: discretize the three-dimensional surface into a voxel grid to show the spatial distribution of plant roots.

[0041] By adopting the above technical solutions, the spatial distribution and density of plant roots can be displayed through a three-dimensional model, providing an intuitive basis for evaluating the impact of building structures; a more accurate three-dimensional model can be generated by combining multimodal data to avoid misjudgment of a single mode; and the weights can be dynamically adjusted according to the accuracy of the detection module to ensure the scientificity and reliability of the model generation.

[0042] Further configuration: the data fusion module includes:

[0043] A data preprocessing unit, used to receive and normalize the data of humidity detection, infrared thermal imaging and acoustic wave detection;

[0044] A position determination unit, used to calculate the center position of the plant growth area and generate a three-dimensional boundary by a weighted distance algorithm;

[0045] Growth status analysis unit, used to calculate the growth activity and root density of plants;

[0046] Impact Assessment Unit for evaluating physical and chemical damage to building structures caused by plants based on the added weight of the root system and the degree of chemical attack.

[0047] By adopting the above technical solutions, the three-dimensional spatial distribution of plant roots can be located through multimodal data fusion; combined with humidity, temperature and sound wave data, the plant growth activity and distribution density can be quantified; thereby accurately assessing the physical and chemical damage of plants to building structures.

[0048] Further configuration: the evaluation module is configured with an impact evaluation model and a life evaluation model, and the algorithm formula of the impact evaluation model is as follows:

[0049] I=α 1 Ax+α 2 D+α 3 W r +α 4 E c

[0050] in,

[0051] I represents the comprehensive impact index (dimensionless);

[0052] α 1 , α 2 , α 3 , α 4 Indicates the weight of each factor;

[0053] Ax represents plant growth activity;

[0054] D represents the density of the root system;

[0055] W r Indicates the additional weight of the root system;

[0056] E c Indicates the risk of chemical attack;

[0057] The remaining life assessment model of the building structure is adjusted based on the comprehensive impact index. The algorithm formula of the life assessment model is as follows:

[0058] L=L 0 -γ·I

[0059] in,

[0060] L represents the remaining life of the building;

[0061] L 0 Indicates the design life of the building;

[0062] Y represents the life loss coefficient.

[0063] By adopting the above technical solutions, introducing additional weight and chemical corrosion risks, combined with activity and density, a multi-dimensional building impact analysis is provided; through the comprehensive impact index, the building life prediction model is dynamically adjusted to enhance scientificity and applicability.

[0064] Another object of the present invention is to provide a multi-sensor fusion building structure detection and evaluation method, comprising the following steps:

[0065] S1: Use the humidity detection module to monitor humidity changes inside the building and identify humidity anomalies caused by plant growth;

[0066] S2: Use infrared thermal imaging modules to detect temperature changes on the surface and inside of buildings and locate plant growth areas;

[0067] S3: The sound wave detection module transmits sound waves and receives reflected signals, analyzes the propagation characteristics of the sound waves, and identifies the growth activities of the plant roots;

[0068] S4: Input the humidity, temperature and sound wave detection data into the data fusion module for comprehensive analysis to identify the plant growth status and its location;

[0069] S5: Evaluate the potential physical and chemical damage of plant growth to the building based on the analysis results and assess the remaining life of the building.

[0070] In summary, the present invention has the following beneficial effects:

[0071] The present invention provides a multi-dimensional detection method that can accurately identify the growth status of plants in buildings. By simulating the growth status of plants, possible damage to buildings can be predicted in advance to avoid potential safety hazards of building structures. Combined with the specific conditions of the building, an assessment of the damage caused by plant growth to the building can be output, which can help formulate scientific building repair and maintenance plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 is a system architecture block diagram of the present invention;

[0073] Figure 2 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0074] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0075] Example:

[0076] A multi-sensor fusion building structure detection and evaluation system, such as Figure 1 As shown, including:

[0077] Humidity detection modules, distributed in multiple locations of the building structure, are used to monitor humidity changes inside the building in real time and identify humidity anomalies caused by plant growth;

[0078] An infrared thermal imaging module, working in conjunction with the humidity detection module, is used to capture the temperature distribution on the surface and inside of the building, and further locate the plant growth area by analyzing the heat changes in the humidity abnormal area;

[0079] An acoustic wave detection module, linked with the humidity detection module and the infrared thermal imaging module, is used to transmit acoustic waves in the humidity abnormality area and the heat abnormality area and receive reflected signals, and analyze the propagation characteristics of the acoustic waves to confirm the growth activities of the plant roots;

[0080] A data fusion module is used to receive the detection data of the humidity detection module, the infrared thermal imaging module and the sound wave detection module, and comprehensively judge the growth position and state of the plant and its impact on the building structure through fusion analysis;

[0081] An evaluation module, connected to the data fusion module, is used to evaluate the physical and chemical damage of plant growth to the building structure based on the fused data and predict the remaining life of the building.

[0082] The humidity detection module includes:

[0083] A humidity sensor network includes humidity sensors arranged at multiple locations of a building structure to collect humidity data in real time. i (t);

[0084] Data acquisition unit, used to obtain the humidity change rate ΔH of each sensor i , and mark the area where the humidity change rate exceeds the threshold θ as the humidity abnormal area;

[0085] The regional analysis unit is used to calculate the average humidity H in the monitoring unit. avg and standard deviation σ H , analyze the humidity distribution characteristics;

[0086] Time series analysis unit, used to perform sliding window calculations on humidity data W t , determine the persistence of the abnormal humidity area;

[0087] The data transmission unit is used to upload the location information and humidity characteristics of the humidity abnormality area to the data fusion module.

[0088] The strategy, steps and algorithm for the humidity detection module to obtain humidity data are as follows:

[0089] Acquisition strategies include:

[0090] Sensor placement: Arrange a humidity sensor network based on the material properties of the building structure, spatial distribution, and possible plant growth paths (such as wall gaps, near pipes, etc.).

[0091] Key monitoring areas: cracks on building surfaces, basements and humid areas to avoid blind spots.

[0092] Real-time monitoring by area: The building structure is divided into several monitoring units, each of which is covered by multiple humidity sensors. The humidity data is monitored in real time and uploaded to the data fusion module.

[0093] The specific steps include:

[0094] Data collection steps:

[0095] Each humidity sensor collects environmental humidity data H in real time i (t), where i represents the sensor number and t represents time; the data sampling frequency is adjusted according to the characteristics of the building environment, such as a higher sampling frequency in high humidity areas.

[0096] Abnormal identification steps:

[0097] By calculating the humidity change rate of each sensor, the specific algorithm formula is as follows:

[0098]

[0099] If the humidity change rate ΔH i If it exceeds the set threshold θ, the area is marked as abnormal.

[0100] Parameter explanation:

[0101] ΔH i : Humidity change rate of the i-th sensor;

[0102] H i (t): humidity value of the i-th sensor at time t;

[0103] H i (t-1): humidity value of the i-th sensor at the previous moment;

[0104] Δt: the time interval between two consecutive samplings;

[0105] θ: Humidity change rate threshold, used to determine whether it is abnormal.

[0106] Steps for analyzing spatial humidity distribution:

[0107] Calculate the average value H of the sensor data in each monitoring unit avg and standard deviation σ H , the specific algorithm formula is as follows:

[0108]

[0109] According to H avg and σ H Analyze humidity distribution characteristics and locate areas with abnormal humidity.

[0110] Parameter explanation:

[0111] H avg : The average humidity in the current monitoring unit;

[0112] σ H : The standard deviation of humidity within the current monitoring unit, used to reflect the unevenness of humidity distribution.

[0113] H i : Humidity data of the i-th sensor;

[0114] n: Number of humidity sensors in the monitoring unit.

[0115] Time series anomaly detection steps:

[0116] The sliding window algorithm is used to perform time series analysis on humidity data to determine whether there are persistent anomalies in the area. The specific algorithm formula is as follows:

[0117]

[0118] If W t >T threshold , then mark the exception.

[0119] Parameter explanation:

[0120] W t : The average humidity value within the time window, used to detect trend anomalies;

[0121] k: The size of the sliding window, indicating the time span of the analysis;

[0122] H i : Humidity data within the sliding window;

[0123] T threshold : Humidity threshold for abnormal judgment.

[0124] Data upload and fusion steps:

[0125] The location information and data features of the humidity anomaly area are passed to the data fusion module and combined with other detection modules for further analysis.

[0126] The infrared thermal imaging module comprises:

[0127] A heat map capture unit, used to scan the abnormal area marked by the humidity detection module and generate a temperature distribution heat map T(x, y) in the area;

[0128] The temperature difference analysis unit is used to calculate the temperature difference ΔT(x, y) of each point in the heat map and calculate the temperature difference according to the set temperature difference threshold θ T Mark temperature anomalies;

[0129] The regional clustering unit is used to cluster adjacent temperature anomaly points to form potential plant growth areas;

[0130] The heat-humidity joint analysis unit is used to cross-validate the temperature anomaly area and the humidity anomaly area, and extract the overlapping part as the plant growth area.

[0131] The positioning strategies of infrared thermal imaging modules include:

[0132] Initial positioning coordination: Based on the abnormal humidity area identified by the humidity detection module, the infrared thermal imaging module limits its detection range to avoid full-area scanning and improve efficiency;

[0133] Thermal anomaly detection: Use thermal imaging equipment to capture thermal images of areas with abnormal humidity, and further locate possible plant growth areas by analyzing temperature distribution and temperature difference changes;

[0134] Multi-point temperature comparison: In the humidity abnormality area, compare and analyze the temperature of different points to find the abnormal temperature rise or drop area (plant growth usually causes local temperature fluctuations).

[0135] The specific positioning steps include:

[0136] Heatmap capture steps:

[0137] Use infrared thermal imaging equipment to scan the humidity abnormality area and generate a temperature distribution heat map T(x, y), where (x, y) is the coordinate point in the heat map.

[0138] Temperature difference calculation steps:

[0139] Compute the temperature difference for each point in the heat map:

[0140] ΔT(x, y) = T(x, y) - T avg

[0141] If |ΔT(x, y)|>θ T (temperature difference threshold), it is marked as a temperature anomaly point.

[0142] Parameter Description:

[0143] T(x, y): the temperature of the point (x, y) in the heat map;

[0144] T avg : Average temperature in the humidity anomaly area;

[0145] θ T : Temperature difference threshold, used to determine temperature anomalies.

[0146] Regional cluster analysis steps:

[0147] Based on the temperature difference results, the abnormal points are regionally clustered to form the potential plant growth region RRR;

[0148] A clustering algorithm is used to classify adjacent temperature anomaly points into the same area. The regional clustering unit is based on the K-Means algorithm to classify adjacent temperature anomaly points into plant growth areas:

[0149]

[0150] Parameter Description:

[0151] k: the number of clustered regions;

[0152] C i : the i-th clustering area;

[0153] μ i |: The center of the i-th region.

[0154] Combined heat and moisture analysis steps:

[0155] The temperature anomaly areas were cross-validated with the humidity anomaly areas, and the overlapping parts were extracted as possible plant growth areas.

[0156] Steps to output positioning results:

[0157] Output the final plant growth area heat map, marking the center coordinates and range of the potential plant growth area.

[0158] The acoustic wave detection module comprises:

[0159] An acoustic wave transmitting unit, used for transmitting acoustic waves in an area with abnormal humidity and temperature;

[0160] The sound wave transmitting unit determines the sound wave propagation path based on the optimized path integral method, as follows:

[0161]

[0162] Parameter Description:

[0163] L min : Optimal propagation path length;

[0164] P 1 , P 2 : The coordinates of the sound wave transmitting point and receiving point;

[0165] z: Path height.

[0166] An acoustic wave receiving unit, used to receive the reflected signal and record the propagation time t and amplitude A of the signal;

[0167] A reflection analysis unit, used to calculate the spatial position and material properties of the reflection point according to the propagation time formula and the attenuation formula;

[0168] The reflection analysis unit determines the acoustic impedance characteristics of the plant root area through the reflection coefficient, and the reflection coefficient is as follows:

[0169]

[0170] Parameter Description:

[0171] R: reflection coefficient;

[0172] Z 1 : Acoustic impedance of sound wave entering the medium;

[0173] Z 2 : Acoustic impedance of sound waves entering the plant root system.

[0174] A feature extraction unit is used to extract the time delay, amplitude change and waveform distortion of the reflected signal to identify the growth activity of the plant root system;

[0175] The joint analysis unit is used to cross-validate the reflection point with the humidity anomaly area and the temperature anomaly area to determine the actual location of the plant roots.

[0176] The steps for locating the acoustic wave detection module include:

[0177] (1) Sound wave transmission and reception

[0178] Arrange multiple acoustic wave transmitting and receiving points (array sensors) in the detection area;

[0179] The emission frequency f and the sound wave wavelength λ are designed according to the acoustic characteristics of building materials and plant roots;

[0180] (2) Signal propagation and reflection analysis

[0181] When sound waves propagate in building materials, their path length L can be described by the following formula:

[0182]

[0183] in,

[0184] (x 1 ,y 1 , z 1 ) represents the coordinates of the sound wave emission point;

[0185] (x 2 ,y 2 , z 2 ) represents the coordinates of the reflection point.

[0186] (3) Calculation of propagation time

[0187] According to the path length L and the speed of sound v, the propagation time formula corresponding to the propagation time t can be expressed as:

[0188]

[0189] Where, v: speed of sound, depends on the acoustic properties of the material.

[0190] (4) Analysis of acoustic wave attenuation characteristics

[0191] The sound wave will experience energy attenuation during propagation, and the attenuation formula is:

[0192] A=A 0 e -αL

[0193] Parameter Description:

[0194] A: The amplitude of the received sound wave;

[0195] A 0 : The amplitude of the sound wave during emission;

[0196] α: Attenuation coefficient, which is material and frequency dependent.

[0197] (5) Reflection signal feature extraction

[0198] The following key features are extracted from the received reflected signal:

[0199] Time delay: The reflection point position is calculated using the propagation time t.

[0200] Amplitude change: Estimate the material properties of the reflection area through the attenuation formula.

[0201] Waveform distortion: Identify the presence of plant roots by analyzing changes in the signal shape.

[0202] (6) Joint analysis

[0203] Compare the reflection points with infrared thermal imaging and areas of humidity anomalies to confirm the actual location and extent of the plant roots.

[0204] Further,

[0205] The joint analysis unit is configured with a plant root distribution model generation strategy, and the plant root distribution model generation strategy includes:

[0206] Data fusion and interpolation processing:

[0207] The coordinate data of humidity, temperature and acoustic wave detection points are fused to define the joint point set P = {(xi, yi, zi, vi)}, where vi represents the multimodal joint weight:

[0208] v i =w h H i +w t Ti +w s S i

[0209] H i : Humidity characteristic value, normalized representation;

[0210] T i : Temperature characteristic value, normalized representation;

[0211] S i : Acoustic wave reflection characteristic value, normalized representation;

[0212] w h 、w t 、w s : The weight parameters of humidity, temperature and acoustic wave features are dynamically adjusted according to the detection accuracy.

[0213] 3D point cloud generation:

[0214] Through data interpolation algorithms (such as cubic spline interpolation or Delaunay triangulation), the detection point set P is converted into a continuous three-dimensional point cloud model:

[0215] PointCloud={(x,y,z,v)}

[0216] Spatial distribution fitting:

[0217] The radial basis function (RBF) is used to perform spatial fitting on the point cloud data to generate the three-dimensional distribution surface S (x, y, z) of the plant root system:

[0218] Voxelization:

[0219] Discretize the 3D surface S(x, y, z) into a voxel grid for visualization and subsequent analysis.

[0220] The data fusion module comprises:

[0221] A data preprocessing unit, used to receive and normalize the data of humidity detection, infrared thermal imaging and acoustic wave detection;

[0222] A position determination unit, used to calculate the center position of the plant growth area and generate a three-dimensional boundary by a weighted distance algorithm;

[0223] Growth status analysis unit, used to calculate the growth activity and root density of plants;

[0224] Impact Assessment Unit for evaluating physical and chemical damage to building structures caused by plants based on the added weight of the root system and the degree of chemical attack.

[0225] Specific steps:

[0226] 1. Data preprocessing:

[0227] Input data:

[0228] Humidity abnormal points output by the humidity detection module;

[0229] The temperature abnormality points output by the infrared thermal imaging module;

[0230] The reflection point and acoustic impedance characteristics output by the acoustic wave detection module;

[0231] Data Normalization:

[0232] Normalize the data of different modes uniformly:

[0233]

[0234] V i : Characteristic value of a point (humidity, temperature, acoustic impedance);

[0235] v min , v max : corresponds to the minimum and maximum values ​​of the mode.

[0236] 2. Data fusion and location determination

[0237] Distance Weighted Algorithm:

[0238] The weighted distance method is used for the abnormal points detected by different modes to calculate the center point of the plant growth area (x c ,y c , z c ):

[0239]

[0240] wi: weight, dynamically adjusted according to the reliability of humidity, temperature and acoustic wave detection modules;

[0241] (X i ,y i , z i ): The coordinates of the i-th outlier point.

[0242] Three-dimensional region fitting:

[0243] The radial basis function (RBF) is used to spatially fit the abnormal points to generate the three-dimensional boundary of the plant root growth area.

[0244] 3. Growth status analysis

[0245] Activity analysis:

[0246] Combine temperature and humidity data to calculate plant growth activity Ax:

[0247] Ax=αH+βT

[0248] H: humidity value;

[0249] T: temperature value;

[0250] α, β: weights of humidity and temperature.

[0251] Root density:

[0252] Based on the density of reflection points detected by acoustic wave, the distribution of roots in the detection area is evaluated D:

[0253]

[0254] N: the number of outliers detected;

[0255] V: region volume.

[0256] 4. Impact assessment on building structures

[0257] Risk of physical damage:

[0258] Calculate the additional root weight W based on root distribution and density r :

[0259] W r =ρ r ·V r

[0260] ρ r : Density of plant roots;

[0261] Vr: volume of plant root system.

[0262] Chemical attack risk:

[0263] Based on the humidity and ultrasonic test results, analyze whether the plant roots may secrete acidic substances (such as oxalic acid) and the concentration of acidic substances C a and distribution range V a :

[0264] E c =C a ·V a β m

[0265] E c : Chemical attack risk;

[0266] C a : Acidic substance concentration;

[0267] V a : Volume of the area affected by acidic substances;

[0268] βm : Tolerance coefficient of building materials (related to material properties).

[0269] The evaluation module is configured with an impact evaluation model and a life evaluation model. The algorithm formula of the impact evaluation model is as follows:

[0270] I=α 1 Ax+α 2 D+α 3 W r +α 4 E c

[0271] in,

[0272] I represents the comprehensive impact index (dimensionless);

[0273] α 1 , α 2 , α 3 , α 4 Indicates the weight of each factor; it is adjusted according to the building type and detection requirements;

[0274] Ax represents plant growth activity;

[0275] D represents the density of the root system;

[0276] W r Indicates the additional weight of the root system;

[0277] E c Indicates the risk of chemical attack;

[0278] The remaining life assessment model of the building structure is adjusted based on the comprehensive impact index. The algorithm formula of the life assessment model is as follows:

[0279] L=L 0 -γ·I

[0280] in,

[0281] L represents the remaining life of the building;

[0282] L 0 Indicates the design life of the building;

[0283] γ represents the life loss factor, (adjusted according to the properties of the building materials and environmental conditions).

[0284] This system is equipped with corresponding detection and evaluation methods, such as Figure 2 As shown, the following steps are included:

[0285] S1: Use the humidity detection module to monitor humidity changes inside the building and identify humidity anomalies caused by plant growth;

[0286] S2: Use infrared thermal imaging modules to detect temperature changes on the surface and inside of buildings and locate plant growth areas;

[0287] S3: The sound wave detection module transmits sound waves and receives reflected signals, analyzes the propagation characteristics of the sound waves, and identifies the growth activities of the plant roots;

[0288] S4: Input the humidity, temperature and sound wave detection data into the data fusion module for comprehensive analysis to identify the plant growth status and its location;

[0289] S5: Evaluate the potential physical and chemical damage of plant growth to the building based on the analysis results and assess the remaining life of the building.

[0290] This solution has the following advantages:

[0291] Precise positioning: Combining humidity, thermal imaging and acoustic wave detection modules, through multimodal data fusion, the specific location of plant growth can be accurately identified.

[0292] Comprehensive status assessment: Comprehensive analysis of plant growth status through activity, density, root weight and chemical erosion risk.

[0293] Scientific Life Prediction: The assessment module scientifically predicts the remaining life of the building structure by quantifying the physical and chemical impact of plants on the building.

[0294] Efficient module linkage: All modules in the system work together to optimize detection efficiency and result accuracy.

[0295] The above-described implementation methods do not constitute a limitation on the protection scope of the technical solution. Any modification, equivalent replacement and improvement made within the spirit and principle of the above-described implementation methods shall be included in the protection scope of the technical solution.

Claims

1. A multi-sensor fusion building structure detection and evaluation system, characterized in that: include: Humidity detection modules, distributed in multiple locations of the building structure, are used to monitor humidity changes inside the building in real time and identify humidity anomalies caused by plant growth; An infrared thermal imaging module, working in conjunction with the humidity detection module, is used to capture the temperature distribution on the surface and inside of the building, and further locate the plant growth area by analyzing the heat changes in the humidity abnormal area; An acoustic wave detection module, linked with the humidity detection module and the infrared thermal imaging module, is used to transmit acoustic waves in the humidity abnormality area and the heat abnormality area and receive reflected signals, and analyze the propagation characteristics of the acoustic waves to confirm the growth activities of the plant roots; A data fusion module is used to receive the detection data of the humidity detection module, the infrared thermal imaging module and the sound wave detection module, and comprehensively judge the growth position and state of the plant and its physical and chemical damage effects on the building structure through fusion analysis; An assessment module, connected to the data fusion module, is used to assess the remaining life of the building based on the physical and chemical damage effects of plant growth on the building structure.

2. The building structure detection and evaluation system of multi-sensor fusion according to claim 1 is characterized in that: The humidity detection module comprises: A humidity sensor network, including humidity sensors arranged at multiple locations of the building structure, for collecting humidity data in real time; A data acquisition unit is used to obtain the humidity change rate of each sensor and mark the area where the humidity change rate exceeds a threshold as an abnormal humidity area; The regional analysis unit is used to calculate the mean and standard deviation of humidity within the monitoring unit and analyze the humidity distribution characteristics; The time series analysis unit is used to perform sliding window calculations on humidity data to determine the persistence of humidity anomaly areas; The data transmission unit is used to upload the location information and humidity characteristics of the humidity abnormality area to the data fusion module.

3. The building structure detection and evaluation system of multi-sensor fusion according to claim 2 is characterized in that: The infrared thermal imaging module comprises: A heat map capture unit, used to scan the abnormal area marked by the humidity detection module and generate a heat map of the temperature distribution in the area; The temperature difference analysis unit is used to calculate the temperature difference of each point in the thermal map and mark the temperature abnormal points according to the set temperature difference threshold; The regional clustering unit is used to cluster adjacent temperature anomaly points to form potential plant growth areas; The heat-humidity joint analysis unit is used to cross-validate the temperature anomaly area and the humidity anomaly area, and extract the overlapping part as the plant growth area.

4. The building structure detection and evaluation system of multi-sensor fusion according to claim 3 is characterized in that: The acoustic wave detection module comprises: An acoustic wave transmitting unit, used for transmitting acoustic waves in an area with abnormal humidity and temperature; An acoustic wave receiving unit, used to receive the reflected signal and record the propagation time and amplitude of the signal; A reflection analysis unit, used to calculate the spatial position and material properties of the reflection point according to the propagation time formula and the attenuation formula; A feature extraction unit is used to extract the time delay, amplitude change and waveform distortion of the reflected signal to identify the growth activity of the plant root system; The joint analysis unit is used to cross-validate the reflection point with the humidity anomaly area and the temperature anomaly area to determine the actual location of the plant roots.

5. The building structure detection and evaluation system of multi-sensor fusion according to claim 4 is characterized in that: The joint analysis unit is configured with a plant root distribution model generation strategy, and the plant root distribution model generation strategy includes: Data fusion step: receiving the data of humidity detection, infrared thermal imaging and acoustic wave detection to form a joint point set; Three-dimensional point cloud generation steps: converting the joint point set into a three-dimensional point cloud model of the plant root system through an interpolation algorithm; Three-dimensional surface fitting steps: spatial fitting of point cloud data based on radial basis function; Voxelization processing step: discretize the three-dimensional surface into a voxel grid to show the spatial distribution of plant roots.

6. The building structure detection and evaluation system of multi-sensor fusion according to claim 4 is characterized in that: The data fusion module comprises: A data preprocessing unit, used to receive and normalize the data of humidity detection, infrared thermal imaging and acoustic wave detection; A position determination unit, used to calculate the center position of the plant growth area and generate a three-dimensional boundary by a weighted distance algorithm; Growth status analysis unit, used to calculate the growth activity and root density of plants; Impact Assessment Unit for evaluating physical and chemical damage to building structures caused by plants based on the added weight of the root system and the degree of chemical attack.

7. The building structure detection and evaluation system of multi-sensor fusion according to claim 6 is characterized in that: The evaluation module is configured with an impact evaluation model and a life evaluation model. The algorithm formula of the impact evaluation model is as follows: I=α1Ax+α2D+α3W r +α4E c in, I represents the comprehensive impact index (dimensionless); α1, α2, α3, and α4 represent the weights of each factor; Ax represents plant growth activity; D represents the density of the root system; W r Indicates the additional weight of the root system; E c Indicates the risk of chemical attack; The remaining life assessment model of the building structure is adjusted based on the comprehensive impact index. The algorithm formula of the life assessment model is as follows: L=L0-γ·I in, L represents the remaining life of the building; L0 represents the building design life; γ represents the life loss coefficient.

8. A multi-sensor fusion building structure detection and evaluation method, applied to the multi-sensor fusion building structure detection and evaluation system according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1: Use the humidity detection module to monitor humidity changes inside the building and identify humidity anomalies caused by plant growth; S2: Use infrared thermal imaging modules to detect temperature changes on the surface and inside of buildings and locate plant growth areas; S3: The sound wave detection module transmits sound waves and receives reflected signals, analyzes the propagation characteristics of the sound waves, and identifies the growth activities of the plant roots; S4: Input the humidity, temperature and sound wave detection data into the data fusion module for comprehensive analysis to identify the plant growth status and its location; S5: Evaluate the potential physical and chemical damage of plant growth to the building based on the analysis results and assess the remaining life of the building.

Images