Damage monitoring auxiliary system based on self-sensing concrete

Abstract

The invention provides a damage monitoring auxiliary system based on self-sensing concrete, which relates to the technical field of concrete damage monitoring and comprises a self-sensing concrete hardware module, a data acquisition and processing module, an environment compensation and correction module, a data decoupling module and a data analysis and prediction module. According to the invention, the layout of the sensing network is optimized, a high-density microelectrode grid is pre-embedded in self-sensing concrete, the microelectrodes can not only increase signal acquisition points, but also effectively reduce the problem of signal attenuation, and at the same time, the electromagnetic suppression sub-module is matched, so that the resistance to local electromagnetic interference is enhanced, and the self-sensing performance of the sensor is improved. Even in a complex construction environment, good signal stability can still be kept, so that high-precision and high-robustness damage monitoring is realized by optimizing the layout of the sensing network, enhancing the anti-electromagnetic interference capability, establishing a temperature and humidity compensation model and combining an intelligent signal processing algorithm; therefore, the monitoring efficiency and the engineering applicability of the self-sensing concrete in a complex environment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of concrete damage monitoring, and in particular to a damage monitoring auxiliary system based on self-sensing concrete. Background Art

[0002] With the continuous development of modern engineering technology, structural health monitoring (SHM) has become one of the key technologies to ensure the safety of infrastructure, extend its service life and reduce maintenance costs. In structural health monitoring systems, traditional damage monitoring methods usually rely on external sensors (such as strain gauges, accelerometers, displacement sensors, etc.) to monitor the force, displacement or deformation of the structure. These sensors usually need to be installed on the surface or inside the structure and connected to the data acquisition system through wires. However, traditional sensors have some inherent disadvantages. Especially in large-scale monitoring and long-term use, the layout, maintenance and calibration of sensors are often very cumbersome and increase the cost and complexity of the system.

[0003] To overcome the limitations of traditional sensor systems, self-sensing concrete (SSC) has emerged in recent years as a new structural health monitoring technology. Self-sensing concrete incorporates conductive materials (such as carbon fibers, carbon nanotubes, and steel fibers) into conventional concrete, imparting self-sensing resistance. This allows the concrete itself to function as a sensor, enabling real-time monitoring of its health. This approach not only improves monitoring sensitivity and efficiency but also reduces the need for external sensor installation, providing a new solution for structural health monitoring.

[0004] However, its practical application still faces many technical bottlenecks and challenges. For example, the uneven dispersion of conductive materials in the concrete matrix causes the resistance signal to decay exponentially with transmission distance, seriously affecting the monitoring range and sensitivity. Secondly, the complex environmental electromagnetic interference on the construction site (such as large machinery, wireless communication equipment, etc.) makes the weak damage signal easily drowned out by noise, and the measured signal-to-noise ratio is generally less than 5dB, which greatly increases the difficulty of signal extraction. In addition, the resistance drift caused by fluctuations in ambient temperature and humidity can reach 300% to 500% of the damage signal, resulting in poor baseline stability and difficulty in distinguishing real damage from environmental interference. To complicate matters further, the coupling of multiple factors such as load, corrosion, and freeze-thaw in actual engineering projects can lead to severe aliasing of characteristic signals. Traditional methods have difficulty in effectively decoupling and identifying damage characteristics. Therefore, the present invention proposes a damage monitoring auxiliary system based on self-sensing concrete to address the problems existing in the prior art. Summary of the Invention

[0005] In response to the above problems, the purpose of the present invention is to propose a damage monitoring auxiliary system based on self-sensing concrete. This damage monitoring auxiliary system based on self-sensing concrete has the advantages of improving the monitoring efficiency of self-sensing concrete in complex environments and can solve the problems existing in the existing technology.

[0006] To achieve the purpose of the present invention, the present invention is implemented through the following technical solutions: a self-sensing concrete-based damage monitoring auxiliary system, including a self-sensing concrete hardware module for collecting, processing and transmitting data related to the interior of concrete and the environment in real time through a high-density micro-electrode grid;

[0007] The data acquisition and processing module is used to collect the data collected in the self-sensing concrete hardware module and perform pre-processing;

[0008] Environmental compensation correction module, used to establish a temperature and humidity compensation model and correct the processed data through the temperature and humidity compensation model;

[0009] Data decoupling module, used to decouple the influence of multiple factors in the corrected data;

[0010] The data analysis and prediction module is used to analyze and predict the decoupled data using a convolutional neural network to achieve self-perceiving concrete damage monitoring.

[0011] A further improvement is that the self-sensing concrete hardware module includes a distributed microelectrode submodule for setting up a high-density microelectrode grid inside the self-sensing concrete, an electromagnetic suppression submodule for electromagnetic shielding and reducing electromagnetic interference, a signal enhancement submodule for enhancing the high-density microelectrode grid signal, a temperature and humidity sensor submodule for collecting environmental parameters inside the self-sensing concrete, and a data transmission submodule for sending data through wireless communication technology.

[0012] A further improvement is that the data acquisition and processing module includes a data processing submodule for collecting and processing data of the self-sensing concrete hardware module and a data storage submodule for storing the processed data.

[0013] A further improvement is that the processing flow of the data processing submodule is as follows:

[0014] S1; The collected data is subjected to denoising processing using wavelet transform method and then normalized;

[0015] S2: Extract features from the data processed by S1 to obtain the required feature data;

[0016] S3: Generate a corresponding timestamp for the extracted feature data, and then transmit it to the data storage submodule for storage.

[0017] A further improvement is that the environmental compensation correction module includes a temperature and humidity compensation model establishment submodule for establishing a compensation model according to changes in environmental temperature and humidity, and a data correction submodule for correcting data using the compensation model.

[0018] A further improvement is that the specific steps of establishing a compensation model according to changes in ambient temperature and humidity are:

[0019] SS1: Collect experimental data on temperature, humidity, and resistance changes to build a training dataset;

[0020] SS2: Clean and standardize the acquired data;

[0021] SS3: Use the support vector regression algorithm to establish a nonlinear mapping relationship between resistance change and temperature and humidity, and use the training data set to train the compensation model;

[0022] SS4: Use the ten-fold cross-validation method to evaluate the model, and complete the establishment of the compensation model after the evaluation is qualified.

[0023] A further improvement is that the data decoupling module includes a decoupling processing submodule for separating different influencing factors mixed in the data by using a variational mode decomposition method and a data fusion submodule for fusing the decoupled data.

[0024] A further improvement is that the data analysis and prediction module includes a machine learning construction submodule for constructing a damage analysis and prediction model based on a convolutional neural network, a damage prediction submodule for predicting damage to self-sensing concrete through the damage analysis and prediction model, and a visualization display submodule for visualizing the analysis and prediction results.

[0025] A further improvement is that it also includes a system management platform module for remote configuration, monitoring, updating and user interaction.

[0026] The beneficial effects of the present invention are:

[0027] (1) The present invention optimizes the sensor network layout and pre-buries a high-density microelectrode grid in the self-sensing concrete. These microelectrodes can not only increase the signal collection points, but also effectively solve the signal attenuation problem. At the same time, in conjunction with the electromagnetic suppression submodule, the present invention enhances its resistance to local electromagnetic interference and can maintain good signal stability even in complex construction environments.

[0028] (2) The present invention provides a high-precision and highly robust damage monitoring solution by combining a temperature and humidity compensation model with a machine learning algorithm. By effectively compensating for environmental factors, accurately analyzing and processing signals, and using machine learning for intelligent damage prediction, the system can provide stable and accurate monitoring results in complex environments, significantly improving the application of self-sensing concrete in various projects. As a result, the present invention not only has high engineering applicability, but also can significantly reduce maintenance costs and improve the efficiency and safety of structural health monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a schematic diagram of the module flow of the present invention.

[0030] Figure 2 2 is a schematic diagram of the damage type classification of the present invention.

[0031] Figure 3 It is a schematic diagram of the analysis process of the data analysis and prediction module of the present invention.

[0032] Figure 4 It is a structural schematic diagram of the distributed microelectrode submodule of the present invention. DETAILED DESCRIPTION

[0033] In order to deepen the understanding of the present invention, the present invention will be further described in detail below with reference to the examples. The examples are only used to explain the present invention and do not constitute a limitation on the scope of protection of the present invention.

[0034] The performance of self-sensing concrete depends on the dispersion of conductive materials. Common conductive materials such as carbon fibers, carbon nanotubes, and steel fibers need to be evenly distributed within the concrete matrix to ensure that the concrete's electrical resistance characteristics accurately reflect the structural health. However, in practical applications, the uneven dispersion of conductive materials within the concrete matrix causes the concrete's electrical resistance signal to decay exponentially with transmission distance. This attenuation phenomenon severely impacts the effective monitoring range and sensitivity of the monitoring system, especially when applied to large areas or complex structures. The attenuation of the resistance signal means that only relatively reliable data can be obtained in a localized area, while areas far from the sensor may not be able to obtain accurate health information due to signal attenuation, thereby reducing the overall effectiveness of the monitoring system.

[0035] In actual engineering, concrete structures are often exposed to complex environments, which include a large number of electromagnetic interference sources, such as large machinery, wireless communication equipment, and power facilities. The operation of these devices generates electromagnetic waves, which interfere with weak resistance signals, causing the signals to be submerged by noise. This problem is particularly prominent in field applications, especially in large-scale projects in urban construction or transportation infrastructure, where signal stability and accuracy are difficult to guarantee. Traditional self-sensing concrete technology has a relatively weak signal itself, coupled with the influence of electromagnetic interference, resulting in a measured signal-to-noise ratio generally lower than 5dB. This means that even if the signal comes from real damage, it may be submerged by noise, making it difficult to perform effective signal extraction and analysis. Therefore, when applying self-sensing concrete for damage monitoring in complex environments, the signal processing capabilities of traditional methods are limited, and efficient damage detection and real-time monitoring cannot be achieved.

[0036] Fluctuations in ambient temperature and humidity are significant factors affecting concrete resistance. As temperature and humidity fluctuate, the electrical conductivity within concrete changes significantly, affecting the resistance signal. This resistance drift caused by temperature and humidity fluctuations can reach 300% to 500% of the damage signal, degrading baseline stability and making it difficult to distinguish resistance changes caused by environmental changes from actual damage signals. Traditional self-sensing concrete monitoring systems often rely on simple calibration or baseline adjustment methods to address this type of environmental drift. These methods are unable to effectively eliminate the impact of environmental changes on the resistance signal, thus compromising the reliability and accuracy of monitoring results.

[0037] In practical engineering, concrete structures are often affected by multiple factors simultaneously, such as load, creep, corrosion, and freeze-thaw cycles. These factors can cause changes in the concrete's electrical resistance signal, and these changes can overlap, resulting in complex signal aliasing. Traditional monitoring methods often struggle to effectively decouple the influence of these multiple factors, complicating signal interpretation and analysis. For damage monitoring systems, signal aliasing makes it difficult to separate damage signatures from environmental and non-damaging factors, thereby reducing the system's ability to accurately identify and locate concrete damage.

[0038] Based on this Figures 1-4 As shown, this embodiment proposes a self-sensing concrete damage monitoring auxiliary system, including a self-sensing concrete hardware module for collecting, processing, and transmitting data related to the interior of concrete and the environment in real time through a high-density micro-electrode grid, including:

[0039] The distributed microelectrode submodule is used to set up a high-density microelectrode grid inside self-sensing concrete. Specifically, it is composed of a high-density microelectrode grid (made of alkali-resistant carbon nanofibers / stainless steel microfilaments) embedded in the concrete, distributed in a matrix form. It can then provide key data for structural health monitoring through real-time measurement of resistance / impedance between electrodes.

[0040] The electromagnetic suppression submodule is used for electromagnetic shielding and reducing electromagnetic interference. It consists of a shielding layer and a filter circuit. The shielding layer is a copper-nickel alloy metal mesh wrapped around the electrode grid. The electrode density is ≥4 / m 2 The filtering circuit is integrated into the signal acquisition unit's bandpass filter (frequency band 1kHz–10MHz), suppressing external power-frequency electromagnetic interference (such as power grids and communication equipment) and stray current noise from steel bars in concrete, ensuring a signal-to-noise ratio of >30dB.

[0041] The signal enhancement submodule, designed to enhance the signal from the high-density microelectrode grid, amplifies weak electrical signals (μV to mV) from the microelectrodes and eliminates ambient noise through phase-sensitive detection. It consists of a preamplifier, a low-noise instrumentation amplifier (gain 100 to 1000 times, input impedance >1 GΩ), and a lock-in amplifier. The preamplifier applies a 10mV–1V AC excitation signal (frequency 1 to 100kHz) to the electrodes and synchronously demodulates the output signal.

[0042] The temperature and humidity sensor module is used to collect self-perceived internal environmental parameters of concrete. It is an integrated temperature and humidity sensor (accuracy: ±0.5°C, ±2%RH). It is co-located with the microelectrode grid and its function is to collect real-time internal concrete temperature (-20 to 80°C) and humidity (0–100%RH) data to provide input for environmental compensation.

[0043] The data transmission submodule sends data through wireless communication technology. It uses a low-power microprocessor (ARMCortex-M4) and has a wireless transmission unit (LoRaWAN / NB-IoT dual-mode module, transmission distance >1km). When working, the microprocessor aggregates electrode data, environmental data and device status (battery voltage, signal quality), compresses the data into binary frames (CRC check), and then uploads it to the cloud gateway through the wireless network.

[0044] The data acquisition and processing module is used to collect data collected by the self-sensing concrete hardware module and perform pre-processing, which includes:

[0045] The data processing submodule is used to collect and process data from the self-sensing concrete hardware module. The processing flow of the data processing submodule is as follows:

[0046] S1: The collected data is denoised using the wavelet transform method and then normalized. Specifically, the collected data includes the resistance / impedance time series signal of the microelectrode grid (the signal collected by the electrode reflects the change in conductivity inside the concrete, which helps to detect damage such as cracks), the original data of the temperature and humidity sensor (humidity and temperature are important factors affecting the performance of concrete and are directly related to its structural health) and the equipment status code (signal quality mark, battery voltage, used to evaluate the working status of the equipment and the reliability of signal acquisition). The collected raw signal is then denoised using the wavelet transform method. The processed data will be standardized through normalization, compressing the electrode signal to [0,1], and retaining the physical units of temperature and humidity (℃, %RH). This unifies the numerical range of each feature to avoid unnecessary deviations in subsequent analysis of data of different dimensions;

[0047] S2: Feature extraction is performed on the data processed by S1 to obtain the required feature data. Feature data that is helpful for evaluating the health status of concrete is extracted from the processed data. Specifically, the data types are electrode spatial signals (featured by regional average resistance value, resistance gradient extreme value, local variance (reflecting cracks)), electrode time series signals (featured by resistance change rate within 1 hour, frequency domain principal component energy (FFT extracts 0.1-10Hz frequency band)), temperature and humidity data (temperature and humidity mean (10-minute window), temperature and humidity change slope) and equipment status (signal packet loss rate, battery voltage drop trend). Furthermore, the spatial features are calculated by Delaunay triangulation based on the electrode coordinates to calculate the grid unit resistance statistics, while the time series features are calculated by sliding window (length 60 seconds, step size 10 seconds) to calculate the change rate and FFT spectrum;

[0048] S3: Generate a corresponding timestamp for the extracted feature data (using GPS / Beidou dual-mode timing (accuracy ±1ms)), and then transmit it to the data storage submodule for storage.

[0049] The average resistance value of the above region can reflect the overall resistance characteristics of the concrete, the extreme value of the resistance gradient can help identify the presence and development of cracks, and the local variance reflects the characteristics of local cracks in the concrete. The above resistance change rate within 1 hour reflects the rate of change of resistance over time. The frequency domain principal component energy is extracted by FFT (Fast Fourier Transform) in the 0.1-10 Hz frequency band to help analyze the frequency characteristics of resistance. The above temperature and humidity mean reflects the average change in temperature and humidity over a short period of time. The temperature and humidity change slope reflects the speed of temperature and humidity change, which helps analyze the environmental response characteristics of concrete. The above signal packet loss rate reflects the stability of data transmission and helps determine whether there is abnormal packet loss. The battery voltage decline trend reflects the health of the battery.

[0050] The data storage submodule used to store processed data is divided into an edge cache unit and a central database unit. The edge cache unit is a data hard disk group column that stores 7 days of original data in a cyclical manner, while the central database unit is deployed in a cloud server to store the extracted time series feature data.

[0051] The environmental compensation correction module is used to establish a temperature and humidity compensation model and correct the processed data using the temperature and humidity compensation model. It includes:

[0052] The temperature and humidity compensation model establishment submodule is used to establish a compensation model according to the changes in ambient temperature and humidity. The specific steps for establishing a compensation model according to the changes in ambient temperature and humidity are as follows:

[0053] SS1: Collect experimental data on temperature, humidity, and resistance variation to construct a training dataset. Specifically, accelerated testing is conducted on standard concrete specimens (with the same mix ratio) in a climate simulation chamber, covering the full range of temperature (-20°C to 60°C) and humidity (20% to 100% RH). Then, in the field, reference nodes are deployed on healthy structures (undamaged areas) to collect resistance baseline data under natural temperature variations. The corresponding dataset structure includes temperature, humidity, resistance variation, and concrete age.

[0054] SS2: The acquired data were cleaned and standardized. The cleaning rule was to remove data with resistance changes exceeding ±3 times the standard deviation (outliers), and then linear interpolation was used to interpolate missing values.

[0055] SS3: Use the support vector regression algorithm to establish a nonlinear mapping relationship between resistance change and temperature and humidity. Use the training data set to train the compensation model. The input data are temperature, humidity, and age, and the output data is the predicted resistance change.

[0056] SS4: Use the ten-fold cross validation method to evaluate the model. After the evaluation is qualified, the compensation model is established. Specifically, the training data set is divided into 10 subsets, 9 training + 1 validation, and the cycle is repeated 10 times. The evaluation indicators are mean absolute error MAE ≤ 0.05Ω and determination coefficient R 2 ≥0.95, based on field conditions, continuous monitoring on healthy structures for 30 days, and the resistance fluctuation rate after compensation is ≤2% (verification of model generalization ability).

[0057] The data correction submodule is used to correct the data through the compensation model. It inputs the real-time value of the resistance after preprocessing, as well as the current temperature and humidity and concrete age, and outputs the corrected resistance value through the compensation model.

[0058] Furthermore, in the present invention, the environmental compensation correction module achieves compensation accuracy of: temperature interference suppression rate ≥ 90% (error < 0.03Ω in the range of -20°C to 60°C); humidity interference suppression rate ≥ 85% (error < 0.05Ω in the range of 20% to 100% RH). The compensated resistance variance of the healthy reference node is calculated monthly. If it exceeds 150% of the initial value for three consecutive times, the model is retrained, automatically collecting data under the new environmental conditions and dynamically updating the model parameters using the online SVR algorithm.

[0059] The above involves the age of concrete. During the concrete pouring process, construction workers will record the time when concrete pouring begins. This is the starting point of the concrete age. When conducting inspections or analyses, the difference between the current time and the pouring time is used to calculate the age.

[0060] The data decoupling module is used to decouple the influence of multiple factors in the corrected data. Its function is to separate the mixed multiple influencing factors (load, creep, corrosion, etc.) in the resistance data after environmental compensation and extract the pure damage characteristic signal. It includes:

[0061] The decoupling submodule uses variational mode decomposition (VMD) to separate the various influencing factors in the data. Specifically, its input data is the environmentally compensated resistance time series signal. It then determines the VMD parameters for decomposition. These parameters include the number of decomposition modes, K, which determines the number of intrinsic mode functions (IMFs) into which the signal is decomposed; a balance parameter, which controls the accuracy and complexity of the signal decomposition; and tolerance and error tolerance level parameters, which control the computational progress during the decomposition process.

[0062] The output is then K intrinsic mode functions (IMFs), each representing a portion of the signal's frequency components or time-domain characteristics. Some IMFs may contain interference signals unrelated to environmental changes (such as corrosion and load), while others may be related to damage characteristics.

[0063] The data fusion submodule, used to fuse the decoupled data, takes as input the K intrinsic mode functions (IMFs) decomposed from the decomposition. It selects the damage-related IMFs, typically exhibiting frequency ranges and temporal fluctuations consistent with damage characteristics, such as resistance changes caused by concrete cracks or corrosion. This function can be one or more, with the remaining components treated as interference. Signal reconstruction is then performed, performing a time-frequency energy-weighted fusion of the selected IMFs to obtain the final, corrected damage signal. This reconstructed signal eliminates interference from factors such as load, creep, or corrosion, extracting a pure damage signature signal.

[0064] Damage characteristic signal x r The calculation formula is:

[0065]

[0066] Where u i is the selected damage-related IMF, w i is the corresponding weighting coefficient, M is the number of IMFs related to the damage, and i = 1 means starting from the first IMF. The reconstructed damage signal can be used for subsequent damage analysis and structural health monitoring.

[0067] The data analysis and prediction module is used to analyze and predict the decoupled data using a convolutional neural network to achieve self-perceiving concrete damage monitoring, including:

[0068] The machine learning submodule for building a damage analysis and prediction model based on a convolutional neural network includes the following steps:

[0069] A1: Construct a data set. The data source of the data set is consistent with the data source used to establish the compensation model for changes in ambient temperature and humidity.

[0070] A2: Convert the distributed electrode signals of the distributed microelectrode submodule into a two-dimensional matrix to form an "electrode signal image" and label it. That is, each piece of data should correspond to a damage status label.

[0071] A3: Build a damage analysis and prediction model based on a convolutional neural network and perform model training using the Adam optimizer.

[0072] The damage prediction submodule is used to predict the damage of self-sensing concrete through the damage analysis and prediction model. Its input receives the damage signal output by the decoupling module, analyzes it through the damage analysis and prediction model, and determines the damage type (no damage, crack, corrosion and freeze-thaw) according to the one with the highest probability. Then, the damage degree is determined by rounding off (1 / 2 / 3 levels, 3 is the highest), that is, Figure 2 shown.

[0073] The visualization display submodule is used to visualize the analysis and prediction results, with the output in the form of a web interface (B / S architecture) and a mobile APP (Android / iOS).

[0074] It also includes a system management platform module for remote configuration, monitoring, updating and user interaction.

[0075] Using a traditional sensor system as a reference, this system was compared to the current one. The model accuracy was tested in a coupled environment of salt-freeze cycling, electrochemical corrosion, and fatigue loading. The system was then deployed in a bridge overpass or tunnel lining section, and compared to the traditional sensor data. The results are shown in Table 1 below:

[0076] Target Traditional sensor systems This system Signal transmission stability >10cm severe attenuation <20cm attenuation≤5% Environmental drift ratio 300％～500％ ≤20％ Impairment SNR <h2 style=";text-align:left;direction:ltr"><5dB<h2 style=";text-align:left;direction:ltr"> >15dB Multi-factor recognition accuracy Unable to decouple >85% (four factors) Crack positioning accuracy ±50cm ±10cm

[0077] Table 1

[0078] As shown in Table 1 above, this system significantly outperforms traditional sensor systems in several key metrics. In particular, it demonstrates significant advantages in signal transmission stability, environmental drift ratio, damage signal-to-noise ratio, multi-factor identification accuracy, and crack location accuracy. These performance improvements enable the self-sensing concrete system to provide more reliable and accurate monitoring results in environments with coupled salt-freeze cycles, electrochemical corrosion, and fatigue loads, significantly improving the efficiency and accuracy of structural health monitoring.

[0079] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above-described embodiments. The above-described embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and improvements may be made to the present invention without departing from the framework and scope of application of the present invention. Such changes and improvements are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A self-sensing concrete damage monitoring auxiliary system, characterized by: Includes a self-sensing concrete hardware module for collecting, processing, and transmitting real-time data related to the concrete interior and its environment through a high-density micro-electrode grid; The data acquisition and processing module is used to collect the data collected in the self-sensing concrete hardware module and perform pre-processing; Environmental compensation correction module, used to establish a temperature and humidity compensation model and correct the processed data through the temperature and humidity compensation model; Data decoupling module, used to decouple the influence of multiple factors in the corrected data; The data analysis and prediction module is used to analyze and predict the decoupled data using a convolutional neural network to achieve self-perceiving concrete damage monitoring.

2. The self-sensing concrete damage monitoring auxiliary system according to claim 1 is characterized by: The self-sensing concrete hardware module includes a distributed microelectrode submodule for setting up a high-density microelectrode grid inside the self-sensing concrete, an electromagnetic suppression submodule for electromagnetic shielding and reducing electromagnetic interference, a signal enhancement submodule for enhancing the high-density microelectrode grid signal, a temperature and humidity sensor submodule for collecting environmental parameters inside the self-sensing concrete, and a data transmission submodule for sending data through wireless communication technology.

3. The self-sensing concrete damage monitoring auxiliary system according to claim 1 is characterized by: The data acquisition and processing module includes a data processing submodule for collecting and processing data of the self-sensing concrete hardware module and a data storage submodule for storing the processed data.

4. The self-sensing concrete damage monitoring auxiliary system according to claim 3 is characterized by: The processing flow of the data processing submodule is as follows: S1; The collected data is subjected to denoising processing using wavelet transform method and then normalized; S2: Extract features from the data processed by S1 to obtain the required feature data; S3: Generate a corresponding timestamp for the extracted feature data, and then transmit it to the data storage submodule for storage.

5. The self-sensing concrete damage monitoring auxiliary system according to claim 1 is characterized in that: The environmental compensation and correction module includes a temperature and humidity compensation model establishment submodule for establishing a compensation model according to changes in environmental temperature and humidity, and a data correction submodule for correcting data using the compensation model.

6. The self-sensing concrete damage monitoring auxiliary system according to claim 5 is characterized by: The specific steps of establishing the compensation model according to the changes in ambient temperature and humidity are as follows: SS1: Collect experimental data on temperature, humidity, and resistance changes to build a training dataset; SS2: Clean and standardize the acquired data; SS3: Use the support vector regression algorithm to establish a nonlinear mapping relationship between resistance change and temperature and humidity, and use the training data set to train the compensation model; SS4: Use the ten-fold cross-validation method to evaluate the model, and complete the establishment of the compensation model after the evaluation is qualified.

7. The self-sensing concrete damage monitoring auxiliary system according to claim 1 is characterized by: The data decoupling module includes a decoupling processing submodule for separating different influencing factors mixed in the data by using a variational mode decomposition method and a data fusion submodule for fusing the decoupled data.

8. The self-sensing concrete damage monitoring auxiliary system according to claim 1 is characterized by: The data analysis and prediction module includes a machine learning construction submodule for constructing a damage analysis and prediction model based on a convolutional neural network, a damage prediction submodule for predicting damage to self-sensing concrete through the damage analysis and prediction model, and a visualization display submodule for visualizing the analysis and prediction results.

9. The self-sensing concrete damage monitoring auxiliary system according to claim 1 is characterized in that: It also includes a system management platform module for remote configuration, monitoring, updating and user interaction.

Images