An effective evaluation method for dynamic repairing effect of microbial self-repairing concrete

Abstract

This invention provides an effective evaluation method for the dynamic repair effect of microbial self-healing concrete, relating to the field of intelligent building material performance monitoring technology. An effective evaluation system for the dynamic repair effect of microbial self-healing concrete includes a monitoring unit, a time-series image acquisition unit, a data acquisition and transmission unit, and a dynamic repair analysis platform. The monitoring unit is a multimodal sensor array embedded inside the concrete specimen. The sensor array components include distributed fiber optic strain sensors, embedded micro-stress sensors, temperature and humidity sensors, pH sensors, and electrochemical impedance spectroscopy probes. Through non-destructive, automated dynamic time-series monitoring of concrete, and by integrating multi-source parameters to construct a multi-dimensional evaluation system, this method avoids the shortcomings of traditional static testing, destructive sampling, single evaluation methods, and human error. It accurately quantifies the repair performance of microbial concrete, resulting in highly reliable evaluation results.

Description

Technical Field

[0001] This invention relates to the field of intelligent building material performance monitoring technology, specifically an effective evaluation method for the dynamic repair effect of microbial self-healing concrete. Background Technology

[0002] Microbial self-healing concrete, as a type of intelligent building composite material with self-healing capabilities, mainly relies on microbial capsules embedded inside the concrete. When cracks are generated and come into contact with external water vapor and oxygen, the microbial activity is activated, inducing calcium ions to mineralize and deposit, forming calcium carbonate deposits. This enables the autonomous closure and repair of micro and macro cracks inside the concrete, significantly improving the durability of concrete structures and extending their service life. It is widely used in complex engineering scenarios such as geotechnical engineering, roads and bridges, and hydraulic structures.

[0003] Currently, the evaluation system for the repair effect of microbial self-healing concrete is not perfect, and existing testing technologies have significant shortcomings, making it difficult to accurately characterize the actual repair evolution process of concrete. Specific technical deficiencies are as follows:

[0004] First, the evaluation methods are static and fixed. Existing evaluations mostly use offline testing methods such as penetration testing and strength testing, which can only obtain the repair endpoint parameters at specific time points. They cannot continuously capture the entire process of microbial mineralization, make it difficult to characterize the temporal changes of stress, cracks, and chemical environment, and cannot accurately determine key dynamic indicators such as repair delay period and repair rate.

[0005] Secondly, the testing methods are destructive. Traditional mechanical testing and cross-section characterization require the specimen to be broken, and the specimen cannot be reused after testing, making it impossible to continuously monitor the same sample throughout the entire cycle. At the same time, there are significant individual differences in the specimens, the testing cost is high, and the repeatability of the monitoring data is poor.

[0006] Third, the evaluation indicators are too simplistic and one-sided. Most existing evaluations rely on single indicators such as crack closure rate and strength recovery rate, ignoring the multi-field coupling effects of internal concrete stress, electrochemistry, temperature and humidity, pH and microbial activity. They do not combine mechanical, microscopic and chemical parameters, resulting in low reliability of the evaluation results.

[0007] Fourth, the level of automation in detection is low. At present, crack observation relies heavily on intermittent manual sampling, with no fixed observation cycle and large human error; moreover, traditional equipment is difficult to collect multi-source data such as mechanical, environmental and image data simultaneously, resulting in poor data synchronization and the inability to build a standardized time-series repair database. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides an effective evaluation method for the dynamic repair effect of microbial self-healing concrete, solving the problems of static and rigid evaluation methods, destructive testing methods, one-sided and limited evaluation indicators, and low degree of automation in testing.

[0009] To achieve the above objectives, the present invention provides the following technical solution: an effective evaluation system for the dynamic repair effect of microbial self-healing concrete, comprising a monitoring unit, a time-series image acquisition unit, a data acquisition and transmission unit, and a dynamic repair analysis platform.

[0010] The monitoring unit is a multimodal sensor array embedded inside the concrete specimen. The sensor array consists of distributed fiber optic strain sensors, embedded micro-stress sensors, temperature and humidity sensors, pH sensors, and electrochemical impedance spectroscopy probes. Multiple types of sensors work together to collect multi-dimensional parameters of the concrete's internal mechanical, environmental, and electrochemical properties.

[0011] The time-series image acquisition unit includes a high-speed camera equipped with a microscope lens. The camera and the macro-crack scanner, two types of image acquisition devices, are mounted together on an adjustable bracket on the surface of the specimen. The angle and height can be flexibly adjusted by the adjustable bracket, so as to realize the timed and automated shooting of micro and macro crack morphology and completely preserve the image data of the entire process of crack evolution.

[0012] The data acquisition and transmission unit operates with low power consumption. The controller is the core control component. The controller's communication terminal is electrically connected to various sensors and a time-series image acquisition unit, integrating the acquired sensor data and image data, and then... The dual communication modules will upload integrated data to the cloud server in real time, ensuring the stability and timeliness of data transmission;

[0013] The dynamic repair analysis platform is deployed inside a cloud server. The platform has a built-in multi-source data fusion engine, which relies on the intelligent algorithms on the engine to complete time series damage inversion calculations and comprehensive evaluation of concrete repair efficiency, thereby realizing dynamic analysis of the repair process.

[0014] Preferably, the embedded micro-stress sensor is equipped with a micro-hole fixing base, which is made of epoxy resin-quartz composite material;

[0015] This micro-stress sensor is embedded in the pre-defined crack prediction path of the concrete using a microporous fixing base, ensuring that the sensor is positioned to fit the damage evolution area. At the same time, the sensor is covered with a microbial-compatible permeable membrane with a pore size of no more than 0.5μm, which allows water vapor and ions to pass through while blocking impurities from entering and avoiding interference with the microbial living environment. This embedded micro-stress sensor has a detection accuracy of ±0.01MPa, accurately capturing subtle stress changes inside the concrete.

[0016] Preferably, the time-series image acquisition unit is equipped with an ultra-depth-of-field microscope lens group, and the lens magnification can be steplessly adjusted within the range of 200× to 1000×;

[0017] This lens assembly, paired with an image processing module featuring a built-in AI crack boundary recognition algorithm, can automatically identify and calibrate crack boundary contours, generate dynamic curves of crack width changes in real time, and achieve crack detection accuracy at the micrometer level, specifically ±1μm, accurately depicting the microscopic evolution of crack opening and closing.

[0018] Preferably, the multi-source data fusion engine includes a multi-parameter time-series database, a coupled repair dynamics module, and an environmental factor compensation unit;

[0019] The multi-parameter time-series database is used to classify and store stress data, strain data, electrochemical impedance spectroscopy data, environmental temperature and humidity data, pH data, and crack image feature vectors throughout the monitoring process, and to construct a standardized time-series dataset.

[0020] The coupled remediation kinetics module establishes a correlation model between microbial activity and mineralization deposition rate. Based on the model, remediation efficiency is dynamically extrapolated, and the formula for calculating remediation efficiency is set as follows: In the formula, K represents the material correction factor, and σ(t) represents the real-time stress recovery rate of concrete. Represents the real-time closing area of ​​the crack;

[0021] The environmental factor compensation unit collects environmental temperature and humidity fluctuation data in real time, dynamically corrects the remediation rate threshold according to the environmental change pattern, and eliminates the impact of environmental interference on the remediation assessment results.

[0022] Preferably, the monitoring unit, time-series image acquisition unit, data acquisition and transmission unit, and dynamic repair analysis platform are integrated into a built-in self-repairing accelerated verification working mode.

[0023] In this mode, a micro-amplitude cyclic stress with a frequency of 0.05Hz to 0.2Hz is applied to the concrete specimen by an electronically controlled loading device to simulate the alternating load under the actual service conditions of concrete. The dynamic mechanical response spectrum of concrete is collected simultaneously during the repair process, and the stress-strain hysteresis curve is extracted. The curve area attenuation rate is used as the core quantitative indicator to evaluate the stability of concrete repair and to quickly verify the self-healing performance.

[0024] An effective evaluation method for the dynamic repair effect of microbial self-healing concrete includes the following steps:

[0025] Step S1: Prepare standard concrete specimens doped with microbial capsules. Initial cracks are machined on the surface of the specimens using mechanical prefabrication. The width of the initial cracks is controlled within the range of 50μm to 300μm to ensure the consistency of the specimens and to provide standard test samples for subsequent repair monitoring.

[0026] Step S2: After starting the monitoring unit, time-series image acquisition unit, data acquisition and transmission unit, and dynamic repair analysis platform, to During the continuous monitoring period, a set of concrete mechanical and environmental physical parameters was continuously collected. Simultaneously acquire crack evolution image set Im to complete multi-source basic data acquisition;

[0027] Step S3: Preprocess and extract features from the acquired crack image set Im, filter the core morphological parameters of the cracks, and construct feature vectors. = (width) ,area fractal dimension ), quantitatively characterizing the evolution features of crack morphology;

[0028] Step S4: Integrate the physical parameter set P and the crack morphology feature vector FV to construct a repair state tensor R(t) that dynamically changes over time. This enables the structured fusion of multi-source data;

[0029] Step S5: Input the repair state tensor R(t) into the pre-trained repair performance evaluation model. The model outputs a dynamic repair index, calculated using the following formula:

[0030]

[0031] In the formula For the mechanical recovery of the specimen, For crack sealing, For chemical stability, These are weighted correction factors;

[0032] Step S6: Based on the monitoring data, plot the DRI(t)-t dynamic change curve, and extract three core evaluation indicators through curve calculation, namely, the repair trigger delay period. Average repair rate and repair saturation threshold A comprehensive evaluation of the repair results was completed.

[0033] Preferably, in step S4, the repair state tensor R(t) adopts a timestamp alignment strategy to complete the synchronization processing of multi-source heterogeneous data;

[0034] Using the image acquisition time tm as a unified time reference, a cubic spline interpolation algorithm is used to complete the intermittent data acquisition from the sensor, ensuring the continuity of the data in the time dimension. At the same time, the electrochemical impedance spectroscopy data is processed by fast Fourier transform to extract the characteristic frequency amplitude, which is used as an evaluation index of the internal electrochemical activity of concrete, thereby optimizing the accuracy of data fusion.

[0035] Preferably, the repair performance evaluation model is constructed using an LSTM-Attention neural network. A training sample set is constructed by selecting concrete samples with known repair performance, and the neural network is iteratively trained. The model input layer is set with a repair state tensor R(t) time series with a time-domain slice length of 6 hours. The model output layer outputs a three-dimensional label vector, which corresponds to mechanical recovery degree ΔM, crack sealing degree ΔS, and chemical stability ΔC, respectively. The model loss function adopts a weighted combination of root mean square error and dynamic time warping to reduce time series data errors and improve model fitting accuracy.

[0036] Preferably, the mechanical restoring degree ΔM is quantitatively calculated using an integral formula, the specific calculation formula being:

[0037]

[0038] in, This is a hardness mapping function for concrete materials. For a single performance evaluation cycle, the degree of recovery of concrete mechanical properties before and after repair is accurately quantified through integral calculation.

[0039] Preferably, steps S1 to S5 further employ a graded evaluation mode to dynamically classify the repair level, with the specific classification criteria as follows:

[0040] In the initial repair stage, the Dynamic Repair Index (DRI) is less than 0.4, at which point the crack surface is initially sealed, and the water permeability of the concrete decreases by 40% to 60%.

[0041] In the intermediate repair stage, the dynamic repair index is 0.4 ≤ DRI < 0.7. At this time, the repair rate of crack depth exceeds 50%, and the recovery rate of concrete impermeability is greater than 75%.

[0042] This invention provides an effective method for evaluating the dynamic repair effect of microbial self-healing concrete. It has the following beneficial effects:

[0043] 1. This invention uses a multimodal sensor array combined with a time-series image acquisition unit to continuously collect mechanical, environmental and image time-series data of the entire process of concrete repair, and capture the evolution law of microbial mineralization deposition in real time. It effectively solves the defects of traditional detection that can only obtain endpoint parameters and cannot characterize the dynamic process of repair, accurately obtain dynamic characteristic indicators such as repair delay period and repair rate, and improve the evaluation time resolution.

[0044] 2. This invention utilizes pre-embedded sensors and a non-contact imaging method, eliminating the need to break concrete specimens and enabling continuous, non-destructive monitoring of the same sample throughout its entire lifecycle. This avoids sample waste caused by traditional destructive testing, eliminates experimental errors due to individual specimen differences, significantly improves the repeatability of monitoring data, reduces testing costs, and is suitable for long-term service tracking observation.

[0045] 3. This invention integrates multiple parameters from mechanical, electrochemical, environmental, and crack morphology sources, combined with... A neural network is used to construct a dynamic repair index, comprehensively considering the coupling effects of multiple fields. This addresses the problems of traditional evaluation indicators being singular and having limited dimensions, taking into account mechanical recovery, crack sealing, and chemical stability, resulting in more comprehensive and reliable evaluation results.

[0046] 4. This invention relies on The controller and cloud platform enable automated data collection, transmission and analysis. Combined with AI crack recognition algorithms, intelligent calibration is completed without the need for frequent and intermittent manual sampling. This overcomes the shortcomings of traditional manual observation, such as large errors and inconsistent cycles, ensures the time synchronization of multi-source data, and builds a standardized time-series database, significantly improving the intelligence and accuracy of the evaluation. Attached Figure Description

[0047] Figure 1 This is an overall framework diagram of the present invention;

[0048] Figure 2 This is a system flowchart of the present invention;

[0049] Figure 3 This is a flowchart of the steps of the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Example 1:

[0052] like Figure 1 and Figure 2 As shown, this embodiment of the invention provides an effective evaluation system for the dynamic repair effect of microbial self-healing concrete, including a monitoring unit, a time-series image acquisition unit, a data acquisition and transmission unit, and a dynamic repair analysis platform:

[0053] The monitoring unit is a multimodal sensor array embedded inside the concrete specimen. The sensor array comprises distributed fiber optic strain sensors, embedded micro-stress sensors, temperature and humidity sensors, pH sensors, and an electrochemical impedance spectroscopy probe. These multiple sensors collaboratively collect multi-dimensional mechanical, environmental, and electrochemical parameters within the concrete. The embedded micro-stress sensor is equipped with a microporous fixing base made of epoxy resin-quartz composite material. This micro-stress sensor is pre-embedded at the predicted crack path location in the concrete, ensuring the sensor's placement closely matches the damage evolution area. Simultaneously, the sensor's outer surface is covered with a microbial-compatible permeable membrane with pore sizes no larger than 0.5 μm, allowing water vapor and ions to pass through while blocking impurities and preventing interference with the microbial survival environment. This embedded micro-stress sensor achieves a detection accuracy of ±0.01 MPa, precisely capturing minute stress changes within the concrete.

[0054] The time-series image acquisition unit includes a high-speed sensor equipped with a microscope lens. The camera and the macro-crack scanner, two types of image acquisition devices, are mounted together on an adjustable bracket on the surface of the specimen. The adjustable bracket allows for flexible adjustment of the angle and height, thereby enabling timed and automated shooting of micro and macro crack morphology and complete preservation of image data of the entire crack evolution process. The time-series image acquisition unit is equipped with an ultra-depth-of-field microscope lens group, and the lens magnification can be steplessly adjusted within the range of 200× to 1000×.

[0055] This lens group is equipped with an image processing module with a built-in AI crack boundary recognition algorithm, which can automatically identify and calibrate crack boundary contours, generate dynamic change curves of crack width in real time, and achieve crack detection accuracy at the micrometer level, with a specific accuracy of ±1μm, accurately depicting the microscopic evolution law of crack opening and closing.

[0056] The data acquisition and transmission unit uses low power consumption. The controller is the core control component. The controller's communication terminal is electrically connected to various sensors and a time-series image acquisition unit, integrating the acquired sensor data and image data, and then... The dual communication modules will upload integrated data to the cloud server in real time, ensuring the stability and timeliness of data transmission;

[0057] The dynamic repair analysis platform is deployed inside a cloud server. The platform has a built-in multi-source data fusion engine. Relying on the intelligent algorithm on the engine, it completes time series damage inversion calculation and comprehensive evaluation of concrete repair efficiency, realizing dynamic analysis of the repair process. The multi-source data fusion engine includes a multi-parameter time series database, a coupled repair dynamics module, and an environmental factor compensation unit.

[0058] The multi-parameter time-series database is used to classify and store stress data, strain data, electrochemical impedance spectroscopy data, environmental temperature and humidity data, pH data, and crack image feature vectors throughout the monitoring process, and to construct a standardized time-series dataset.

[0059] The coupled remediation kinetics module establishes a correlation model between microbial activity and mineralization deposition rate. Based on the model, remediation efficiency is dynamically extrapolated, and the formula for calculating remediation efficiency is set as follows: In the formula, K represents the material correction factor, and σ(t) represents the real-time stress recovery rate of concrete. Represents the real-time closing area of ​​the crack;

[0060] The environmental factor compensation unit collects real-time data on environmental temperature and humidity fluctuations, dynamically adjusts the remediation rate threshold based on environmental change patterns, and eliminates the impact of environmental interference on the remediation assessment results.

[0061] The integrated self-healing accelerated verification working mode is achieved by combining the monitoring unit, time-series image acquisition unit, data acquisition and transmission unit, and dynamic repair analysis platform.

[0062] In this mode, a micro-amplitude cyclic stress with a frequency of 0.05Hz to 0.2Hz is applied to the concrete specimen by an electronically controlled loading device to simulate the alternating load under the actual service conditions of concrete. The dynamic mechanical response spectrum of concrete is collected simultaneously during the repair process, and the stress-strain hysteresis curve is extracted. The curve area attenuation rate is used as the core quantitative indicator to evaluate the stability of concrete repair and to quickly verify the self-healing performance.

[0063] Furthermore, the monitoring unit is embedded inside the concrete specimen using a pre-embedded deployment method, forming a multi-modal sensor array. This array specifically includes distributed fiber optic strain sensors, embedded micro-stress sensors, temperature and humidity sensors, pH sensors, and an electrochemical impedance spectroscopy probe. Through the collaborative operation of multiple sensor types, it simultaneously collects multi-dimensional fundamental parameters such as the internal mechanical response of the concrete, environmental conditions, and electrochemical characteristics. The embedded micro-stress sensor is equipped with a microporous fixing base, which is integrally fabricated from an epoxy resin-quartz composite material. This composite material possesses high strength, corrosion resistance, and excellent compatibility with the concrete matrix. The micro-stress sensor, with the aid of the microporous fixing base, is precisely pre-embedded at the predicted crack path location in the concrete, ensuring that the sensor deployment area accurately matches the high-incidence area of ​​damage evolution, thus improving the targeting of data acquisition. Meanwhile, the sensor is coated with a microbial-compatible permeable membrane with a maximum pore size of no more than 0.5 μm. This pore size range ensures the free passage of water vapor and free ions and meets the survival and metabolic needs of microorganisms, while effectively blocking external solid impurities and large particulate pollutants from entering the sensing and detection area. This avoids impurities interfering with the microbial growth environment and ensures the stable progress of microbial mineralization reactions. The embedded micro-stress sensor has a stress detection accuracy of ±0.01 MPa, which can accurately capture the subtle stress fluctuations generated during the damage and repair process inside concrete.

[0064] The time-series image acquisition unit consists of a high-speed CMOS camera equipped with a microscope lens and a macroscopic crack scanner. Both types of image acquisition devices are uniformly mounted on an adjustable metal bracket on the surface of the specimen. The adjustable bracket can achieve stepless adjustment of height, horizontal angle, and pitch angle to adapt to the imaging needs of different observation positions and different crack orientations. This enables timed and automated imaging of crack morphology at both the microscopic and macroscopic scales, completely preserving the original image data of the entire process of concrete crack initiation, expansion, and closure. The time-series image acquisition unit is equipped with an ultra-depth-of-field microscope lens group, with a lens magnification that can be steplessly adjusted within the range of 200× to 1000×, taking into account both macroscopic crack morphology observation and microscopic crystal deposition observation. The lens group is equipped with a dedicated image processing module with a built-in AI crack boundary recognition algorithm. Relying on pixel threshold segmentation and edge contour extraction technology, it automatically identifies and accurately calibrates the crack boundary contour, generating a dynamic evolution curve of crack width over time in real time. The crack detection accuracy is stably at the micrometer level, with a specific detection accuracy of ±1μm, which can finely characterize the microscopic evolution law of crack opening and closing during the self-healing cycle.

[0065] The data acquisition and transmission unit uses a low-power MCU controller as its core control component. The data acquisition and communication terminals of the MCU controller are electrically connected to various sensors and time-series image acquisition units within the monitoring unit. It summarizes and integrates sensor detection data and high-definition image data in real time, and completes data noise reduction, format unification, and preliminary screening and preprocessing. The controller is equipped with a LoRa / 5G dual communication module, which takes into account the requirements of long-distance low-power transmission and high-speed large-capacity transmission. It wirelessly uploads the integrated multi-source heterogeneous data to the cloud server in real time, ensuring the stability, continuity and timeliness of data transmission under complex working conditions.

[0066] The dynamic repair analysis platform is deployed on a cloud server. The platform incorporates a multi-source data fusion engine, which, based on time-series data analysis algorithms, damage inversion algorithms, and repair efficiency evaluation algorithms, performs time-series damage inversion calculations and comprehensive quantitative assessments of repair efficiency for concrete. This enables dynamic, visual, and quantitative analysis of the entire microbial self-repair process. The multi-source data fusion engine includes a multi-parameter time-series database, a coupled repair kinetics module, and an environmental factor compensation unit. The multi-parameter time-series database is used to classify and archive stress data, strain data, electrochemical impedance spectroscopy data, environmental temperature and humidity data, pH data, and crack image feature vectors from the entire monitoring process, constructing a standardized, traceable, and time-aligned concrete repair time-series dataset. The coupled repair kinetics module internally builds a mathematical model relating microbial activity to mineralization deposition rate. Combining microbial metabolic patterns and calcium carbonate deposition mechanisms, it dynamically extrapolates the real-time repair efficiency of concrete. The repair efficiency calculation formula is set as follows: In the formula, K is the material correction coefficient, used to adapt to the differences in concrete matrix with different mix proportions; σ(t) is the real-time stress recovery rate of concrete; σ0 is the initial intact stress of the specimen; This represents the real-time closure area of ​​the crack. The environmental factor compensation unit collects real-time data on environmental temperature and humidity fluctuations, dynamically adjusts the remediation rate determination threshold based on the influence of temperature and humidity on microbial activity, eliminates evaluation biases caused by environmental disturbances, and improves the accuracy of the assessment results.

[0067] In addition, the monitoring unit, time-series image acquisition unit, data acquisition and transmission unit, and dynamic repair analysis platform are integrated into a self-healing accelerated verification working mode for rapid verification of the long-term repair stability of concrete. In this working mode, a micro-amplitude cyclic alternating stress with a frequency of 0.05Hz to 0.2Hz is applied to the concrete specimen by an external electrically controlled loading device to simulate the repeated stress conditions caused by vehicle loads and temperature stresses in actual engineering. The dynamic mechanical response spectrum of concrete during the repair process is collected simultaneously, and the stress-strain hysteresis curve is extracted. The attenuation rate of the hysteresis curve area is used as the core quantitative indicator for evaluating the compactness and structural stability of the repaired concrete, so as to achieve rapid verification of the self-healing performance of microorganisms.

[0068] Example 2:

[0069] like Figure 3 As shown, an effective evaluation method for the dynamic repair effect of microbial self-healing concrete includes the following steps:

[0070] Step S1: Prepare standard concrete specimens doped with microbial capsules. Initial cracks are machined on the surface of the specimens using mechanical prefabrication. The width of the initial cracks is controlled within the range of 50μm to 300μm to ensure the consistency of the specimens and to provide standard test samples for subsequent repair monitoring.

[0071] Step S2: After starting the monitoring unit, time-series image acquisition unit, data acquisition and transmission unit, and dynamic repair analysis platform, to During the continuous monitoring period, a set of concrete mechanical and environmental physical parameters was continuously collected. Simultaneously acquire crack evolution image set Im to complete multi-source basic data acquisition;

[0072] Step S3: Preprocess and extract features from the acquired crack image set Im, filter the core morphological parameters of the cracks, and construct feature vectors. = (width) ,area fractal dimension ), quantitatively characterizing the evolution features of crack morphology;

[0073] Step S4: Integrate the physical parameter set P and the crack morphology feature vector FV to construct a repair state tensor R(t) that dynamically changes over time. To achieve structured fusion of multi-source data, the state tensor R(t) is repaired by using a timestamp alignment strategy to complete the synchronous processing of multi-source heterogeneous data;

[0074] Using the image acquisition time tm as a unified time reference, a cubic spline interpolation algorithm is used to complete the intermittent sensor acquisition data to ensure the continuity of the data in the time dimension. At the same time, the electrochemical impedance spectroscopy data is processed by fast Fourier transform to extract the characteristic frequency amplitude, which is used as an evaluation index of the internal electrochemical activity of concrete to optimize the accuracy of data fusion.

[0075] Step S5: Input the repair state tensor R(t) into the pre-trained repair performance evaluation model. The model outputs a dynamic repair index, calculated using the following formula:

[0076]

[0077] In the formula For the mechanical recovery of the specimen, For crack sealing, For chemical stability, The weighted correction coefficients are used in the repair effectiveness assessment model. The neural network was trained by constructing a training sample set using concrete samples with known repair performance. Iterative training of the neural network was then completed, with the model input layer set to a repair state tensor with a time-domain slice length of 6 hours. The time series model outputs three-dimensional label vectors, each corresponding to the mechanical restoring degree. Crack sealing Chemical stability ΔC; the model loss function adopts a weighted combination of root mean square error and dynamic time warping to reduce time series data errors, improve model fitting accuracy, and enhance mechanical restitution. The quantitative calculation is performed using an integral formula. The specific calculation formula is as follows:

[0078]

[0079] in, This is a hardness mapping function for concrete materials. For a single performance evaluation cycle, the degree of recovery of concrete mechanical properties before and after repair is accurately quantified through integral calculation;

[0080] Step S6: Draw based on monitoring data The dynamic change curve is used to extract three core evaluation indicators through curve calculation, namely, the repair trigger delay period. Average repair rate and repair saturation threshold A comprehensive evaluation of the repair results was completed.

[0081] Steps S1 to S5 further employ a tiered evaluation model to dynamically classify the repair level. The specific classification criteria are as follows:

[0082] In the initial repair stage, the Dynamic Repair Index (DRI) is less than 0.4, at which point the crack surface is initially sealed, and the water permeability of the concrete decreases by 40% to 60%.

[0083] In the intermediate repair stage, the dynamic repair index is 0.4 ≤ DRI < 0.7. At this time, the repair rate of crack depth exceeds 50%, and the recovery rate of concrete impermeability is greater than 75%.

[0084] Furthermore: First, standardized concrete specimens with embedded microbial capsules were prepared. Initial cracks were machined on the surface of the specimens using a mechanical prefabrication and cutting process. The width of the initial cracks was strictly controlled to be between 50 μm and 300 μm. This crack width range covers common micro-cracks and medium-sized cracks in engineering, closely reflecting actual concrete cracking conditions. During the specimen preparation process, the mix proportions and curing conditions were standardized, and the differences in the basic physical properties of the specimens were strictly controlled to ensure the consistency of the test samples. This provides a standardized test substrate for full-cycle repair monitoring and avoids interference with the evaluation results caused by individual specimen deviations.

[0085] After specimen preparation and curing are completed, the monitoring unit, time-series image acquisition unit, data acquisition and transmission unit, and dynamic repair analysis platform are started sequentially, with the monitoring time interval set as follows: ~ During the continuous monitoring period, concrete mechanical parameters and environmental parameters are continuously collected to construct a physical parameter set P, which includes strain data. Stress data Electrochemical impedance data The system monitors ambient temperature (T), ambient humidity (H), and pH; it also synchronously and periodically acquires images of crack evolution, and compiles these images into a crack evolution image set. This enables the synchronous acquisition of multi-source heterogeneous basic data, ensuring the integrity and synchronization of data timing.

[0086] The acquired crack evolution image set Im underwent image preprocessing operations such as noise reduction, grayscale processing, and threshold segmentation to remove interference from shooting light and shadow and surface impurities, and accurately extracted key morphological parameters of the cracks. The instantaneous crack width Wt, effective crack area At, and crack fractal dimension Dt were selected as core characterization indicators to construct a three-dimensional crack morphological feature vector. = (width) ,area fractal dimension The fractal dimension is used to characterize the roughness of the crack and the tortuosity of the internal pores, which directly reflects the filling effect of mineralized materials, thereby quantifying the morphological evolution law of the crack throughout the entire process;

[0087] The collected physical parameter set P is coupled and integrated with the crack morphology feature vector FV to construct a repair state tensor R(t) with time dynamic attributes. This completes the structured fusion of multi-source heterogeneous data. To eliminate data misalignment caused by different sampling frequencies of various sensors, a timestamp alignment strategy is adopted to complete data synchronization calibration, using the image acquisition time t... mTo unify the benchmark, cubic spline interpolation algorithm is used to supplement missing data generated by intermittent sensor acquisition, maintaining the continuity of time-series data. At the same time, fast Fourier transform is performed on electrochemical impedance spectroscopy data to convert time-domain signals into frequency-domain signals, and characteristic frequency amplitudes are extracted as electrochemical evaluation indicators of the internal pore structure and ion activity of concrete, effectively improving the accuracy of multi-source data fusion.

[0088] The calibrated repair state tensor Import the pre-trained repair efficacy evaluation model and calculate the dynamic repair index. The parameters were calibrated based on concrete strength and microbial capsule dosage. This evaluation model is based on... The neural network was constructed using a gradient iteration method for model training. The training sample set consisted of multiple sets of concrete sample data with known repair performance gradients. The input layer of the model was set to a time-domain slice length of 6 hours to accommodate the slow mineralization reaction characteristics of microorganisms. The output layer corresponds to three-dimensional evaluation vectors of mechanical, sealing, and chemical stability. The model employed a weighted combination loss function combining root mean square error and dynamic time warping to reduce time-series data offset errors and improve model fitting accuracy. The mechanical resilience was precisely calculated using integral operations to integrate mechanical fluctuation data within the period and accurately determine the structural mechanical recovery level.

[0089] Based on full-cycle monitoring data The dynamic evolution curve is analyzed using differential calculus and feature inflection point extraction to obtain the repair trigger delay period. Average repair rate and repair saturation threshold Three core evaluation indicators are used to comprehensively determine the concrete repair effect. Furthermore, based on the aforementioned monitoring and evaluation process, a grading mechanism is added to dynamically classify the repair levels: The initial repair stage requires a DRI < 0.4, at which point the surface of the crack is sealed by mineralized deposits, water permeability decreases by 40%–60%, and only surface sealing is achieved; the intermediate repair stage requires 0.4 ≤ DRI < 0.7, the deep crack filling ratio is higher than 50%, the concrete's impermeability recovery rate is greater than 75%, and the internal pore structure is significantly optimized. This grading standard is clearly quantified, aligns with engineering application logic, and can accurately distinguish the performance differences of microbial concrete at different repair stages, achieving standardized grading evaluation of the repair degree.

[0090] Example 3:

[0091]

[0092] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An effective evaluation system for the dynamic repair effect of microbial self-healing concrete, comprising a monitoring unit, a time-series image acquisition unit, a data acquisition and transmission unit, and a dynamic repair analysis platform, characterized in that: The monitoring unit is a multimodal sensor array embedded inside the concrete specimen. The sensor array consists of distributed fiber optic strain sensors, embedded micro-stress sensors, temperature and humidity sensors, pH sensors, and electrochemical impedance spectroscopy probes. Multiple types of sensors work together to collect multi-dimensional parameters of the concrete's internal mechanical, environmental, and electrochemical properties. The time-series image acquisition unit includes a high-speed camera equipped with a microscope lens. The camera and the macro-crack scanner, two types of image acquisition devices, are mounted together on an adjustable bracket on the surface of the specimen. The angle and height can be flexibly adjusted by the adjustable bracket, so as to realize the timed and automated shooting of micro and macro crack morphology and completely preserve the image data of the entire process of crack evolution. The data acquisition and transmission unit operates with low power consumption. The controller is the core control component. The controller's communication terminal is electrically connected to various sensors and a time-series image acquisition unit, integrating the acquired sensor data and image data, and then... The dual communication modules will upload integrated data to the cloud server in real time, ensuring the stability and timeliness of data transmission; The dynamic repair analysis platform is deployed inside a cloud server. The platform has a built-in multi-source data fusion engine, which relies on the intelligent algorithms on the engine to complete time series damage inversion calculations and comprehensive evaluation of concrete repair efficiency, thereby realizing dynamic analysis of the repair process.

2. The effective evaluation system for the dynamic repair effect of microbial self-healing concrete according to claim 1, characterized in that: The embedded micro-stress sensor is equipped with a micro-hole fixing base, which is made of epoxy resin-quartz composite material; This micro-stress sensor is embedded in the pre-defined crack prediction path of the concrete using a microporous fixing base, ensuring that the sensor is positioned to fit the damage evolution area. At the same time, the sensor is covered with a microbial-compatible permeable membrane with a pore size of no more than 0.5μm, which allows water vapor and ions to pass through while blocking impurities from entering and avoiding interference with the microbial living environment. This embedded micro-stress sensor has a detection accuracy of ±0.01MPa, accurately capturing subtle stress changes inside the concrete.

3. The effective evaluation system for the dynamic repair effect of microbial self-healing concrete according to claim 1, characterized in that: The time-series image acquisition unit is equipped with an ultra-depth-of-field microscope lens group, and the lens magnification can be steplessly adjusted within the range of 200× to 1000×. This lens assembly, paired with an image processing module featuring a built-in AI crack boundary recognition algorithm, can automatically identify and calibrate crack boundary contours, generate dynamic curves of crack width changes in real time, and achieve crack detection accuracy at the micrometer level, specifically ±1μm, accurately depicting the microscopic evolution of crack opening and closing.

4. The effective evaluation system for the dynamic repair effect of microbial self-healing concrete according to claim 1, characterized in that: The multi-source data fusion engine includes a multi-parameter time-series database, a coupled repair dynamics module, and an environmental factor compensation unit. The multi-parameter time-series database is used to classify and store stress data, strain data, electrochemical impedance spectroscopy data, environmental temperature and humidity data, pH data, and crack image feature vectors throughout the monitoring process, and to construct a standardized time-series dataset. The coupled remediation kinetics module establishes a correlation model between microbial activity and mineralization deposition rate. Based on the model, remediation efficiency is dynamically extrapolated, and the formula for calculating remediation efficiency is set as follows: In the formula, K represents the material correction factor, and σ(t) represents the real-time stress recovery rate of concrete. Represents the real-time closing area of ​​the crack; The environmental factor compensation unit collects environmental temperature and humidity fluctuation data in real time, dynamically corrects the remediation rate threshold according to the environmental change pattern, and eliminates the impact of environmental interference on the remediation assessment results.

5. The effective evaluation system for the dynamic repair effect of microbial self-healing concrete according to claim 1, characterized in that: The monitoring unit, time-series image acquisition unit, data acquisition and transmission unit, and dynamic repair analysis platform are integrated into a built-in self-repairing accelerated verification working mode. In this mode, a micro-amplitude cyclic stress with a frequency of 0.05Hz to 0.2Hz is applied to the concrete specimen by an electronically controlled loading device to simulate the alternating load under the actual service conditions of concrete. The dynamic mechanical response spectrum of concrete is collected simultaneously during the repair process, and the stress-strain hysteresis curve is extracted. The curve area attenuation rate is used as the core quantitative indicator to evaluate the stability of concrete repair and to quickly verify the self-healing performance.

6. An effective evaluation method for the dynamic repair effect of microbial self-healing concrete, using an effective evaluation system for the dynamic repair effect of microbial self-healing concrete as described in any one of claims 1 to 5, characterized in that, Includes the following steps: Step S1: Prepare standard concrete specimens doped with microbial capsules. Initial cracks are machined on the surface of the specimens using mechanical prefabrication. The width of the initial cracks is controlled within the range of 50μm to 300μm to ensure the consistency of the specimens and to provide standard test samples for subsequent repair monitoring. Step S2: After starting the monitoring unit, time-series image acquisition unit, data acquisition and transmission unit, and dynamic repair analysis platform, to During the continuous monitoring period, a set of concrete mechanical and environmental physical parameters was continuously collected. Simultaneously acquire crack evolution image set Im to complete multi-source basic data acquisition; Step S3: Preprocess and extract features from the acquired crack image set Im, filter the core morphological parameters of the cracks, and construct feature vectors. = (width) ,area fractal dimension ), quantitatively characterizing the evolution features of crack morphology; Step S4: Integrate the physical parameter set P and the crack morphology feature vector FV to construct a repair state tensor R(t) that dynamically changes over time. This enables the structured fusion of multi-source data; Step S5: Input the repair state tensor R(t) into the pre-trained repair performance evaluation model. The model outputs a dynamic repair index, calculated using the following formula: ； In the formula For the mechanical recovery of the specimen, For crack sealing, For chemical stability, These are weighted correction factors; Step S6: Based on the monitoring data, plot the DRI(t)-t dynamic change curve, and extract three core evaluation indicators through curve calculation, namely, the repair trigger delay period. Average repair rate and repair saturation threshold A comprehensive evaluation of the repair results was completed.

7. The effective evaluation method for the dynamic repair effect of microbial self-healing concrete according to claim 6, characterized in that: In step S4, the repair state tensor R(t) uses a timestamp alignment strategy to complete the synchronization processing of multi-source heterogeneous data. Using the image acquisition time tm as a unified time reference, a cubic spline interpolation algorithm is used to complete the intermittent data acquisition from the sensor, ensuring the continuity of the data in the time dimension. At the same time, the electrochemical impedance spectroscopy data is processed by fast Fourier transform to extract the characteristic frequency amplitude, which is used as an evaluation index of the internal electrochemical activity of concrete, thereby optimizing the accuracy of data fusion.

8. The effective evaluation method for the dynamic repair effect of microbial self-healing concrete according to claim 6, characterized in that: The repair performance evaluation model is constructed using an LSTM-Attention neural network. A training sample set is constructed using concrete samples with known repair performance. The neural network is iteratively trained. The model input layer is set with a repair state tensor R(t) time series with a time-domain slice length of 6 hours. The model output layer outputs a three-dimensional label vector, which corresponds to mechanical recovery degree ΔM, crack sealing degree ΔS, and chemical stability ΔC, respectively. The model loss function adopts a weighted combination of root mean square error and dynamic time warping to reduce time series data errors and improve model fitting accuracy.

9. The effective evaluation method for the dynamic repair effect of microbial self-healing concrete according to claim 8, characterized in that: The mechanical restoring degree ΔM is quantitatively calculated using an integral formula, specifically: ； in, This is a hardness mapping function for concrete materials. For a single performance evaluation cycle, the degree of recovery of concrete mechanical properties before and after repair is accurately quantified through integral calculation.

10. The effective evaluation method for the dynamic repair effect of microbial self-healing concrete according to claim 6, characterized in that: Steps S1 to S5 further employ a tiered evaluation model to dynamically classify the repair level. The specific classification criteria are as follows: In the initial repair stage, the Dynamic Repair Index (DRI) is less than 0.4, at which point the crack surface is initially sealed, and the water permeability of the concrete decreases by 40% to 60%. In the intermediate repair stage, the dynamic repair index is 0.4 ≤ DRI < 0.

7. At this time, the repair rate of crack depth exceeds 50%, and the recovery rate of concrete impermeability is greater than 75%.

Images