System and method for multi-field coupling monitoring and life prediction of recycled concrete deterioration

Abstract

The application provides a system and method for monitoring and predicting the life of deteriorated recycled concrete, comprising: an embedded sensor network, which is arranged in layers along the depth direction of carbonized recycled concrete in the area susceptible to erosion, including a mechanical sensor and a chemical sensor network arranged in space in cooperation, for synchronously collecting mechanical field data reflecting mechanical properties and chemical field data reflecting chemical ion concentration; a data acquisition and transmission unit connected with the embedded sensor network, for collecting and uploading time series data from the mechanical field and the chemical field; a cloud analysis and prediction platform configured to receive and fuse multi-field data, evaluate the degree of structural erosion deterioration based on the fused data, and predict the remaining service life of the structure based on the time evolution trend of the degree of deterioration. The application solves the technical problems of incomplete single physical field information, lack of targeted analysis of the erosion process and insufficient life prediction capability in existing concrete monitoring technology.

Description

Technical Field

[0001] This invention relates to the technical field of structural health monitoring, and in particular to a multi-field coupled monitoring and life prediction system and method for the deterioration of recycled concrete. Background Technology

[0002] Carbonation curing technology is an effective means to improve the early performance of recycled concrete, but its long-term service performance, especially its resistance to corrosion, is the key to its success in major infrastructure applications. The corrosion process is a slow, gradual deterioration, starting with damage to the material's internal microstructure and gradually progressing to a decline in macroscopic mechanical properties. Traditional concrete structure testing methods, such as rebound hammer testing and ultrasonic testing, typically only detect damage when it accumulates to a certain extent, exhibiting a significant lag and making early warning difficult. While core sampling and other methods are accurate, they are damaging to the structure and cannot achieve continuous monitoring.

[0003] In existing technologies, embedded sensors offer the possibility of real-time monitoring of concrete conditions. However, current monitoring systems often rely on single physical field (e.g., mechanical) information, making it difficult to accurately distinguish and quantify specific deterioration caused by dissolution. Furthermore, they lack predictive models that dynamically correlate real-time monitoring data with the remaining service life of the structure. This results in the inability to achieve early warning and predictive maintenance for dissolution deterioration.

[0004] Therefore, developing an intelligent monitoring system that can integrate information from multiple fields, specifically target the corrosion and degradation mechanism, and has accurate life prediction capabilities is of great significance for ensuring the safety of major projects using new concrete materials and for achieving condition-based and predictive maintenance. Summary of the Invention

[0005] This invention aims to address the problems of incomplete single-physical field information, lack of targeted analysis of the dissolution process, and insufficient life prediction capability in existing concrete monitoring technologies, and provides a comprehensive solution integrating data perception, analysis, and prediction.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: In a first aspect, the present invention provides a multi-field coupled monitoring and life prediction system for the deterioration of recycled concrete, comprising: An embedded sensor network is deployed in layers along the depth direction of the carbonized recycled concrete structure in critical areas susceptible to dissolution. It includes a spatially coordinated array of mechanical sensors and a network of chemical sensors to synchronously acquire mechanical field data reflecting the mechanical properties of the structure and chemical field data reflecting the concentration of chemical ions. A data acquisition and transmission unit, which is connected to the embedded sensor network, is used to acquire and upload time-series data from mechanical and chemical fields; A cloud-based analysis and prediction platform is configured to receive and fuse multi-field data, assess the degree of corrosion degradation of the structure based on the fused data, and predict the remaining service life of the structure based on the time evolution trend of the degree of corrosion degradation.

[0007] Furthermore, the mechanical sensor array includes multiple cement-based piezoelectric sensors, which operate in an active sensing mode and acquire data related to the elastic modulus of concrete as the mechanical field data by analyzing the propagation characteristics of elastic waves.

[0008] Furthermore, the chemical sensor network includes multiple solid-state calcium ion selective electrodes for in-situ measurement of calcium ion concentration in concrete pore fluid as chemical field data.

[0009] Furthermore, the cloud-based analytics and prediction platform includes: The data fusion module is used to perform time synchronization, spatial matching, and data association fusion processing on the received mechanical field data and chemical field data. The spatial matching establishes the correspondence between the data and the spatial location of the structure based on the layered deployment depth of the sensors. The degradation status assessment module integrates a pre-trained machine learning model, which is configured to output a quantitative index characterizing the degree of dissolution degradation based on the fused mechanical field data and chemical field data.

[0010] Furthermore, the machine learning model is a recurrent neural network or a long short-term memory network model, which is trained using historical dissolution test data to learn the coupling relationship between the decay of local elastic modulus and the change of calcium ion concentration gradient.

[0011] Furthermore, the cloud-based analysis and prediction platform also includes a life prediction module, which is configured to dynamically calculate the remaining time for the concrete structure to reach a preset performance threshold based on the degree of degradation and historical degradation rate, using a time series prediction algorithm or degradation evolution model.

[0012] Furthermore, the system also includes an early warning module, which is configured to trigger an early warning signal when the real-time degradation level exceeds a first threshold or the remaining service life is lower than a second threshold.

[0013] Secondly, the present invention also provides a multi-field coupled monitoring and lifetime prediction method for the deterioration of recycled concrete, comprising the following steps: S1. During the pouring of carbonized recycled concrete structure, a sensor network is installed in layers in key areas susceptible to erosion. The sensor network includes a corresponding cement-based piezoelectric sensor array and a solid-state calcium ion selective electrode network. Each monitoring node is equipped with at least one cement-based piezoelectric sensor and one solid-state calcium ion selective electrode. S2. The data acquisition and transmission unit excites the cement-based piezoelectric sensor array to emit sweeping ultrasonic guided waves according to a preset cycle, and simultaneously acquires the guided wave propagation characteristic data and the potential signal of the solid calcium ion selective electrode network. After filtering and amplification preprocessing, the potential signal is converted into calcium ion concentration data to form a standardized mechanical field and chemical field dataset. S3. Synchronize, spatially match and fused the elastic modulus-related guided wave parameters in the mechanical field dataset with the calcium ion concentration data in the chemical field dataset to generate a multi-field fusion dataset. Input the dataset into the trained machine learning coupling analysis model and output a normalized dissolution degradation index in the range of 0 to 1 to complete the assessment of the current dissolution degradation state. S4. Based on the historical degradation rate and the current corrosion degradation index, combined with the preset structural performance limit threshold, the degradation evolution curve is fitted by a time series prediction algorithm or degradation evolution model to solve the time corresponding to the curve reaching the performance limit threshold, thus obtaining the remaining service life of the structure, and triggering an automatic warning when the warning conditions are met.

[0014] Compared with the prior art, the present invention has at least the following beneficial effects: (1) Completeness and accuracy of monitoring information are achieved: By deploying a cement-based piezoelectric sensor array and a solid-state calcium ion selective electrode network in a layered and coordinated manner along the depth of the structure, the system can simultaneously and in situ acquire "mechanical field data" reflecting macroscopic mechanical properties (such as elastic modulus) and "chemical field data" reflecting internal chemical processes (calcium ion concentration). This simultaneous acquisition and coupling of multi-field information fundamentally overcomes the one-sidedness of monitoring a single physical field. For example, monitoring only mechanical changes cannot distinguish whether the damage originates from load or corrosion, but this system can specifically identify and quantify corrosion as a specific degradation mechanism by correlating the chemical signals of calcium ion loss, making the condition assessment results more accurate and reliable.

[0015] (2) It possesses early warning capability for corrosion degradation: corrosion is a gradual damage process that begins with a chemical process. The calcium ion selective electrode deployed in this application can keenly capture changes in the concentration of calcium ions in the pore fluid. Changes in this chemical indicator often precede significant decreases in macroscopic mechanical properties (such as elastic modulus). By monitoring this early chemical signal in real time and combining it with the analysis of its gradient changes in the depth direction, the system can achieve early warning of the corrosion degradation process, enabling maintenance personnel to take intervention measures before the structural load-bearing capacity decreases significantly.

[0016] (3) Achieving a functional leap from condition monitoring to lifespan prediction, supporting predictive maintenance: One of the core advantages of this application is that it can not only "diagnose the condition" but also "predict the lifespan." The cloud-based analysis and prediction platform uses machine learning models (such as LSTM) to intelligently analyze the fused multi-field data and output a normalized degradation index. Subsequently, the lifespan prediction module, based on the historical evolution trend of this degradation index over time, dynamically predicts the remaining time required for the structural performance to drop to a preset threshold through algorithms such as curve fitting. This enables infrastructure management to transform from the traditional periodic inspection or post-maintenance model to a more economical and efficient "predictive maintenance" model, providing key data support for making scientific maintenance decisions and optimizing asset lifecycle management.

[0017] (4) Provides safety assurance and technical verification for the engineering application of new materials such as carbonized recycled concrete: The long-term durability of carbonized recycled concrete, especially its resistance to corrosion, is a focus of attention in the engineering field. The system in this application provides real-time, online data monitoring and evaluation methods for its long-term service performance. By obtaining long-term performance evolution data in actual structures, concerns about the durability of new materials in engineering applications can be effectively eliminated, providing a solid technical guarantee for its safe promotion and reliable application in major infrastructure, and possessing significant engineering value and environmental significance. Attached Figure Description

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the overall system architecture provided in this embodiment. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0021] The following detailed description of some embodiments of the present invention is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0022] Example 1

[0023] This embodiment provides a multi-field coupled monitoring and life prediction system for the deterioration of recycled concrete. Please refer to [link / reference]. Figure 1 As shown, the system integrates an embedded cement-based piezoelectric sensor array and a solid-state calcium ion selective electrode network to monitor in real time the degradation of mechanical properties (such as elastic modulus) and changes in the chemical environment (such as calcium ion concentration) of concrete caused by dissolution, respectively. A cloud-based analysis and prediction platform is responsible for fusing the data from these two fields and using a pre-trained machine learning model to accurately assess the real-time deterioration state of the structure. Based on time-series analysis of the deterioration state, the system can further predict the remaining service life of the structure and provide early warning. This invention solves the problem of incomplete information from a single physical field monitoring, realizing a leap from condition monitoring to service life prediction, and providing key technical support for the safe application of new concrete materials in major infrastructure.

[0024] Specifically, the system mainly consists of three parts: Embedded sensor network: A hybrid sensor network is pre-embedded in critical areas of the concrete structure. This network comprises two types of sensors working in tandem. (a) Cement-based piezoelectric sensor array: Composed of multiple sensors made of piezoelectric ceramics (such as PZT), encapsulated in cement-based material and integrated with the main concrete. The array operates in an "active sensing" mode (one or more sensors emit swept-frequency ultrasonic guided waves, and other sensors receive the signals) to monitor parameters such as wave velocity and attenuation of the guided waves in real time, thereby calculating the dynamic changes in the local elastic modulus of the concrete.

[0025] (b) Solid-state calcium ion selective electrode (Ca-ISE) network: Solid-state calcium ion selective electrodes are deployed at different depths near the piezoelectric sensors. Each electrode is used for in-situ, real-time measurement of the calcium ion concentration in the concrete pore fluid at its location.

[0026] Data Acquisition and Transmission Unit (DAQ): Connected to the sensor network, it is responsible for periodically stimulating the piezoelectric sensor, acquiring signals from all sensors, performing preliminary signal filtering and amplification, and transmitting the acquired time-series data to the cloud server via wireless (such as LoRa, 5G) or wired means.

[0027] Cloud-based analytics and prediction platform: The core software deployed on a server. It includes the following functional modules: (a) Data fusion module: Receives and synchronizes mechanical field data (elastic modulus E(t)) from the piezoelectric sensor array and chemical field data (Ca) from the Ca-ISE network. 2+ Concentration C(x, t), where x is depth.

[0028] (b) Deterioration Status Assessment Module: This module incorporates a coupled analysis model based on machine learning (such as a Recurrent Neural Network (RNN) or a Long Short-Term Memory (LSTM) network. This model is trained using extensive indoor accelerated corrosion test data to learn and establish the intrinsic correlation between "local elastic modulus decay" and "calcium ion concentration gradient change." Based on real-time uploaded data, the model can accurately assess the degree of corrosion deterioration D(t) of the current structure.

[0029] (c) Life Prediction Module: Based on the current degradation level D(t) and historical degradation rate, combined with the preset structural performance limit state threshold (e.g., a 30% decrease in elastic modulus), the module uses time series prediction algorithms (such as the ARIMA model) or a trained degradation evolution model to dynamically predict the time required for the structure to reach the performance limit state, i.e., the remaining service life (RSL).

[0030] (d) Visualization and early warning interface: Display the structural health status, degradation trend graph, and RSL prediction results to users through a web interface, and automatically trigger an alarm when the predicted RSL is lower than the safety threshold.

[0031] Example 2

[0032] This embodiment also provides a multi-field coupled monitoring and life prediction method for the deterioration of recycled concrete. Please refer to [link / reference]. Figure 1 As shown, it specifically includes four consecutive steps: S1 is a system deployment. During the pouring of carbonized recycled concrete structures, a sensor network is buried in layers in key areas susceptible to erosion to ensure that each monitoring node is equipped with at least one cement-based piezoelectric sensor and one solid-state calcium ion selective electrode, thus ensuring the synchronous acquisition of data from multiple fields. S2 is for data acquisition and preprocessing. The data acquisition and transmission unit excites the piezoelectric sensor to emit swept-frequency ultrasonic guided waves according to a preset cycle, and simultaneously acquires the guided wave propagation characteristic data and the potential signal of the calcium ion selective electrode. After filtering and amplification preprocessing, the potential signal is converted into calcium ion concentration data to form a standardized dataset. S3 is for assessing the deterioration state. It performs time synchronization, spatial matching, and correlation fusion of mechanical and chemical field data to generate a multi-field fusion dataset. The dataset is input into the trained machine learning model and outputs a normalized dissolution deterioration index. S4 is for life prediction and early warning. Based on historical degradation rate and current degradation index, combined with performance limit threshold, it fits degradation evolution curve through relevant algorithms to solve the remaining service life and triggers automatic early warning when the early warning conditions are met.

[0033] This method comprehensively describes the entire operational process from hardware deployment, data acquisition, signal processing, data fusion, intelligent assessment to lifespan prediction and early warning, ensuring the feasibility of the technical solution. Specifically, the layered deployment design in S1 ensures spatial representativeness of the data; the standardized preprocessing in S2 improves data quality; the multi-field fusion and machine learning assessment in S3 ensures the accuracy of degradation levels; and the lifespan prediction and early warning in S4 enables proactive risk management. Its core advantage lies in forming a logically rigorous and operationally feasible complete technical process, implementing the system's hardware components and software functions in a step-by-step manner, ensuring the integrity of monitoring data, the accuracy of assessment results, and the timeliness of prediction and early warning, comprehensively supporting the safety monitoring and maintenance of carbonized recycled concrete structures.

[0035] Scenario: Long-term health monitoring of a bridge pier using carbonized recycled concrete. (1) System Deployment: During the construction and pouring of the bridge piers, the sensor network of this invention is buried in layers in key areas susceptible to erosion, such as the water-facing side and the water level change zone. For example, at depths of 5cm, 10cm, and 20cm from the surface, a monitoring node consisting of 3 cement-based piezoelectric sensors and 1 solid-state Ca-ISE is deployed. All sensors are connected to a waterproof DAQ box installed on the surface of the bridge piers via pre-embedded shielded cables.

[0036] (2) Data acquisition: The DAQ system is set to perform fully automatic data acquisition every 6 hours.

[0037] First, the DAQ sequentially excites each piezoelectric sensor to emit a set of chirp signals with frequencies ranging from 10 kHz to 200 kHz. Other piezoelectric sensors within the same node are responsible for receiving these signals. Raw waveform data is then acquired.

[0038] Then, the DAQ acquires the potential signal of each Ca-ISE electrode and converts it into a calcium ion concentration value (mmol / L) according to the built-in calibration curve.

[0039] The collected data packets (including timestamps, sensor IDs, waveform data, and concentration data) are uploaded to a designated cloud server via a 5G network.

[0040] (3) Cloud processing Elastic modulus calculation: The signal processing program of the cloud-based analysis and prediction platform analyzes the received waveform data, calculates the propagation time of the ultrasonic guided wave on different transmission and reception paths, and calculates the real-time elastic modulus E(t) of the concrete in the area by combining the precise distance between the sensors.

[0041] Degradation assessment: A pre-trained LSTM model receives real-time input E(t) sequences and C(x, t) sequences of different depths. The model outputs a normalized degradation index D(t) between 0 and 1, where 0 represents an initial healthy state and 1 represents complete failure. For example, a D(t) of 0.08 calculated from the input data of the day indicates that the structure is in a slightly degraded state.

[0042] Lifetime prediction: The lifetime prediction module fits the degradation evolution curve D(t) = f(t) based on the time series data of D(t) over the past month. Assuming the failure threshold is set to D_fail = 0.4, f(t) = 0.4 is solved by extrapolation, and the predicted remaining service life (RSL) is 15.2 years.

[0043] (4) Results presentation and early warning Bridge management engineers can log in to the monitoring platform through a browser and intuitively see the real-time elastic modulus distribution cloud map, calcium ion concentration profile map, and the overall structural health index D(t) change curve over time and dynamically updated RSL value.

[0044] The system is configured with early warning rules: when the D(t) of any monitoring node is greater than 0.2, or the predicted RSL is less than 5 years, the system will automatically send early warning information to the management personnel via email and SMS, prompting them to pay attention to the area and arrange a detailed manual inspection or maintenance plan.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-field coupled monitoring and life prediction system for the deterioration of recycled concrete, characterized in that, include: An embedded sensor network is deployed in layers along the depth direction of the carbonized recycled concrete structure in critical areas susceptible to dissolution. It includes a spatially coordinated array of mechanical sensors and a network of chemical sensors to synchronously acquire mechanical field data reflecting the mechanical properties of the structure and chemical field data reflecting the concentration of chemical ions. A data acquisition and transmission unit, which is connected to the embedded sensor network, is used to acquire and upload time-series data from mechanical and chemical fields; A cloud-based analysis and prediction platform is configured to receive and fuse multi-field data, assess the degree of corrosion degradation of the structure based on the fused data, and predict the remaining service life of the structure based on the time evolution trend of the degree of corrosion degradation.

2. The multi-field coupled monitoring and life prediction system for the deterioration of recycled concrete according to claim 1, characterized in that, The mechanical sensor array includes multiple cement-based piezoelectric sensors, which operate in an active sensing mode and acquire data related to the elastic modulus of concrete as the mechanical field data by analyzing the propagation characteristics of elastic waves.

3. The multi-field coupled monitoring and life prediction system for the deterioration of recycled concrete according to claim 2, characterized in that, The chemical sensor network includes multiple solid-state calcium ion selective electrodes for in-situ measurement of calcium ion concentration in concrete pore fluid as chemical field data.

4. The multi-field coupled monitoring and life prediction system for the deterioration of recycled concrete according to any one of claims 1-3, characterized in that, The cloud-based analytics and prediction platform includes: The data fusion module is used to perform time synchronization, spatial matching and data association fusion processing on the received mechanical field data and chemical field data. The spatial matching establishes the correspondence between data and structural spatial location based on the layered deployment depth of the sensors. The degradation status assessment module integrates a pre-trained machine learning model, which is configured to output a quantitative index characterizing the degree of dissolution degradation based on the fused mechanical field data and chemical field data.

5. The multi-field coupled monitoring and life prediction system for the deterioration of recycled concrete according to claim 4, characterized in that, The machine learning model is a recurrent neural network or a long short-term memory network model, which is trained using historical dissolution test data to learn the coupling relationship between the decay of local elastic modulus and the change of calcium ion concentration gradient.

6. The multi-field coupled monitoring and life prediction system for the deterioration of recycled concrete according to claim 4, characterized in that, The cloud-based analysis and prediction platform also includes a life prediction module, which is configured to dynamically calculate the remaining time for the concrete structure to reach a preset performance threshold based on the degree of degradation and historical degradation rate, using a time series prediction algorithm or degradation evolution model.

7. The multi-field coupled monitoring and life prediction system for the deterioration of recycled concrete according to claim 6, characterized in that, The system also includes an early warning module, which is configured to trigger an early warning signal when the real-time degradation level exceeds a first threshold or the remaining service life is lower than a second threshold.

8. A multi-field coupled monitoring and life prediction method for the deterioration of recycled concrete, characterized in that, Includes the following steps: S1. During the pouring of carbonized recycled concrete structure, a sensor network is installed in layers in key areas susceptible to erosion. The sensor network includes a corresponding cement-based piezoelectric sensor array and a solid-state calcium ion selective electrode network. Each monitoring node is equipped with at least one cement-based piezoelectric sensor and one solid-state calcium ion selective electrode. S2. The data acquisition and transmission unit excites the cement-based piezoelectric sensor array to emit sweeping ultrasonic guided waves according to a preset cycle, and simultaneously acquires the guided wave propagation characteristic data and the potential signal of the solid calcium ion selective electrode network. After filtering and amplification preprocessing, the potential signal is converted into calcium ion concentration data to form a standardized mechanical field and chemical field dataset. S3. Synchronize, spatially match and fuse the elastic modulus-related guided wave parameters in the mechanical field dataset with the calcium ion concentration data in the chemical field dataset to generate a multi-field fusion dataset. Input the dataset into the trained machine learning coupling analysis model and output a normalized dissolution degradation index in the range of 0 to 1 to complete the assessment of the current degree of dissolution degradation. S4. Based on the historical degradation rate and the current corrosion degradation index, combined with the preset structural performance limit threshold, the degradation evolution curve is fitted by a time series prediction algorithm or degradation evolution model to solve the time corresponding to the curve reaching the performance limit threshold, thus obtaining the remaining service life of the structure, and triggering an automatic warning when the warning conditions are met.

Images