Real-time monitoring and data processing method for reinforced concrete structure

Abstract

The invention discloses a real-time monitoring and data processing method for a reinforced concrete structure, and particularly relates to the technical field of data processing, and the method comprises the steps: building a data anomaly detection influence model through obtaining the saline corrosion information, the dynamic load information of waves and inrush current and the attachment information of underwater organisms during the operation of an external sensor; measuring the influence of the dynamic load of saline corrosion, waves and inrush current and the attachment of underwater organisms on the existing data anomaly detection, comparing a data anomaly detection influence index with a preset data anomaly detection influence index threshold value, and when an anomaly detection influence signal is generated, constructing a data anomaly detection machine learning model, the data anomaly detection is optimized according to the influence degree of the data anomaly detection and the difference of the reinforced concrete structure data transmitted by the external sensor and the internal sensor, so that too frequent false report and missing report are avoided, the detection strategy is adjusted in time, and the accurate monitoring of the health state of the structure is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and more specifically, to a method for real-time monitoring and data processing of reinforced concrete structures. Background Art

[0002] With the acceleration of urbanization and the continuous expansion of infrastructure construction, reinforced concrete structures are widely used in buildings, bridges, tunnels, water conservancy projects and other fields. Although reinforced concrete structures have good bearing capacity and durability, they still face many challenges in long-term use. In marine and underwater structure monitoring, it is usually necessary to deploy most sensors on the outside and inside of reinforced concrete structures. Compared with sensors deployed inside, sensors deployed outside are more seriously affected by salt water corrosion, dynamic loads of waves and surges, and attachment of underwater organisms. The existing real-time monitoring system of reinforced concrete structures usually uses the same standard data anomaly detection algorithm to detect anomalies in the reinforced concrete structure monitoring data transmitted by external and internal sensors for external and internal sensors with different degrees of influence, which may cause the normal data of external sensors to be mistakenly identified as abnormalities and cause false alarms. Based on inaccurate anomaly detection results, managers may make wrong maintenance decisions. For example, responding to false alarms from external sensors too frequently may lead to unnecessary inspections and maintenance, wasting resources; while missed alarms may lead to neglect of the true health status of the structure, thereby increasing the safety risk of the structure. When the monitoring system frequently issues false alarms or fails to identify real abnormal conditions in a timely manner, it may cause users to lose trust in the system, affecting the adoption of monitoring data and subsequent decision-making. Summary of the invention

[0003] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a real-time monitoring and data processing method for reinforced concrete structures to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] The method for real-time monitoring and data processing of reinforced concrete structure comprises the following steps:

[0006] Step S1, monitoring, obtaining salt water corrosion information, dynamic load information of waves and currents, and attachment information of underwater organisms when the external sensor is in operation;

[0007] Step S2, constructing a data anomaly detection impact model, generating a data anomaly detection impact index, and measuring the impact of salt water corrosion, dynamic loads of waves and surges, and attachment of underwater organisms on existing data anomaly detection;

[0008] Step S3, comparing the data anomaly detection impact index with a preset data anomaly detection impact index threshold, and evaluating the impact of salt water corrosion, dynamic loads of waves and currents, and attachment of underwater organisms on existing data anomaly detection;

[0009] Step S4, when an anomaly detection impact signal is generated, a data anomaly detection machine learning model is constructed, and the data anomaly detection is optimized according to the impact degree of the data anomaly detection and the difference in reinforced concrete structure data transmitted by the external sensor and the internal sensor.

[0010] In a preferred embodiment, the salt water corrosion information includes the salt water corrosion abnormal rise coefficient Ysfs, the wave and current dynamic load information includes the wave and current turbulence abnormal coefficient Blyl, and the underwater organism attachment information includes the underwater organism attachment coefficient Sxsf.

[0011] In a preferred embodiment, by acquiring salt water corrosion information when the external sensor is running, analyzing the abnormal increase of salt water corrosion on the external sensor, and acquiring the salt water corrosion abnormal increase coefficient, the abnormal increase degree of salt water corrosion on the external sensor is measured;

[0012] The logic for obtaining the abnormal rise coefficient of salt water corrosion is as follows:

[0013] Obtain the working current, working voltage, electrochemical impedance, salt water concentration, and ambient temperature of the external sensor within a fixed time period T, and calculate the corrosion rate Fsx at time t t , the expression is as follows Where cs = {1, 2, ..., 5}, TZ1 represents the working current, TZ2 represents the working voltage, TZ3 represents the electrochemical impedance, TZ4 represents the salt water concentration, TZ5 represents the ambient temperature, β0, β cs represents the regression coefficient;

[0014] Calculate the baseline corrosion rate jzs, the expression is as follows Where t={1,2,...,T}, T is a positive integer;

[0015] The corrosion rate at time t+1 is compared with the corrosion rate at time t. When the corrosion rate at time t+1 is greater than the corrosion rate at time t, the corrosion rate at time t+1 is marked as an abnormal corrosion rate, and the corrosion rate rise value ssz at adjacent times is calculated. The expression is as follows: Among them Fsx t+1 represents the corrosion rate at time t+1;

[0016] Calculate the abnormal rise coefficient of salt water corrosion Ysfs, the expression is as follows Among them ssz jh represents the jhth occurrence of an event in which the corrosion rate at time t+1 is greater than the corrosion rate at time t, jh = {1, 2, ..., J}, J is a positive integer, YFsx n It represents the corrosion rate marked as abnormal corrosion rate for the nth time, na={1,2,...,N}, where N is a positive integer.

[0017] In a preferred embodiment, by acquiring dynamic load information of waves and surges, the influence of the dynamic load of waves and surges on the external sensor is analyzed, and the wave surge turbulence abnormality coefficient is obtained to measure the degree to which the external sensor is affected by the dynamic load of waves and surges;

[0018] The logic for obtaining the wave surge and turbulence anomaly coefficient is as follows:

[0019] The wave frequency pl at the location of the external sensor is obtained through the ocean radar, and the wave energy S(pl) at different wave frequencies is calculated. The expression is as follows Where g represents the acceleration due to gravity, c1 and c2 are constant factors;

[0020] The wave load calculation expression is as follows: Where Fpl(st) represents the wave load at time st, represents the phase angle;

[0021] Obtain the fluid density at the location of the external sensor, the cross-sectional area of ​​the exposed area of ​​the external sensor, and the inrush current velocity, and calculate the inrush current load expression as follows: Fyl(st)=ρ*MJ*sd(st), where Fyl(st) represents the inrush current load at time st, ρ represents the fluid density at the location of the external sensor, MJ represents the cross-sectional area of ​​the exposed area of ​​the external sensor, and sd(st) represents the inrush current velocity at time st;

[0022] Calculate the wave surge anomaly coefficient Blyl, the expression is as follows Where a1 and a2 represent the preset proportional coefficients of wave load and surge load respectively.

[0023] In a preferred embodiment, by acquiring the attachment information of underwater organisms when the external sensor is running, the influence of the underwater organism attachment on the external sensor is analyzed, and the underwater organism attachment coefficient is obtained to measure the degree to which the external sensor is affected by the underwater organism attachment;

[0024] The logic for obtaining the underwater biological adhesion coefficient is as follows:

[0025] Obtain the biomass attached to the surface of the external sensor at time sk, the biomass concentration in the underwater area where the external sensor is located, and the water flow velocity at time sk, and calculate the bioattachment rate Fz, which is expressed as follows: Where zl(sk) represents the biomass attached at time sk, zl max represents the expected maximum attached biomass on the surface of the external sensor, sl(sk) represents the water velocity at time sk, nd(sk) represents the biomass concentration in the underwater area where the external sensor is located, and the value range of sk is [0, SK], where SK is a positive integer; the bioshedding rate Tl is calculated as follows: Calculate the underwater biological attachment coefficient Sxsf, the expression is as follows Sxsf where Fzc represents the initial attached biomass on the external sensor surface.

[0026] In a preferred embodiment, a data anomaly detection impact model is constructed based on the salt water corrosion abnormal rise coefficient, the wave surge turbulence abnormal coefficient, and the underwater biological attachment coefficient to generate a data anomaly detection impact index Sjyx. The model is based on the following formula: Wherein b1, b2, and b3 represent the preset proportional coefficients of the abnormal rise coefficient of salt water corrosion, the abnormal coefficient of wave surge turbulence, and the underwater biological attachment coefficient, respectively, and b1, b2, and b3 are all greater than 0.

[0027] In a preferred embodiment, the data anomaly detection impact index is compared with a preset data anomaly detection impact index threshold to evaluate the impact of salt water corrosion, dynamic loads of waves and currents, and attachment of underwater organisms on existing data anomaly detection, as follows:

[0028] If the data anomaly detection impact index is greater than the data anomaly detection impact index threshold, an anomaly detection impact signal is generated;

[0029] If the data anomaly detection impact index is less than or equal to the data anomaly detection impact index threshold, there is no need to generate an anomaly detection impact signal.

[0030] In a preferred embodiment, when an abnormality detection influencing signal is generated, a data abnormality detection machine learning model is constructed to perform abnormality detection on reinforced concrete structure data transmitted by an external sensor, as follows:

[0031] Step S4-1, preprocessing the reinforced concrete structure data transmitted by the external sensor, and dividing the data set into a training set and a test set;

[0032] Step S4-2, selecting a radial basis kernel function as a kernel function to map the reinforced concrete structure data to a high-dimensional space, constructing an optimization objective function to obtain a hyperplane that maximizes the interval, and separating the data in the high-dimensional space;

[0033] Step S4-3, obtain the difference CAS between the data anomaly detection impact index and the data anomaly detection impact index threshold, and the difference FAK of the reinforced concrete structure data transmitted by the external sensor and the internal sensor to adjust the width parameter σ and the penalty parameter C of the radial basis kernel, as follows where σ new represents the width parameter of the new radial basis kernel, C new represents a new penalty parameter, p1 and p2 represent the difference CAS between the data anomaly detection influence index and the data anomaly detection influence index threshold when the width parameter of the radial basis kernel is adjusted, and the preset proportional coefficients of the reinforced concrete structure data difference value FAK transmitted by the external sensor and the internal sensor, respectively; q1 and q2 represent the difference CAS between the data anomaly detection influence index and the data anomaly detection influence index threshold when the penalty parameter is adjusted, and the preset proportional coefficients of the reinforced concrete structure data difference value FAK transmitted by the external sensor and the internal sensor, respectively;

[0034] Step S4-3, obtaining a decision function by solving the optimization objective function, and performing anomaly detection on the reinforced concrete structure data transmitted by the external sensor according to the decision function.

[0035] Technical effects and advantages of the present invention:

[0036] 1. The present invention acquires salt water corrosion information, dynamic load information of waves and currents, and attachment information of underwater organisms when external sensors are running, constructs a data anomaly detection impact model, generates a data anomaly detection impact index, measures the impact of salt water corrosion, dynamic load of waves and currents, and attachment of underwater organisms on existing data anomaly detection, distinguishes the different impacts of data transmitted by external sensors and internal sensors, compares the data anomaly detection impact index with a preset data anomaly detection impact index threshold, evaluates the impact of salt water corrosion, dynamic load of waves and currents, and attachment of underwater organisms on existing data anomaly detection, and when an anomaly detection impact signal is generated, constructs a data anomaly detection machine learning model, and optimizes data anomaly detection according to the impact of data anomaly detection and the difference in reinforced concrete structure data transmitted by external sensors and internal sensors, avoids too frequent false alarms and missed alarms, timely adjusts the detection strategy, improves the accurate monitoring of the health status of the structure, enhances the stability and reliability of the system, truly reflects the health status of the structure, improves the trust of managers and decision makers in the system, promotes the adoption of monitoring data and the formulation of subsequent decisions, and enhances the application of intelligent monitoring systems in actual engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0038] Figure 1 Schematic diagram of the structure of the method of the embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] Example: Figure 1 The present invention provides a method for real-time monitoring and data processing of reinforced concrete structures, comprising the following steps:

[0041] Step S1, monitoring, obtaining salt water corrosion information, dynamic load information of waves and currents, and attachment information of underwater organisms when the external sensor is in operation;

[0042] Step S2, constructing a data anomaly detection impact model, generating a data anomaly detection impact index, and measuring the impact of salt water corrosion, dynamic loads of waves and surges, and attachment of underwater organisms on existing data anomaly detection;

[0043] Step S3, comparing the data anomaly detection impact index with a preset data anomaly detection impact index threshold, and evaluating the impact of salt water corrosion, dynamic loads of waves and currents, and attachment of underwater organisms on existing data anomaly detection;

[0044] Step S4, when an anomaly detection impact signal is generated, a data anomaly detection machine learning model is constructed, and the data anomaly detection is optimized according to the impact degree of the data anomaly detection and the difference in reinforced concrete structure data transmitted by the external sensor and the internal sensor;

[0045] Step S1, monitoring, obtaining salt water corrosion information, dynamic load information of waves and currents, and attachment information of underwater organisms when the external sensor is in operation;

[0046] The salt water corrosion information includes the salt water corrosion abnormal rise coefficient Ysfs, the wave and surge dynamic load information includes the wave surge turbulence abnormal coefficient Blyl, and the underwater organism attachment information includes the underwater organism attachment coefficient Sxsf;

[0047] The salt water corrosion abnormal rise coefficient is used to measure the abnormal rise degree of salt water corrosion of sensors deployed outside reinforced concrete structures, reflecting the corrosion resistance of sensor materials and the impact of environmental factors on their performance, and helping to evaluate potential structural safety risks. By calculating the salt water corrosion abnormal rise coefficient, it is helpful to timely discover the trend of sensor corrosion, capture the noise caused by increased sensor signal distortion due to corrosion abnormality, and avoid erroneously identifying normal fluctuations as abnormalities due to abnormal increase in noise in sensor signals, resulting in frequent false alarms. The larger the salt water corrosion abnormal rise coefficient, the deeper the abnormal rise degree of salt water corrosion of external sensors, and the greater its impact on the existing data anomaly detection; conversely, the smaller the salt water corrosion abnormal rise coefficient, the shallower the abnormal rise degree of salt water corrosion of external sensors, and the smaller its impact on the existing data anomaly detection;

[0048] Therefore, by obtaining the salt water corrosion information of the external sensor during operation, the abnormal increase of the external sensor due to salt water corrosion is analyzed, and the salt water corrosion abnormal increase coefficient is obtained to measure the abnormal increase degree of the external sensor due to salt water corrosion;

[0049] The logic for obtaining the abnormal rise coefficient of salt water corrosion is as follows:

[0050] Obtain the working current, working voltage, electrochemical impedance, salt water concentration, and ambient temperature of the external sensor within a fixed time period T, and calculate the corrosion rate Fsx at time t t , the expression is as follows Where cs = {1, 2, ..., 5}, TZ1 represents the working current, TZ2 represents the working voltage, TZ3 represents the electrochemical impedance, TZ4 represents the salt water concentration, TZ5 represents the ambient temperature, β0, β cs represents the regression coefficient;

[0051] It should be noted that before calculating the corrosion rate, it is necessary to ensure that the working current, working voltage, electrochemical impedance, salt water concentration, and ambient temperature are all normalized. Common normalization methods include Min-Max normalization and Z-Score standardization. The regression coefficient can be obtained by substituting existing historical data, passing statistical tests, and fitting with the matlab program according to the principle of least squares, which will not be described in detail here.

[0052] Calculate the baseline corrosion rate jzs, the expression is as follows Where t={1,2,...,T}, T is a positive integer;

[0053] The corrosion rate at time t+1 is compared with the corrosion rate at time t. When the corrosion rate at time t+1 is greater than the corrosion rate at time t, the corrosion rate at time t+1 is marked as an abnormal corrosion rate, and the corrosion rate rise value ssz at adjacent times is calculated. The expression is as follows: Among them Fsx t+1 represents the corrosion rate at time t+1;

[0054] Calculate the abnormal rise coefficient of salt water corrosion Ysfs, the expression is as follows Among them ssz jh represents the jhth occurrence of an event in which the corrosion rate at time t+1 is greater than the corrosion rate at time t, jh = {1, 2, ..., J}, J is a positive integer, YFsx n represents the corrosion rate marked as abnormal corrosion rate for the nth time, na={1,2,...,N}, N is a positive integer;

[0055] The wave surge and turbulence anomaly coefficient is used to measure the degree to which sensors deployed outside reinforced concrete structures are affected by the dynamic loads of waves and surges. These dynamic loads usually come from waves and surges in the ocean or other water environments, which may cause mechanical vibration or displacement of external sensors, thereby affecting their normal operation and even causing data anomalies or distortion. By calculating the wave surge and turbulence anomaly coefficient, the performance of sensors under wave and surge loads can be identified in a timely manner, and whether there are anomalies can be assessed, and further measures can be taken to protect the equipment or improve the data analysis method to avoid false alarms or missed alarms. The larger the wave surge and turbulence anomaly coefficient, the more deeply the external sensors are affected by the dynamic loads of waves and surges, and the greater the impact on the detection of existing data anomalies. Conversely, the smaller the wave surge and turbulence anomaly coefficient, the less deeply the external sensors are affected by the dynamic loads of waves and surges, and the smaller the impact on the detection of existing data anomalies.

[0056] Therefore, by obtaining the dynamic load information of waves and surges, the influence of the dynamic loads of waves and surges on the external sensors is analyzed, and the wave surge turbulence anomaly coefficient is obtained to measure the degree to which the external sensors are affected by the dynamic loads of waves and surges;

[0057] The logic for obtaining the wave surge and turbulence anomaly coefficient is as follows:

[0058] The wave frequency pl at the location of the external sensor is obtained through the ocean radar, and the wave energy S(pl) at different wave frequencies is calculated. The expression is as follows Where g represents the acceleration due to gravity, c1 and c2 are constant factors, which can be set by technicians in this field according to the marine environment conditions;

[0059] The wave load calculation expression is as follows: Where Fpl(st) represents the wave load at time st, represents the phase angle;

[0060] Obtain the fluid density at the location of the external sensor, the cross-sectional area of ​​the exposed area of ​​the external sensor, and the inrush current velocity, and calculate the inrush current load expression as follows: Fyl(st)=ρ*MJ*sd(st), where Fyl(st) represents the inrush current load at time st, ρ represents the fluid density at the location of the external sensor, MJ represents the cross-sectional area of ​​the exposed area of ​​the external sensor, and sd(st) represents the inrush current velocity at time st;

[0061] Calculate the wave surge anomaly coefficient Blyl, the expression is as follows Where a1 and a2 represent the preset proportional coefficients of wave load and surge load respectively;

[0062] It should be noted that before calculating the wave surge and turbulence anomaly coefficient, it is necessary to ensure that the wave load and surge load are normalized, and a1 and a2 are set according to the actual situation. For example, the expert weighting method is adopted, that is, experts in related fields are invited to determine the preset proportional coefficients of various indicators through professional opinion surveys and comprehensive evaluations;

[0063] The underwater biological adhesion coefficient is used to measure the degree to which sensors deployed outside reinforced concrete structures are affected by underwater biological adhesion. Specifically, it reflects the degree of attachment of underwater organisms (such as marine algae, shellfish, marine microorganisms, etc.) to the sensor surface, as well as the potential impact of these attachments on sensor performance, measurement accuracy and sensor service life. The impact of biological adhesion is not only reflected in the contamination and wear of the sensor surface itself, but may also affect the working performance of the sensor by increasing surface friction, changing thermal conductivity, electrical interference, etc. Calculating the underwater biological adhesion coefficient helps to identify and capture the impact of the degree of attachment of underwater organisms on external sensors in a timely manner, and to deal with it before biological adhesion has a significant impact on sensor performance, thereby extending the service life of the sensor; the larger the underwater biological adhesion coefficient, the deeper the external sensor is affected by underwater biological adhesion, and the greater its impact on the existing data anomaly detection; conversely, the smaller the underwater biological adhesion coefficient, the shallower the external sensor is affected by underwater biological adhesion, and the smaller its impact on the existing data anomaly detection;

[0064] Therefore, by obtaining the attachment information of underwater organisms when the external sensor is running, the influence of underwater organism attachment on the external sensor is analyzed, and the underwater organism attachment coefficient is obtained to measure the degree to which the external sensor is affected by underwater organism attachment;

[0065] The logic for obtaining the underwater biological adhesion coefficient is as follows:

[0066] Obtain the biomass attached to the surface of the external sensor at time sk, the biomass concentration in the underwater area where the external sensor is located, and the water flow velocity at time sk, and calculate the bioattachment rate Fz, which is expressed as follows: Where zl(sk) represents the biomass attached at time sk, zl max represents the expected maximum attached biomass on the surface of the external sensor, sl(sk) represents the water velocity at time sk, nd(sk) represents the biomass concentration in the underwater area where the external sensor is located, and the value range of sk is [0, SK], where SK is a positive integer; the bioshedding rate Tl is calculated as follows: Calculate the underwater biological attachment coefficient Sxsf, the expression is as follows Where Fzc represents the initial attached biomass on the external sensor surface;

[0067] It should be noted that before calculating the biological attachment rate and biological shedding rate, it is necessary to ensure that the attached biomass, biological concentration, and water flow rate are all normalized. If there is no attached biomass on the external sensor surface at the beginning, Fzc is a minimum value to ensure that the denominator is not 0;

[0068] Step S2, constructing a data anomaly detection impact model, generating a data anomaly detection impact index, and measuring the impact of salt water corrosion, dynamic loads of waves and surges, and attachment of underwater organisms on existing data anomaly detection;

[0069] According to the salt water corrosion abnormal rise coefficient, wave surge turbulence abnormal coefficient, and underwater biological attachment coefficient, a data anomaly detection impact model is constructed to generate the data anomaly detection impact index Sjyx. The formula based on the model is as follows Where b1, b2, and b3 represent the preset proportional coefficients of the salt water corrosion abnormal rise coefficient, the wave surge turbulence abnormal coefficient, and the underwater biological attachment coefficient, respectively, and b1, b2, and b3 are all greater than 0;

[0070] It should be noted that before constructing the data anomaly detection impact model, it is necessary to ensure that the salt water corrosion abnormal rise coefficient, wave surge turbulence abnormal coefficient, and underwater biological attachment coefficient are all normalized, and b1, b2, and b3 are set according to actual conditions. For example, the expert empowerment method is adopted, that is, experts in related fields are invited to determine the preset proportional coefficients of various indicators through professional opinion surveys and comprehensive evaluations;

[0071] It can be seen from the above calculation expression that the larger the salt water corrosion abnormal rise coefficient, the larger the wave surge turbulence abnormal coefficient, and the larger the underwater biological attachment coefficient, the larger the data anomaly detection influence index, indicating that the external sensor is more deeply affected by salt water corrosion, wave surge, and underwater organisms, and the greater its influence on the existing data anomaly detection; conversely, the smaller the salt water corrosion abnormal rise coefficient, the smaller the wave surge turbulence abnormal coefficient, and the smaller the underwater biological attachment coefficient, the smaller the data anomaly detection influence index, indicating that the external sensor is less affected by salt water corrosion, wave surge, and underwater organisms, and the smaller its influence on the existing data anomaly detection;

[0072] Step S3, comparing the data anomaly detection impact index with a preset data anomaly detection impact index threshold, and evaluating the impact of salt water corrosion, dynamic loads of waves and currents, and attachment of underwater organisms on existing data anomaly detection;

[0073] If the data anomaly detection impact index is greater than the data anomaly detection impact index threshold, it indicates that salt water corrosion, dynamic loads of waves and surges, and attachment of underwater organisms have a great impact on the existing data anomaly detection, and an anomaly detection impact signal is generated;

[0074] If the data anomaly detection impact index is less than or equal to the data anomaly detection impact index threshold, it indicates that the impact of salt water corrosion, dynamic loads of waves and currents, and attachment of underwater organisms on the existing data anomaly detection is small, and there is no need to generate an anomaly detection impact signal;

[0075] Step S4, when an anomaly detection impact signal is generated, a data anomaly detection machine learning model is constructed, and the data anomaly detection is optimized according to the impact degree of the data anomaly detection and the difference in reinforced concrete structure data transmitted by the external sensor and the internal sensor;

[0076] When an anomaly detection impact signal is generated, a data anomaly detection machine learning model is constructed to perform anomaly detection on the reinforced concrete structure data transmitted by the external sensor, as follows:

[0077] Step S4-1, preprocessing the reinforced concrete structure data transmitted by the external sensor, including denoising, missing value filling, data standardization, feature extraction, and dividing the data set into a training set and a test set;

[0078] The reinforced concrete structure data include but are not limited to temperature data, strain data, humidity data, and vibration data;

[0079] Temperature data: records temperature changes of the structure (e.g., external temperature, internal temperature, etc.);

[0080] Strain data: monitors the strain and deformation of the structure under external loads;

[0081] Humidity data: monitor humidity fluctuations in the structure environment, especially in humid environments;

[0082] Vibration data: records vibration responses caused by external environments (such as waves, earthquakes, etc.);

[0083] Step S4-2, select the radial basis kernel function as the kernel function to map the reinforced concrete structure data to the high-dimensional space. The radial basis kernel function optimizes the objective function as follows: where x i 、x j Represents the data point, σ represents the width parameter of the radial basis kernel; construct the optimization objective function to obtain the hyperplane that maximizes the interval and separate the data in the high-dimensional space. The optimization objective function is as follows Where w represents the normal vector of the hyperplane, b represents the bias of the hyperplane, ξ represents the slack variable used to tolerate misclassification, and C represents the penalty parameter to control the degree of penalty for misclassification;

[0084] Step S4-3, obtain the difference CAS between the data anomaly detection impact index and the data anomaly detection impact index threshold, and the difference FAK of the reinforced concrete structure data transmitted by the external sensor and the internal sensor to adjust the width parameter σ and the penalty parameter C of the radial basis kernel, as follows where σ new represents the width parameter of the new radial basis kernel, C new represents a new penalty parameter, p1 and p2 represent the difference CAS between the data anomaly detection influence index and the data anomaly detection influence index threshold when the width parameter of the radial basis kernel is adjusted, and the preset proportional coefficients of the reinforced concrete structure data difference value FAK transmitted by the external sensor and the internal sensor, respectively; q1 and q2 represent the difference CAS between the data anomaly detection influence index and the data anomaly detection influence index threshold when the penalty parameter is adjusted, and the preset proportional coefficients of the reinforced concrete structure data difference value FAK transmitted by the external sensor and the internal sensor, respectively;

[0085] It should be noted that p1, p2, q1, and q2 are set according to actual conditions. For example, an expert weighting method is adopted, that is, experts in related fields are invited to determine the preset proportional coefficients of various indicators through professional opinion surveys and comprehensive evaluations. The difference value of reinforced concrete structure data transmitted by external sensors and internal sensors can be calculated by the Euclidean distance calculation method, which will not be elaborated here.

[0086] Step S4-3, a decision function is obtained by solving the optimization objective function, and anomaly detection is performed on the reinforced concrete structure data transmitted by the external sensor according to the decision function. The obtained decision function is as follows: Among them, f(x) represents the output value of the decision function, αi represents the Lagrange multiplier, y i represents the label of the training sample, K(x i , x) represents the kernel function, b represents the bias term; the output value of the decision function is compared with the preset abnormal threshold, if the output value of the decision function is greater than the abnormal threshold, the reinforced concrete structure data point is determined to be normal, if the output value of the decision function is less than or equal to the abnormal threshold, the reinforced concrete structure data point is determined to be abnormal;

[0087] The present invention acquires salt water corrosion information, dynamic load information of waves and currents, and attachment information of underwater organisms when an external sensor is running, constructs a data anomaly detection impact model, generates a data anomaly detection impact index, measures the impact of salt water corrosion, dynamic load of waves and currents, and attachment of underwater organisms on existing data anomaly detection, distinguishes different impacts of data transmitted by external sensors and internal sensors, compares the data anomaly detection impact index with a preset data anomaly detection impact index threshold, evaluates the impact of salt water corrosion, dynamic load of waves and currents, and attachment of underwater organisms on existing data anomaly detection, and when an anomaly detection impact signal is generated, constructs a data anomaly detection machine learning model, and optimizes data anomaly detection according to the impact of data anomaly detection and the difference in reinforced concrete structure data transmitted by external sensors and internal sensors, avoids too frequent false alarms and missed alarms, timely adjusts the detection strategy, improves the accurate monitoring of the health status of the structure, enhances the stability and reliability of the system, truly reflects the health status of the structure, improves the trust of managers and decision makers in the system, promotes the adoption of monitoring data and the formulation of subsequent decisions, and enhances the application of intelligent monitoring systems in actual projects.

[0088] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0089] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0090] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for real-time monitoring and data processing of reinforced concrete structures, characterized in that: The steps include: Step S1, monitoring, obtaining salt water corrosion information, dynamic load information of waves and currents, and attachment information of underwater organisms when the external sensor is in operation; Step S2, constructing a data anomaly detection impact model, generating a data anomaly detection impact index, and measuring the impact of salt water corrosion, dynamic loads of waves and surges, and attachment of underwater organisms on existing data anomaly detection; Step S3, comparing the data anomaly detection impact index with a preset data anomaly detection impact index threshold, and evaluating the impact of salt water corrosion, dynamic loads of waves and currents, and attachment of underwater organisms on existing data anomaly detection; Step S4, when an anomaly detection impact signal is generated, a data anomaly detection machine learning model is constructed, and the data anomaly detection is optimized according to the impact degree of the data anomaly detection and the difference in reinforced concrete structure data transmitted by the external sensor and the internal sensor.

2. The method for real-time monitoring and data processing of reinforced concrete structure according to claim 1, characterized in that: The salt water corrosion information includes the salt water corrosion abnormal rise coefficient Ysfs, the wave and surge dynamic load information includes the wave surge turbulence abnormal coefficient Blyl, and the underwater organism attachment information includes the underwater organism attachment coefficient Sxsf.

3. The method for real-time monitoring and data processing of reinforced concrete structure according to claim 2 is characterized in that: By acquiring the salt water corrosion information of the external sensor during operation, analyzing the abnormal increase of the external sensor due to salt water corrosion, and obtaining the salt water corrosion abnormal increase coefficient, the abnormal increase degree of the external sensor due to salt water corrosion is measured; The logic for obtaining the abnormal rise coefficient of salt water corrosion is as follows: Obtain the working current, working voltage, electrochemical impedance, salt water concentration, and ambient temperature of the external sensor within a fixed time period T, and calculate the corrosion rate Fsx at time t t , the expression is as follows Where cs = {1, 2, ..., 5}, TZ1 represents the working current, TZ2 represents the working voltage, TZ3 represents the electrochemical impedance, TZ4 represents the salt water concentration, TZ5 represents the ambient temperature, β0, β cs represents the regression coefficient; Calculate the baseline corrosion rate jzs, the expression is as follows Where t={1,2,...,T}, T is a positive integer; The corrosion rate at time t+1 is compared with the corrosion rate at time t. When the corrosion rate at time t+1 is greater than the corrosion rate at time t, the corrosion rate at time t+1 is marked as an abnormal corrosion rate, and the corrosion rate rise value ssz at adjacent times is calculated. The expression is as follows: Among them Fsx t+1 represents the corrosion rate at time t+1; Calculate the abnormal rise coefficient of salt water corrosion Ysfs, the expression is as follows Among them ssz jh represents the jhth occurrence of an event in which the corrosion rate at time t+1 is greater than the corrosion rate at time t, jh = {1, 2, ..., J}, J is a positive integer, YFsx n It represents the corrosion rate marked as abnormal corrosion rate for the nth time, na={1,2,...,N}, where N is a positive integer.

4. The method for real-time monitoring and data processing of reinforced concrete structure according to claim 2 is characterized in that: By acquiring the dynamic load information of waves and surges, the influence of the dynamic loads of waves and surges on the external sensors is analyzed, and the wave surge turbulence anomaly coefficient is obtained to measure the degree to which the external sensors are affected by the dynamic loads of waves and surges; The logic for obtaining the wave surge and turbulence anomaly coefficient is as follows: The wave frequency pl at the location of the external sensor is obtained through the ocean radar, and the wave energy S(pl) at different wave frequencies is calculated. The expression is as follows Where g represents the acceleration due to gravity, c1 and c2 are constant factors; The wave load calculation expression is as follows: Where Fpl(st) represents the wave load at time st, represents the phase angle; Obtain the fluid density at the location of the external sensor, the cross-sectional area of ​​the exposed area of ​​the external sensor, and the inrush current velocity, and calculate the inrush current load expression as follows: Fyl(st)=ρ*MJ*sd(st), where Fyl(st) represents the inrush current load at time st, ρ represents the fluid density at the location of the external sensor, MJ represents the cross-sectional area of ​​the exposed area of ​​the external sensor, and sd(st) represents the inrush current velocity at time st; Calculate the wave surge anomaly coefficient Blyl, the expression is as follows Where a1 and a2 represent the preset proportional coefficients of wave load and surge load respectively.

5. The method for real-time monitoring and data processing of reinforced concrete structure according to claim 2, characterized in that: By acquiring the attachment information of underwater organisms when the external sensor is running, the influence of underwater organism attachment on the external sensor is analyzed, and the underwater organism attachment coefficient is obtained to measure the degree to which the external sensor is affected by underwater organism attachment; The logic for obtaining the underwater biological adhesion coefficient is as follows: Obtain the biomass attached to the surface of the external sensor at time sk, the biomass concentration in the underwater area where the external sensor is located, and the water flow velocity at time sk, and calculate the bioattachment rate Fz, which is expressed as follows: Where zl(sk) represents the biomass attached at time sk, zl max represents the expected maximum attached biomass on the surface of the external sensor, sl(sk) represents the water velocity at time sk, nd(sk) represents the biomass concentration in the underwater area where the external sensor is located, and the value range of sk is [0, SK], where SK is a positive integer; the bioshedding rate Tl is calculated as follows: Calculate the underwater biological attachment coefficient Sxsf, the expression is as follows where Fzc represents the initial attached biomass on the external sensor surface.

6. The method for real-time monitoring and data processing of reinforced concrete structure according to claim 2, characterized in that: According to the salt water corrosion abnormal rise coefficient, wave surge turbulence abnormal coefficient, and underwater biological attachment coefficient, a data anomaly detection impact model is constructed to generate the data anomaly detection impact index Sjyx. The formula based on the model is as follows Wherein b1, b2, and b3 represent the preset proportional coefficients of the abnormal rise coefficient of salt water corrosion, the abnormal coefficient of wave surge turbulence, and the underwater biological attachment coefficient, respectively, and b1, b2, and b3 are all greater than 0.

7. The method for real-time monitoring and data processing of reinforced concrete structure according to claim 6, characterized in that: The data anomaly detection impact index is compared with the preset data anomaly detection impact index threshold to evaluate the impact of salt water corrosion, dynamic loads of waves and currents, and attachment of underwater organisms on the existing data anomaly detection, as follows: If the data anomaly detection impact index is greater than the data anomaly detection impact index threshold, an anomaly detection impact signal is generated; If the data anomaly detection impact index is less than or equal to the data anomaly detection impact index threshold, there is no need to generate an anomaly detection impact signal.

8. The method for real-time monitoring and data processing of reinforced concrete structure according to claim 7 is characterized in that: When an anomaly detection impact signal is generated, a data anomaly detection machine learning model is constructed to perform anomaly detection on the reinforced concrete structure data transmitted by the external sensor, as follows: Step S4-1, preprocessing the reinforced concrete structure data transmitted by the external sensor, and dividing the data set into a training set and a test set; Step S4-2, selecting a radial basis kernel function as a kernel function to map the reinforced concrete structure data to a high-dimensional space, constructing an optimization objective function to obtain a hyperplane that maximizes the interval, and separating the data in the high-dimensional space; Step S4-3, obtain the difference CAS between the data anomaly detection impact index and the data anomaly detection impact index threshold, and the difference FAK of the reinforced concrete structure data transmitted by the external sensor and the internal sensor to adjust the width parameter σ and the penalty parameter C of the radial basis kernel, as follows where σ new represents the width parameter of the new radial basis kernel, C new represents a new penalty parameter, p1 and p2 represent the difference CAS between the data anomaly detection influence index and the data anomaly detection influence index threshold when the width parameter of the radial basis kernel is adjusted, and the preset proportional coefficients of the reinforced concrete structure data difference value FAK transmitted by the external sensor and the internal sensor, respectively; q1 and q2 represent the difference CAS between the data anomaly detection influence index and the data anomaly detection influence index threshold when the penalty parameter is adjusted, and the preset proportional coefficients of the reinforced concrete structure data difference value FAK transmitted by the external sensor and the internal sensor, respectively; Step S4-3, obtaining a decision function by solving the optimization objective function, and performing anomaly detection on the reinforced concrete structure data transmitted by the external sensor according to the decision function.

Images