Concrete pouring homogeneity detection system and method based on sensing data

Abstract

The invention relates to the technical field of concrete pouring, in particular to a concrete pouring homogeneity detection system and method based on sensing data, and aims at achieving efficient and accurate concrete pouring homogeneity evaluation and defect detection. The system comprises a multi-sensor sensing module, a multi-sensor data conversion and transmission module and a concrete homogeneity evaluation module, the multi-sensor sensing module comprises a temperature sensor, an ultrasonic sensor and a vibration sensor; the multi-sensing data conversion and transmission module is used for acquiring multi-sensing data and performing data type conversion and transmission on the multi-sensing data; and the concrete homogeneity evaluation module is used for constructing a concrete pouring homogeneity evaluation model to detect and evaluate the homogeneity and defects in the concrete pouring process. According to the invention, detection and evaluation of the concrete pouring homogeneity are realized through combination of multi-sensing data, and the accuracy of detection of the concrete pouring homogeneity is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of concrete pouring, and in particular to a concrete pouring homogeneity detection system and method based on sensor data. Background Art

[0002] Concrete is one of the most basic and important structural materials in modern construction projects. It is widely used in various buildings, infrastructure and road projects due to its excellent physical properties, customizability, cost-effectiveness and sustainability. However, during the pouring process, concrete may have voids, cracks and material separation defects, which may lead to local strength reduction and insufficient durability, and even cause safety hazards. These defects usually need to be remedied by later repairs or rework, which not only increases time and economic costs, but may also affect the construction progress. Therefore, there is an urgent need for a fast and accurate concrete pouring homogeneity detection system to detect problems in time and take remedial measures, so as to help the construction team immediately adjust the process to ensure the quality and safety of the project.

[0003] At present, there are some intelligent methods for testing the homogeneity of concrete, including ultrasonic testing technology, machine vision testing technology and artificial neural network methods. However, these methods mainly detect the homogeneity of concrete by collecting single sensor data, which may have defects such as insufficient data interpretability, sensitivity to environmental conditions, limited detection range and poor positioning accuracy, resulting in insufficient accuracy and reliability of concrete homogeneity detection.

[0004] Therefore, a concrete pouring homogeneity detection system and method based on sensor data is proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a concrete pouring homogeneity detection system and method based on sensor data, which realizes efficient and accurate concrete pouring homogeneity evaluation and defect detection through the collaborative work of multiple modules. First, the multi-sensor perception module of the system comprehensively monitors the temperature change, sound velocity change and vibration uniformity of concrete during the pouring process by setting temperature sensors, ultrasonic sensors and vibration sensors; secondly, the multi-sensor data conversion and transmission module collects multi-sensor data through the multi-sensor perception module, converts the data type of the multi-sensor data and transmits it to the concrete homogeneity evaluation module; finally, the concrete homogeneity evaluation module constructs a concrete pouring homogeneity evaluation model based on the data transmitted by the multi-sensor data conversion and transmission module, and detects and evaluates the homogeneity and defects in the concrete pouring process.

[0006] To achieve the above object, the present invention provides the following technical solutions: A concrete pouring homogeneity detection system based on sensor data, comprising: A multi-sensor sensing module is used to comprehensively monitor the homogeneity and state changes of concrete during the pouring process through multiple sensors; the multiple sensors include: a temperature sensor, an ultrasonic sensor and a vibration sensor; the temperature sensor is used to monitor the temperature changes of concrete during the pouring process; the ultrasonic sensor is used to use ultrasonic waves to penetrate concrete during the pouring process to analyze the changes in sound velocity in real time, detect internal defects and density; the vibration sensor is used to monitor the vibration frequency and amplitude during the pouring process, and evaluate the vibration uniformity of the concrete; A multi-sensor data conversion and transmission module, used for collecting multi-sensor data through the multi-sensor perception module, converting the data type of the multi-sensor data and transmitting it to the concrete homogeneity evaluation module; The concrete homogeneity assessment module is used to construct a concrete pouring homogeneity assessment model based on the data transmitted by the multi-sensor data conversion and transmission module, and to detect and conduct quality assessment on the homogeneity and defects during the concrete pouring process; it includes: a data preprocessing layer, a feature extraction layer, a homogeneity assessment and a defect detection layer.

[0007] Preferably, the temperature sensors are located at different heights in the concrete pouring area and are arranged along the vertical direction of the concrete pouring area, including the surface, middle layer and bottom of the pouring area, respectively. The vertical distance between each temperature sensor is set. The ultrasonic sensor is located in the concrete pouring area, including the middle position of the four vertical sides of the pouring area and the center of the pouring area. The vibration sensor is located at the upper left corner, the center position and the lower right corner of the surface of the concrete pouring area and on the vibrating equipment, including on the head of the vibrator and the bottom of the vibrator in contact with the concrete.

[0008] Preferably, the multi-sensor data acquisition and transmission module includes: a data acquisition layer, a data conversion layer and a data transmission layer; the data acquisition layer is used to acquire the multi-sensor data through the multi-sensor; the multi-sensor data includes: initial temperature sensor data, ultrasonic sensor data and initial vibration sensor data; wherein the initial temperature sensor data is an analog signal, which is obtained by the temperature sensor acquiring the temperature value of concrete at different positions; the ultrasonic sensor data is a digital signal, which is obtained by the ultrasonic sensor emitting and receiving ultrasonic signals and acquiring the propagation time, sound speed and echo intensity of the ultrasonic wave; the initial vibration sensor data is an analog signal, which is obtained by the vibration sensor acquiring the vibration frequency and amplitude; The data conversion layer is used to process and convert the multi-sensor data to ensure that the multi-sensor data can be subsequently analyzed in a unified format; for the analog signal data collected by the temperature sensor and the vibration sensor, the analog-to-digital converter is used to convert it into a digital signal to obtain the temperature sensor data and the vibration sensor data; The data transmission layer is used to transmit the ultrasonic sensing data and the temperature sensing data and the vibration sensing data obtained by the data conversion layer to the concrete homogeneity evaluation module through wireless communication technology, so as to facilitate subsequent concrete homogeneity analysis and defect detection.

[0009] Preferably, the data preprocessing layer includes: filtering and denoising, data smoothing and interpolation, data normalization and time synchronization to obtain standard temperature sensor data, standard ultrasonic sensor data and standard vibration sensor data; wherein, the filtering and denoising includes: denoising the multi-sensor data by filtering technology to eliminate the noise of the multi-sensor due to the concrete pouring site during collection; the data smoothing and interpolation includes: smoothing the multi-sensor data by linear interpolation algorithm and filling the missing data; the data normalization includes: uniformly scaling the signals of different sensors to ensure that the data collected by the multi-sensor are in the same numerical range; the time synchronization includes: timestamp matching the multi-sensor data collected by the multi-sensor to ensure that the multi-sensor data are in the same time window.

[0010] Preferably, the feature extraction layer includes: extracting features related to concrete casting homogeneity and potential defects, so as to establish the concrete casting homogeneity evaluation model to evaluate the concrete building homogeneity and detect defects, and obtain concrete features; the concrete features include: temperature features, ultrasonic features and vibration features; wherein the temperature features include the standard deviation and temperature gradient of concrete temperature distribution; the ultrasonic features include the sound velocity standard deviation and sound velocity variation range; the vibration features include the coefficient of variation of vibration frequency and the coefficient of variation of vibration amplitude; The weight coefficient is set according to the weight of the feedback data of concrete temperature, concrete internal defects and vibration uniformity of different sensors, and the temperature feature, the ultrasonic feature and the vibration feature are integrated by weighted average method to obtain a comprehensive feature index; the comprehensive feature index is: ; in, is the comprehensive characteristic index; is the standard deviation of the temperature distribution; is the temperature gradient; is the temperature feature weight; is the standard deviation of the speed of sound; is the range of sound speed variation; is the ultrasonic feature weight; is the coefficient of variation of the vibration frequency; is the coefficient of variation of the vibration frequency; is the vibration feature weight; ; According to the comprehensive characteristic index, the homogeneity level and defect type are divided by setting thresholds, including: When , it is judged as high homogeneity; when and When , it is judged that the homogeneity is poor and crack defects are detected; when and When , it is judged that the homogeneity is poor and a void defect is detected; when and When , it is judged that the homogeneity is poor and material separation defects are detected; among them, is the high homogeneity threshold; is the threshold for detecting crack defects; is the threshold for detecting void defects; The threshold for detecting material separation defects.

[0011] Preferably, the homogeneity assessment and defect detection layer includes: selecting the best model as the concrete casting homogeneity assessment model by comparing various machine learning models to perform concrete homogeneity assessment and defect detection, wherein the best model is a decision tree model; training the concrete casting homogeneity assessment model to detect and evaluate the homogeneity and defects of concrete casting according to the temperature characteristics, ultrasonic characteristics, vibration characteristics and comprehensive characteristic index obtained by the feature extraction layer and the homogeneity grade and defect type divided by the comprehensive characteristic index; the specific training process is: dividing the temperature characteristics, the ultrasonic characteristics, the vibration characteristics and the comprehensive characteristic index into a training set, a validation set and a test set; using the training set as an input variable, inputting it into the concrete casting homogeneity assessment model to train the decision tree model; splitting the nodes of the decision tree model according to information gain; outputting the homogeneity grade of the training set; optimizing the parameters of the decision tree model by cross-validation through the validation set; outputting the homogeneity assessment and defect detection results of concrete by using the optimized decision tree model through the test set, and evaluating the accuracy of the defect detection results.

[0012] A method for detecting homogeneity of concrete pouring based on sensor data comprises the following steps: The uniformity and state changes of concrete during the pouring process are comprehensively monitored by multiple sensors; the multiple sensors include: temperature sensors, ultrasonic sensors and vibration sensors; the temperature sensor is used to monitor the temperature changes of concrete during the pouring process; the ultrasonic sensor is used to use ultrasonic waves to penetrate concrete during the pouring process to analyze the changes in sound speed in real time, detect internal defects and density; the vibration sensor is used to monitor the vibration frequency and amplitude during the pouring process and evaluate the vibration uniformity of the concrete.

[0013] Preferably, the multi-sensor data is collected by the multi-sensor, and the data type is converted and transmitted.

[0014] Preferably, a concrete pouring homogeneity assessment model is constructed based on the multi-sensor data after the data type conversion, and the homogeneity and defects in the concrete pouring process are detected and the quality is evaluated; it includes: a data preprocessing layer, a feature extraction layer, a homogeneity evaluation and defect detection layer; wherein the feature extraction layer includes: extracting features related to the concrete pouring homogeneity and potential defects, so as to establish the concrete pouring homogeneity assessment model to evaluate the concrete building homogeneity and detect defects, and obtain concrete characteristics; the concrete characteristics include: temperature characteristics, ultrasonic characteristics and vibration characteristics; according to the weights of the feedback data of different sensors on the concrete temperature, concrete internal defects and vibration uniformity, a weighting coefficient is set, and the temperature characteristics, the ultrasonic characteristics and the vibration characteristics are fused by the weighted average method to obtain a comprehensive characteristic index.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention proposes a concrete data collection method in the concrete pouring process based on multiple sensors, which comprehensively monitors the temperature change, sound velocity change and vibration uniformity of concrete during the pouring process by combining temperature sensors, ultrasonic sensors and vibration sensors. This method can monitor the internal temperature change of concrete in real time through temperature sensors, identify the risk of cracks caused by temperature gradients; evaluate the internal density and locate potential voids through the sound velocity change of ultrasonic sensors; and detect the defects of concrete material separation by capturing the frequency and amplitude variation during the vibration process through vibration sensors. Compared with a single data source, multi-sensor data fusion can realize comprehensive and accurate detection of concrete pouring homogeneity and defects from three aspects: temperature change, sound velocity change and vibration uniformity; this method of multi-dimensional data collaborative analysis provides multi-dimensional data for the subsequent extraction of temperature characteristics, ultrasonic characteristics, vibration characteristics and comprehensive characteristic indexes to construct a concrete pouring homogeneity evaluation model, and provides a scientific basis for actual construction adjustments, greatly improving the accuracy and efficiency of construction quality control.

[0016] 2. The present invention proposes a method for dividing the homogeneity grade and defect type of concrete pouring based on a comprehensive characteristic index. The temperature characteristics are obtained by calculating the standard deviation and temperature gradient of the concrete temperature distribution, the ultrasonic characteristics are obtained by calculating the standard deviation and sound velocity variation range of the ultrasonic sound velocity, and the vibration characteristics are obtained by calculating the coefficient of variation of the vibration frequency and the coefficient of variation of the vibration amplitude. The temperature characteristics, ultrasonic characteristics and vibration characteristics are integrated by constructing a comprehensive characteristic index, and the homogeneity grade and defect type are divided according to the comprehensive characteristic index. The homogeneity of concrete pouring can be comprehensively quantified from the temperature, internal defects of concrete and vibration uniformity, and defect types such as cracks, voids and material separation can be identified. In the actual construction process, this method can accurately evaluate the overall homogeneity of the concrete pouring process and quickly detect a variety of defect types, so that appropriate adjustments can be made during the production and construction process to optimize the quality of concrete pouring.

[0017] 3. The present invention proposes a concrete pouring homogeneity assessment model, which combines multiple features including temperature features, ultrasonic features, vibration features and comprehensive feature indexes to conduct multi-dimensional comprehensive assessment and precise detection of homogeneity and defects in the concrete pouring process. By comprehensively considering temperature distribution, sound velocity uniformity and vibration parameters, the temperature difference, vibration uniformity and density of the pouring area can be monitored in real time, so as to dynamically analyze the concrete pouring process from multiple dimensions and avoid the one-sidedness of single indicator evaluation. By combining temperature features, ultrasonic features, vibration features and comprehensive feature indexes and applying them to the concrete pouring homogeneity assessment model, it is possible to automatically identify crack problems caused by temperature differences inside large volumes of concrete, timely locate potential voids or areas of insufficient density and identify material separation phenomena. This concrete pouring homogeneity assessment model can greatly improve the efficiency and accuracy of construction quality control through comprehensive detection and analysis capabilities, reduce the error rate of manual detection, and provide a scientific basis for optimizing construction technology and long-term maintenance, ensuring the structural safety and durability of construction projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flow chart of a method for detecting homogeneity of concrete pouring based on sensor data provided by an embodiment of the present invention; Figure 2 A structural diagram of a concrete pouring homogeneity detection system based on sensor data provided by an embodiment of the present invention; Figure 3 A schematic diagram of the arrangement of temperature sensors provided in an embodiment of the present invention; Figure 4 A schematic diagram of the arrangement of ultrasonic sensors provided in an embodiment of the present invention; Figure 5 A schematic diagram of the arrangement of vibration sensors provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] 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.

[0020] Concrete is one of the most basic and important structural materials in modern construction projects. It is widely used in various buildings, infrastructure and road projects due to its excellent physical properties, customizability, cost-effectiveness and sustainability. However, during the pouring process, concrete may have defects such as voids, cracks and material separation, which may lead to local strength reduction and insufficient durability, and even cause safety hazards. Therefore, a fast and accurate concrete pouring homogeneity detection system can help operators detect problems in time and take remedial measures, thereby helping the construction team to immediately adjust the process to ensure project quality and safety.

[0021] The present invention proposes a concrete pouring homogeneity detection system based on sensor data, which can accurately evaluate the homogeneity of concrete during pouring and quickly detect defects. This system is implemented by a concrete pouring homogeneity detection method based on sensor data. For a specific method flowchart and system structure diagram, please refer to Figure 1 and Figure 2 In order to illustrate that the system and method of the present invention can accurately and quickly evaluate the homogeneity of concrete and detect defects during the pouring process, the effectiveness of the present invention will be described in two embodiments below.

[0022] See also Figures 1 to 5 The present invention provides a concrete pouring homogeneity detection system and method based on sensor data, and the technical solution is as follows:

[0023] Embodiment 1 In the embodiment of the present application, the concrete pouring homogeneity analysis process is described in detail using the system and method proposed in the present invention. In the embodiment of the present application, the concrete pouring homogeneity analysis is aimed at the concrete pouring of the foundation pedestal of Building A in a certain building construction project. The total amount of concrete pouring of the foundation pedestal is 60×60×3 cubic meters. Figure 1 and Figure 2 The content describes in detail the concrete pouring homogeneity analysis and defect detection process in the concrete pouring area; among them, Figure 1The specific process of the method proposed in the present invention includes: S10. arranging multiple sensors to collect multiple sensor data during concrete pouring; S20. collecting multiple sensor data according to the multiple sensors, converting the data type of the multiple sensor data and transmitting it; S30. preprocessing the multi-sensor data after the data type conversion to obtain standard multi-sensor data; S40. extracting features from the standard multi-sensor data, and fusing the features through weighted averaging to obtain a comprehensive feature index; S50. constructing a concrete pouring homogeneity assessment model to perform homogeneity assessment and defect detection on concrete pouring; S60. visually displaying the homogeneity assessment and defect detection results of concrete pouring by the concrete pouring homogeneity assessment model, and providing real-time feedback to the operator. Figure 2 The structure diagram of the system proposed by the present invention includes: a multi-sensor perception module, a multi-sensor data conversion and transmission module and a concrete homogeneity evaluation module; Figure 1 and Figure 2 The following is a description of the contents: A concrete pouring homogeneity detection system based on sensor data, comprising: A multi-sensor sensing module is used to comprehensively monitor the homogeneity and state changes of concrete during the pouring process through multiple sensors, corresponding to the above-mentioned step S10; the multiple sensors include: a temperature sensor, an ultrasonic sensor and a vibration sensor; the temperature sensor is used to monitor the temperature change of concrete during the pouring process; the ultrasonic sensor is used to use ultrasonic waves to penetrate concrete during the pouring process to analyze the change of sound velocity in real time, detect internal defects and density; the vibration sensor is used to monitor the vibration frequency and amplitude during the pouring process, and evaluate the vibration uniformity of the concrete; Specifically, see Figure 3 The schematic diagram of the arrangement of temperature sensors is shown, in which the triangles are temperature sensors; the arrangement of the temperature sensors includes: arranging 5 temperature sensors at the four corners and the center of the surface, middle layer and bottom of the concrete pouring area along the vertical direction of the concrete pouring area, and ensuring that each temperature sensor covers an area of ​​every 10 meters, ensuring that the temperature sensors can fully cover the 60×60×3 concrete pouring area to obtain the full range of temperature changes of the concrete during the pouring process; See also Figure 4 The schematic diagram of the arrangement of ultrasonic sensors is shown, in which the circles are ultrasonic sensors; the arrangement of the ultrasonic sensors includes: arranging 6 ultrasonic sensors in the concrete pouring area, wherein 4 ultrasonic sensors are respectively located in the middle of the four vertical sides of the pouring area, 1.5 meters away from the bottom, and the other 2 ultrasonic sensors are arranged in the center area of ​​the surface and bottom of the pouring area, ensuring that the ultrasonic sensors can penetrate the entire concrete area; See also Figure 5 The schematic diagram of the arrangement of vibration sensors is shown, in which the squares are vibration sensors; the arrangement of the vibration sensors includes: arranging one vibration sensor at the upper left corner, the center and the lower right corner of the 60×60 area on the surface of the concrete pouring area, respectively, to detect the uniformity of vibration transmission; and setting two vibration sensors on the vibrating equipment, respectively located at the head of the vibrator and the bottom where the vibrator contacts the concrete, to ensure the uniformity of vibration.

[0024] In the embodiment of the present application, comprehensive monitoring of concrete parameters during the pouring process is achieved through the arrangement of multiple sensors; temperature sensors can monitor heat flow changes at different heights, ultrasonic sensors can evaluate the density of the internal structure, and vibration sensors ensure uniform vibration. The data collected by multiple sensors can help to fully characterize the homogeneity characteristics of concrete, avoid the information blind spots caused by a single sensor, and thus accurately evaluate the homogeneity of the concrete pouring area, which helps to quickly discover problems such as abnormal edge temperature gradients and uneven center vibration, provide a scientific basis for subsequent construction, and improve the analysis capability and reliability of concrete pouring homogeneity.

[0025] Preferably, a multi-sensor data conversion and transmission module is used to collect multi-sensor data through the multi-sensor perception module, convert the data type of the multi-sensor data and transmit it to the concrete homogeneity evaluation module, corresponding to the above-mentioned step S20; the multi-sensor data collection and transmission module includes: a data collection layer, a data conversion layer and a data transmission layer; wherein the data collection layer is used to collect the multi-sensor data through the multi-sensor; the multi-sensor data includes: initial temperature sensor data, ultrasonic sensor data and initial vibration sensor data; wherein the initial temperature sensor data is an analog signal, which is obtained by the temperature sensor collecting the temperature value of the concrete at different positions once per second; the ultrasonic sensor data is a digital signal, which is obtained by the ultrasonic sensor by transmitting and receiving ultrasonic signals , the propagation time, sound speed and echo intensity of the ultrasonic wave are collected; the initial vibration sensing data is an analog signal, which is obtained by the vibration sensor recording the vibration frequency and amplitude in real time; the data conversion layer is used to process the multi-sensing data and convert the data type to ensure that the data of the multi-sensor can be subsequently analyzed in a unified format; for the analog signal data collected by the temperature sensor and the vibration sensor, they are converted into digital signals by using an analog-to-digital converter to obtain temperature sensing data and vibration sensing data; the data transmission layer is used to transmit the ultrasonic sensing data and the temperature sensing data and the vibration sensing data obtained by the data conversion layer to the concrete homogeneity evaluation module through wireless communication technology, so as to facilitate subsequent concrete homogeneity analysis and defect detection.

[0026] In the embodiment of the present application, multiple temperature, ultrasonic and vibration sensors are used to collect data inside the concrete, and these data are converted into a unified digital signal type to facilitate subsequent model analysis and fusion. Subsequently, the data is transmitted to the cloud analysis system via industrial Ethernet, which helps to generate homogeneity assessment results in real time and feedback to the construction team to guide operators to adjust the vibration equipment and pouring process, ensuring the accuracy and completeness of subsequent analysis. At the same time, data conversion and accurate data transmission enhance the reliability of concrete homogeneity assessment results.

[0027] Preferably, the concrete homogeneity assessment module is used to construct a concrete pouring homogeneity assessment model according to the data transmitted by the multi-sensor data conversion and transmission module, and detect and evaluate the homogeneity and defects in the concrete pouring process; it includes: a data preprocessing layer, a feature extraction layer, a homogeneity assessment and defect detection layer; the data preprocessing layer includes: filtering and denoising, data smoothing and interpolation, data normalization and time synchronization, and obtains standard temperature sensor data, standard ultrasonic sensor data and standard vibration sensor data, corresponding to the above-mentioned step S30; wherein the filtering and denoising includes: denoising the multi-sensor data by filtering technology to eliminate the noise of the multi-sensor due to the concrete pouring site during collection; the data smoothing and interpolation includes: smoothing the multi-sensor data by a linear interpolation algorithm and filling the missing data; the data normalization includes: uniformly adjusting the scale of the signals of different sensors to ensure that the data collected by the multi-sensor is in the same numerical range; the time synchronization includes: matching the timestamps of the multi-sensor data collected by the multi-sensor to ensure that the multi-sensor data is in the same time window.

[0028] In an embodiment of the present application, the multi-sensor data collected by the temperature sensor, ultrasonic sensor and vibration sensor are preprocessed by the data preprocessing layer, the vibration noise caused by the environmental interference at the concrete pouring construction site or the interference wave in the ultrasonic signal during multi-sensor collection is removed, and the multi-sensor data is time-unified; this step improves the quality and reliability of the multi-sensor data, ensures that the model of the present invention can run based on real and reliable data, and facilitates subsequent feature extraction and model training.

[0029] Preferably, the feature extraction layer includes: extracting features related to concrete pouring homogeneity and potential defects, so as to establish the concrete pouring homogeneity evaluation model to evaluate the concrete building homogeneity and detect defects, and obtain concrete features, corresponding to the above-mentioned step S40; the concrete features include: temperature features, ultrasonic features and vibration features; wherein the temperature features include: temperature uniformity and crack risk score; the temperature uniformity is evaluated by calculating the standard deviation of the temperature distribution in the temperature sensor acquisition area, and the specific evaluation criteria are: the smaller the standard deviation of the temperature distribution, the more uniform the temperature distribution, and the more consistent the solidification speed inside and outside the concrete; the standard deviation formula of the temperature distribution is: ; in, is the standard deviation of the temperature distribution; is the total number of temperature sensors; For temperature sensor The temperature at each location; is the average temperature; The crack risk score is evaluated by calculating the temperature gradient to detect the crack risk caused by the temperature difference. The specific evaluation criteria are: the larger the temperature gradient and the larger the temperature difference, the higher the risk of crack defects. The calculation formula of the temperature gradient is: ; in, is the temperature gradient; is the maximum temperature; is the minimum temperature; The ultrasonic characteristics include: sound velocity uniformity and cavity risk score; the sound velocity uniformity is evaluated by calculating the sound velocity standard deviation, and the specific evaluation criteria are: the smaller the sound velocity standard deviation, the more uniform the density inside the concrete; the sound velocity standard deviation formula is: ; in, is the standard deviation of the speed of sound; is the number of ultrasonic sensors; Ultrasonic sensor The speed of sound at a position; is the average value of the speed of sound; The cavity risk score is evaluated by calculating the sound velocity variation range. The specific evaluation criteria are: the larger the sound velocity variation range, the greater the possibility of the existence of cavities or loose areas; the formula for the sound velocity variation range is: ; in, is the range of sound speed variation; is the maximum speed of sound; is the minimum speed of sound; The vibration characteristics include: vibration uniformity and material separation risk score; the vibration uniformity is evaluated by calculating the coefficient of variation of the vibration frequency, and the specific evaluation criteria are: the smaller the coefficient of variation of the vibration frequency, the more uniform the vibration; the formula for the coefficient of variation of the vibration frequency is: ; in, is the coefficient of variation of the vibration frequency; is the standard deviation of vibration frequency; is the average vibration frequency; is the number of vibration sensors; For vibration sensor The vibration frequency of each position; The material separation risk score is evaluated by calculating the coefficient of variation of the vibration amplitude. The specific evaluation criteria are: the larger the coefficient of variation of the vibration amplitude, the greater the possibility of material separation defects. The coefficient of variation formula of the vibration amplitude is: ; in, is the coefficient of variation of the vibration amplitude; is the standard deviation of vibration frequency; is the average value of vibration amplitude; For vibration sensor The vibration amplitude at each position; The weight coefficient is set according to the weight of the feedback data of concrete temperature, concrete internal defects and vibration uniformity of different sensors, and the temperature feature, the ultrasonic feature and the vibration feature are integrated by weighted average method to obtain a comprehensive feature index; the comprehensive feature index is: ; in, is the comprehensive characteristic index; is the temperature feature weight; is the ultrasonic feature weight; is the vibration feature weight; ; According to the comprehensive characteristic index, the homogeneity level and defect type are divided by setting thresholds, including: When , it is judged as high homogeneity; when and When , it is judged that the homogeneity is poor and crack defects are detected; when and When , it is judged that the homogeneity is poor and a void defect is detected; when and When , it is judged that the homogeneity is poor and material separation defects are detected; among them, is the high homogeneity threshold; is the threshold for detecting crack defects; is the threshold for detecting void defects; is the threshold value for detecting material separation defects. In this embodiment, is 0.9, 4.5℃, is 20m / s, is 15Hz.

[0030] In the embodiment of the present application, the temperature difference change is monitored by the temperature sensor, and the standard deviation of the temperature distribution and the temperature variation range are extracted, which can provide a basis for subsequent detection and avoidance of the risk of internal cracks in the concrete; the density is evaluated by the ultrasonic sensor, and the standard deviation of the sound velocity is extracted, which can help to identify the void area in the future; the vibration frequency variation coefficient is extracted by the vibration sensor, which can help the model identify the problem of uneven vibration. Through the homogeneity grade classification, the homogeneity quality status of the concrete pouring process can be intuitively reflected, providing clear guidance for the subsequent homogeneity evaluation and defect detection.

[0031] Preferably, the homogeneity assessment and defect detection layer is used to construct a concrete pouring homogeneity assessment model to perform homogeneity assessment and defect detection on concrete pouring, corresponding to the above-mentioned step S50; specifically comprising: dividing the temperature characteristics, the ultrasonic characteristics, the vibration characteristics and the comprehensive characteristic index into a 75% training set, a 15% validation set and a 15% test set; training and comparing each machine learning model according to the training set using the "Regression Learner" tool of MATLAB, and selecting the best model for the concrete pouring homogeneity assessment model to perform concrete homogeneity assessment and defect detection; please refer to Table 1, which shows the concrete homogeneity assessment performance comparison results of each machine learning model.

[0032] Table 1 Comparison of concrete homogeneity evaluation performance of various machine learning models

[0033] According to the comparison results of concrete homogeneity evaluation performance of various machine learning models shown in Table 1, it is concluded that the best model is the decision tree model; The optimal model is applied to train the concrete casting homogeneity assessment model to detect and evaluate the homogeneity and defects of concrete casting; the specific training process is: taking the training set as an input variable, inputting it into the decision tree model to train the concrete casting homogeneity assessment model; splitting the nodes of the decision tree model according to information gain; outputting the homogeneity level of the training set; optimizing the parameters of the decision tree model by cross-validation using the validation set; outputting the homogeneity assessment and defect detection results of concrete using the optimized concrete casting homogeneity assessment model through the test set, and evaluating the performance of the concrete casting homogeneity assessment model according to the homogeneity assessment and defect detection results.

[0034] Table 2 shows the results of the concrete pouring homogeneity assessment model for homogeneity assessment and defect detection during concrete pouring and the performance evaluation of the concrete pouring homogeneity assessment model.

[0035] Table 2 Concrete pouring homogeneity evaluation results and performance evaluation of the concrete pouring homogeneity evaluation model

[0036] In an embodiment of the present invention, a concrete pouring homogeneity assessment model is constructed through a decision tree model to perform concrete pouring homogeneity assessment and defect detection. The decision tree model can effectively combine the multi-sensor data of temperature, ultrasonic and vibration sensors to comprehensively evaluate the homogeneity of concrete from multiple dimensions, such as detecting crack risks through temperature gradients, identifying voids or defects through the standard deviation of sound velocity, and judging vibration uniformity through vibration frequency and amplitude variation coefficient. This fusion analysis method significantly improves the concrete pouring homogeneity assessment model's assessment ability and reliability for concrete pouring homogeneity. At the same time, the decision tree model can quickly process data collected in real time by multiple sensors and quickly generate homogeneity assessment results; in large-volume concrete pouring, the temperature, sound velocity and vibration data can be updated in real time, and the current construction status can be dynamically evaluated to ensure the homogeneity and overall quality of the concrete pouring process.

[0037] The concrete pouring homogeneity detection system design of the present invention realizes the full-process concrete pouring homogeneity detection from multi-dimensional, data analysis to real-time detection and visual feedback: the multi-sensor perception module comprehensively collects key parameters such as temperature change, internal density, vibration uniformity, etc. of concrete during pouring through temperature sensors, ultrasonic sensors and vibration sensors, ensuring comprehensive coverage of factors that may affect homogeneity; the multi-sensor data conversion and transmission module efficiently converts and transmits the collected data to ensure the integrity and adaptability of the data; the concrete homogeneity evaluation module combines multi-dimensional data to build a model, accurately quantify the homogeneity of concrete pouring, and quickly detect potential defects such as cracks, voids and material separation. The whole process not only significantly improves the accuracy of concrete pouring homogeneity evaluation and defect detection, but also effectively reduces quality risks and post-construction repair costs, providing scientific guarantees for concrete pouring homogeneity detection in large areas, and is suitable for efficient construction of large-scale and complex structural projects.

[0038] Embodiment 2 In Example 1, the method and system proposed in the present invention successfully implement the homogeneity detection of the concrete pouring process. To further verify the effectiveness of the present invention, another concrete pouring area B is also tested for concrete homogeneity in the present application example.

[0039] In order to better carry out subsequent concrete homogeneity assessment and defect detection, a concrete pouring homogeneity detection method based on sensor data is applied to the concrete pouring area B, including the following steps: The uniformity and state changes of concrete during the pouring process are fully monitored by multiple sensors; the multiple sensors include: temperature sensors, ultrasonic sensors and vibration sensors; the temperature sensor is used to monitor the temperature changes of concrete during the pouring process; the ultrasonic sensor is used to use ultrasonic waves to penetrate concrete during the pouring process to analyze the changes in sound velocity in real time, detect internal defects and density; the vibration sensor is used to monitor the vibration frequency and amplitude during the pouring process and evaluate the vibration uniformity of the concrete; The multi-sensor data is collected by the multi-sensor, and the multi-sensor data is converted into a data type and transmitted; the multi-sensor data collection and transmission module includes: a data collection layer, a data conversion layer and a data transmission layer; According to the multi-sensor data after the data type conversion, a concrete pouring homogeneity assessment model is constructed to detect and evaluate the homogeneity and defects in the concrete pouring process; it includes: a data preprocessing layer, a feature extraction layer, a homogeneity assessment and defect detection layer; wherein the feature extraction layer includes: extracting features related to the homogeneity and potential defects of concrete pouring, so as to establish the concrete pouring homogeneity assessment model to evaluate the homogeneity of concrete construction and detect defects, and obtain concrete features; the concrete features include: temperature features, ultrasonic features and vibration features; according to the weights of the feedback data of different sensors on concrete temperature, concrete internal defects and vibration uniformity, a weighting coefficient is set, and the temperature features, the ultrasonic features and the vibration features are integrated by the weighted average method to obtain a comprehensive feature index; according to the comprehensive feature index, the homogeneity level and defect type are divided by setting a threshold.

[0040] The homogeneity assessment and defect detection layer includes: selecting the best model as the concrete casting homogeneity assessment model by comparing various machine learning models to perform concrete homogeneity assessment and defect detection, wherein the best model is a decision tree model; training the concrete casting homogeneity assessment model to perform detection and quality assessment on concrete casting homogeneity and defects according to the temperature features, ultrasonic features, vibration features and comprehensive feature indexes obtained by the feature extraction layer and the homogeneity grades and defect types divided by the comprehensive feature indexes; The specific training process is as follows: the temperature feature, the ultrasonic feature, the vibration feature and the comprehensive feature index are divided into a training set, a validation set and a test set; the training set is used as an input variable and input into the concrete pouring homogeneity evaluation model to train the decision tree model; the nodes of the decision tree model are split according to information gain; the homogeneity level of the training set is output; the parameters of the decision tree model are optimized by cross-validation through the validation set; the concrete homogeneity evaluation and defect detection results are output by the optimized decision tree model through the test set, and the accuracy of the defect detection results is evaluated. Table 3 shows the results of the concrete pouring homogeneity evaluation model for the homogeneity evaluation and defect detection of the concrete pouring area B and the performance evaluation of the concrete pouring homogeneity evaluation model.

[0041] Table 3 Homogeneity evaluation results and performance evaluation of concrete pouring area B of the concrete pouring homogeneity evaluation model

[0042] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A concrete pouring homogeneity detection system based on sensor data, characterized in that: include: Multi-sensor perception module, multi-sensor data conversion and transmission module and concrete homogeneity assessment module; The multi-sensor sensing module is used to comprehensively monitor the homogeneity and state changes of concrete during the pouring process through multiple sensors; the multiple sensors include: a temperature sensor, an ultrasonic sensor and a vibration sensor; the temperature sensor is used to monitor the temperature changes of concrete during the pouring process; the ultrasonic sensor is used to use ultrasonic waves to penetrate concrete during the pouring process to analyze the changes in sound velocity in real time, detect internal defects and density; the vibration sensor is used to monitor the vibration frequency and amplitude during the pouring process, and evaluate the vibration uniformity of the concrete; The multi-sensor data conversion and transmission module is used to collect multi-sensor data through the multi-sensor perception module, perform data type conversion on the multi-sensor data and transmit it to the concrete homogeneity evaluation module; The concrete homogeneity assessment module is used to construct a concrete pouring homogeneity assessment model based on the data transmitted by the multi-sensor data conversion and transmission module, and to detect and conduct quality assessment on the homogeneity and defects during the concrete pouring process; it includes: a data preprocessing layer, a feature extraction layer, a homogeneity assessment and defect detection layer.

2. A concrete pouring homogeneity detection system based on sensor data according to claim 1, characterized in that: The temperature sensors are located at different heights in the concrete pouring area and are arranged along the vertical direction of the concrete pouring area, including the surface, middle layer and bottom of the pouring area, respectively. The vertical distance between each temperature sensor is set. The ultrasonic sensor is located in the concrete pouring area, including the middle position of the four vertical sides of the pouring area and the center of the pouring area. The vibration sensor is located at the upper left corner, the center position and the lower right corner of the surface of the concrete pouring area and the vibrating equipment, including on the head of the vibrator and the bottom of the vibrator in contact with the concrete.

3. The concrete pouring homogeneity detection system based on sensor data according to claim 1 is characterized in that: The multi-sensor data acquisition and transmission module includes: a data acquisition layer, a data conversion layer and a data transmission layer; the data acquisition layer is used to collect the multi-sensor data through the multi-sensor; the multi-sensor data includes: initial temperature sensor data, ultrasonic sensor data and initial vibration sensor data; wherein, the initial temperature sensor data is an analog signal, which is obtained by the temperature sensor collecting the temperature value of concrete at different positions; the ultrasonic sensor data is a digital signal, which is obtained by the ultrasonic sensor through transmitting and receiving ultrasonic signals, collecting the propagation time, sound speed and echo intensity of the ultrasonic wave; the initial vibration sensor data is an analog signal, which is obtained by the vibration sensor collecting the vibration frequency and amplitude; The data conversion layer is used to process and convert the multi-sensor data to ensure that the multi-sensor data can be subsequently analyzed in a unified format; for the analog signal data collected by the temperature sensor and the vibration sensor, the analog-to-digital converter is used to convert the analog signal data into digital signals to obtain temperature sensor data and vibration sensor data; The data transmission layer is used to transmit the ultrasonic sensing data and the temperature sensing data and the vibration sensing data obtained by the data conversion layer to the concrete homogeneity evaluation module through wireless communication technology, so as to facilitate subsequent concrete homogeneity analysis and defect detection.

4. The concrete pouring homogeneity detection system based on sensor data according to claim 1 is characterized in that: The data preprocessing layer includes: filtering and denoising, data smoothing and interpolation, data normalization and time synchronization to obtain standard temperature sensor data, standard ultrasonic sensor data and standard vibration sensor data; wherein, the filtering and denoising includes: denoising the multi-sensor data by filtering technology to eliminate the noise of the multi-sensor due to the concrete pouring site during collection; the data smoothing and interpolation includes: smoothing the multi-sensor data by linear interpolation algorithm and filling the missing data; the data normalization includes: uniformly adjusting the scale of signals of different sensors to ensure that the data collected by the multi-sensor are in the same numerical range; the time synchronization includes: timestamp matching the multi-sensor data collected by the multi-sensor to ensure that the multi-sensor data are in the same time window.

5. The concrete pouring homogeneity detection system based on sensor data according to claim 1 is characterized in that: The feature extraction layer includes: extracting features related to concrete casting homogeneity and potential defects, so as to establish the concrete casting homogeneity evaluation model to evaluate the concrete building homogeneity and detect defects, and obtain concrete features; the concrete features include: temperature features, ultrasonic features and vibration features; wherein the temperature features include the standard deviation and temperature gradient of concrete temperature distribution; the ultrasonic features include the sound velocity standard deviation and sound velocity variation range; the vibration features include the coefficient of variation of vibration frequency and the coefficient of variation of vibration amplitude; The weight coefficient is set according to the weight of the feedback data of concrete temperature, concrete internal defects and vibration uniformity of different sensors, and the temperature feature, the ultrasonic feature and the vibration feature are integrated by weighted average method to obtain a comprehensive feature index; the comprehensive feature index is: ； in, is the comprehensive characteristic index; is the standard deviation of the temperature distribution; is the temperature gradient; is the temperature feature weight; is the standard deviation of the speed of sound; is the range of sound speed variation; is the ultrasonic feature weight; is the coefficient of variation of the vibration frequency; is the coefficient of variation of the vibration frequency; is the vibration feature weight; ; According to the comprehensive characteristic index, the homogeneity level and defect type are divided by setting thresholds, including: When , it is judged as high homogeneity; when and When , it is judged that the homogeneity is poor and crack defects are detected; when and When , it is judged that the homogeneity is poor and a void defect is detected; when and When , it is judged that the homogeneity is poor and material separation defects are detected; among them, is the high homogeneity threshold; is the threshold for detecting crack defects; is the threshold for detecting void defects; The threshold for detecting material separation defects.

6. The concrete pouring homogeneity detection system based on sensor data according to claim 1 is characterized in that: The homogeneity assessment and defect detection layer includes: selecting the best model as the concrete casting homogeneity assessment model by comparing various machine learning models to perform concrete homogeneity assessment and defect detection, wherein the best model is a decision tree model; training the concrete casting homogeneity assessment model to perform detection and quality assessment on concrete casting homogeneity and defects according to the temperature features, ultrasonic features, vibration features and comprehensive feature indexes obtained by the feature extraction layer and the homogeneity grades and defect types divided by the comprehensive feature indexes; The specific training process is: dividing the temperature characteristics, the ultrasonic characteristics, the vibration characteristics and the comprehensive characteristic index into a training set, a validation set and a test set; using the training set as an input variable, inputting it into the concrete pouring homogeneity evaluation model to train the decision tree model; splitting the nodes of the decision tree model according to information gain; outputting the homogeneity level of the training set; optimizing the parameters of the decision tree model by cross-validation through the validation set; outputting the concrete homogeneity evaluation and defect detection results by using the optimized decision tree model through the test set, and evaluating the accuracy of the defect detection results.

7. A method for detecting the homogeneity of concrete pouring based on sensor data, characterized in that: include: The uniformity and state changes of concrete during the pouring process are fully monitored by multiple sensors; the multiple sensors include: temperature sensors, ultrasonic sensors and vibration sensors; the temperature sensor is used to monitor the temperature changes of concrete during the pouring process; the ultrasonic sensor is used to use ultrasonic waves to penetrate concrete during the pouring process to analyze the changes in sound velocity in real time, detect internal defects and density; the vibration sensor is used to monitor the vibration frequency and amplitude during the pouring process and evaluate the vibration uniformity of the concrete; Collecting multi-sensor data through the multi-sensor, converting the data type of the multi-sensor data and transmitting the data; According to the multi-sensor data after the data type conversion, a concrete pouring homogeneity evaluation model is constructed to detect and evaluate the homogeneity and defects in the concrete pouring process; including: a data preprocessing layer, a feature extraction layer, a homogeneity evaluation and a defect detection layer.

8. The method for detecting concrete pouring homogeneity based on sensor data according to claim 7, characterized in that: The feature extraction layer includes: extracting features related to concrete casting homogeneity and potential defects so as to establish the concrete casting homogeneity evaluation model and obtain concrete features; the concrete features include: temperature features, ultrasonic features and vibration features; wherein the temperature features include the standard deviation and temperature gradient of concrete temperature distribution; the ultrasonic features include the sound velocity standard deviation and the sound velocity variation range; the vibration features include the coefficient of variation of vibration frequency and the coefficient of variation of vibration amplitude; the temperature features, the ultrasonic features and the vibration features are fused by weighted averaging method to obtain a comprehensive feature index; according to the comprehensive feature index, the homogeneity level and defect type are divided by setting a threshold.

9. The method for detecting concrete pouring homogeneity based on sensor data according to claim 7, characterized in that: The classification of homogeneity levels and defect types includes: When , it is judged as high homogeneity; when and When , it is judged that the homogeneity is poor and crack defects are detected; when and When , it is judged that the homogeneity is poor and a void defect is detected; when and When , it is judged that the homogeneity is poor and material separation defects are detected; among them, is the high homogeneity threshold; is the threshold for detecting crack defects; is the threshold for detecting void defects; The threshold for detecting material separation defects.

10. The method for detecting concrete pouring homogeneity based on sensor data according to claim 7, characterized in that: The homogeneity assessment and defect detection layer includes: selecting the best model as the concrete casting homogeneity assessment model by comparing various machine learning models to perform concrete homogeneity assessment and defect detection, wherein the best model is a decision tree model; training the concrete casting homogeneity assessment model to perform detection and quality assessment on concrete casting homogeneity and defects according to the temperature features, ultrasonic features, vibration features and comprehensive feature indexes obtained by the feature extraction layer and the homogeneity grades and defect types divided by the comprehensive feature indexes; The specific training process is: dividing the temperature characteristics, the ultrasonic characteristics, the vibration characteristics and the comprehensive characteristic index into a training set, a validation set and a test set; using the training set as an input variable, inputting it into the decision tree model to train the concrete pouring homogeneity evaluation model; splitting the nodes of the decision tree model according to information gain; outputting the homogeneity level of the training set; optimizing the parameters of the decision tree model by cross-validation using the validation set; outputting the concrete homogeneity evaluation and defect detection results by using the optimized decision tree model through the test set, and evaluating the accuracy of the defect detection results.

Images