Intelligent control system and method for concrete vibration

Abstract

According to the concrete vibration intelligent control system and method, a closed-loop control system integrating a multi-mode vibration execution terminal, a distributed sensing network and an intelligent decision-making module is constructed, and comprehensive sensing, accurate evaluation and self-adaptive regulation and control of the concrete vibration process are achieved. The system can identify the compact state of concrete in real time based on deep fusion of multi-source sensor data, automatically optimizes vibration parameters through an intelligent algorithm, effectively overcomes the limitation of traditional judgment depending on artificial experience, remarkably improves the uniformity and consistency of concrete vibration quality, and improves the working efficiency. Quality defects such as insufficient vibration or excessive vibration are reduced, meanwhile, transparent management and accurate control of the construction process are achieved through remote monitoring and digital twinning technologies, and reliable technical guarantee is provided for improving the concrete construction quality.

Description

Technical Field

[0001] This invention relates to the field of concrete construction quality control technology, and in particular to an intelligent control system for concrete vibration. Background Technology

[0002] Concrete vibration is a crucial step in concrete pouring, directly impacting the final strength and durability of the concrete structure. Traditional concrete vibration quality control relies heavily on the experience and judgment of construction workers, assessing concrete density by observing air bubbles escaping from the surface and feeling the resistance transmitted by the vibrator. This manual method is inherently subjective and time-sensitive, failing to accurately reflect the actual internal density of the concrete in real time, often leading to under- or over-vibration defects. Under-vibration results in defects such as honeycombing and voids, reducing the structure's load-bearing capacity; while over-vibration can cause aggregate segregation and cement paste floating, similarly affecting concrete uniformity and final strength. This is especially true in large structures, complex reinforcement, or high-performance concrete construction, where traditional methods struggle to guarantee uniformity and consistency. With advancements in sensor technology, data analysis, and intelligent control theory, the concrete construction field has begun exploring more scientific and quantitative quality control methods. However, existing technologies are often limited to single-parameter monitoring or open-loop control, lacking deep integration of multi-source information and closed-loop control capabilities based on real-time quality assessment, thus failing to achieve precise optimization and control of the vibration process. Summary of the Invention

[0003] In view of this, the present invention provides an intelligent control system for concrete vibration. One or more embodiments of this specification also relate to an intelligent control method for concrete vibration, in order to address the technical deficiencies existing in the prior art.

[0004] According to a first aspect of the present invention, a smart control system for concrete vibration is provided, comprising: A multimodal vibration execution terminal is used to apply mechanical vibration to concrete and collect vibration state data; Distributed sensor networks are used to acquire information about the internal state of concrete. The central control unit is used to coordinate the operation of various components of the system; The data processing module is used to process the raw data collected by the multimodal vibration actuator and the distributed sensor network; The compactness identification module is used to determine the compactness of concrete based on preprocessed data; An adaptive control module is used to adjust the vibration parameters based on the compaction identification results; A remote monitoring platform is used to visually display the vibration process and quality status; The multimodal vibration execution terminal transmits the collected vibration state data to the data processing module. The data processing module filters and extracts features from the vibration state data before inputting it into the compaction identification module. The compaction level output by the compaction identification module is transmitted to the adaptive control module. The adaptive control module generates vibration parameter adjustment instructions and sends them to the multimodal vibration execution terminal, forming a closed-loop control.

[0005] In some implementations, the multimodal vibration execution terminal includes a vibrator body, a triaxial accelerometer, a dynamic pressure sensor array, a temperature sensor, and an edge computing gateway; A triaxial accelerometer is used to measure the vibration acceleration of a vibrator in three orthogonal directions; A dynamic pressure sensor array is used to detect the reaction pressure of concrete on the vibrator; Temperature sensors are used to monitor concrete temperature; Edge computing gateways are used to perform preliminary processing of sensor data.

[0006] In some implementations, the distributed sensor network includes fiber Bragg grating sensors embedded in concrete and UWB positioning systems deployed at the construction site. Fiber optic grating sensors are used to measure internal strain in concrete; The UWB positioning system is used to track the spatial position of the vibrator in real time.

[0007] In some implementations, the compactness identification module employs a deep convolutional neural network model, which takes preprocessed time-series data from multiple sensors as input and outputs a concrete compactness level classification.

[0008] In some implementations, the adaptive control module employs a reinforcement learning algorithm to automatically adjust the vibration frequency and amplitude based on the compaction level, concrete temperature, and steel reinforcement density.

[0009] In some implementations, the remote monitoring platform constructs a digital twin model synchronized with the physical site, displaying the working status of each vibrating terminal and the distribution of concrete density in real time.

[0010] In some implementations, the system determines the vibration efficiency coefficient using the following formula: in, α is the vibration efficiency coefficient, which is a dimensionless comprehensive evaluation index output by the system. It is used to quantify the vibration effect. The value range is [0,1]. The larger the value, the better the vibration effect. α, β, and δ are weighting coefficients used to balance the contribution of different physical quantities to the comprehensive index. They satisfy α+β+δ=1 and are determined based on experimental data. The pressure-stiffness sensitivity coefficient is used to adjust the sensitivity of the effect of the pressure-stiffness matching degree on the efficiency coefficient, and is determined based on experimental data; N is the total number of dynamic pressure sensors, representing the number of pressure monitoring points, which is derived from the system configuration parameters. The dynamic pressure sensor reading of the i-th dynamic pressure sensor reflects the reaction pressure of concrete on the vibrator, and is derived from the dynamic pressure sensor array. The effective working area of ​​the i-th dynamic pressure sensor is used for pressure-to-force conversion calculations and is derived from the sensor design parameters. The nominal compressive strength of concrete serves as a benchmark value for pressure comparison and is derived from the design parameters of concrete materials. The nominal area of ​​the vibratory rod's working head serves as a benchmark for area comparison and is derived from the mechanical design parameters of the vibrator; M represents the total number of fiber optic grating sensors, indicating the number of internal status monitoring points, and is derived from the system configuration parameters. The dynamic elastic modulus of concrete is the value of the j-th fiber grating sensor, which reflects the stiffness characteristics of concrete and is calculated based on the strain measurement value of the fiber grating sensor. The nominal modulus of elasticity of concrete serves as a benchmark for comparison of moduli of elasticity and is derived from the design parameters of concrete materials. The temperature sensitivity coefficient is used to adjust the sensitivity of the effect of temperature deviation on the efficiency coefficient, and is determined based on experimental data; K is the total number of temperature sensors, representing the number of temperature monitoring points, which is derived from the system configuration parameters. The temperature measurement value of the kth temperature sensor reflects the temperature state of the concrete. The optimal vibration reference temperature for concrete is used as the target value for temperature control and is determined based on the concrete mix proportion and curing requirements; L is the total number of acceleration sensors, representing the number of vibration state monitoring points, which is derived from the system configuration parameters. The dimensionless damping ratio of the l-th accelerometer reflects the energy dissipation characteristics of concrete. The maximum allowable damping ratio is determined based on the properties of concrete materials and is used to normalize the effect of the damping ratio.

[0011] In some implementations, the damping ratio Determined by the following formula: in, The maximum amplitude of the acceleration of the l-th accelerometer reflects the peak value of the vibration intensity and is derived from the measurement data of the triaxial accelerometer. The minimum acceleration amplitude of the l-th accelerometer reflects the valley value of vibration intensity and is derived from the measurement data of the triaxial accelerometer. is the equivalent concrete resistance of the l-th accelerometer, reflecting the concrete's resistance to vibration, calculated based on data from the dynamic pressure sensor array; F0 is the rated excitation force of the vibrator, used as a benchmark for resistance comparison, derived from the vibrator's performance parameters; tanh(·) is the hyperbolic tangent function, used to limit the influence of the resistance ratio within a reasonable range.

[0012] In some implementations, the system also includes an alarm module that generates an early warning signal and sends it to the remote monitoring platform when the vibration efficiency coefficient is detected to be lower than the threshold or the compaction identification module outputs an abnormal state for multiple consecutive cycles.

[0013] According to a second aspect of the present invention, a method for intelligent control of concrete vibration is provided, the method being applied to the intelligent control system for concrete vibration of the preceding claims, the method comprising: The concrete is subjected to mechanical vibration by a multimodal vibration execution terminal, and vibration status data is collected. Information about the internal state of concrete is obtained through a distributed sensor network. The data processing module processes the raw data collected by the multimodal vibration execution terminal and the distributed sensor network. The density identification module determines the density of concrete based on preprocessed data. The vibration parameters are adjusted by the adaptive control module based on the compaction identification results. The vibration process and quality status are visualized through a remote monitoring platform; The multimodal vibration execution terminal transmits the collected vibration state data to the data processing module. The data processing module filters and extracts features from the vibration state data before inputting it into the compaction identification module. The compaction level output by the compaction identification module is transmitted to the adaptive control module. The adaptive control module generates vibration parameter adjustment instructions and sends them to the multimodal vibration execution terminal, forming a closed-loop control.

[0014] At least one embodiment of this invention achieves comprehensive perception, precise evaluation, and adaptive control of the concrete vibration process by constructing a closed-loop control system integrating a multimodal vibration execution terminal, a distributed sensor network, and an intelligent decision-making module. This system can identify the concrete compaction state in real time based on deep fusion of multi-source sensor data and automatically optimize vibration parameters through intelligent algorithms. This effectively overcomes the limitations of traditional methods relying on manual experience, significantly improving the uniformity and consistency of concrete vibration quality and reducing quality defects such as insufficient or excessive vibration. Furthermore, through remote monitoring and digital twin technology, it achieves transparent management and precise control of the construction process, providing a reliable technical guarantee for improving concrete construction quality. Attached Figure Description

[0015] Figure 1 This is a simplified structural diagram of an intelligent control system for concrete vibration provided by the present invention; Figure 2 This is a flowchart of an intelligent control method for concrete vibration provided by the present invention. Detailed Implementation

[0016] Many specific details are set forth in the following description to provide a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this specification. Therefore, this specification is not limited to the specific implementations disclosed below.

[0017] The terminology used in one or more embodiments of this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the one or more embodiments of this specification. The singular forms “a” and “the” as used in one or more embodiments of this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items. The modifications “a” and “a plurality” as used in this disclosure are illustrative and not restrictive, and those skilled in the art will understand that they should be understood as “one or more” unless the context clearly indicates otherwise.

[0018] It should be understood that although the terms first, second, etc., may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first may also be referred to as second without departing from the scope of one or more embodiments of this specification, and similarly, second may also be referred to as first. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0019] See Figure 1 , Figure 1A simplified structural diagram of a concrete vibration intelligent control system according to some embodiments of this specification is shown. Specifically, it includes: a multimodal vibration execution terminal for applying mechanical vibration to concrete and collecting vibration state data; a distributed sensor network for acquiring internal state information of the concrete; a central control unit for coordinating the operation of various system components; a data processing module for processing the raw data collected by the multimodal vibration execution terminal and the distributed sensor network; a density identification module for determining the density of the concrete based on preprocessed data; an adaptive control module for adjusting vibration parameters according to the density identification results; and a remote monitoring platform for visually displaying the vibration process and quality status. The multimodal vibration execution terminal transmits the collected vibration state data to the data processing module. The data processing module filters and extracts features from the vibration state data before inputting it to the density identification module. The density level output by the density identification module is transmitted to the adaptive control module. The adaptive control module generates vibration parameter adjustment instructions and sends them to the multimodal vibration execution terminal, forming a closed-loop control.

[0020] A multimodal vibration execution terminal can refer to a device that integrates vibration execution and data acquisition functions. For example, it may include a vibrator body, sensors, and an edge computing gateway. It applies mechanical vibration and collects data such as vibration acceleration and pressure to vibrate concrete and monitor the vibration status in real time. Mechanical vibration can refer to the periodic force generated by a mechanical device, such as high-frequency vibration generated by an eccentric block driven by a vibrator motor. The frequency can be adjusted from tens to hundreds of hertz, used to rearrange concrete particles, reduce voids, and increase density. Vibration status data can refer to a series of measurements reflecting vibration characteristics, such as acceleration time-series data collected by a triaxial accelerometer, including amplitude, frequency, and phase information, used to analyze the vibration effect and concrete response.

[0021] Distributed sensor networks refer to network systems composed of multiple sensor nodes. For example, deploying fiber optic grating sensors and UWB positioning systems allows for the measurement of strain by the fiber optic grating sensors and the tracking of location by the UWB positioning system, thus acquiring information about the internal state of concrete and the spatial characteristics of the vibrator. Concrete internal state information refers to data describing the physical properties of the concrete, such as strain distribution and temperature data measured by fiber optic grating sensors, reflecting the stiffness and temperature field of the concrete, and used to assess the concrete's curing state and density.

[0022] The central control unit can refer to the core processing and control component of the system, such as an embedded processor or industrial PLC, which runs control algorithms, coordinates data flow and instruction transmission, manages the cooperation of various modules, and ensures stable system operation.

[0023] A data processing module can refer to a software or hardware unit responsible for collecting and initially processing raw data. For example, it might use a digital signal processor for filtering and feature extraction. Filtering uses a low-pass filter to remove noise, and feature extraction includes calculating statistical features to clean the data and extract useful information for subsequent modules. Raw data can refer to unprocessed data directly collected from sensors, such as voltage signals or digital readings output by an accelerometer. This data may contain noise and interference and is used as input to a preprocessing module to generate clean data.

[0024] The compaction identification module refers to an intelligent analysis unit used to determine the degree of concrete compaction. For example, it might employ a deep convolutional neural network model, inputting time-series data from multiple sensors and outputting compaction classification results to automatically assess the concrete's compaction status and guide vibration operations. Concrete compaction can refer to the degree of tightness of solid particle filling within the concrete, which can be indirectly measured, for example, through density, porosity, or acoustic properties. In this system, the compaction identification module outputs a grade to quantify the vibration effect and ensure project quality.

[0025] An adaptive control module can refer to an intelligent control unit that automatically adjusts operating parameters based on feedback. For example, it might employ reinforcement learning algorithms to calculate the optimal vibration frequency and amplitude based on the compaction level and environmental parameters, dynamically optimizing vibration parameters for efficient compaction. The compaction identification result can refer to the evaluation conclusion output by the compaction identification module, expressed in grades such as low, medium, and high compaction, or as a continuous numerical score. This serves as input to the adaptive control module, triggering parameter adjustments. Vibration parameters can refer to adjustable variables controlling the vibrator's operation, such as vibration frequency, amplitude, and vibration time. The adaptive control module generates adjustment commands to optimize the compaction process and improve compaction uniformity.

[0026] A remote monitoring platform can refer to a software interface that provides remote visualization and management functions. For example, it may be built using web technology, displaying a digital twin model and updating vibration status and density distribution in real time, allowing operators to remotely monitor and control the vibration process. The vibration process can refer to the entire sequence of concrete vibration operations, such as the cycle from start to stop of the vibrator, including parameter adjustment and data acquisition stages. This serves as an object for monitoring and optimization to ensure quality. The quality status can refer to the current state of concrete vibration quality, for example, through comprehensive evaluation using indicators such as density and uniformity, and visualized on the remote monitoring platform for real-time feedback on vibration effects to support decision-making. The density level can refer to a classification of the concrete's density, such as three levels: insufficient, normal, and excessive, determined based on the probability distribution of a neural network output. This simplifies the status representation and guides control strategies. Vibration parameter adjustment commands can refer to control commands used to modify vibration parameters. For example, these commands are sent to the vibration execution terminal via communication protocols to adjust the motor speed or eccentric block position to change the frequency and amplitude, enabling real-time optimization of vibration parameters.

[0027] As a concrete example: In the vibration compaction of bridge pier concrete, a multimodal vibration execution terminal applies mechanical vibration through a vibrator at a frequency of 50 Hz and an amplitude of 2 mm. A triaxial accelerometer collects vibration acceleration data at a sampling rate of 100 Hz. Fiber optic sensors in a distributed sensor network are embedded in the concrete to measure internal strain changes. A UWB (Ultra-Wideband) positioning system tracks the vibrator's three-dimensional coordinates in real time with an update rate of 10 Hz. The central control unit coordinates data flow via Ethernet protocol to ensure synchronization among modules. The data processing module performs low-pass filtering on the raw acceleration data, with a cutoff frequency of 20 Hz, and extracts time-domain features such as the root mean square value and frequency-domain features such as the dominant frequency. The compaction degree identification module uses a pre-trained deep convolutional neural network model, taking a 10-second time-series data segment as input, and outputs a compaction degree classification as low, medium, or high. The adaptive control module adjusts the vibration parameters based on the compaction degree and temperature data using a reinforcement learning algorithm, such as increasing the frequency to 60 Hz. The remote monitoring platform uses WebGL technology to build a digital twin model, enabling real-time visualization of the vibration process and quality status. The entire system forms a closed-loop control, improving the uniformity and efficiency of vibration.

[0028] The beneficial effects of one of the embodiments in this specification include at least the following: by constructing a closed-loop control system integrating a multimodal vibration execution terminal, a distributed sensor network, and an intelligent decision-making module, comprehensive perception, accurate evaluation, and adaptive control of the concrete vibration process are achieved. This system can identify the concrete compaction state in real time based on the deep fusion of multi-source sensor data and automatically optimize vibration parameters through intelligent algorithms. This effectively overcomes the limitations of traditional reliance on manual experience, significantly improves the uniformity and consistency of concrete vibration quality, and reduces the occurrence of quality defects such as insufficient or excessive vibration. Simultaneously, through remote monitoring and digital twin technology, transparent management and precise control of the construction process are achieved, providing reliable technical support for improving concrete construction quality.

[0029] In some implementations, the multimodal vibration execution terminal includes a vibrator body, a triaxial accelerometer, a dynamic pressure sensor array, a temperature sensor, and an edge computing gateway; the triaxial accelerometer is used to measure the vibration acceleration of the vibrator in three orthogonal directions; the dynamic pressure sensor array is used to detect the reaction pressure of concrete on the vibrator; the temperature sensor is used to monitor the concrete temperature; and the edge computing gateway is used to perform preliminary processing on the sensor data.

[0030] The vibrator body refers to the physical structure of the vibrator, used to directly apply mechanical vibration to concrete to promote compaction. A triaxial accelerometer refers to a sensor capable of measuring acceleration in three orthogonal directions, outputting acceleration data along the X, Y, and Z axes. A dynamic pressure sensor array refers to an array of multiple pressure sensors used for dynamic pressure measurement. For example, based on piezoelectric or strain gauge principles, these sensors are arranged on the vibrator surface to measure the concrete's reaction pressure, outputting pressure distribution data to detect the concrete's resistance to vibration and reflect its compaction status. An edge computing gateway refers to a device located at the network edge that processes data. For example, it uses an embedded system to run lightweight algorithms to filter, compress, or perform preliminary analysis on sensor data, reducing data transmission and improving system response speed. Vibration acceleration refers to the rate of change of velocity of an object during vibration, which can include amplitude and frequency components. It can quantify vibration intensity and analyze the vibration effect. Reaction pressure refers to the reverse pressure exerted by the concrete on the vibrator, measured, for example, by a dynamic pressure sensor array. It reflects the concrete's fluidity and compaction, used to assess the concrete's condition and guide the adjustment of vibration parameters. Sensor data refers to raw measurements collected from various sensors, such as data streams of acceleration, pressure, and temperature, transmitted as digital signals to serve as system input for status recognition and control. Preliminary processing refers to the initial processing and analysis of the raw data, such as filtering, denoising, or feature extraction at edge computing gateways, reducing data volume, improving data quality, and alleviating the burden on central processing.

[0031] As a concrete example: In the vibration compaction of foundation concrete for high-rise buildings, the vibrator of the multimodal vibration execution terminal applies mechanical vibration at a frequency of 120 Hz and an amplitude of 2 mm. A triaxial accelerometer measures the vibration acceleration in three orthogonal directions with a sampling rate of 800 Hz, outputting acceleration time-series data. A dynamic pressure sensor array, based on the piezoelectric principle, detects the concrete reaction pressure and is arranged on the surface of the vibrator head, outputting a pressure distribution map. A PT100 temperature sensor monitors the concrete temperature, and the data is transmitted via the I2C protocol. The edge computing gateway uses an ARM processor to run lightweight algorithms, performing low-pass filtering and compression on the sensor data, with the filter cutoff frequency set to 50 Hz. The processed data is uploaded to the central control unit via Ethernet protocol, coordinating the work of the compaction identification module and the adaptive control module. The remote monitoring platform, based on web technology, displays the real-time vibration status and temperature distribution, forming a closed-loop control to improve vibration uniformity.

[0032] By integrating multiple sensors and edge computing technology, real-time data acquisition and preliminary processing of the vibration process are achieved, enhancing the system's response speed and data reliability, thereby improving the uniformity and control accuracy of concrete vibration quality.

[0033] In some implementations, the distributed sensor network includes fiber Bragg grating sensors embedded in the concrete and a UWB positioning system deployed at the construction site; the fiber Bragg grating sensors are used to measure the internal strain of the concrete; and the UWB positioning system is used to track the spatial position of the vibrator in real time.

[0034] Fiber Bragg grating sensors refer to sensors based on the principle of fiber Bragg gratings. For example, a grating is embedded in the core of an optical fiber, and strain changes are measured by the reflection wavelength shift. Sampling rates can reach 100 Hz, used to monitor the strain distribution inside concrete and assess structural deformation and compaction. UWB positioning systems refer to positioning systems based on ultra-wideband technology. UWB stands for Ultra-Wideband. For example, by emitting nanosecond-level pulse signals and using a time-of-arrival algorithm to calculate the position, positioning accuracy can reach the centimeter level, used for real-time tracking of the vibrator's three-dimensional spatial coordinates. Concrete internal strain refers to the deformation per unit length within the concrete. This can be measured using fiber Bragg grating sensors, and strain values ​​are calculated based on wavelength shifts, with resolution down to the micro-strain level. This reflects the stress state of the concrete and assesses compaction and structural integrity. The spatial position of the vibrator refers to its coordinate information in three-dimensional space, obtained through a UWB positioning system, including X, Y, and Z axis coordinates, updated at a frequency of 10 Hz. This is used to determine the vibration position and guide vibration path planning.

[0035] As a concrete example: In the vibration compaction of dam concrete, fiber optic grating sensors of a distributed sensor network are embedded in the concrete, measuring internal strain at a 50 Hz sampling rate. Strain values ​​are calculated based on wavelength offset, with an accuracy of ±2 micro-strain. A UWB positioning system, using four base stations and a TOA algorithm, tracks the spatial position of the vibrator in real time at a 20 Hz update rate, achieving a positioning accuracy of 5 cm. Strain data is transmitted via fiber optic cable to a central control unit for calculating the dynamic elastic modulus of the concrete. Location data is transmitted via Wi-Fi and recorded synchronously with vibration parameters. This integrated data is input into a compaction identification module, which, combined with sensor data, comprehensively assesses the concrete compaction state and optimizes the vibration path and parameters through an adaptive control module.

[0036] By accurately measuring the internal strain of concrete and positioning the vibrator in real time, more comprehensive condition monitoring data is provided, enhancing the accuracy of compaction assessment and optimizing vibration trajectory control, thereby improving the uniformity and reliability of concrete vibration quality.

[0037] In some implementations, the compactness identification module employs a deep convolutional neural network model, which takes preprocessed time-series data from multiple sensors as input and outputs a concrete compactness level classification.

[0038] A deep convolutional neural network (CNN) model can refer to a deep learning-based CNN architecture, such as employing multiple convolutional layers, pooling layers, and fully connected layers. It takes multi-channel time-series data as input, calculates features through forward propagation, and automatically learns the spatial and temporal characteristics of sensor data to achieve density classification. Preprocessed multi-sensor data time-series segments can refer to cleaned and normalized multi-source sensor data time series. For example, acceleration, pressure, and strain data are aligned with timestamps, normalized to the same units, and divided into fixed-length segments to ensure the consistency and comparability of model inputs. Concrete density level classification can refer to the discretized classification results of concrete density, such as taking the category corresponding to the highest probability based on the model output probability as the final classification, dividing it into three levels: insufficient, normal, and excessive. This simplifies state representation and supports control decisions.

[0039] As a concrete example: During the vibration of tunnel lining concrete, the density identification module employs a deep convolutional neural network model. This model contains three convolutional layers and two fully connected layers. The input is a preprocessed time-series data segment from multiple sensors, including acceleration, pressure, and strain data. The segment length is 5 seconds, and the sampling rate is 100 Hz. The model receives the aligned multi-channel time-series data, extracts local features through the convolutional layers, and integrates global information through the fully connected layers. Finally, it outputs a concrete density level classification, categorized into low, medium, and high levels. This classification result is passed to the adaptive control module, which, combined with strain and location data, dynamically adjusts the vibration parameters to form a closed-loop control, ensuring uniform and dense compaction of the lining concrete.

[0040] By employing a deep convolutional neural network model to automatically analyze multi-sensor time-series data, the system achieves accurate identification and classification of concrete density, reduces human judgment errors, and improves the adaptability and reliability of vibration quality control.

[0041] In some implementations, the adaptive control module employs a reinforcement learning algorithm to automatically adjust the vibration frequency and amplitude based on the compaction level, concrete temperature, and steel reinforcement density.

[0042] Reinforcement learning algorithms can refer to machine learning methods that learn optimal strategies through interaction with the environment. For example, Q-learning or deep Q-networks can be used to select actions based on states such as compaction level and temperature, and optimize the strategy through a reward function. This allows for the automatic adjustment of vibration parameters, improving vibration efficiency and quality. Reinforcement density refers to the distribution of reinforcing steel bars per unit volume of concrete. It can be estimated, for example, using design drawings or sensor data to calculate the proportion of steel bars in the concrete volume. Inputting this as a 3D model or scanned data can guide the adjustment of vibration parameters, avoiding interference from reinforcing steel bars and ensuring uniform compaction. Vibration frequency refers to the number of vibrations per unit time of the vibrator. For example, by controlling the motor speed, the frequency can be adjusted from tens to hundreds of hertz to optimize vibration intensity, affecting concrete fluidity and improving compaction.

[0043] As a concrete example: In the vibration compaction of industrial plant floor concrete, the adaptive control module employs a reinforcement learning algorithm. The state inputs include the compaction level (e.g., medium grade), concrete temperature (20 degrees Celsius), and steel reinforcement density (0.15 cubic meters per cubic meter). The algorithm is based on a Q-learning update strategy, and the reward function is designed according to changes in compaction and energy consumption. The output adjusts the vibration frequency from 50 Hz to 65 Hz and the amplitude from 1.5 mm to 2 mm. This adjustment command is sent to the multimodal vibration execution terminal via the CAN bus protocol, and combined with closed-loop control, optimizes the vibration process in real time. The remote monitoring platform displays parameter changes and effects, ensuring uniform compaction of the floor concrete and reducing defects.

[0044] By applying reinforcement learning algorithms to dynamically adjust vibration parameters, the system can adapt to changes in concrete state, improve the adaptability and accuracy of the vibration process, thereby optimizing the compaction effect, reducing quality fluctuations, and enhancing project reliability.

[0045] In some implementations, the remote monitoring platform constructs a digital twin model synchronized with the physical site, displaying the working status of each vibrating terminal and the distribution of concrete density in real time.

[0046] The physical site can refer to the actual concrete construction site or environment. A digital twin model can be a virtual replica of the physical entity, mapping the actual system state in real time. For example, it can be built based on WebGL or Unity engines, integrating multi-source sensor data streams, updating at a 5Hz frequency, displaying dynamic changes, visually monitoring the vibration process, and supporting remote decision-making and intervention. The working status of each vibration terminal can refer to the current operating conditions and performance indicators of each vibration execution terminal. For example, vibration frequency, amplitude, and temperature data can be collected through an edge computing gateway and transmitted as status parameters in JSON format for real-time monitoring of terminal operation and timely detection of anomalies or faults. Concrete density distribution can refer to the spatial variation of density in the concrete structure. For example, a heat map can be generated based on the output of a density identification module and UWB location data, using the Kriging interpolation algorithm to fill in unmeasured points, graphically displaying density uniformity and guiding local vibration adjustments.

[0047] As a concrete example: In the vibration compaction of highway pavement concrete, a remote monitoring platform uses WebGL technology to build a digital twin model synchronized with the physical site. The model integrates UWB positioning system data (update rate 10 Hz) and density identification module output (categorized as low, medium, and high). The platform displays the real-time operating status of each vibration terminal, including vibration frequency (50 to 100 Hz range), amplitude (1 to 3 mm), and temperature (20 to 30 degrees Celsius), and visualizes the concrete density distribution through color-coded heat maps. Data is transmitted via the MQTT protocol. Combined with closed-loop control, operators can remotely monitor vibration uniformity and adjust parameters promptly to ensure consistent pavement quality.

[0048] By constructing a digital twin model to synchronize physical site data in real time, the visualization and monitoring capabilities of the vibration process are enhanced, and the efficiency of remote management is improved, thereby optimizing the compaction and uniformity of concrete and reducing quality defects.

[0049] In some implementations, the system determines the vibration efficiency coefficient using the following formula: in, α is the vibration efficiency coefficient, which is a dimensionless comprehensive evaluation index output by the system. It is used to quantify the vibration effect. The value range is [0,1]. The larger the value, the better the vibration effect. α, β, and δ are weighting coefficients used to balance the contribution of different physical quantities to the comprehensive index. They satisfy α+β+δ=1 and are determined based on experimental data. The pressure-stiffness sensitivity coefficient is used to adjust the sensitivity of the effect of the pressure-stiffness matching degree on the efficiency coefficient, and is determined based on experimental data; N is the total number of dynamic pressure sensors, representing the number of pressure monitoring points, which is derived from the system configuration parameters. The dynamic pressure sensor reading of the i-th dynamic pressure sensor reflects the reaction pressure of concrete on the vibrator, and is derived from the dynamic pressure sensor array. The effective working area of ​​the i-th dynamic pressure sensor is used for pressure-to-force conversion calculations and is derived from the sensor design parameters. The nominal compressive strength of concrete serves as a benchmark value for pressure comparison and is derived from the design parameters of concrete materials. The nominal area of ​​the vibratory rod's working head serves as a benchmark for area comparison and is derived from the mechanical design parameters of the vibrator; M represents the total number of fiber optic grating sensors, indicating the number of internal status monitoring points, and is derived from the system configuration parameters. The dynamic elastic modulus of concrete is the value of the j-th fiber grating sensor, which reflects the stiffness characteristics of concrete and is calculated based on the strain measurement value of the fiber grating sensor. The nominal modulus of elasticity of concrete serves as a benchmark for comparison of moduli of elasticity and is derived from the design parameters of concrete materials. The temperature sensitivity coefficient is used to adjust the sensitivity of the effect of temperature deviation on the efficiency coefficient, and is determined based on experimental data; K is the total number of temperature sensors, representing the number of temperature monitoring points, which is derived from the system configuration parameters. The temperature measurement value of the kth temperature sensor reflects the temperature state of the concrete. The optimal vibration reference temperature for concrete is used as the target value for temperature control and is determined based on the concrete mix proportion and curing requirements; L is the total number of acceleration sensors, representing the number of vibration state monitoring points, which is derived from the system configuration parameters. The dimensionless damping ratio of the l-th accelerometer reflects the energy dissipation characteristics of concrete. The maximum allowable damping ratio is determined based on the properties of concrete materials and is used to normalize the effect of the damping ratio.

[0050] The vibration efficiency coefficient can refer to a comprehensive evaluation index that quantifies the vibration effect of concrete. For example, it can be calculated by a central control unit, with inputs including pressure, strain, and temperature sensor data, using weighting coefficients α, β, and δ to balance contributions, and an output value range of [0,1], used to quantify the vibration effect and as feedback input for adaptive control. The pressure-stiffness sensitivity coefficient can refer to a parameter that measures the sensitivity of the efficiency coefficient to the degree of pressure-stiffness matching. For example, it can be calibrated based on concrete material tests; a larger value indicates greater sensitivity to pressure-stiffness deviations, which can optimize the response of vibration parameters to the concrete state. The total number of dynamic pressure sensors can refer to the number of dynamic pressure sensors deployed in the system. For example, it can be configured according to the vibrator design and monitoring requirements. The number N is set during system initialization and is used to determine the sample size for pressure averaging. The dynamic pressure sensor reading can refer to the reaction pressure value measured by the dynamic pressure sensor, which can be used to calculate the efficiency coefficient of pressure-related terms. The effective working area of ​​the dynamic pressure sensor can refer to the effective area of ​​the pressure sensor used for force conversion. For example, it is determined by the sensor design and is used to convert pressure readings into force values, in square meters, to standardize pressure measurements in calculations. The nominal compressive strength of concrete refers to the standard compressive strength value in concrete design, such as that determined by mix proportions and age, derived from material design parameters, and measured in megapascals (MPa). It serves as a benchmark value for pressure comparison. The nominal area of ​​the vibrator head refers to the standard area of ​​contact between the vibrator head and the concrete, such as that given by the vibrator's mechanical design, used to calculate the nominal force, measured in square meters (m²), and used as a benchmark for area comparison. The total number of fiber Bragg grating sensors refers to the number of fiber Bragg grating sensors deployed in the system, such as those set according to the distribution of monitoring points. The number M is defined in the system and used to determine the sample size for strain averaging. The dynamic modulus of elasticity of concrete refers to the modulus of elasticity of concrete under dynamic loading, such as that calculated based on strain measurements using fiber Bragg grating sensors, reflecting stiffness characteristics, measured in gigapascals (GPa), and used to assess the internal state of concrete. The nominal modulus of elasticity of concrete also refers to the standard modulus of elasticity value in concrete design, such as that determined by mix proportions, derived from design parameters, and measured in gigapascals (GPa), used as a benchmark for modulus of elasticity comparison. The temperature sensitivity coefficient refers to a parameter that indicates the sensitivity of the efficiency coefficient to temperature deviations. It is determined, for example, through experimental data; a higher value indicates greater sensitivity to temperature changes and is used to optimize temperature-related adjustments. The total number of temperature sensors refers to the number of temperature sensors deployed in the system, configured according to monitoring needs. The number, K, is defined in the system settings and is used to determine the sample size for averaging temperatures. The optimal vibration reference temperature for concrete refers to the ideal temperature reference value for concrete vibration, set based on concrete mix proportions and curing requirements, derived from materials science data, and used as a target value for temperature control. The total number of acceleration sensors refers to the number of acceleration sensors deployed in the system, set according to vibration monitoring points. The number, L, is defined in the system and is used to determine the sample size for averaging damping ratios.Dimensionless damping ratio refers to a dimensionless parameter representing the energy attenuation characteristics of concrete vibration, such as that calculated from acceleration data based on the logarithmic decay method, with a value range of [0, 0.05], used to assess the energy dissipation of concrete. Maximum allowable damping ratio refers to the upper limit of the damping ratio allowed, for example, set based on the material properties of concrete, with a typical value of 0.05, used to limit the damping ratio to normalize its influence in calculations.

[0051] By integrating multi-sensor data and intelligently calculating the vibration efficiency coefficient, the vibration effect can be quantitatively evaluated and optimized in real time, thereby improving the density and uniformity of concrete, reducing quality defects, and enhancing the controllability and reliability of construction quality control.

[0052] In some implementations, the damping ratio Determined by the following formula: in, The maximum amplitude of the acceleration of the l-th accelerometer reflects the peak value of the vibration intensity and is derived from the measurement data of the triaxial accelerometer. The minimum acceleration amplitude of the l-th accelerometer reflects the valley value of vibration intensity and is derived from the measurement data of the triaxial accelerometer. is the equivalent concrete resistance of the l-th accelerometer, reflecting the concrete's resistance to vibration, calculated based on data from the dynamic pressure sensor array; F0 is the rated excitation force of the vibrator, used as a benchmark for resistance comparison, derived from the vibrator's performance parameters; tanh(·) is the hyperbolic tangent function, used to limit the influence of the resistance ratio within a reasonable range.

[0053] The maximum amplitude of acceleration refers to the maximum value of vibration acceleration within a specific time period, such as that extracted from triaxial accelerometer data and determined using a peak detection algorithm. It is measured in meters per square second (m / s²) and reflects the peak vibration intensity, used to calculate the damping ratio. The minimum amplitude of acceleration refers to the minimum value of vibration acceleration within a specific time period, such as that obtained from accelerometer data and found using a valley detection algorithm. It is measured in meters per square second (m / s²) and can be combined with the maximum amplitude to calculate the attenuation rate, used to determine the damping ratio. The equivalent resistance of concrete refers to the equivalent resistance force of concrete to the vibration of the vibrator. It is calculated, for example, based on dynamic pressure sensor array data, obtained by integrating or multiplying the average pressure value by the area of ​​action. It is measured in Newtons (N) and reflects the resistance of concrete to vibration, used to assess the concrete condition. The rated excitation force of the vibrator refers to the standard excitation force value specified in the vibrator design, such as that obtained from the vibrator performance parameter table. It is measured in Newtons and serves as a benchmark for resistance comparison, supporting damping ratio calculations.

[0054] By accurately calculating the damping ratio, the assessment of vibration energy attenuation characteristics is enhanced, and the accuracy of the vibration efficiency coefficient is improved, thereby optimizing concrete density control, reducing quality fluctuations, and improving construction reliability.

[0055] In some implementations, the system also includes an alarm module that generates an early warning signal and sends it to the remote monitoring platform when the vibration efficiency coefficient is detected to be lower than the threshold or the compaction identification module outputs an abnormal state for multiple consecutive cycles.

[0056] An alarm module can refer to a component used to generate and send alarms. For example, it can be implemented through software algorithms to monitor the vibration efficiency coefficient and compaction status. When conditions are met, it triggers an alarm via a digital signal to promptly notify the operator of potential problems and prevent quality defects. Multiple consecutive cycles can refer to a series of continuous time periods or operation cycles. For example, a cycle length of several seconds can be defined to count the number of consecutive abnormal outputs, such as three times, implemented through a counter to confirm the persistence of the abnormal state and reduce false alarms. An abnormal state can refer to a state that does not meet normal or expected standards. For example, a low-compactness classification output by a compaction identification module, based on probability threshold judgment and detected through a state machine, can indicate potential quality problems and trigger an early warning.

[0057] As a concrete example: In the vibration compaction of bridge piers and abutments, the alarm module monitors the output status of the vibration efficiency coefficient and the density identification module in real time. When the vibration efficiency coefficient falls below the threshold of 0.8 or the density identification module outputs an abnormal state (such as low density level) for three consecutive cycles (5 seconds each), the alarm module generates an early warning signal. The signal is sent to the remote monitoring platform in JSON format via the MQTT protocol. Upon receiving the signal, the platform triggers a visual alarm, notifying the operator to check the vibration parameters. Combined with closed-loop control and adaptive adjustment, problems are corrected promptly to ensure uniform and dense concrete compaction. By monitoring key parameters in real time and generating early warning signals, the system's ability to respond promptly to abnormal states is enhanced, improving the reliability and controllability of concrete vibration quality and reducing the risk of construction defects.

[0058] Corresponding to the above system embodiments, this specification also provides embodiments of intelligent control methods for concrete vibration. Figure 2 A flowchart of a smart control method for concrete vibration provided in some embodiments of this specification is shown. Figure 2 As shown, the specific steps include: The concrete is subjected to mechanical vibration by a multimodal vibration execution terminal, and vibration status data is collected. Information about the internal state of concrete is obtained through a distributed sensor network. The data processing module processes the raw data collected by the multimodal vibration execution terminal and the distributed sensor network. The density identification module determines the density of concrete based on preprocessed data. The vibration parameters are adjusted by the adaptive control module based on the compaction identification results. The vibration process and quality status are visualized through a remote monitoring platform; The multimodal vibration execution terminal transmits the collected vibration state data to the data processing module. The data processing module filters and extracts features from the vibration state data before inputting it into the compaction identification module. The compaction level output by the compaction identification module is transmitted to the adaptive control module. The adaptive control module generates vibration parameter adjustment instructions and sends them to the multimodal vibration execution terminal, forming a closed-loop control.

[0059] The above is a schematic scheme of an intelligent control method for concrete vibration according to this embodiment. It should be noted that the technical solution of this intelligent control method for concrete vibration belongs to the same concept as the technical solution of the intelligent control system for concrete vibration described above. For details not described in detail in the technical solution of the intelligent control method for concrete vibration, please refer to the description of the technical solution of the intelligent control system for concrete vibration described above.

[0060] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0061] The preferred embodiments disclosed above are merely illustrative of this specification. The optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this invention. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize this specification. This specification is limited only by the claims and their full scope and equivalents.

Claims

1. A smart control system for concrete vibration, characterized in that, include: A multimodal vibration execution terminal is used to apply mechanical vibration to concrete and collect vibration state data; Distributed sensor networks are used to acquire information about the internal state of concrete. The central control unit is used to coordinate the operation of various components of the system; The data processing module is used to process the raw data collected by the multimodal vibration execution terminal and the distributed sensor network; The compactness identification module is used to determine the compactness of concrete based on preprocessed data; An adaptive control module is used to adjust the vibration parameters based on the compaction identification results; A remote monitoring platform is used to visually display the vibration process and quality status; The multimodal vibration execution terminal transmits the collected vibration state data to the data processing module. The data processing module filters and extracts features from the vibration state data and then inputs it into the compaction identification module. The compaction level output by the compaction identification module is transmitted to the adaptive control module. The adaptive control module generates vibration parameter adjustment instructions and sends them to the multimodal vibration execution terminal, forming a closed-loop control.

2. The system according to claim 1, characterized in that, The multimodal vibration execution terminal includes a vibrator body, a triaxial accelerometer, a dynamic pressure sensor array, a temperature sensor, and an edge computing gateway; The triaxial accelerometer is used to measure the vibration acceleration of the vibrator in three orthogonal directions; The dynamic pressure sensor array is used to detect the reaction pressure of concrete on the vibrator. The temperature sensor is used to monitor the concrete temperature; The edge computing gateway is used to perform preliminary processing on sensor data.

3. The system according to claim 2, characterized in that, The distributed sensor network includes fiber optic grating sensors embedded in concrete and UWB positioning systems deployed at the construction site. The fiber optic grating sensor is used to measure the internal strain of concrete. The UWB positioning system is used to track the spatial position of the vibrator in real time.

4. The system according to claim 1, characterized in that, The density identification module adopts a deep convolutional neural network model. The input of the model is a preprocessed time-series segment of multi-sensor data, and the output is a classification of concrete density level.

5. The system according to claim 1, characterized in that, The adaptive control module uses a reinforcement learning algorithm to automatically adjust the vibration frequency and amplitude based on the compaction level, concrete temperature, and steel reinforcement density.

6. The system according to claim 3, characterized in that, The remote monitoring platform constructs a digital twin model synchronized with the physical site, displaying the working status of each vibration terminal and the distribution of concrete density in real time.

7. The system according to claim 2, characterized in that, The system determines the vibration efficiency coefficient using the following formula: in, α is the vibration efficiency coefficient, which is a dimensionless comprehensive evaluation index output by the system. It is used to quantify the vibration effect. The value range is [0,1]. The larger the value, the better the vibration effect. α, β, and δ are weighting coefficients used to balance the contribution of different physical quantities to the comprehensive index. They satisfy α+β+δ=1 and are determined based on experimental data. The pressure-stiffness sensitivity coefficient is used to adjust the sensitivity of the effect of the pressure-stiffness matching degree on the efficiency coefficient, and is determined based on experimental data; N is the total number of dynamic pressure sensors, representing the number of pressure monitoring points, which is derived from the system configuration parameters. The dynamic pressure sensor reading of the i-th dynamic pressure sensor reflects the reaction pressure of concrete on the vibrator, and is derived from the dynamic pressure sensor array. The effective working area of ​​the i-th dynamic pressure sensor is used for pressure-to-force conversion calculations and is derived from the sensor design parameters. The nominal compressive strength of concrete serves as a benchmark value for pressure comparison and is derived from the design parameters of concrete materials. The nominal area of ​​the vibratory rod's working head serves as a benchmark for area comparison and is derived from the vibratory machine's mechanical design parameters; M represents the total number of fiber optic grating sensors, indicating the number of internal status monitoring points, and is derived from the system configuration parameters. The dynamic elastic modulus of concrete is the value of the j-th fiber grating sensor, which reflects the stiffness characteristics of concrete and is calculated based on the strain measurement value of the fiber grating sensor. The nominal modulus of elasticity of concrete serves as a benchmark for comparison of moduli of elasticity and is derived from the design parameters of concrete materials. The temperature sensitivity coefficient is used to adjust the sensitivity of the effect of temperature deviation on the efficiency coefficient, and is determined based on experimental data; K is the total number of temperature sensors, representing the number of temperature monitoring points, which is derived from the system configuration parameters. The temperature measurement value of the kth temperature sensor reflects the temperature state of the concrete. The optimal vibration reference temperature for concrete is used as the target value for temperature control and is determined based on the concrete mix proportion and curing requirements; L is the total number of acceleration sensors, representing the number of vibration state monitoring points, which is derived from the system configuration parameters. The dimensionless damping ratio of the l-th accelerometer reflects the energy dissipation characteristics of concrete. The maximum allowable damping ratio is determined based on the properties of concrete materials and is used to normalize the effect of the damping ratio.

8. The system according to claim 7, characterized in that, The damping ratio Determined by the following formula: in, The maximum amplitude of the acceleration of the l-th accelerometer reflects the peak value of the vibration intensity and is derived from the measurement data of the triaxial accelerometer. The minimum acceleration amplitude of the l-th accelerometer reflects the valley value of vibration intensity and is derived from the measurement data of the triaxial accelerometer. is the equivalent concrete resistance of the l-th accelerometer, reflecting the concrete's resistance to vibration, calculated based on data from the dynamic pressure sensor array; F0 is the rated excitation force of the vibrator, used as a benchmark for resistance comparison, derived from the vibrator's performance parameters; tanh(·) is the hyperbolic tangent function, used to limit the influence of the resistance ratio within a reasonable range.

9. The system according to claim 1, characterized in that, The system also includes an alarm module, which generates an early warning signal and sends it to the remote monitoring platform when the vibration efficiency coefficient is detected to be lower than the threshold or the compaction identification module outputs an abnormal state for multiple consecutive cycles.

10. A method for intelligent control of concrete vibration based on the system of claim 1, characterized in that, The method is applied to the intelligent concrete vibration control system according to any one of claims 1 to 9, and the method includes: The concrete is subjected to mechanical vibration by a multimodal vibration execution terminal, and vibration status data is collected. Information about the internal state of concrete is obtained through a distributed sensor network. The data processing module processes the raw data collected by the multimodal vibration execution terminal and the distributed sensor network. The density identification module determines the density of concrete based on preprocessed data. The vibration parameters are adjusted by the adaptive control module based on the compaction identification results. The vibration process and quality status are visualized through a remote monitoring platform; The multimodal vibration execution terminal transmits the collected vibration state data to the data processing module. The data processing module filters and extracts features from the vibration state data and then inputs it into the compaction identification module. The compaction level output by the compaction identification module is transmitted to the adaptive control module. The adaptive control module generates vibration parameter adjustment instructions and sends them to the multimodal vibration execution terminal, forming a closed-loop control.

Images