Concrete working performance monitoring method and system based on coarse aggregate slurry wrapping state recognition

Abstract

The invention discloses a concrete working performance monitoring method and system based on coarse aggregate slurry wrapping state recognition, and relates to a method and system for monitoring the working performance of premixed concrete based on machine vision and machine learning. The invention discloses a concrete working performance monitoring method based on coarse aggregate slurry wrapping state recognition. The concrete working performance monitoring method comprises the following steps: acquiring a video stream in a stirring process through an image acquisition device arranged in a concrete stirrer; extracting a key frame from the video stream, and identifying coarse aggregate particles in the key frame; calculating characteristic parameters reflecting the wrapping state of the slurry on the surfaces of the coarse aggregate particles; inputting the characteristic parameters into a pre-trained concrete working performance evaluation model to obtain a working performance evaluation result; and outputting the evaluation result. According to the method, direct and quantitative on-line monitoring of the microstructure state of the concrete in the stirring process is achieved, a more essential and more accurate quality evaluation means is provided by revealing the internal relation between the slurry wrapping state and the working performance, and the lagging, indirect and macroscopic defects of a traditional method are overcome.

Description

Technical Field

[0001] This invention relates to the fields of building materials and industrial automation technology, and in particular to a method and system for real-time, online intelligent monitoring of the workability of ready-mixed concrete by analyzing the slurry coating state on the surface of coarse aggregate in a concrete mixer, based on machine vision and machine learning. Background Technology

[0002] The workability of ready-mixed concrete (such as fluidity, cohesiveness, and water retention) is crucial in determining its construction quality and final structural performance. Currently, monitoring workability mainly relies on operator experience and offline slump tests, which are highly subjective and subject to delays.

[0003] Several automated monitoring solutions have emerged in the prior art. For example, Chinese invention patent (publication number CN103042603A) discloses a method for indirectly determining the consistency of concrete by monitoring the current of the mixer. However, this method is easily affected by equipment status and cannot reflect the component state of the concrete (such as segregation). Chinese patent CN117788202A discloses a slump detection method based on image recognition, but it detects static concrete in a specific container and cannot achieve real-time online monitoring during the production process.

[0004] More importantly, existing visual methods primarily focus on the overall flow pattern or color texture of concrete, lacking the technology to penetrate deep into the mixer and quantitatively analyze the microscopic characteristic of the paste coating state on the surface of coarse aggregate—a characteristic closely related to concrete workability. The paste coating state directly reflects the lubrication and cohesion of the cement paste, and is a core factor determining the fluidity and segregation resistance of concrete. Therefore, developing a technology capable of online and accurate assessment of the paste coating state is of great significance for improving concrete quality control. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method and system for monitoring the workability of concrete based on the identification of the coating state of coarse aggregate slurry, so as to solve the problem that the existing technology cannot directly and quantitatively evaluate the micro-state of concrete during the mixing process.

[0006] One of the technical solutions provided by this invention is a method for monitoring the workability of concrete based on the identification of the coating state of coarse aggregate paste, comprising the following steps:

[0007] Step S1: During the mixing process, video stream data containing coarse aggregate is continuously collected using a wear-resistant endoscope fixed inside the concrete mixer.

[0008] Step S2: Extract keyframes containing clear coarse aggregate images from the video stream data, perform image analysis on the keyframes, and identify and segment the coarse aggregate particles.

[0009] Step S3: Calculate the characteristic parameters reflecting the slurry coating state; the characteristic parameters include at least the slurry coating thickness ratio and / or slurry coating uniformity.

[0010] Step S4: Input the feature parameters into the pre-trained concrete workability evaluation model and output the evaluation result of the current concrete workability. The evaluation result includes quantitative indicators and / or state classification.

[0011] Step S5: Based on the evaluation results, generate production guidance information or control instructions.

[0012] This invention relates to a method for monitoring the workability of concrete based on the identification of the slurry coating state of coarse aggregate. In step S3, the method for calculating the slurry coating thickness ratio includes:

[0013] Step S31: For each individual coarse aggregate particle that has been separated, calculate its minimum bounding rectangle or equivalent ellipse;

[0014] Step S32: Calculate the ratio of the actual pixel area A_actual of the coarse aggregate particle to its minimum bounding contour range, such as the pixel area A_bounding of a rectangle or equivalent ellipse.

[0015] Step S33: The slurry coating thickness ratio R_coating is defined as: R_coating=(A_bounding-A_actual) / A_actual, which reflects the thickness of the slurry coating layer relative to the aggregate core.

[0016] This invention relates to a concrete workability monitoring method based on the identification of coarse aggregate slurry coating state. In step S3, the method for calculating the uniformity of slurry coating is as follows: statistically analyze the dispersion of the ratio of slurry coating thickness of all identified coarse aggregate particles in a single frame image, specifically including the standard deviation or coefficient of variation. The larger the value, the worse the coating uniformity.

[0017] This invention relates to a method for monitoring the workability of concrete based on the identification of the coarse aggregate slurry encapsulation state. In step S4, the method for constructing a pre-trained concrete workability evaluation model includes:

[0018] Step A1: Collect a large number of endoscopic video samples with different mixing ratios and different stirring conditions, and simultaneously measure their corresponding standard working performance indicators as labels;

[0019] Step A2: Perform steps S2 and S3 for each video sample to calculate the feature parameters of its slurry encapsulation state and construct a labeled training sample library;

[0020] Step A3: Use a machine learning model to learn from the training sample library and establish a mapping model from the slurry encapsulation state characteristic parameters to the working performance indicators.

[0021] This invention relates to a method for monitoring the workability of concrete based on the identification of the coarse aggregate slurry encapsulation state, wherein in step A3, the machine learning model is one of gradient boosting decision tree, random forest, or neural network model.

[0022] This invention relates to a concrete workability monitoring method based on the identification of coarse aggregate slurry encapsulation state. In step S5, the control instructions include: triggering an alarm signal, such as an audible and visual alarm, when the evaluation result indicates abnormal workability; or feeding back the evaluation result to the concrete production control system for automatically adjusting batching parameters, such as water content or admixture dosage.

[0023] Another technical solution provided by the present invention is a monitoring system for implementing the above method, comprising:

[0024] The data acquisition module includes a wear-resistant endoscope fixed to the mixing arm of a mixer or the inner wall of a tank via a mounting base. The mounting base has a cooling channel inside, and the endoscope tip has an anti-adhesion mechanism.

[0025] The edge computing module communicates with the data acquisition module to receive video data, identify coarse aggregates, and perform image processing, feature parameter extraction, and model evaluation calculations.

[0026] The results output and interaction module includes an audible and visual alarm and / or a data interface, used to display evaluation results, trigger alarms, and / or send control commands to the host computer system.

[0027] It also includes a cloud platform that communicates with multiple edge computing modules to receive data from multiple mixers, aggregate the data for big data analysis, optimize the evaluation model, and provide a remote monitoring interface.

[0028] The present invention, a method and system for monitoring the workability of concrete based on the identification of the coarse aggregate slurry encapsulation state, differs from existing technologies in that:

[0029] (1) The essentiality and innovation of feature selection: This invention is the first to use the microscopic physical feature of the slurry coating state on the surface of coarse aggregate as the core monitoring indicator. This feature has a direct and clear physical relationship with the workability of concrete, such as lubricity, cohesiveness and water retention. Compared with macroscopic dynamic or texture features, it can better reflect the intrinsic quality of concrete, and the evaluation results are more scientific and accurate.

[0030] (2) Advanced and quantifiable monitoring methods: Through advanced image segmentation and machine learning technologies, the subjective visual judgment of “wrapping sensation” is transformed into objective and quantifiable parameters such as the slurry wrapping thickness ratio and uniformity, thereby realizing precise measurement of the workability of concrete at the micro level.

[0031] (3) The early warning is advanced and highly instructive: Poor slurry coating is an early sign of concrete segregation and bleeding. This invention can issue an early warning at the early stage of the problem (while it is still in the mixer) and can provide direct and specific operational guidance for adjusting the material ratio (such as increasing the sand ratio or the amount of slurry), which is incomparable to indirect measurement methods and macroscopic visual methods.

[0032] (4) System practicality and reliability: The entire system is built on industrial endoscope and edge computing hardware, and is designed for the harsh environment inside the mixer, ensuring the long-term stability and reliability of online monitoring. Detailed Implementation

[0033] The following examples are used to illustrate the present invention, but are not intended to limit the scope of the invention.

[0034] Example 1

[0035] This invention relates to a concrete workability monitoring method based on the identification of coarse aggregate slurry encapsulation state, and slump prediction based on slurry encapsulation thickness.

[0036] In the mixing plant, the system continuously acquires video containing coarse aggregate through a data acquisition module, i.e., an image acquisition device, such as an endoscope, fixed on the mixing arm inside the concrete mixer. Then, the edge computing module identifies and segments the coarse aggregate (particle size greater than 5mm) in the keyframes, calculates the paste coating thickness ratio R_coating for each aggregate, and calculates the average value R_avg of this ratio for all aggregates in the entire frame. R_avg is then input into a pre-trained XGBoost regression model, which directly outputs the slump prediction value. When the prediction value is lower than a set threshold, the system alarms to prompt the operator to check the water consumption or paste content.

[0037] Example 2

[0038] This invention relates to a concrete workability monitoring method based on the identification of coarse aggregate paste encapsulation state, and a segregation risk warning based on encapsulation uniformity:

[0039] For high-flowability concrete, the system focuses on monitoring the uniformity of the slurry coating. In the later stages of mixing, the system calculates the coefficient of variation (CV) of the aggregate coating thickness ratio in multiple consecutive frames of images. The CV is then input into a pre-trained concrete workability evaluation model. When the CV value is consistently higher than a preset risk threshold, the system determines that there is a high risk of segregation, triggering a special alarm for "uneven coating, risk of segregation," and recommends checking the water-reducing agent dosage and aggregate gradation.

[0040] Example 3

[0041] This invention relates to a concrete workability monitoring system based on the identification of coarse aggregate slurry encapsulation state and a closed-loop control system for the batching system:

[0042] In the automated mixing plant, the system's edge computing module transmits the calculated average slurry coating thickness ratio R_avg to the central controller in real time. The central controller has PID control logic: if R_avg remains below the target range, the output and interaction module generate production guidance information or control commands, such as automatically increasing the amount of fine aggregate (sand) and / or water in the next batch to increase the slurry volume, thus achieving closed-loop quality control based on micro-state conditions.

[0043] Example 4:

[0044] This invention relates to a comprehensive evaluation method for monitoring the workability of concrete based on the identification of the coarse aggregate paste encapsulation state, using multi-feature fusion:

[0045] The data acquisition module simultaneously extracts the slurry coating thickness ratio, coating uniformity, and traditional aggregate movement speed to construct a multi-dimensional feature vector. This vector is then input into the neural network model of the edge computing module. The output module, along with the interaction module, simultaneously outputs the slump prediction, cohesiveness score, and segregation risk level, providing the operator with a more comprehensive performance dashboard view.

[0046] The present invention provides a concrete workability monitoring method and system based on the identification of coarse aggregate slurry coating state. This method uses the microscopic physical characteristic of the slurry coating state on the surface of coarse aggregate as the core monitoring indicator. This characteristic has a direct and clear physical correlation with the workability of concrete, such as lubricity, cohesiveness and water retention. Compared with macroscopic dynamic or texture characteristics, this method can better reflect the intrinsic quality of concrete, and the evaluation results are more scientific and accurate.

[0047] This invention relates to a method and system for monitoring the workability of concrete based on the identification of the coarse aggregate slurry coating state. Through advanced image segmentation and machine learning technologies, it transforms the subjective visual judgment of "coating sensation" into objective and quantifiable parameters such as the slurry coating thickness ratio and uniformity, thereby achieving precise measurement of the workability of concrete at the microscopic level.

[0048] Poor slurry coating is an early sign of concrete segregation and bleeding. This invention can provide early warnings at the early stage of the problem (while it is still in the mixer) and provide direct and specific operational guidance for adjusting the material ratio (such as increasing the sand ratio or slurry volume), which is unmatched by indirect measurement methods and macroscopic visual methods.

[0049] This invention relates to a concrete workability monitoring system based on the identification of coarse aggregate slurry encapsulation state. Built on industrial endoscopes and edge computing hardware, it is specifically designed for the harsh environment inside mixers, ensuring long-term stability and reliability of online monitoring.

[0050] Although the present invention has been described in detail above with general descriptions and specific embodiments, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the present invention fall within the scope of protection claimed by the present invention.

Claims

1. A method for monitoring the workability of concrete based on the identification of the coating state of coarse aggregate paste, characterized in that: Includes the following steps: Video streams during the mixing process are captured by an image acquisition device installed inside the concrete mixer. Extract keyframes from the video stream and identify coarse aggregate particles in the keyframes; Calculate the characteristic parameters that reflect the slurry coating state on the surface of the coarse aggregate particles; The feature parameters are input into a pre-trained concrete workability evaluation model to obtain the workability evaluation results; Output the evaluation results.

2. The method according to claim 1, characterized in that: The feature parameters include the slurry coating thickness ratio. For a single coarse aggregate particle, calculate its actual pixel area A_actual and the pixel area A_bounding of its outer contour range, then the slurry coating thickness ratio R_coating=(A_bounding-A_actual) / A_actual.

3. The method according to claim 1 or 2, characterized in that: The feature parameters also include slurry coating uniformity, which is the statistical dispersion of the ratio of slurry coating thickness of multiple coarse aggregate particles in a single frame image.

4. The method according to claim 1, characterized in that: The method for constructing the pre-trained concrete workability evaluation model includes the following steps: Acquire video training samples from multiple image acquisition devices and their corresponding measured standard operating performance values; Feature parameters of the slurry encapsulation state are extracted from each training sample to construct a feature vector; Using the feature vector as input and the measured standard working performance value as the output target, a machine learning model is trained to obtain the evaluation model.

5. The method according to claim 4, characterized in that: The machine learning model is a gradient boosting decision tree or a neural network model.

6. The method according to claim 1, characterized in that: The output of the evaluation results specifically includes: generating an alarm signal when the evaluation results indicate abnormal performance; or feeding the evaluation results back to the concrete production control system for automatic adjustment of batching parameters.

7. A concrete workability monitoring system for implementing the method of any one of claims 1-6, characterized in that, include: The data acquisition module includes an endoscope installed inside the mixer; The edge computing module is communicatively connected to the data acquisition module and is used to perform video stream reception, coarse aggregate identification, calculation of slurry encapsulation state characteristic parameters, and model evaluation. The results output and interaction module is communicatively connected to the edge computing module and is used to output the evaluation results.

8. The system according to claim 7, characterized in that: The endoscope is fixed to the mixing arm of the mixer or the inner wall of the tank via a mounting base. The mounting base has a cooling channel inside, and the endoscope tip is provided with an anti-adhesion mechanism.

9. The system according to claim 7, characterized in that: The result output and interaction module includes an audible and visual alarm and a data interface, used to send control commands to the host computer control system. The data acquisition module includes a wear-resistant endoscope, which is fixed inside the mixer by a special mounting base. The endoscope is equipped with a wear-resistant lens and an anti-adhesion mechanism at its end. The edge computing module communicates with the data acquisition module to receive video data and perform image processing, feature parameter extraction, and model evaluation calculations. The results output and interaction module is used to display evaluation results, trigger alarms, and / or send control commands to the host computer system.

10. The system according to claim 7, characterized in that: It also includes a cloud platform that communicates with multiple edge computing modules to aggregate data, optimize the evaluation model, and provide a remote monitoring interface.