Concrete accelerator production quality detection system and method
By adopting intelligent data processing, robustness enhancement and comprehensive error analysis detection algorithms in the concrete rapid set agent production quality detection system, the problem of insufficient accuracy of production data feature processing and model robustness is solved, and efficient, accurate and Reliable quality inspection results.
Patent Information
- Application Number
- CN202510011885.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-06
- Publication Date
- 2025-02-07
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
During the quality inspection of concrete rapid settling agent production, the accuracy of the production data is poor, the model is poor, and the detection results are inaccurate.
A concrete rapid settling agent production quality detection system and method is adopted, including real-time acquisition of raw data and intelligent processing, real-time analysis using existing pre-trained neural network models, and the introduction of a robust enhancement function optimization model. Through redundant feature elimination algorithms and data augmentation technology, ensure the independence of feature sets and the accuracy of analysis results. Based on the analysis results, a comprehensive error analysis detection algorithm is used to perform quality inspection, and early warning information and adjustment suggestions are generated.
It improves the accuracy of the characteristic processing of concrete rapid settling agent production data and the robustness of the model, ensures the stability and reliability of the test results, and meets the needs of modern production for efficient, accurate and real-time quality inspection.
Smart Images
Figure CN119398623A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of quality detection, and in particular to a concrete accelerating setting agent production quality detection system and method. Background Art
[0002] Quality inspection is a crucial link in the production process of concrete accelerators. Traditional quality inspection methods usually rely on manual inspection and simple instrument measurement, which is not only inefficient, but also difficult to ensure accuracy and consistency. These traditional methods cannot meet the needs of modern concrete accelerator production for efficient, accurate and real-time quality inspection.
[0003] With the development of industrial automation and intelligent technology, more and more manufacturing companies have begun to introduce advanced sensor technology, data analysis technology and automatic control technology to improve the automation level of the production process and product quality.
[0004] However, the above technology has the following technical problems: in the process of concrete accelerator production quality inspection, the accuracy of feature processing of production data is poor, the model robustness is poor and the inspection results are inaccurate. Summary of the invention
[0005] The present invention provides a concrete accelerating setting agent production quality detection system and method, so as to solve the technical problems of poor accuracy, poor model robustness and inaccurate detection results when performing feature processing on production data during the concrete accelerating setting agent production quality detection process.
[0006] A method for detecting the production quality of a concrete accelerating setting agent comprises the following steps: S1. Real-time collection of original concrete accelerator production data, intelligent processing of the original concrete accelerator production data, and obtaining intelligently processed data; real-time analysis of the intelligently processed data using an existing pre-trained neural network model, and introduction of a robustness enhancement function to optimize the pre-trained neural network model, obtaining an optimized pre-trained neural network model, and using the output of the optimized pre-trained neural network model as the analysis result, the analysis result includes quality indicator analysis results, trend analysis results, and anomaly detection results; wherein the specific formula of the robustness enhancement function is: , in, It is data after intelligent processing; is the output of an existing pre-trained neural network model; and is an adjustment parameter used to control the strength of the nonlinear transformation; , , All are weight matrices; is the result of the analysis; S2. Based on the analysis results, the concrete accelerator is subjected to quality inspection using a comprehensive error analysis detection algorithm; in the process of implementing the comprehensive error analysis detection algorithm, a concrete accelerator sample is collected and tested to obtain an actual detection value; based on the actual detection value, an actual detection value matrix is constructed; based on the quality index analysis results in the analysis results and the actual detection value matrix, a detection error matrix is calculated; based on the detection error matrix, a comprehensive error scoring mechanism is introduced to obtain a comprehensive error score; the specific formula for the comprehensive error score is: , in, and is the weight coefficient; is the comprehensive error score; is the number of concrete accelerator samples; is the number of detection indicators; is the first Concrete Accelerator Sample No. The error value of each detection indicator; It is Anomaly detection scores of concrete accelerator samples; It is The trend value of each detection indicator; Based on the comprehensive error score, the test results are obtained; based on the test results, early warning information and adjustment suggestions in the production process are obtained; based on the adjustment suggestions, production parameter adjustment instructions are generated to adjust the production parameters.
[0007] Preferably, the S1 specifically includes: In the process of realizing intelligent processing, the original concrete accelerator production data is preprocessed to obtain preprocessed data; the preprocessed data is subjected to feature extraction to obtain a feature set; and a redundant feature elimination algorithm is introduced to update the feature set.
[0008] Preferably, the S1 specifically includes: The specific implementation steps of the redundant feature elimination algorithm are: the first step is to calculate the correlation between the features in the feature set; the second step is to screen the collinear features to obtain the screened feature set; the third step is to use the screened feature set to update the feature set.
[0009] Preferably, the S1 specifically includes: Perform data enhancement processing on the updated feature set to obtain enhanced data; merge the enhanced data, the updated feature set and the pre-processed data to obtain intelligently processed data.
[0010] Preferably, the S1 specifically includes: The optimization objective function is introduced to optimize the parameters of the optimized pre-trained neural network model.
[0011] A concrete accelerating agent production quality detection system includes the following parts: Data acquisition module, data intelligent processing module, intelligent analysis module, automatic detection module, intelligent control module; The data acquisition module collects the production data of the original concrete accelerating agent in real time and sends the production data of the original concrete accelerating agent to the data intelligent processing module; The data intelligent processing module pre-processes the original concrete accelerator production data to obtain pre-processed data; extracts features from the pre-processed data to obtain a feature set; introduces a redundant feature elimination algorithm to update the feature set; performs data enhancement processing on the updated feature set to obtain enhanced data; merges the enhanced data, the updated feature set and the pre-processed data to obtain intelligently processed data; and sends the intelligently processed data to the intelligent analysis module; The intelligent analysis module performs real-time analysis on the intelligently processed data and performs robustness enhancement processing to obtain analysis results; the analysis results are sent to the automated detection module; The automated detection module performs quality detection on the concrete accelerating agent based on the analysis results to obtain the detection results; and sends the detection results to the intelligent control module; The intelligent control module obtains early warning information and adjustment suggestions in the production process based on the detection results, generates production parameter adjustment instructions based on the adjustment suggestions, and adjusts the production parameters according to the production parameter adjustment instructions.
[0012] The beneficial effects of the technical solution of the present invention are: 1. In the process of intelligently processing the original concrete accelerator production data, the present invention uses a redundant feature elimination algorithm to calculate the correlation between the features in the feature set, solves the problem of feature redundancy and collinearity, and ensures the independence and effectiveness of the feature set; uses the existing pre-trained neural network model to perform real-time analysis on the data after intelligent processing to ensure the accuracy and timeliness of the analysis; introduces a robustness enhancement function to perform robustness enhancement processing to enhance the resistance of the existing pre-trained neural network model to abnormal data, optimizes the existing pre-trained neural network model, and outputs the optimized pre-trained neural network model as the analysis result; by optimizing parameters and weight matrices, the performance of the optimized pre-trained neural network model is further improved, the over-fitting of the optimized pre-trained neural network model is prevented, and the stability and reliability of the analysis result are ensured.
[0013] 2. The present invention introduces a comprehensive error analysis detection algorithm, calculates the detection error between the actual detection value matrix and the quality index analysis result in the analysis result, and introduces a comprehensive error scoring mechanism to conduct a comprehensive analysis of various detection indicators to ensure the accuracy and comprehensiveness of the detection results, effectively evaluate the quality of the concrete accelerator, and determine whether it meets the standards. This method can comprehensively consider factors such as detection error, trend value and abnormal detection score, and improve the comprehensiveness and reliability of quality detection. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Figure 1 This is a structural diagram of a concrete accelerating agent production quality detection system according to the present invention; Figure 2 The present invention is a flow chart of a method for detecting the production quality of a concrete accelerating setting agent. DETAILED DESCRIPTION
[0015] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all 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.
[0016] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0017] The specific scheme of the concrete accelerating setting agent production quality detection system and method provided by the present invention is described in detail below with reference to the accompanying drawings.
[0018] Refer to the attached Figure 1 , which shows a structural diagram of a concrete accelerating agent production quality detection system provided by an embodiment of the present invention, the system includes the following parts: Data acquisition module, data intelligent processing module, intelligent analysis module, automatic detection module, intelligent control module; The data acquisition module collects the physical and chemical index data of the concrete accelerator in real time through sensors, obtains the original concrete accelerator production data, and sends the original concrete accelerator production data to the data intelligent processing module; the physical and chemical index data include temperature, humidity, pH value, concentration, etc.; The data intelligent processing module pre-processes the original concrete accelerator production data to obtain pre-processed data; extracts features from the pre-processed data to obtain a feature set; introduces a redundant feature elimination algorithm to eliminate redundant features and update the feature set; performs data enhancement processing on the updated feature set to obtain enhanced data; merges the enhanced data, the updated feature set and the pre-processed data to obtain intelligently processed data; and sends the intelligently processed data to the intelligent analysis module; The intelligent analysis module performs real-time analysis on the intelligently processed data and performs robustness enhancement processing to obtain analysis results; the analysis results are sent to the automated detection module; The automated detection module performs quality detection on the concrete accelerating agent based on the analysis results to obtain the detection results; and sends the detection results to the intelligent control module; The intelligent control module obtains early warning information and adjustment suggestions in the production process based on the detection results, generates production parameter adjustment instructions based on the adjustment suggestions, adjusts the production parameters according to the production parameter adjustment instructions, and realizes the production quality detection of concrete accelerator.
[0019] Refer to the attached Figure 2 , which shows a flow chart of a method for detecting the production quality of a concrete accelerating setting agent provided by an embodiment of the present invention, the method comprising the following steps: S1. Real-time collection of original concrete accelerator production data, intelligent processing of the original concrete accelerator production data, and obtaining intelligently processed data; real-time analysis of the intelligently processed data to obtain analysis results; Expert experience is used to select and install appropriate sensors according to detection requirements, such as temperature sensors, humidity sensors, pH sensors, and concentration sensors; the sensors are used to collect concrete accelerator production data in real time to obtain original concrete accelerator production data, such as temperature, humidity, pH value, and concentration.
[0020] The original concrete accelerator production data is intelligently processed to obtain the intelligently processed data. The specific process is as follows: First, preprocessing is performed based on the original concrete accelerator production data. The specific implementation process is as follows: The original concrete accelerator production data is cleaned to obtain cleaned data, and the cleaned data is processed for outliers using statistical analysis methods (such as Z score, IQR method) to obtain outlier processed data; the outlier processed data is processed for data synchronization using time series analysis method to obtain synchronized data, i.e. preprocessed data; the technologies used in the preprocessing process are all existing technologies and will not be elaborated here; Furthermore, the preprocessed data is subjected to feature extraction using existing feature engineering techniques to obtain a feature set; In order to solve the problem of feature redundancy and collinearity, a redundant feature elimination algorithm is introduced to update the feature set; the specific implementation process is as follows: The first step is to calculate the correlation between the features in the feature set; For the feature set extracted using feature engineering technology, calculate the mean and standard deviation, and then construct the correlation coefficient matrix: Calculate features and Correlation coefficient : , in, It is The mean of the features, indicating The average value of the features reflects the The center position of a feature; It is The feature of eigenvalues; The number of eigenvalues representing the feature; It is The standard deviation of the feature; It is The feature of eigenvalues; It is The standard deviation of the feature; It is The mean of the features, indicating The average value of the features; according to the correlation coefficient Get the correlation coefficient matrix ; The second step is to screen collinear features and obtain the screened feature set; Setting correlation coefficient threshold based on expert experience , used to determine the collinearity between features; for the correlation coefficient matrix To traverse, for The characteristic pairs ( , ), retain one of the features and delete the redundant features by defining a feature selection function. The specific implementation formula of the feature selection function is as follows: , in, It is a feature selection function used to determine whether to retain the feature; Select function based on features The output of retains non-redundant features to form a filtered feature set : , The third step is to use the filtered feature set to update the feature set: Based on the filtered feature set , recalculate the correlation coefficient matrix between the features in the filtered feature set to ensure that the filtered feature set The features in the filter are highly independent, and the collinearity problem is effectively solved. Output and update the feature set for subsequent feature extraction and analysis.
[0021] Further, the updated feature set is subjected to data enhancement processing using data enhancement technology to obtain enhanced data; the data enhancement technology is an existing technology and will not be described in detail here; Finally, the enhanced data, updated feature set and preprocessed data are merged to obtain intelligently processed data. .
[0022] The data after intelligent processing is analyzed in real time using the existing pre-trained neural network model, and then the robustness enhancement function is introduced to perform robustness enhancement processing to enhance the resistance of the existing pre-trained neural network model to abnormal data, and to optimize the existing pre-trained neural network model to obtain the optimized pre-trained neural network model, and output the optimized pre-trained neural network model as the analysis result; the analysis result may include quality indicator analysis results, trend analysis results and anomaly detection results. The specific implementation is as follows: , in, It is data after intelligent processing; is the output of an existing pre-trained neural network model; and It is to adjust the parameters to control the intensity of nonlinear transformation and ensure that the optimized pre-trained neural network model can flexibly adjust its response to abnormal data; , It is a weight matrix used for nonlinear transformation, mapping the output of the existing pre-trained neural network model to a higher-dimensional space, and improving the expressiveness of the optimized pre-trained neural network model; is the weight matrix; is the output of the robustness enhancement function, that is, the output of the optimized pre-trained neural network model, which represents the analysis result obtained after multiple layers of nonlinear transformation; Furthermore, the optimized pre-trained neural network model is parameter optimized; by optimizing the parameters , and the weight matrix , , , improve the robustness of the optimized pre-trained neural network model, and the optimization objective function is: , Use gradient descent method to optimize parameters: , , , , , in, is the learning rate, which controls the step size during the gradient descent process and ensures the stability and effectiveness of parameter updates; , , It is a regularization parameter used to control the size of the weight matrix to prevent overfitting of the optimized pre-trained neural network model; It is the optimization objective function, which measures the error between the output of the optimized pre-trained neural network model and the expected result, and contains regularization terms to prevent overfitting; , , , , It is Parameters after optimization; , , , , It is Parameters after optimization.
[0023] The present invention uses a redundant feature elimination algorithm to calculate the correlation between features in a feature set during the intelligent processing of original concrete accelerator production data, solves feature redundancy and collinearity problems, and ensures the independence and effectiveness of the feature set; uses an existing pre-trained neural network model to perform real-time analysis on the data after intelligent processing to ensure the accuracy and timeliness of the analysis; introduces a robustness enhancement function to perform robustness enhancement processing to enhance the resistance of the existing pre-trained neural network model to abnormal data, optimizes the existing pre-trained neural network model, and outputs the optimized pre-trained neural network model as an analysis result; and further improves the performance of the optimized pre-trained neural network model by optimizing parameters and weight matrices, prevents the optimized pre-trained neural network model from overfitting, and ensures the stability and reliability of the analysis result.
[0024] S2. Conduct quality inspection on concrete accelerating agent based on the analysis results to obtain inspection results; obtain early warning information and adjustment suggestions in the production process based on the inspection results; generate production parameter adjustment instructions based on the adjustment suggestions to adjust the production parameters.
[0025] Based on the analysis results, the concrete accelerator is automatically tested using a comprehensive error analysis detection algorithm to obtain the test results. The specific implementation process is as follows: Firstly, concrete accelerator samples are collected from the production line, and the concrete accelerator samples are tested using a testing instrument to obtain actual test values; based on the actual test values, an actual test value matrix is constructed; Secondly, based on the quality index analysis results in the analysis results and the actual detection value matrix, the detection error matrix is calculated. The specific formula is as follows: , in, is the actual detection value matrix; It is the quality indicator analysis result in the analysis result; is the detection error matrix; Furthermore, based on the detection error matrix , calculate the comprehensive error score , the specific formula is as follows: , in, and is the weight coefficient; is the comprehensive error score; is the number of concrete accelerator samples; is the number of detection indicators; is the first Concrete Accelerator Sample No. The error value of each detection indicator; It is Anomaly detection scores of concrete accelerator samples; It is The trend value of each detection indicator; Finally, according to the comprehensive error score , judge whether the quality of concrete accelerator meets the standard and obtain the test result. Set the threshold value according to the expert experience method. If the comprehensive error score is lower than the threshold value, the quality is considered qualified; otherwise, issue an early warning message, and determine the level of the early warning message according to the industry standard and the expert experience method, and then find the corresponding adjustment suggestion from the adjustment suggestion rule library according to the level of the early warning message, wherein the adjustment suggestion rule library is constructed according to the expert experience method and the historical data in the existing database; generate specific production parameter adjustment instructions based on the adjustment suggestions; the instruction format should meet the control requirements of the production equipment to ensure that it can be correctly identified and executed by the production equipment, and transmit the production parameter adjustment instructions to the relevant production equipment through the standard communication protocol. After the production equipment receives the production parameter adjustment instructions, it automatically executes the production parameter adjustment; the adjustment content includes temperature control, humidity adjustment, and additive ratio adjustment, and finally realizes the production quality detection of concrete accelerator.
[0026] The present invention introduces a comprehensive error analysis detection algorithm, calculates the detection error between the actual detection value matrix and the quality index analysis result in the analysis result, and introduces a comprehensive error scoring mechanism to conduct a comprehensive analysis of various detection indicators, thereby ensuring the accuracy and comprehensiveness of the detection results, effectively evaluating the quality of the concrete accelerator, and judging whether it meets the standards. This method can comprehensively consider factors such as detection error, trend value, and abnormal detection score, and improve the comprehensiveness and reliability of quality detection.
[0027] In summary, a concrete accelerating agent production quality detection system and method are completed.
[0028] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0029] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.
[0030] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.
Claims
1. A method for detecting the production quality of a concrete accelerating agent, characterized in that: The following steps are involved: S1. Real-time collection of original concrete accelerator production data, intelligent processing of the original concrete accelerator production data, and obtaining intelligently processed data; The existing pre-trained neural network model is used to perform real-time analysis on the intelligently processed data, and the robustness enhancement function is introduced to optimize the pre-trained neural network model to obtain the optimized pre-trained neural network model. The output of the optimized pre-trained neural network model is used as the analysis result, which includes the quality indicator analysis result, trend analysis result and anomaly detection result; among them, the specific formula of the robustness enhancement function is: ， in, It is data after intelligent processing; is the output of an existing pre-trained neural network model; and is an adjustment parameter used to control the strength of the nonlinear transformation; , , are all weight matrices; is the result of the analysis; S2. Based on the analysis results, the concrete accelerator is subjected to quality inspection using a comprehensive error analysis detection algorithm; in the process of implementing the comprehensive error analysis detection algorithm, a concrete accelerator sample is collected and tested to obtain an actual detection value; based on the actual detection value, an actual detection value matrix is constructed; based on the quality index analysis results in the analysis results and the actual detection value matrix, a detection error matrix is calculated; based on the detection error matrix, a comprehensive error scoring mechanism is introduced to obtain a comprehensive error score; the specific formula for the comprehensive error score is: ， in, and is the weight coefficient; is the comprehensive error score; is the number of concrete accelerator samples; is the number of detection indicators; is the first Concrete Accelerator Sample No. The error value of each detection indicator; It is Anomaly detection scores of concrete accelerator samples; It is The trend value of each detection indicator; Based on the comprehensive error score, the test results are obtained; based on the test results, early warning information and adjustment suggestions in the production process are obtained; based on the adjustment suggestions, production parameter adjustment instructions are generated to adjust the production parameters.
2. A method for detecting the production quality of a concrete accelerating agent according to claim 1, characterized in that: The S1 specifically includes: In the process of realizing intelligent processing, the original concrete accelerator production data is preprocessed to obtain preprocessed data; the preprocessed data is subjected to feature extraction to obtain a feature set; and a redundant feature elimination algorithm is introduced to update the feature set.
3. A method for detecting the production quality of a concrete accelerating agent according to claim 2, characterized in that: The S1 specifically includes: The specific implementation steps of the redundant feature elimination algorithm are: the first step is to calculate the correlation between the features in the feature set; the second step is to screen the collinear features to obtain the screened feature set; the third step is to use the screened feature set to update the feature set.
4. A method for detecting the production quality of a concrete accelerating agent according to claim 3, characterized in that: The S1 specifically includes: Perform data enhancement processing on the updated feature set to obtain enhanced data; merge the enhanced data, the updated feature set and the pre-processed data to obtain intelligently processed data.
5. A method for detecting the production quality of a concrete accelerating agent according to claim 1, characterized in that: The S1 specifically includes: The optimization objective function is introduced to optimize the parameters of the optimized pre-trained neural network model.
6. A concrete accelerating agent production quality detection system, applied to a concrete accelerating agent production quality detection method as claimed in claim 1, characterized in that: Includes the following sections: Data acquisition module, data intelligent processing module, intelligent analysis module, automatic detection module, intelligent control module; The data acquisition module collects the production data of the original concrete accelerating agent in real time and sends the production data of the original concrete accelerating agent to the data intelligent processing module; The data intelligent processing module pre-processes the original concrete accelerating agent production data to obtain pre-processed data; Extract features from the preprocessed data to obtain a feature set; introduce Redundant feature elimination algorithm is used to update the feature set; data enhancement processing is performed on the updated feature set to obtain enhanced data; The enhanced processed data, the updated feature set and the pre-processed data are merged to obtain the intelligently processed data; Send the intelligently processed data to the intelligent analysis module; The intelligent analysis module performs real-time analysis on the intelligently processed data and performs robustness enhancement processing to obtain analysis results; Send the analysis results to the automated detection module; The automated testing module performs quality testing on the concrete accelerating agent based on the analysis results to obtain the test results; Send the test results to the intelligent control module; The intelligent control module obtains early warning information and adjustment suggestions in the production process based on the detection results, generates production parameter adjustment instructions based on the adjustment suggestions, and adjusts the production parameters according to the production parameter adjustment instructions.