Cement strength prediction method and system based on multi-source data analysis

Abstract

The invention discloses a cement strength prediction method and system based on multi-source data analysis, and relates to the technical field of building material production. Data from different sources are integrated through a data acquisition and processing module, various production related data in the prefabricated part production process are monitored in real time, a cement strength value Snqz, a production difference value Sccz and an actual difference value Sjcz are calculated and evaluated, production process parameters are adjusted in time, and it is ensured that the production process is transparent and controllable; and the feedback adjustment module formulates a specific production adjustment strategy according to the evaluation result of the production difference value through secondary verification, so that abnormal conditions occurring in the production process can be solved in a targeted manner, and the stability and consistency of the cement strength are ensured. The real-time prediction and control module can dynamically respond to changes in the production process by continuously monitoring and adjusting production parameters, timely optimize the parameters and improve the production efficiency and the product quality so as to promote reduction of human intervention in the production process and improve the automation level.

Description

Technical Field

[0001] The present invention relates to the technical field of building material production, and in particular to a cement strength prediction method and system based on multi-source data analysis. Background Art

[0002] The cement strength prediction system uses big data and artificial intelligence technology, combined with historical data and production parameters, to make real-time predictions of cement strength. Through data collection, data preprocessing, model training and optimization, the system can achieve precise control of the cement production process, improve product quality, reduce resource waste and production costs, and thus enhance corporate competitiveness.

[0003] Although the existing cement strength prediction system has achieved certain results in improving production efficiency and product quality, due to the wide application of cement strength prediction system, including prefabricated component manufacturing; however, the current cement strength prediction system still has some shortcomings when applied to prefabricated component manufacturing:

[0004] 1. Incomplete data collection: Due to unreasonable sensor layout and insufficient number, key production parameters cannot be fully covered, resulting in insufficient accuracy of the prediction model.

[0005] 2. The prediction results are unstable. Some prediction models have poor generalization ability when dealing with data from different batches and under different environmental conditions, resulting in unstable prediction results. Summary of the invention

[0006] Technical issues solved

[0007] In view of the deficiencies in the prior art, the present invention provides a cement strength prediction method and system based on multi-source data analysis, which solves the technical problems of incomplete data collection and unstable prediction results mentioned in the background technology.

[0008] Technical Solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions: a cement strength prediction system based on multi-source data analysis, including a data acquisition and processing module, a prediction model building module, a real-time prediction and control module, a feedback adjustment module and a data storage and management module;

[0010] The data acquisition and processing module is used to integrate data from different sources during the production of prefabricated components, including sensor data, historical production data and data uploaded by users; and to pre-process the integrated data;

[0011] The prediction model building module is used to build a special model for predicting the cement strength of prefabricated components, and to evaluate and optimize the model performance through a cross-validation method;

[0012] The real-time prediction and control module is used to monitor various production-related data in the prefabricated component production process in real time, and to make a real-time prediction of cement strength in combination with the prediction model; to calculate and evaluate the production difference Sccz of the cement strength of the current prefabricated component, and finally to adjust the production process parameters according to the evaluation results;

[0013] The feedback adjustment module is used to combine the actual cement data and cement strength test results in the prefabricated component production process, and after adjusting the production process parameters, perform a secondary verification on the cement strength of the prefabricated component; finally, adjust the prediction model parameters and model algorithm based on the results of the secondary verification;

[0014] The data storage and management module is used to store and manage multi-source data and establish an efficient database system; it also provides data query and analysis tools to verify decision analysis during the production process.

[0015] Preferably, the data acquisition and processing module includes a data acquisition unit and a data processing unit; the data acquisition unit adjusts the sensor network during the deployment of sensors to collect data related to temperature, humidity, ingredient ratio and mixing time; secondly, a collection and storage mechanism for historical production data is established to extract relevant data from past production records; secondly, the user uploads production data through the set API interface, including laboratory test data and manually recorded data; finally, data synchronization technology is used to integrate data from different sources into a unified data platform.

[0016] Preferably, the data processing unit is used to preprocess the integrated data. First, the collected data is cleaned to remove noise and outliers; second, the data format is standardized and unified, and data from different sources are converted into a consistent format and unit; then, feature engineering methods are used to extract and construct features of the data, and key features are selected and reduced in dimension; finally, all data are dimensionlessly processed, and a production data set and an environmental data set are constructed.

[0017] Preferably, the prediction model building module collects and organizes the data used for modeling through the data acquisition and processing module, including real-time production data, historical production data and data uploaded by users; secondly, the support vector machine algorithm is selected for modeling, and an initial prediction model is established based on the processed production data set; thirdly, the data set is divided into a training set and a validation set using a cross-validation method, and the prediction performance of the model is evaluated through repeated experiments, and the accuracy, precision and recall of the model are calculated; then, the model is adjusted according to the cross-validation results, including parameter tuning, feature selection and algorithm improvement; finally, the optimized model is used to verify the actual production data of prefabricated components.

[0018] Preferably, the real-time prediction and control module includes a real-time prediction unit and a control unit; the real-time prediction unit is used to monitor the real-time production-related data of cement in real time, extract relevant data from the production data set and input it into the prediction model, perform real-time data analysis, and calculate the predicted strength value Yqdz of cement, then fit the predicted strength value Yqdz with the preset target strength value Mqdz of cement, and calculate the current production difference Sccz of cement strength, and the specific calculation formula is as follows;

[0019]

[0020] Sccz=Yqdz(t)-Mqdz(t);

[0021] Where Wd i represents the temperature of the i-th sample, Sd i represents the humidity of the i-th sample, Plb i Indicates the ingredient ratio of the i-th sample, Jsj i represents the stirring time of the i-th sample, Gsj i represents the curing time of the i-th sample;

[0022] t represents the time point, Yqdz(t) represents the cement strength prediction value provided by the prediction model at the time point t; Mqdz(t represents the cement strength target value set at the time point t in the production plan;

[0023] The real-time prediction unit compares the production difference Sccz by preset production difference threshold Q1 and production difference threshold Q2, and the production difference threshold Q1>production difference threshold Q2. The specific evaluation content is as follows:

[0024] If the production difference Sccz ≥ the production difference threshold Q1, it means that the production process is abnormal, and the first difference result is generated;

[0025] If the production difference threshold Q1> the production difference Sccz> the production difference threshold Q2, it means that the production process is normal, and the second difference result is generated;

[0026] If the production difference threshold Q2 ≥ the production difference Sccz, it indicates that the production process is abnormal, and a third difference result is generated.

[0027] Preferably, the control unit adjusts specific production process parameters based on the first difference result and the third difference result, including temperature control, humidity control and ingredient ratio adjustment; secondly, sends an adjustment instruction to the production equipment to implement corresponding process parameter adjustment operations, and the specific adjustment operations are:

[0028] If the production difference Sccz is evaluated as the first difference result, reduce the amount of cement and other ingredients by 10%, shorten the mixing time by 5%, reduce the curing temperature by 5%, or reduce the humidity by 3%;

[0029] If the production difference Sccz is evaluated as the third difference result, then increase the amount of cement and other ingredients by 10%, extend the mixing time by 5%, increase the curing temperature by 5%, and increase or decrease the humidity by 3%;

[0030] Finally, the adjusted production data is continuously monitored and collected, and the adjusted production data is transmitted to the feedback adjustment module.

[0031] Preferably, the feedback adjustment module includes a real-time data acquisition unit, a secondary verification unit and an adjustment unit; the real-time data acquisition unit is used to collect real-time cement production data in actual production, including water-cement ratio, concrete vibration time and mixing uniformity, by deploying sensors on the production line and manually recording by staff during the production of prefabricated components, and upload the collected data to the secondary verification unit;

[0032] The secondary verification unit extracts the environment-related data from the environmental data set, fits it with the data uploaded by the real-time data acquisition unit, and obtains the cement strength value Snqz by calculation. The specific acquisition formula is as follows:

[0033]

[0034] In the formula, Hjzs represents the environmental index, Qy represents the air pressure, and Fhl represents the dust content;

[0035] Z s Indicates the quality of water, Z n Indicates the quality of cement, represents water-cement ratio; Hnt represents concrete vibration time, and Hjy represents mixing uniformity;

[0036] And K is a constant, obtained by fitting the experimental data; e is the base of the natural logarithm; a, b, c and d are the weight values ​​of water-cement ratio, concrete vibration time Hnt, mixing uniformity Hjy and environmental index Hjzs respectively;

[0037] Where a+b+c+d=1, 0<a≤1, 0<b≤1, 0<c≤1, 0<d≤1;

[0038] The secondary verification unit performs secondary verification by comparing the predicted strength value Yqdz with the cement strength value Snqz. The specific comparison content is as follows:

[0039] Sjcz=Yqdz-Snqz

[0040] Where Sjcz represents the actual difference value;

[0041] By comparing the preset actual difference threshold with the actual difference value Sjcz, and the preset actual difference threshold is equal to 1, the following results are generated:

[0042] If the actual difference value Sjcz=0, a first verification result is generated;

[0043] If the actual difference value Sjcz≠0 and is between ±1, a second verification result is generated;

[0044] If the actual difference value Sjcz≠0 and exceeds the range of ±1, a third verification result is generated.

[0045] Preferably, the adjustment unit is used to adjust the production process parameters and the parameters and algorithms of the prediction model according to the secondary verification results; specifically including:

[0046] When the first verification result is generated, it means that the actual production process is consistent with the predicted result; at this time, the existing production process parameters are maintained unchanged, and the existing model and its parameters are maintained;

[0047] When the second verification result is generated, it means that there is a deviation, but it is within an acceptable range; at this time, according to the specific deviation direction, the production parameters and the control parameters of the maintenance environment are adjusted, or according to the deviation data, the weight parameters in the model are adjusted or the model is corrected;

[0048] When the third verification result is generated, there is a deviation, and the deviation is abnormal; at this time, the production parameters are reset and the control parameters of the maintenance environment are adjusted; the prediction model is retrained, the model algorithm is optimized, and the performance of different models is evaluated through cross-validation methods, and the hyperparameters of the model are adjusted.

[0049] Preferably, the data storage and management module includes a storage management unit, a database establishment unit and a data query analysis unit; the storage management unit is used to receive and store data from different sources, including sensor data, historical production data and user uploaded data; then the stored data is classified, labeled and archived, and the storage efficiency is adjusted through data compression and optimization algorithms;

[0050] The database establishment unit is used to build a database system that supports rapid retrieval and processing of large-scale data; and adopts a distributed database architecture; the data query and analysis unit is used to provide data query tools, build a multi-condition combination query mechanism, and integrate data analysis functions for verifying and optimizing decisions in the production process.

[0051] A cement strength prediction method based on multi-source data analysis comprises the following steps:

[0052] Step 1: Integrate data from different sources during the production of prefabricated components through a data acquisition and processing module, including sensor data, historical production data, and data uploaded by users; and pre-process the integrated data;

[0053] Step 2: Establish a special model for predicting cement strength of prefabricated components through the prediction model building module, and evaluate and optimize the model performance through the cross-validation method;

[0054] Step 3: Monitor various production-related data in the prefabricated component production process in real time through the real-time prediction and control module, and make real-time prediction of cement strength in combination with the prediction model; calculate and evaluate the production difference Sccz of the cement strength of the current prefabricated component, and finally adjust the production process parameters according to the evaluation results;

[0055] Step 4: After adjusting the production process parameters, the cement strength of the prefabricated components is re-verified by combining the actual cement data and cement strength test results in the prefabricated component production process through the feedback adjustment module; finally, the prediction model parameters and model algorithm are adjusted based on the results of the secondary verification;

[0056] Step 5: Store and manage multi-source data through the data storage and management module, and establish an efficient database system; at the same time, provide data query and analysis tools to verify decision analysis in the production process.

[0057] Beneficial Effects

[0058] The present invention provides a communication network fault early warning method and system based on artificial intelligence, which has the following beneficial effects:

[0059] (1) This method and system for predicting cement strength based on multi-source data analysis integrates data from different sources in the prefabricated component production process through a data acquisition and processing module, and performs preprocessing, feature extraction and dimensionality reduction on these data to ensure high data quality and consistency. This comprehensive data acquisition and processing method solves the problem of incomplete data and provides a reliable data basis for subsequent predictions. The prediction model establishment module adopts a support vector machine algorithm and uses a cross-validation method to evaluate and optimize the model to ensure the prediction performance of the model. By continuously adjusting parameters, selecting features and improving algorithms, the accuracy, precision and recall rate of the model are improved. This method ensures the stability and reliability of the prediction results. The real-time prediction and control module monitors the data in the production process in real time and combines the prediction model for real-time prediction, calculates the production difference Sccz of cement strength, and adjusts the production process parameters in a timely manner. This real-time monitoring and control mechanism ensures the transparency and controllability of the production process, timely discovers and corrects abnormal situations in production, and improves product quality.

[0060] (2) A cement strength prediction method and system based on multi-source data analysis, wherein the feedback adjustment module further adjusts the production process parameters and model parameters by secondary verification of actual production data and prediction data, thereby ensuring continuous improvement and optimization of the model. This closed-loop feedback mechanism can effectively solve the deviation problem in the production process, ensure the accuracy of the prediction results and the rationality of the production process; the data storage and management module constructs an efficient database system and adopts a distributed database architecture to achieve rapid retrieval and processing of large-scale data, and provides powerful data query and analysis tools. This efficient data storage and management method ensures the security and accessibility of the data, supports decision-making analysis in the production process, and improves the level of intelligent production management. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A schematic diagram of the framework structure of a cement strength prediction system based on multi-source data analysis according to the present invention;

[0062] Figure 2 The present invention is a schematic diagram of the steps of a cement strength prediction method based on multi-source data analysis. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] Example 1

[0065] See also Figure 1 , a cement strength prediction system based on multi-source data analysis, including a data acquisition and processing module, a prediction model building module, a real-time prediction and control module, a feedback adjustment module and a data storage and management module;

[0066] The data acquisition and processing module is used to integrate data from different sources during the production of prefabricated components, including sensor data, historical production data and data uploaded by users; and to pre-process the integrated data;

[0067] The prediction model building module is used to build a special model for predicting the cement strength of prefabricated components, and to evaluate and optimize the model performance through a cross-validation method;

[0068] The real-time prediction and control module is used to monitor various production-related data in the prefabricated component production process in real time, and to make a real-time prediction of cement strength in combination with the prediction model; to calculate and evaluate the production difference Sccz of the cement strength of the current prefabricated component, and finally to adjust the production process parameters according to the evaluation results;

[0069] The feedback adjustment module is used to combine the actual cement data and cement strength test results in the prefabricated component production process, and after adjusting the production process parameters, perform a secondary verification on the cement strength of the prefabricated component; finally, adjust the prediction model parameters and model algorithm based on the results of the secondary verification;

[0070] The data storage and management module is used to store and manage multi-source data and establish an efficient database system; it also provides data query and analysis tools to verify decision analysis during the production process.

[0071] In this embodiment, the data acquisition and processing module effectively integrates multi-source data from sensors, historical production records, and user uploads to ensure the comprehensiveness and consistency of the data. Through the preprocessing steps, the data quality is improved, providing a solid foundation for subsequent analysis and model building;

[0072] The prediction model building module evaluates and optimizes model performance through cross-validation methods to ensure the accuracy and reliability of the prediction model; the cross-validation method can identify and reduce the over-fitting and under-fitting problems of the model, improve the generalization ability of the model, and adapt to different production batches and environmental conditions;

[0073] The real-time prediction and control module realizes the real-time monitoring of various parameters in the production process, and combines the prediction model to make real-time predictions of cement strength. By calculating and evaluating the production difference Sccz, the system can quickly identify deviations in production and adjust the production process parameters according to the evaluation results to ensure that the cement strength of prefabricated components always meets the expected standards.

[0074] After adjusting the production process parameters, the feedback adjustment module conducts secondary verification of the cement strength of prefabricated components based on actual cement data and cement strength test results. Based on the results of the secondary verification, the prediction model parameters and algorithms are further optimized to continuously improve the prediction accuracy and adaptability of the system.

[0075] The data storage and management module achieves storage and management of multi-source data by establishing an efficient database system. The data query and analysis tools it provides can verify and optimize decisions in the production process, ensure data security and availability, and support scientific decision-making.

[0076] Example 2

[0077] See also Figure 1, the data acquisition and processing module includes a data acquisition unit and a data processing unit; the data acquisition unit adjusts the sensor network in the process of deploying sensors to collect temperature, humidity, ingredient ratio and mixing time related data; secondly, a collection and storage mechanism for historical production data is established to extract relevant data from past production records; secondly, users upload production data through the set API interface, including laboratory test data and manually recorded data; finally, data synchronization technology is used to integrate data from different sources into a unified data platform.

[0078] In this embodiment, by optimizing the deployment of the sensor network, the data acquisition unit can collect parameters in the production process in real time and accurately. In addition, the systematic historical data collection mechanism and the user-friendly API interface ensure the comprehensive integration of historical production data and user-uploaded data. The application of data synchronization technology ensures the real-time and efficient integration of multi-source data to form a unified data platform. The data quality and analysis efficiency are improved, providing reliable data support for the subsequent cement strength prediction model.

[0079] Example 3

[0080] See also Figure 1 The data processing unit is used to pre-process the integrated data. First, the collected data is cleaned to remove noise and outliers; second, the data format is standardized and unified, and data from different sources are converted into consistent formats and units; then, feature engineering methods are used to extract and construct features of the data, select and reduce the dimension of key features; finally, all data are dimensionlessly processed, and production data sets and environmental data sets are constructed.

[0081] In this embodiment, the data processing unit performs comprehensive preprocessing on the integrated data through a series of detailed steps and methods; removes noise and outliers through data cleaning to ensure data quality; standardizes data formats and unifies units to ensure that data from different sources are processed under a consistent format and unit system; extracts and selects key features through feature engineering methods, and uses dimensionality reduction technology to reduce model complexity; dimensionless processing ensures that all data are compared at the same scale; finally, constructs production data sets and environmental data sets to ensure data continuity, comprehensiveness and consistency, providing high-quality data support for subsequent cement strength prediction models, and significantly improving the prediction accuracy and reliability of the system.

[0082] Example 4

[0083] See also Figure 1The prediction model building module collects and organizes the data used for modeling through the data acquisition and processing module, including real-time production data, historical production data and data uploaded by users; secondly, the support vector machine algorithm is selected for modeling, and an initial prediction model is established based on the processed production data set; thirdly, the data set is divided into a training set and a validation set by a cross-validation method, and the prediction performance of the model is evaluated through repeated experiments, and the accuracy, precision and recall of the model are calculated; then, the model is adjusted according to the cross-validation results, including parameter tuning, feature selection and algorithm improvement; finally, the optimized model is used to verify the actual production data of prefabricated components.

[0084] In this embodiment, the prediction model building module selects the support vector machine algorithm (SVM) for modeling, and establishes an initial prediction model based on the processed production data set; then, the data set is divided into a training set and a validation set by a cross-validation method, and multiple repeated experiments are performed to evaluate the prediction performance of the model, and the accuracy, precision and recall of the model are calculated; according to the cross-validation results, the model is adjusted, including selecting appropriate kernel functions and penalty parameters, screening out features that have a significant impact on the prediction effect, and using kernel techniques or combining other machine learning algorithms; finally, the optimized model is used to verify the actual production data of prefabricated components to ensure the applicability and stability of the model in a real production environment; this process not only improves the prediction accuracy and generalization ability of the model, but also effectively guides actual production to ensure that the cement strength of prefabricated components meets the expected standards.

[0085] Example 5

[0086] See also Figure 1 , the real-time prediction and control module includes a real-time prediction unit and a control unit; the real-time prediction unit is used to monitor the real-time production-related data of cement in real time, extract relevant data from the production data set and input it into the prediction model, perform real-time data analysis, and calculate the predicted strength value Yqdz of cement, then fit the predicted strength value Yqdz with the preset target strength value Mqdz of cement, and calculate the current production difference Sccz of cement strength. The specific calculation formula is as follows;

[0087]

[0088] Sccz=Yqdz(t)-Mqdz(t);

[0089] Where Wd i represents the temperature of the i-th sample, Sd i represents the humidity of the i-th sample, Plb i Indicates the ingredient ratio of the i-th sample, Jsj i represents the stirring time of the i-th sample, Gsj irepresents the curing time of the i-th sample;

[0090] t represents the time point, Yqdz(t) represents the cement strength prediction value provided by the prediction model at the time point t; Mqdz(t represents the cement strength target value set at the time point t in the production plan;

[0091] The real-time prediction unit compares the production difference Sccz by preset production difference threshold Q1 and production difference threshold Q2, and the production difference threshold Q1>production difference threshold Q2. The specific evaluation content is as follows:

[0092] If the production difference Sccz ≥ the production difference threshold Q1, it means that the production process is abnormal, and the first difference result is generated;

[0093] If the production difference threshold Q1> the production difference Sccz> the production difference threshold Q2, it means that the production process is normal, and the second difference result is generated;

[0094] If the production difference threshold Q2 ≥ the production difference Sccz, it indicates that the production process is abnormal, and a third difference result is generated.

[0095] The control unit adjusts specific production process parameters based on the first difference result and the third difference result, including temperature control, humidity control and ingredient ratio adjustment; secondly, sends an adjustment instruction to the production equipment to implement corresponding process parameter adjustment operations, and the specific adjustment operations are:

[0096] If the production difference Sccz is evaluated as the first difference result, reduce the amount of cement and other ingredients by 10%, shorten the mixing time by 5%, reduce the curing temperature by 5%, or reduce the humidity by 3%;

[0097] If the production difference Sccz is evaluated as the third difference result, then increase the amount of cement and other ingredients by 10%, extend the mixing time by 5%, increase the curing temperature by 5%, and increase or decrease the humidity by 3%;

[0098] Finally, the adjusted production data is continuously monitored and collected, and the adjusted production data is transmitted to the feedback adjustment module.

[0099] In this embodiment, the temperature Wd directly affects the hydration reaction speed and strength development process of cement. Within an appropriate temperature range, the strength of cement can be well developed, while too high or too low a temperature may lead to insufficient strength or uneven hardening. Changes in the temperature Wd lead to inaccurate prediction of cement strength, which must be monitored and adjusted in real time to ensure that the strength meets the standard.

[0100] Humidity Sd affects the hardening and final strength of cement. Too low humidity leads to too fast evaporation of water, affecting the hydration reaction; too high humidity leads to surface water absorption, affecting the strength. Controlling humidity Sd is to ensure that cement solidifies under appropriate humidity conditions and avoid strength fluctuations caused by changes in ambient humidity Sd.

[0101] The batching ratio Plb determines the mixing ratio of cement and other materials, and has a direct impact on the final cement strength. The correct batching ratio Plb can ensure that the strength and performance of cement meet expectations; accurate control of the batching ratio Plb is the key to ensuring the consistency of cement strength and quality stability. Any deviation may lead to insufficient or excessive strength.

[0102] The mixing time Jsj affects the mixing uniformity of cement and other materials. A mixing time Jsj that is too short may result in uneven mixing, while a mixing time Jsj that is too long may affect the properties of the materials. An appropriate mixing time Jsj ensures a uniform mixture, thereby ensuring the consistency and reliability of cement strength.

[0103] The curing time Gsj refers to the time from when cement is formed to when it is completely hardened. Appropriate curing time Gsj can ensure that the cement reaches its designed strength. Different curing times Gsj will affect the ultimate strength and durability of the cement. Controlling the curing time Gsj is to ensure that the cement has enough time to undergo hydration reaction during the curing process, thereby achieving the designed strength requirements.

[0104] The solution of real-time monitoring and adjustment of production process parameters not only improves the accuracy of cement strength prediction, but also ensures the stability and consistency of the production process, reduces the error of human intervention, improves the accuracy and reliability of production, and ensures that the cement strength of prefabricated components meets the expected standards. Through the comprehensive operation of continuous monitoring and feedback adjustment modules, the system can continuously optimize the prediction model and production process, further improving production efficiency and product quality.

[0105] Example 6

[0106] The feedback adjustment module includes a real-time data acquisition unit, a secondary verification unit and an adjustment unit; the real-time data acquisition unit is used to collect real-time cement production data in actual production, including water-cement ratio, concrete vibration time and mixing uniformity, by deploying sensors on the production line and manually recording by staff during the prefabricated component production process, and upload the collected data to the secondary verification unit;

[0107] The secondary verification unit extracts the environment-related data from the environmental data set, fits it with the data uploaded by the real-time data acquisition unit, and obtains the cement strength value Snqz by calculation. The specific acquisition formula is as follows:

[0108]

[0109] In the formula, Hjzs represents the environmental index, Qy represents the air pressure, and Fhl represents the dust content;

[0110] Z s Indicates the quality of water, Z n Indicates the quality of cement, represents water-cement ratio; Hnt represents concrete vibration time, and Hjy represents mixing uniformity;

[0111] And K is a constant, obtained by fitting the experimental data; e is the base of the natural logarithm; a, b, c and d are the weight values ​​of water-cement ratio, concrete vibration time Hnt, mixing uniformity Hjy and environmental index Hjzs respectively;

[0112] Where a+b+c+d=1, 0<a≤1, 0<b≤1, 0<c≤1, 0<d≤1;

[0113] The secondary verification unit performs secondary verification by comparing the predicted strength value Yqdz with the cement strength value Snqz. The specific comparison content is as follows:

[0114] Sjcz=Yqdz-Snqz

[0115] Where Sjcz represents the actual difference value;

[0116] By comparing the preset actual difference threshold with the actual difference value Sjcz, and the preset actual difference threshold is equal to 1, the following results are generated:

[0117] If the actual difference value Sjcz=0, a first verification result is generated;

[0118] If the actual difference value Sjcz≠0 and is between ±1, a second verification result is generated;

[0119] If the actual difference value Sjcz≠0 and exceeds the range of ±1, a third verification result is generated.

[0120] The adjustment unit is used to adjust the production process parameters and the parameters and algorithms of the prediction model according to the secondary verification results; specifically including:

[0121] When the first verification result is generated, it means that the actual production process is consistent with the predicted result; at this time, the existing production process parameters are maintained unchanged, and the existing model and its parameters are maintained;

[0122] When the second verification result is generated, it means that there is a deviation, but it is within an acceptable range; at this time, according to the specific deviation direction, the production parameters and the control parameters of the maintenance environment are adjusted, or according to the deviation data, the weight parameters in the model are adjusted or the model is corrected;

[0123] When the third verification result is generated, there is a deviation, and the deviation is abnormal; at this time, the production parameters are reset and the control parameters of the maintenance environment are adjusted; the prediction model is retrained, the model algorithm is optimized, and the performance of different models is evaluated through cross-validation methods, and the hyperparameters of the model are adjusted.

[0124] In this embodiment, the air pressure Qy has a certain influence on the hydration reaction and curing process of cement. Different air pressure conditions may affect the evaporation rate of water and the hardening rate of concrete; the dust content Fhl refers to the dust concentration in the air in the production environment. A high dust environment may affect the bonding performance of cement and aggregate, and thus affect the strength;

[0125] Water-cement ratio It refers to the ratio of the weight of water to the weight of cement. It has an important influence on the workability and strength of cement. Too high a water-cement ratio may lead to insufficient strength of concrete, while too low a water-cement ratio may affect the workability of concrete and make it difficult to construct. The correct water-cement ratio It can ensure the appropriate fluidity and final strength of cement paste; concrete vibration time Hnt refers to the time used to compact the concrete using a vibrator during the pouring process. Appropriate vibration time can eliminate bubbles in the concrete, increase density, and thus improve strength and durability. Too short a vibration time will cause voids inside the concrete, affecting strength. Too long a vibration time may cause stratification or segregation, affecting overall quality. Mixing uniformity Hjy refers to the uniformity of materials such as cement, sand, gravel and water during the mixing process. Uniform mixing can ensure that the composition of each part of the concrete is consistent, ensuring strength and durability. Uneven mixing may lead to insufficient strength in local areas or waste of materials, affecting the overall quality of prefabricated components.

[0126] Example 7

[0127] The data storage and management module includes a storage management unit, a database establishment unit and a data query and analysis unit; the storage management unit is used to receive and store data from different sources, including sensor data, historical production data and user uploaded data; then the stored data is classified, labeled and archived, and the storage efficiency is adjusted through data compression and optimization algorithms;

[0128] The database establishment unit is used to build a database system that supports rapid retrieval and processing of large-scale data; and adopts a distributed database architecture; the data query and analysis unit is used to provide data query tools, build a multi-condition combination query mechanism, and integrate data analysis functions for verifying and optimizing decisions in the production process.

[0129] In this embodiment, the storage management unit adopts advanced data compression and optimization algorithms to improve storage efficiency and data access speed; the database establishment unit is used to build a database system that supports rapid retrieval and processing of large-scale data; by adopting a distributed database architecture, high availability and high scalability of data are ensured, and efficient processing of big data is achieved through load balancing and data sharding technology; the data query and analysis unit is used to provide powerful data query tools and build a multi-condition combination query mechanism, so that users can flexibly query according to different needs; in addition, advanced data analysis functions are integrated to conduct in-depth analysis of various data in the production process and provide decision support. Through real-time monitoring of production data and comprehensive analysis of historical data, production decisions can be verified and optimized, thereby improving production efficiency and product quality; through the integration of these functions, multi-source data can be effectively managed and utilized, and the overall performance and intelligence level of the system can be improved.

[0130] Example 8

[0131] See also Figure 2 , a cement strength prediction method based on multi-source data analysis, comprising the following steps:

[0132] Step 1: Integrate data from different sources during the production of prefabricated components through a data acquisition and processing module, including sensor data, historical production data, and data uploaded by users; and pre-process the integrated data;

[0133] Step 2: Establish a special model for predicting cement strength of prefabricated components through the prediction model building module, and evaluate and optimize the model performance through the cross-validation method;

[0134] Step 3: Monitor various production-related data in the prefabricated component production process in real time through the real-time prediction and control module, and make real-time prediction of cement strength in combination with the prediction model; calculate and evaluate the production difference Sccz of the cement strength of the current prefabricated component, and finally adjust the production process parameters according to the evaluation results;

[0135] Step 4: After adjusting the production process parameters, the cement strength of the prefabricated components is re-verified by combining the actual cement data and cement strength test results in the prefabricated component production process through the feedback adjustment module; finally, the prediction model parameters and model algorithm are adjusted based on the results of the secondary verification;

[0136] Step 5: Store and manage multi-source data through the data storage and management module, and establish an efficient database system; at the same time, provide data query and analysis tools to verify decision analysis in the production process.

[0137] In this embodiment, the data from different sources during the production of prefabricated components are integrated in step 1 to ensure the comprehensiveness and diversity of the data. The integrated data is preprocessed to improve the data quality, remove noise and outliers, and make subsequent analysis and modeling more accurate and reliable; a special model for predicting the cement strength of prefabricated components is established in step 2 to accurately predict the cement strength. The model performance is evaluated and optimized by the cross-validation method to improve the generalization ability and prediction accuracy of the model, and to ensure the stability and reliability of the model under different batches and different environmental conditions; various production-related data in the production process of prefabricated components are monitored in real time in step 3, and the cement strength is predicted in real time in combination with the prediction model to ensure the transparency and controllability of the production process. The production difference Sccz of the cement strength of the current prefabricated component is calculated and evaluated, and the production process parameters are adjusted according to the evaluation results to optimize the production process and improve product quality and production efficiency; the adjustment of the production process parameters is verified twice in combination with the actual cement data and cement strength test results in the production process of prefabricated components in step 4 to ensure the effectiveness and accuracy of the adjustment. Through the results of the secondary verification, the prediction model parameters and model algorithms are further adjusted to continuously improve the model performance, prediction accuracy and adaptability; through step 5, multi-source data is stored and managed to establish an efficient database system to ensure data security and accessibility. Provide data query and analysis tools to support decision-making analysis in the production process, improve the scientificity and accuracy of decision-making through comprehensive analysis and verification of data, and enhance the intelligent level of production management.

[0138] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A cement strength prediction system based on multi-source data analysis, Features: It includes data acquisition and processing module, prediction model building module, real-time prediction and control module, feedback adjustment module and data storage and management module; The data acquisition and processing module is used to integrate data from different sources during the production of prefabricated components, including sensor data, historical production data and data uploaded by users; and to pre-process the integrated data; The prediction model building module is used to build a special model for predicting the cement strength of prefabricated components, and to evaluate and optimize the model performance through a cross-validation method; The real-time prediction and control module is used to monitor various production-related data in the prefabricated component production process in real time, and to perform real-time prediction of cement strength in combination with the prediction model; Calculate and evaluate the production difference Sccz of the cement strength of the current prefabricated components, and finally adjust the production process parameters according to the evaluation results; The feedback adjustment module is used to combine the actual cement data and cement strength test results in the prefabricated component production process, and after adjusting the production process parameters, perform a secondary verification on the cement strength of the prefabricated component; finally, adjust the prediction model parameters and model algorithm based on the results of the secondary verification; The data storage and management module is used to store and manage multi-source data and establish an efficient database system; it also provides data query and analysis tools to verify decision analysis during the production process.

2. A cement strength prediction system based on multi-source data analysis according to claim 1, Features: The data acquisition and processing module includes a data acquisition unit and a data processing unit; the data acquisition unit adjusts the sensor network in the process of deploying sensors to collect data related to temperature, humidity, ingredient ratio and mixing time; secondly, a collection and storage mechanism for historical production data is established to extract relevant data from past production records; Secondly, users upload production data, including laboratory test data and manually recorded data, through the set API interface; finally, data synchronization technology is used to integrate data from different sources into a unified data platform.

3. A cement strength prediction system based on multi-source data analysis according to claim 2, Features: The data processing unit is used to pre-process the integrated data. First, the collected data is cleaned to remove noise and outliers; Secondly, standardize and unify the data format, convert data from different sources into consistent formats and units; then use feature engineering methods to extract and construct features of the data, select and reduce the dimension of key features; finally, all data are dimensionlessly processed, and production data sets and environmental data sets are constructed.

4. A cement strength prediction system based on multi-source data analysis according to claim 3, Features: The prediction model building module collects and organizes the data used for modeling through the data acquisition and processing module, including real-time production data, historical production data and data uploaded by users; secondly, the support vector machine algorithm is selected for modeling, and an initial prediction model is established based on the processed production data set; thirdly, the data set is divided into a training set and a validation set by a cross-validation method, and the prediction performance of the model is evaluated through repeated experiments, and the accuracy, precision and recall rate of the model are calculated; then, the model is adjusted according to the cross-validation results, including parameter tuning, feature selection and algorithm improvement; finally, the optimized model is used to verify the actual production data of prefabricated components.

5. A cement strength prediction system based on multi-source data analysis according to claim 4, Features: The real-time prediction and control module includes a real-time prediction unit and a control unit; the real-time prediction unit is used to monitor the real-time production-related data of cement in real time, extract relevant data from the production data set and input it into the prediction model, perform real-time data analysis, and calculate the predicted strength value Yqdz of cement, then fit the predicted strength value Yqdz with the preset target strength value Mqdz of cement, and calculate the current production difference Sccz of cement strength. The specific calculation formula is as follows; ScCz=Yqdz(t)-Mqdz(t); Where Wd i represents the temperature of the i-th sample, Sd i represents the humidity of the i-th sample, Plb i Indicates the ingredient ratio of the i-th sample, Jsj i represents the stirring time of the i-th sample, Gsj i represents the curing time of the i-th sample; t represents the time point, Yqdz(t) represents the cement strength prediction value provided by the prediction model at the time point t; Mqdz(t represents the cement strength target value set at the time point t in the production plan; The real-time prediction unit compares the production difference Sccz by preset production difference threshold Q1 and production difference threshold Q2, and the production difference threshold Q1>production difference threshold Q2. The specific evaluation content is as follows: If the production difference Sccz ≥ the production difference threshold Q1, it means that the production process is abnormal, and the first difference result is generated; If the production difference threshold Q1> the production difference Sccz> the production difference threshold Q2, it means that the production process is normal, and the second difference result is generated; If the production difference threshold Q2 ≥ the production difference Sccz, it indicates that the production process is abnormal, and a third difference result is generated.

6. A cement strength prediction system based on multi-source data analysis according to claim 5, Features: The control unit adjusts specific production process parameters based on the first difference result and the third difference result, including temperature control, humidity control and ingredient ratio adjustment; secondly, sends adjustment instructions to the production equipment to implement corresponding process parameter adjustment operations, and the specific adjustment operations are: If the production difference Sccz is evaluated as the first difference result, reduce the amount of cement and other ingredients by 10%, shorten the mixing time by 5%, reduce the curing temperature by 5%, or reduce the humidity by 3%; If the production difference Sccz is evaluated as the third difference result, then increase the amount of cement and other ingredients by 10%, extend the mixing time by 5%, increase the curing temperature by 5%, and increase or decrease the humidity by 3%; Finally, the adjusted production data is continuously monitored and collected, and the adjusted production data is transmitted to the feedback adjustment module.

7. A cement strength prediction system based on multi-source data analysis according to claim 6, Features: The feedback adjustment module includes a real-time data acquisition unit, a secondary verification unit and an adjustment unit; the real-time data acquisition unit is used to collect real-time cement production data in actual production, including water-cement ratio, concrete vibration time and mixing uniformity, by deploying sensors on the production line and manually recording by staff during the prefabricated component production process, and upload the collected data to the secondary verification unit; The secondary verification unit extracts the environment-related data from the environmental data set, fits it with the data uploaded by the real-time data acquisition unit, and obtains the cement strength value Snqz by calculation. The specific acquisition formula is as follows: In the formula, Hjzs represents the environmental index, Qy represents the air pressure, and Fhl represents the dust content; Z s Indicates the quality of water, Z n Indicates the quality of cement, represents water-cement ratio; Hnt represents concrete vibration time, and Hjy represents mixing uniformity; And K is a constant, obtained by fitting the experimental data; e is the base of the natural logarithm; a, b, c and d are the weight values ​​of water-cement ratio, concrete vibration time Hnt, mixing uniformity Hjy and environmental index Hjzs respectively; Where a+b+c+d=1, 0<a≤1, 0<b≤1, 0<c≤1, 0<d≤1; The secondary verification unit performs secondary verification by comparing the predicted strength value Yqdz with the cement strength value Snqz. The specific comparison content is as follows: Sjcz=Yqdz-Snqz Where Sjcz represents the actual difference value; By comparing the preset actual difference threshold with the actual difference value Sjcz, and the preset actual difference threshold is equal to 1, the following results are generated: If the actual difference value Sjcz=0, a first verification result is generated; If the actual difference value Sjcz≠0 and is between ±1, a second verification result is generated; If the actual difference value Sjcz≠0 and exceeds the range of ±1, a third verification result is generated.

8. A cement strength prediction system based on multi-source data analysis according to claim 7, Features: The adjustment unit is used to adjust the production process parameters and the parameters and algorithms of the prediction model according to the secondary verification results; specifically including: When the first verification result is generated, it means that the actual production process is consistent with the predicted result; at this time, the existing production process parameters are maintained unchanged, and the existing model and its parameters are maintained; When the second verification result is generated, it means that there is a deviation, but it is within an acceptable range; at this time, according to the specific deviation direction, the production parameters and the control parameters of the maintenance environment are adjusted, or according to the deviation data, the weight parameters in the model are adjusted or the model is corrected; When the third verification result is generated, there is a deviation, and the deviation is abnormal; at this time, the production parameters are reset and the control parameters of the maintenance environment are adjusted; the prediction model is retrained, the model algorithm is optimized, and the performance of different models is evaluated through cross-validation methods, and the hyperparameters of the model are adjusted.

9. A cement strength prediction system based on multi-source data analysis according to claim 1, Features: The data storage and management module includes a storage management unit, a database establishment unit and a data query and analysis unit; the storage management unit is used to receive and store data from different sources, including sensor data, historical production data and user uploaded data; then the stored data is classified, labeled and archived, and the storage efficiency is adjusted through data compression and optimization algorithms; The database establishment unit is used to build a database system that supports rapid retrieval and processing of large-scale data; and adopts a distributed database architecture; the data query and analysis unit is used to provide data query tools, build a multi-condition combination query mechanism, and integrate data analysis functions for verifying and optimizing decisions in the production process.

10. A cement strength prediction method based on multi-source data analysis, applied to a cement strength prediction system based on multi-source data analysis as claimed in any one of claims 1 to 9, Features: The following steps are involved: Step 1: Integrate data from different sources during the production of prefabricated components through a data acquisition and processing module, including sensor data, historical production data, and data uploaded by users; and pre-process the integrated data; Step 2: Establish a special model for predicting cement strength of prefabricated components through the prediction model building module, and evaluate and optimize the model performance through the cross-validation method; Step 3: Monitor various production-related data in the prefabricated component production process in real time through the real-time prediction and control module, and make real-time prediction of cement strength in combination with the prediction model; Calculate and evaluate the production difference Sccz of the cement strength of the current prefabricated components, and finally adjust the production process parameters according to the evaluation results; Step 4: After adjusting the production process parameters, the cement strength of the prefabricated components is re-verified by combining the actual cement data and cement strength test results in the prefabricated component production process through the feedback adjustment module; finally, the prediction model parameters and model algorithm are adjusted based on the results of the secondary verification; Step 5: Store and manage multi-source data through the data storage and management module, and establish an efficient database system; at the same time, provide data query and analysis tools to verify decision analysis in the production process.

Images