Intelligent mix proportion design method for beam field

Abstract

The invention relates to the technical field of big data artificial intelligence, and discloses a beam field intelligent mix proportion design method which comprises the following steps: S1, collecting historical production mix proportion data of various raw materials and concrete used in a beam field, transmitting the historical production mix proportion data to an intelligent mix proportion design system in real time, and establishing a raw material and concrete mix proportion database; s2, according to a bridge design drawing, inputting structural design requirements of a beam body into an intelligent mix proportion design system, and preliminarily determining a design target of a concrete mix proportion through system analysis; s3, performing optimization calculation on the concrete mix proportion through an intelligent algorithm; and S4, performing laboratory trial mixing by utilizing the optimized mix proportion, manufacturing a concrete test piece and testing the performance, performing comparative analysis on a test result and a design target by the intelligent mix proportion design system, and automatically adjusting mix proportion parameters until an optimal production mix proportion is output. Through intelligent mix proportion design, the dosage and proportion of the raw materials can be more accurately controlled.

Description

Technical Field

[0001] The present invention relates to the technical field of big data artificial intelligence, and specifically relates to an intelligent mix design method for beam yards. Background Art

[0002] At present, the mix design of precast concrete in beam yards still adopts traditional methods, based on the Bolomy formula, and uses the mass method or volume method to design the concrete mix according to experience, with little use of modern computer means. However, due to the many influencing factors of concrete performance, the traditional concrete mix design method still has problems of large experimental workload and long time consumption in the actual application process, and can no longer fully meet the requirements of modern concrete design, production and application. Therefore, it is necessary to explore new, fast and efficient or more intelligent design methods. Summary of the Invention

[0003] Aiming at the deficiencies of the prior art, the technical solution adopted by the present invention to solve its technical problems is: an intelligent mix design method for beam yards, including the following steps:

[0004] S1. Collect various raw materials used in the beam yard and the historical production mix data of concrete, and transmit them to the intelligent mix design system in real time to establish a raw material and concrete mix database;

[0005] S2. According to the bridge design drawings, input the structural design requirements of the beam body into the intelligent mix design system, and initially determine the design target of the concrete mix through system analysis;

[0006] S3. Optimize and calculate the concrete mix through intelligent algorithms;

[0007] S4. Use the optimized mix for laboratory trial mixing, make concrete specimens and test their performance. The intelligent mix design system compares the test results with the design target for analysis, and automatically adjusts the mix parameters according to the deviation situation until the optimal production mix is output.

[0008] Preferably, the data collection in step S1 includes the physical properties, chemical composition and mechanical properties of raw materials, which are comprehensively collected by intelligent detection equipment and sensors, and the quality stability and variability of raw materials are evaluated through data analysis software.

[0009] Preferably, the evaluation of the quality stability and variability of the raw materials in step S1 is to statistically analyze the strength data of multiple batches of cement to determine its strength fluctuation range, providing a basic basis for mix design.

[0010] Preferably, the intelligent mix design system described in step S1 includes a data management module, a mix calculation module, an intelligent optimization module, a data analysis module, a report generation module, and a database management module. The intelligent optimization module, by means of intelligent algorithms and big data analysis, analyzes a large amount of historical data and actual engineering cases, considers the quality fluctuations of raw materials and environmental factors, and automatically adjusts the mix parameters.

[0011] Preferably, the structural design requirements of the beam body described in step S2 include the structural type, size, concrete strength grade requirements, durability requirements, and construction process characteristics.

[0012] Preferably, the design objectives of the concrete mix include the target slump range, the target compressive strength value, and the minimum cement dosage limit.

[0013] Preferably, the intelligent algorithms described in step S3 include genetic algorithms and neural network algorithms. The algorithms use the raw material data and structural design requirements as constraint conditions and the concrete performance indicators as the objective function for multi-parameter optimization. The optimization process also considers the compatibility between cement and admixtures and the rationality of aggregate gradation.

[0014] Preferably, the genetic algorithm continuously screens out better mix combinations through "crossover" and "mutation" operations on different mix ratio schemes.

[0015] Preferably, the neural network algorithm uses MATLAB software to train the BP neural network model and uses the trained model to predict the concrete compressive strength. The input variables of the neural network are eight factors affecting the concrete compressive strength, and the concrete compressive strength value is the result of the output layer.

[0016] Preferably, the test performances described in step S4 include compressive strength tests, impermeability tests, and elastic modulus tests.

[0017] The beneficial effects of the present invention are as follows:

[0018] 1. Through intelligent mix design, the usage and proportion of raw materials can be more accurately controlled, reducing the interference of human factors, so that the performance of concrete is more stable and reliable. In large-scale beam yard production, it can ensure that the concrete quality of each beam body meets the design requirements, improving the overall quality of the bridge structure.

[0019] 2. The optimized mix can reasonably utilize raw materials on the premise of meeting the design requirements, reduce the usage of high-cost materials such as cement, lower the production cost. At the same time, since the number of trial mixes and the generation of unqualified products are reduced, the test cost and material waste are also indirectly reduced.

[0020] 3. Intelligent algorithms and automated data processing have significantly shortened the cycle of mix proportion design. The process that previously required a large amount of manual calculations and trial mixes can now be completed in a shorter time, enabling rapid response to the progress requirements of engineering construction and improving the production efficiency of the beam yard.

[0021] 4. In the future construction of intelligent transportation infrastructure, the intelligent mix proportion design technology for beam yards can also be combined with the bridge health monitoring system. According to the actual stress conditions and performance changes of the bridge during use, the mix proportion of subsequent beam bodies can be dynamically adjusted and optimized, further enhancing the safety and durability of the bridge structure and promoting the intelligent development process of bridge engineering construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is the flow chart of the present invention;

[0023] Figure 2 is the structural block diagram of the intelligent mix proportion design system in the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0024] The present invention will be further described in detail below in conjunction with the drawings and specific embodiments:

[0025] As Figure 1 shown, a method for intelligent mix proportion design of a beam yard includes the following steps:

[0026] S1. Collect the historical production mix proportion data of various raw materials (such as cement, aggregates, admixtures, mineral admixtures, etc.) used in the beam yard and concrete, and transmit them to the intelligent mix proportion design system in real time. Establish a raw material and concrete mix proportion database, and evaluate the quality stability and variability of the raw materials through data analysis software (MATLAB). For example, by statistically analyzing the strength data of multiple batches of cement, determine its strength fluctuation range, providing a basic basis for mix proportion design;

[0027] S2. According to the bridge design drawings, input the structural design requirements of the beam body into the intelligent mix proportion design system, and preliminarily determine the design objectives of the concrete mix proportion through system analysis;

[0028] S3. Optimize and calculate the concrete mix proportion through intelligent algorithms;

[0029] S4. Use the optimized mix proportion for laboratory trial mixes, make concrete specimens and cure them according to standard curing conditions, and then conduct various performance tests on the specimens (compressive strength test, impermeability test, and elastic modulus test). The intelligent mix proportion design system compares and analyzes the test results with the design objectives. If it is found that there is a deviation between the actual performance and the target, the system will automatically adjust the mix proportion parameters according to the deviation situation, conduct trial mixes and verification again until the optimal production mix proportion is output.

[0030] In step S1, data collection includes the physical properties of raw materials (such as particle size distribution, density, specific surface area, etc.), chemical compositions (such as the chemical components of cement, the chemical structure of admixtures, etc.), and mechanical properties (such as the compressive strength of cement, the crushing value of aggregates, etc.). By establishing a large database of raw materials and concrete mix proportions, storing historical data of various raw materials, different mix proportion schemes and their corresponding concrete performance data, etc., through data mining techniques, potential laws between the properties of raw materials and concrete performance can be discovered from the database, providing reference for the mix proportion design of new projects; for example, when encountering a new cement brand or aggregate origin, the usage of similar raw materials in previous mix proportions can be queried through the database to quickly determine the preliminary mix proportion scheme and reduce the number of trial mixes.

[0031] In the raw material inspection link and the concrete mixing process, intelligent detection equipment (such as FBT-9A full-automatic specific surface area analyzer, Ca-5A free calcium oxide analyzer, ZBSX-92A shock-type standard vibrating sieve machine and other test instruments) and sensors (weight sensors, temperature and humidity sensors, pressure sensors) are used for comprehensive collection. Laser particle size analyzers are used to detect the particle size distribution of aggregates, and automatic pressure testing machines are used to detect the strength of cement, etc. These devices can automatically collect data and transmit it to the system, improving the detection efficiency and data accuracy. The sensors will monitor parameters such as the slump, temperature, and mixing time of concrete in real time, and timely feedback the changes in the workability of concrete, so as to fine-tune the mix proportion during the mixing process to ensure the quality stability of concrete.

[0032] Such as Figure 2As shown in the figure, the intelligent mix design system includes a data management module, a mix ratio calculation module, an intelligent optimization module, a data analysis module, a report generation module, and a database management module. Among them, the data management module can conveniently input, edit, query, and delete the basic information of raw materials, such as the variety, specification, performance index, etc. of cement, sand and gravel, admixtures, etc.; the mix ratio calculation module automatically calculates the preliminary mix ratio according to the input raw material information and design requirements, in accordance with relevant standards and specifications, such as calculating the dosage ratio of materials such as cement, water, sand, and stone based on the strength grade and durability requirements of concrete; the intelligent optimization module uses intelligent algorithms and big data analysis to automatically adjust the mix ratio parameters by analyzing a large amount of historical data and actual engineering cases, considering the quality fluctuations of raw materials and environmental factors; report generation function: it can automatically generate a detailed mix design report, including design basis, raw material information, mix ratio calculation process, performance prediction results, etc., and can be output and printed in a standardized format for technicians to review and file; the data analysis module is used to analyze various data generated during the production process of the beam yard, such as the compressive strength test data of concrete, the inspection data of raw materials, etc. Through data analysis, the rationality and accuracy of the mix design can be evaluated, potential problems and laws can be discovered, and a basis for further optimizing the mix ratio can be provided. Common data analysis software includes Excel, SPSS, SAS, etc., which can perform operations such as statistical analysis, chart drawing, and correlation analysis of data; the database management module is responsible for managing and organizing a large amount of mix design data, raw material data, test data, etc. The database management system can achieve efficient storage, retrieval, update, and sharing of data, ensuring the integrity and consistency of data. For example, by establishing a database, it is convenient to query the quality information of raw materials in different batches and the usage effects of the corresponding mix ratios, providing a reference for new mix designs.

[0033] To further improve the intelligent mix design system, it is equipped with a conventional operating system, a network communication part, and auxiliary equipment. The operating system includes Windows, Linux, etc.; the network communication part includes a network interface and communication protocols. The network interface can be used to connect to the information systems of raw material suppliers to obtain the latest raw material quality data, and can also conduct data interaction with the production management system and quality inspection system of the beam yard to achieve information sharing and collaborative work. The communication protocols ensure accurate and reliable data transmission and communication between different devices and systems. Common communication protocols include TCP / IP, HTTP, FTP, etc. Through these protocols, the system can send, receive, and process data with other devices to achieve functions such as remote monitoring and data synchronization; the auxiliary equipment includes a data acquisition device interface, which is used to connect various data acquisition devices, such as sensors, testing machines, etc., to collect the physical and chemical property data of raw materials in real time, as well as the mechanical property data of concrete test blocks, etc., and transmit the data to the computer system to provide accurate and timely data support for mix design and optimization. At the same time, to ensure the security and confidentiality of mix design data, prevent data leakage and tampering, a data encryption device may be equipped to encrypt the data so that it exists in ciphertext form during transmission and storage, and only authorized users and devices can decrypt and use the data.

[0034] In step S2, the structural design requirements of the beam body include the structural type (such as box girder, T-beam, etc.), dimensions (length, width, height, cross-sectional shape, etc.), concrete strength grade requirements, durability requirements (such as impermeability, frost resistance, etc.), and construction process characteristics (such as pouring method, vibration conditions, etc.); the design objectives of the concrete mix include the target slump range, target compressive strength value, and minimum cement dosage limit.

[0035] In step S3, the intelligent algorithms include genetic algorithms and neural network algorithms. The algorithms use the raw material data and structural design requirements as constraint conditions and the concrete performance indicators (such as strength, durability indicators, etc.) as the objective function for multi-parameter optimization. The optimization process also considers the compatibility between cement and admixtures and the rationality of aggregate gradation; among them, the genetic algorithm simulates the biological evolution process and continuously screens out better mix combinations through "crossover" and "mutation" operations on different mix ratio schemes.

[0036] The neural network algorithm uses MATLAB software to train the BP neural network model. The BP training algorithm is an error backpropagation algorithm based on the gradient descent method and is mostly used for training multi-layer neural networks. A complete BP network consists of an input layer, a hidden layer, and an output layer. The first layer of the network is the input layer of variables, the last layer is called the output layer of predicted variables, and all the layers from the second layer to the second-to-last layer in the middle are called hidden layers. In a neural network, each neuron receives all the outputs of the neurons in the previous layer and obtains the output of this neuron through weight parameters and activation functions. Specifically, as follows:

[0037]

[0038] In the formula: ω i represents the weight parameter corresponding to the i-th input x i of this neuron, b represents the bias of this neuron, and f(x) is the activation function of the neuron. The commonly used activation function is the Sigmoid activation function

[0039]

[0040] Use the trained model to predict the concrete compressive strength. The input variables of the neural network are eight factors that affect the concrete compressive strength, and the concrete compressive strength value is the result of the output layer.

[0041] Using concrete microstructure simulation software, the microstructure of concrete with different mix ratios can be simulated and analyzed to predict the cement hydration process, the structure of the interfacial transition zone between aggregates and cement paste, etc., so as to understand the influence of mix ratio parameters on the macroscopic properties of concrete from the microscopic level. For example, by simulating the distribution and growth of cement hydration products under different water-cement ratios, the optimal range of water-cement ratio can be determined to improve the strength and durability of concrete. At the same time, the simulation analysis of the concrete pouring process can also be carried out to predict the flow state and filling effect of concrete in the beam formwork, providing a reference basis for construction technology in mix ratio design.

[0042] Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art and related fields without creative efforts shall fall within the protection scope of the present invention. The structures, devices, and operation methods not specifically described and explained in the present invention shall be implemented according to the conventional means in the art without special instructions and limitations.

Claims

1. A beam yard intelligent mix design method, characterized in that: The following steps are involved: S1. Collect historical production mix data of various raw materials and concrete used in the beam yard, transmit them to the intelligent mix design system in real time, and establish a raw material and concrete mix database; S2. According to the bridge design drawings, the structural design requirements of the beam body are input into the intelligent mix design system, and the design target of the concrete mix is ​​preliminarily determined through system analysis; S3, optimize the concrete mix ratio through intelligent algorithm; S4. Use the optimized mix ratio to conduct laboratory trials, make concrete specimens and test performance. The intelligent mix ratio design system compares and analyzes the test results with the design goals, and automatically adjusts the mix ratio parameters according to the deviation until the optimal production mix ratio is output.

2. The beam field intelligent mix design method according to claim 1 is characterized by: The data collected in step S1 includes the physical properties, chemical composition and mechanical properties of the raw materials, which are comprehensively collected using intelligent detection equipment and sensors, and the quality stability and variability of the raw materials are evaluated through data analysis software.

3. The beam field intelligent mix design method according to claim 1 is characterized by: The evaluation of the quality stability and variability of the raw materials in step S1 is performed by statistically analyzing the strength data of multiple batches of cement to determine the strength fluctuation range, thereby providing a basic basis for mix design.

4. The beam field intelligent mix design method according to claim 1 is characterized by: The intelligent mix design system described in step S1 includes a data management module, a mix calculation module, an intelligent optimization module, a data analysis module, a report generation module, and a database management module. The intelligent optimization module uses intelligent algorithms and big data analysis to analyze a large amount of historical data and actual engineering cases, taking into account quality fluctuations of raw materials and environmental factors, and automatically adjusts the mix parameters.

5. The beam field intelligent mix design method according to claim 1 is characterized by: The structural design requirements of the beam body in step S2 include structural type, size, concrete strength grade requirements, durability requirements and construction process characteristics.

6. The beam field intelligent mix design method according to claim 5 is characterized by: The design objectives of the concrete mix ratio include a target slump range, a target compressive strength value, and a minimum cement dosage limit.

7. The beam field intelligent mix design method according to claim 4 is characterized by: The intelligent algorithm described in step S3 includes a genetic algorithm and a neural network algorithm. The algorithm uses raw material data and structural design requirements as constraints and concrete performance indicators as objective functions to perform multi-parameter optimization. The optimization process also considers the compatibility of cement and admixtures and the rationality of aggregate grading.

8. The beam field intelligent mix design method according to claim 7 is characterized by: The genetic algorithm continuously screens out better mix ratio combinations through "crossover" and "mutation" operations on different mix ratio schemes.

9. The beam field intelligent mix design method according to claim 7, characterized in that: The neural network algorithm uses MATLAB software to train the BP neural network model, and uses the trained model to predict the compressive strength of concrete. The input variables of the neural network are eight factors that affect the compressive strength of concrete, and the compressive strength value of concrete is the output layer result.

10. The beam field intelligent mix design method according to claim 1, characterized in that: The test performance in step S4 includes a compressive strength test, a permeability test and an elastic modulus test.

Images