Prediction method, device and equipment for predicting compressive strength of medium and high strength concrete based on GA-BP neural network model and storage medium
By optimizing the GA-BP neural network model using a genetic algorithm, the problems of insufficient accuracy of the traditional water-cement ratio model and poor convergence stability of the BP neural network are solved. This enables efficient and accurate prediction of the compressive strength of medium- and high-strength concrete, simplifies the model structure, and improves computational efficiency and applicability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GANSU ARCHITECTURE DESIGN ACAD CO LTD
- Filing Date
- 2026-01-23
- Publication Date
- 2026-05-08
AI Technical Summary
In existing technologies, traditional cement-water ratio models have insufficient prediction accuracy under complex mix proportions, BP neural networks are prone to getting trapped in local optima and have poor convergence stability, and GA-BP neural network models have complex structures and low computational efficiency, making it difficult to meet the needs of modern engineering for refined control of concrete material properties.
A genetic algorithm was used to optimize the GA-BP neural network model. The quadratic error criterion function was used as the objective function. The optimization process of weights and thresholds was simplified by using the genetic algorithm for initialization, selection, crossover, mutation, and individual evaluation steps. The GA-BP neural network model was then constructed for predicting the compressive strength of medium and high strength concrete.
It significantly improves the model's nonlinear fitting ability and convergence stability, with 88% of samples having a prediction error within 2MPa, meeting engineering accuracy requirements. It simplifies the program structure, improves computational efficiency, and is suitable for strength prediction under different mix proportions.
Smart Images
Figure CN121997737A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of concrete strength prediction technology, and in particular to a method, device, equipment and storage medium for predicting the compressive strength of medium and high strength concrete based on a GA-BP neural network model. Background Technology
[0002] High-strength concrete, due to its excellent mechanical properties, is increasingly widely used in high-rise buildings, long-span structures, and complex engineering projects. The compressive strength of concrete is a key parameter for evaluating its mechanical properties and engineering quality, directly affecting the safety and reliability of the structure. Therefore, establishing an accurate and reliable predictive model for concrete compressive strength is of great significance for guiding concrete mix design and optimizing construction techniques.
[0003] Currently, concrete mix design mainly follows the "Specification for Mix Design of Ordinary Concrete" (JGJ55-2019), which is based on the empirical relationship between the cement-water ratio and compressive strength. However, concrete strength is complexly affected by various components such as cement, fly ash, slag, water, water-reducing agents, sand, and aggregate, as well as their interactions. When the components change irregularly, linear models based on the cement-water ratio often fail to accurately reflect the strength variation, resulting in significant prediction deviations and affecting the accuracy and reliability of engineering design.
[0004] In recent years, artificial neural networks have been introduced into the field of concrete strength prediction due to their powerful nonlinear fitting capabilities. For example, backpropagation (BP) neural networks have been used to predict the strength of medium- and low-strength concrete. However, BP neural networks use a gradient-based backpropagation algorithm for optimization, which is prone to getting stuck in local optima during training, has a slow convergence speed, and is sensitive to initial weights and thresholds, resulting in insufficient model stability and large fluctuations in prediction results.
[0005] To overcome the aforementioned problems, researchers attempted to introduce genetic algorithms to optimize the initial parameters of neural networks, such as the GA-BP neural network used in existing technologies to predict the compressive strength of planted porous concrete. However, this model still requires secondary optimization of weights and thresholds during training, resulting in a complex program structure and room for improvement in computational efficiency.
[0006] Therefore, it is necessary to provide a more efficient, stable and accurate method for predicting the compressive strength of concrete to better meet the needs of modern engineering for refined control of concrete material properties. Summary of the Invention
[0007] This application provides a method, apparatus, equipment, and storage medium for predicting the compressive strength of medium-high strength concrete based on a GA-BP neural network model, in order to solve the following problems existing in the prior art: 1) The traditional ash-water ratio model has the problem of insufficient prediction accuracy and large error under complex mix proportion conditions; 2) Backpropagation (BP) neural networks are prone to getting stuck in local optima and have poor convergence stability during training; 3) Existing GA-BP neural network models suffer from complex structures, cumbersome optimization steps, and low computational efficiency.
[0008] Firstly, this application provides a method for predicting the compressive strength of medium-high strength concrete based on a GA-BP neural network model, including: Multiple sets of medium- and high-strength concrete test samples with different mix proportions were obtained. The high-strength concrete test samples included concrete composition parameters and corresponding measured values of compressive strength. The concrete composition parameters included the amount of cement, slag, fly ash, water, water-reducing agent, stone, and sand. Constructing a GA-BP neural network model to predict the compressive strength of high-strength concrete, including: Set the number of input layer nodes M, the number of hidden layer nodes q, and the number of output layer nodes L. The number of input layer nodes M is consistent with the number of types of concrete composition parameters. Using the quadratic error criterion function as the objective function, the optimal solutions for weights and thresholds are obtained through the initialization, selection, crossover, mutation, and individual evaluation steps of the genetic algorithm, thus constructing a GA-BP neural network model. After normalizing the high-strength concrete test samples, they were input into the constructed GA-BP neural network model to complete the model training. The composition parameters of the high-strength concrete to be predicted are normalized and then input into the trained GA-BP neural network model, which outputs the predicted value of the concrete compressive strength.
[0009] In one possible design, the first hidden layer i Input of each node Calculated using the following formula: (1) In the formula, This is the input of the j-th node in the input layer, namely the normalized value of the concrete composition parameters; j For node sequence number, M The number of nodes; For the hidden layer i The node to the input layer j The weights between nodes; For the hidden layer i Threshold for each node; The first hidden layer i Output of each node Calculated using the following formula: (2) In the formula, This represents the activation function of the hidden layer.
[0010] In one possible design, the output layer is... k Output of each node , which is the normalized value of the predicted concrete compressive strength, is calculated using the following formula: (3) In the formula, This represents the activation function of the output layer; Indicates the output layer number k The node to the hidden layer i The weights between nodes; Indicates the output layer number k Threshold for each node.
[0011] In one possible design, the quadratic error criterion function is expressed as: (4) In the formula, It is a quadratic error criterion function; The target value for the sample; This represents the number of nodes in the output layer.
[0012] In one possible design, the normalization process for the high-strength concrete test samples is as follows: (5) In the formula, These are the normalized values of the concrete composition parameters. These are the original values of the concrete composition parameters. The minimum value of the concrete composition parameters among all test samples. This represents the maximum value of the concrete composition parameters across all test samples.
[0013] In one possible design, the activation functions of the hidden layer and the output layer are both logarithmic S-functions.
[0014] In one possible design, the weights of the input layer and the hidden layer, the weights of the hidden layer and the output layer, the threshold of the hidden layer, and the threshold of the output layer satisfy the following: the weights of the input layer and the hidden layer are a 7-row, 30-column matrix, where the 7 rows correspond to the input nodes of cement, slag, fly ash, water, water-reducing agent, stone, and sand, respectively; the weights of the hidden layer and the output layer are a 1-row, 30-column matrix; and the threshold of the hidden layer is a 1-row, 30-column matrix.
[0015] Secondly, this application provides a device for predicting the compressive strength of medium-high strength concrete based on a GA-BP neural network model, the device comprising: The data acquisition module is configured to acquire multiple sets of medium- and high-strength concrete test samples with different mix proportions. The high-strength concrete test samples include concrete composition parameters and corresponding measured values of compressive strength. The concrete composition parameters include cement content, slag content, fly ash content, water content, water-reducing agent content, stone content, and sand content. The model building module is configured to build a GA-BP neural network model to predict the compressive strength of high-strength concrete, including: Set the number of input layer nodes M, the number of hidden layer nodes q, and the number of output layer nodes L. The number of input layer nodes M is consistent with the number of types of concrete composition parameters. Using the quadratic error criterion function as the objective function, the optimal solutions for weights and thresholds are obtained through the initialization, selection, crossover, mutation, and individual evaluation steps of the genetic algorithm, thus constructing a GA-BP neural network model. After normalizing the high-strength concrete test samples, they were input into the constructed GA-BP neural network model to complete the model training. The composition parameters of the high-strength concrete to be predicted are normalized and then input into the trained GA-BP neural network model, which outputs the predicted value of the concrete compressive strength.
[0016] Thirdly, embodiments of this application provide an electronic device, including: at least one processor and a memory; the memory stores computer execution instructions; the at least one processor executes the computer execution instructions stored in the memory, causing the at least one processor to perform the method for predicting the compressive strength of high-strength concrete based on the GA-BP neural network model as described in the first aspect and various possible designs of the first aspect.
[0017] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions. When a processor executes the computer-executable instructions, it implements the method for predicting the compressive strength of high-strength concrete based on the GA-BP neural network model as described in the first aspect and various possible designs of the first aspect.
[0018] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the method for predicting the compressive strength of high-strength concrete based on the GA-BP neural network model as described in the first aspect and various possible designs of the first aspect.
[0019] The method, apparatus, equipment, and storage medium for predicting the compressive strength of medium-high strength concrete based on the GA-BP neural network model provided in this application have at least the following beneficial effects: (1) This application uses a genetic algorithm to optimize the initial weights and thresholds of the neural network, and combines multi-component mix proportion data to significantly improve the nonlinear fitting ability of the model. After verification with 425 sets of samples, the prediction error of about 88% of the samples is within 2MPa, which meets the accuracy requirements of conventional engineering and is far superior to the traditional ash-water ratio model (root mean square error of about 8MPa).
[0020] (2) This application utilizes the global search characteristics of genetic algorithms to effectively avoid the problem of BP neural networks getting trapped in local optima, improves the convergence stability and generalization ability of the model, and is suitable for intensity prediction under different mixing ratios.
[0021] (3) This application improves the traditional GA-BP model by simplifying the secondary optimization of weights and thresholds into a single optimization, reducing training steps, simplifying the program structure, and improving the efficiency of model construction and computation.
[0022] (4) The model input established in this application covers seven key components, including cement, slag, fly ash, water, water-reducing agent, sand, and stone, which more comprehensively reflects the characteristics of concrete mix proportions. It can be directly used for mix proportion design and strength prediction in actual projects, which helps to improve the level of concrete quality control and promote the efficient and economical use of building materials.
[0023] (5) This application can be implemented using common programming tools such as MATLAB. The model parameters are clear and the steps are clear, which makes it highly operable and has great application value. Attached Figure Description
[0024] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.
[0025] Figure 1 A flowchart illustrating a method for predicting the compressive strength of medium-high strength concrete based on a GA-BP neural network model, provided in this application embodiment; Figure 2 A diagram of a BP neural network model with a single hidden layer provided in an embodiment of this application; Figure 3 A network structure diagram provided for embodiments of this application; Figure 4 Distribution diagram of test and predicted values of concrete provided for embodiments of this application; Figure 5 A graph showing the difference between the calculated and measured values of the concrete compressive strength provided in the embodiments of this application; Figure 6 A graph showing the relationship between the cement-water ratio and the compressive strength of concrete provided in the embodiments of this application; Figure 7 The diagram shows the structure of the device for predicting the compressive strength of medium-high strength concrete based on the GA-BP neural network model, which is provided in the embodiments of this application.
[0026] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation
[0027] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.
[0028] The collection, storage, use, processing, transmission, provision, and disclosure of financial data or user data involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.
[0029] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.
[0030] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.
[0031] This application provides a method for predicting the compressive strength of medium-high strength concrete based on a GA-BP neural network model, such as... Figure 1 As shown, the method includes the following steps S10 and S20.
[0032] The method is characterized in that it includes: S10: Obtain multiple sets of medium- and high-strength concrete test samples with different mix proportions. The high-strength concrete test samples include concrete composition parameters and corresponding measured values of compressive strength. The concrete composition parameters include the amount of cement, slag, fly ash, water, water-reducing agent, stone, and sand.
[0033] S20: Construct a GA-BP neural network model to predict the compressive strength of high-strength concrete, including: S201: Set the number of input layer nodes M, the number of hidden layer nodes q, and the number of output layer nodes L, wherein the number of input layer nodes M is consistent with the number of types of concrete composition parameters; S202: Using the quadratic error criterion function as the objective function, the optimal solution for weights and thresholds is obtained through the initialization, selection, crossover, mutation, and individual evaluation steps of the genetic algorithm, thus constructing a GA-BP neural network model. S203: After normalizing the high-strength concrete test samples, input them into the constructed GA-BP neural network model to complete model training; S204: The composition parameters of the high-strength concrete to be predicted are normalized and then input into the trained GA-BP neural network model, which outputs the predicted value of the concrete compressive strength.
[0034] Step S20 is the key to the method of this application. Specifically, Jiao Chujie et al. established a prediction model for the compressive strength of planted porous concrete based on GA-BP neural network (refer to "Jiao Chujie, Tan Siqi, Cui Lishi et al. Prediction model for compressive strength of planted porous concrete based on neural network [J]. Concrete, 2022, (01): 7-10, 16"). In this network model, the weights and thresholds in the BP neural network are first optimized using a genetic algorithm, and then the weights and thresholds are corrected in the backoff algorithm of the BP neural network, so that the weights and thresholds are optimized twice. The network model in this embodiment will improve the above model: the forward algorithm still adopts the BP neural network algorithm, and the backoff algorithm adopts the genetic algorithm, turning the secondary optimization of weights and thresholds into a single optimization, thereby simplifying the program structure.
[0035] A backpropagation (BP) neural network is a feedforward network consisting of an input layer, an output layer, and several hidden layers. Figure 2 This is a BP neural network model with a single hidden layer.
[0036] Figure 1 middle, Indicates the input layer's first... The input of each node, j =1,…, M ; Indicates the hidden layer number 1 iThe node to the input layer j The weights between nodes; Indicates the hidden layer number 1 i The threshold of each node, i =1,…, q ; This represents the activation function of the hidden layer; Indicates the output layer number k The node to the hidden layer i The weights between nodes; Indicates the output layer number k The threshold of each node, k =1,…, L ; This represents the activation function of the output layer; Indicates the output layer number k The output of each node.
[0037] according to Figure 1 The forward algorithm for a BP neural network can be obtained from relevant literature, as shown in equations (1) to (3) below: Hidden layer i Input of each node for: (1) Hidden layer i Output of each node for: (2) Similarly, we can obtain the output layer's... k Output of each node for: (3) When there are many hidden layers, the output layer can be calculated sequentially according to formulas (1) to (2).
[0038] To find the optimal solution for the weights and thresholds, the quadratic error criterion function E is used as the objective function. [5] As shown in equation (4): (4) In the formula, The target value of the sample is denoted as . When there are many neuron nodes, the unknowns in equation (4) increase exponentially with geometric order, and gradient-based optimization algorithms may encounter difficulties such as difficulty in convergence.
[0039] Genetic algorithms are general-purpose large-scale algorithms for solving search problems, including steps such as initialization, selection, crossover, mutation, individual evaluation, and termination conditions. Genetic algorithms mainly involve two operations: first, encoding the initial population, often using binary encoding, which can be achieved by completing the number system conversion; second, defining a suitable fitness function, which can be expressed as equation (4). Thus, the optimal solution for weights and thresholds can be obtained from the genetic algorithm, thus establishing the GA-BP neural network model.
[0040] The compressive strength of concrete was taken as the 28-day strength. The concrete composition included cement, fly ash, slag, water, sand, stone and water-reducing agent, etc. A total of 425 test samples were collected, and the sample statistics are shown in Table 1.
[0041] Table 1. Statistical table of 425 test samples
[0042] The variance in Table 1 is relatively large and the dispersion is high. Using this sample, the model has a wide range of applications.
[0043] The number of input layer nodes in the network structure is consistent with the composition of concrete, therefore M=7 (M, see...). Figure 1 (The same applies below). The number of nodes in the output layer is equal to the compressive strength of the concrete, L=1. Hidden layer design includes the number of layers and the number of neurons. For general problems, when the number of hidden layer nodes is sufficient, one hidden layer can be used; the number of hidden layer nodes q=30, and the network structure is as follows: Figure 3 As shown.
[0044] The activation function is a mapping between nodal inputs and outputs; currently, the sigmoid type function is commonly used. Since the concrete strength is positive, logarithmic sigmoid functions are chosen for Ф and Ψ.
[0045] To avoid overfitting, the input samples are normalized. Considering that the logarithmic sigmoid function curve is relatively flat and changes very slowly near the points 0 and 1, the following formula is used for normalization: (5) In the formula, These are the normalized values of the concrete composition parameters. These are the original values of the concrete composition parameters. The minimum value of the concrete composition parameters among all test samples. This represents the maximum value of the concrete composition parameters across all test samples.
[0046] To determine the GA-BP neural network structure designed in this embodiment (see...), Figure 1If so, the weights and thresholds need to be known. The weights of the input layer and the hidden layer, and the weights of the hidden layer and the output layer are shown in Table 2 and Table 3, respectively; the threshold of the hidden layer is shown in Table 4, and the threshold of the output layer is -0.72.
[0047] Table 2 Weights of Input Layer and Hidden Layer
[0048] Table 3 Weights of Hidden Layers and Output Layers
[0049] Table 4 Thresholds of Hidden Layers
[0050] The weights in Table 2 are 7 (rows) and 30 (columns), corresponding to the number of input layer nodes and the number of hidden layer nodes, respectively. Rows 1-7 correspond to cement, slag, fly ash, water, water-reducing agent, stone, and sand, respectively. In Tables 3 and 4, row 1 represents the number of output layer nodes; column 30 represents the number of hidden layer nodes.
[0051] Concrete compressive strength prediction analysis: Using the established GA-BP neural network model, predicted values of concrete compressive strength for 425 sets of test samples can be obtained. The test values and predicted values of concrete are shown below. Figure 4 As shown, the difference between the predicted value and the experimental value is as follows: Figure 5 As shown.
[0052] Figure 5 Among the samples, 317 groups (74.6% of the total samples) had an overall error within 1 MPa, 377 groups (88% of the total samples) had an error within 2 MPa, and 391 groups (92% of the total samples) had an error within 3 MPa. Considering the discrete nature of concrete compressive strength, this model can be considered to meet the requirements of engineering accuracy.
[0053] The effect of cement-to-water ratio on compressive strength: The relationship between cement-to-water ratio and concrete compressive strength can be established using 425 sets of test samples, such as... Figure 6 As shown. Figure 5 The compressive strength of concrete is mostly distributed within a narrow band, showing a certain regularity. Through linear fitting analysis, the root mean square variance is 8.675 MPa. Therefore, using the cement-water ratio to guide concrete construction and design has a large error.
[0054] This embodiment, based on 425 sets of experimental samples, uses a genetic algorithm and a backpropagation neural network to predict the compressive strength of concrete under different mix proportions. The main conclusions are as follows: 1) The existing prediction model for concrete compressive strength based on GA-BP neural network was improved and verified. The forward algorithm still uses the BP neural network algorithm, while the backward algorithm uses the genetic algorithm, which transforms the secondary optimization of weights and thresholds into a primary optimization, thus simplifying the program structure.
[0055] 2) A linear fitting relationship between cement-water ratio and compressive strength was established using 425 concrete samples. The root mean square variance was about 8 MPa. Using cement-water ratio to guide concrete design and construction will result in a large deviation.
[0056] 3) The established model was used to predict the compressive strength of concrete. Except for a few samples with large errors, the errors of 317 samples (accounting for 74.6% of the total samples) were within 1 MPa, and the errors of 377 samples (accounting for 88% of the total samples) were within 2 MPa. Considering the dispersion of concrete compressive strength, it can be considered that the model established by the method of this application can meet the engineering accuracy requirements.
[0057] This application also provides a device for predicting the compressive strength of medium-high strength concrete based on a GA-BP neural network model, used to implement the methods described in any of the above embodiments, such as... Figure 7 As shown, the device for predicting the compressive strength of medium-strength concrete based on the GA-BP neural network model includes: The data acquisition module 701 is configured to acquire multiple sets of medium- and high-strength concrete test samples with different mix proportions. The high-strength concrete test samples include concrete composition parameters and corresponding measured values of compressive strength. The concrete composition parameters include cement content, slag content, fly ash content, water content, water-reducing agent content, stone content, and sand content. Model building module 702 is configured to build a GA-BP neural network model to predict the compressive strength of high-strength concrete, including: Set the number of input layer nodes M, the number of hidden layer nodes q, and the number of output layer nodes L. The number of input layer nodes M is consistent with the number of types of concrete composition parameters. Using the quadratic error criterion function as the objective function, the optimal solutions for weights and thresholds are obtained through the initialization, selection, crossover, mutation, and individual evaluation steps of the genetic algorithm, thus constructing a GA-BP neural network model. After normalizing the high-strength concrete test samples, they were input into the constructed GA-BP neural network model to complete the model training. The composition parameters of the high-strength concrete to be predicted are normalized and then input into the trained GA-BP neural network model, which outputs the predicted value of the concrete compressive strength.
[0058] This application provides an electronic device. The electronic device may include a processor and a memory, wherein the processor and the memory can communicate; exemplarily, the processor and the memory communicate via a communication bus.
[0059] The processor executes computer execution instructions stored in memory, causing the processor to perform the scheme in the above embodiments. The processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.
[0060] The communication bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus can be divided into address bus, data bus, control bus, etc. Transceivers are used to enable communication between database access devices and other computers (e.g., clients, read-write libraries, and read-only libraries). Memory may include random access memory (RAM) and may also include non-volatile memory.
[0061] The electronic device provided in this application embodiment can be the terminal device described in the above embodiments.
[0062] This application also provides a computer-readable storage medium storing computer instructions. When the computer instructions are executed on a computer, the computer performs the technical solution of the above-described embodiment of the method for predicting the compressive strength of high-strength concrete based on the GA-BP neural network model.
[0063] This application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When the at least one processor executes the computer program, it can implement the technical solution of the method for predicting the compressive strength of high-strength concrete based on the GA-BP neural network model in the above embodiments.
[0064] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or modules, and may be electrical, mechanical, or other forms.
[0065] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to implement the solution of this embodiment according to actual needs.
[0066] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The unit composed of the above modules can be implemented in hardware or in the form of hardware plus software functional units.
[0067] The integrated modules described above, implemented as software functional modules, can be stored in a computer-readable storage medium. These software functional modules, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods of the various embodiments of this application.
[0068] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.
[0069] The memory may include high-speed RAM, and may also include non-volatile storage (NVM), such as at least one disk storage device, and may also be a USB flash drive, external hard drive, read-only memory, disk or optical disc, etc.
[0070] Buses can be Industry Standard Architecture (ISA) buses, Peripheral Component Interconnect (PCI) buses, or Extended Industry Standard Architecture (EISA) buses, etc. Buses can be categorized into address buses, data buses, control buses, etc.
[0071] The aforementioned storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.
[0072] An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Alternatively, the storage medium can be an integral part of the processor. The processor and storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and storage medium can exist as discrete components in an electronic control unit or main control device.
[0073] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.
[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.
Claims
1. A method for predicting the compressive strength of medium-high strength concrete based on a GA-BP neural network model, characterized in that, The method includes: Multiple sets of medium- and high-strength concrete test samples with different mix proportions were obtained. The high-strength concrete test samples included concrete composition parameters and corresponding measured values of compressive strength. The concrete composition parameters included the amount of cement, slag, fly ash, water, water-reducing agent, stone, and sand. Constructing a GA-BP neural network model to predict the compressive strength of high-strength concrete, including: Set the number of input layer nodes M, the number of hidden layer nodes q, and the number of output layer nodes L. The number of input layer nodes M is consistent with the number of types of concrete composition parameters. Using the quadratic error criterion function as the objective function, the optimal solutions for weights and thresholds are obtained through the initialization, selection, crossover, mutation, and individual evaluation steps of the genetic algorithm, thus constructing a GA-BP neural network model. After normalizing the high-strength concrete test samples, they were input into the constructed GA-BP neural network model to complete the model training. The composition parameters of the high-strength concrete to be predicted are normalized and then input into the trained GA-BP neural network model, which outputs the predicted value of the concrete compressive strength.
2. The method for predicting the compressive strength of medium- and high-strength concrete based on the GA-BP neural network model according to claim 1, characterized in that, The first hidden layer i Input of each node Calculated using the following formula: （1） In the formula, This is the input of the j-th node in the input layer, namely the normalized value of the concrete composition parameters; j For node sequence number, M The number of nodes; For the hidden layer i The node to the input layer j The weights between nodes; For the hidden layer i Threshold for each node; The first hidden layer i Output of each node Calculated using the following formula: （2） In the formula, This represents the activation function of the hidden layer.
3. The method for predicting the compressive strength of medium- and high-strength concrete based on the GA-BP neural network model according to claim 2, characterized in that, Output layer k Output of each node , which is the normalized value of the predicted concrete compressive strength, is calculated using the following formula: （3） In the formula, This represents the activation function of the output layer; Indicates the output layer number k The node to the hidden layer i The weights between nodes; Indicates the output layer number k Threshold for each node.
4. The method for predicting the compressive strength of medium- and high-strength concrete based on the GA-BP neural network model according to claim 3, characterized in that, The quadratic error criterion function is expressed as follows: （4） In the formula, It is a quadratic error criterion function; The target value for the sample; This represents the number of nodes in the output layer.
5. The method for predicting the compressive strength of medium- and high-strength concrete based on the GA-BP neural network model according to claim 1, characterized in that, The normalization process for the high-strength concrete test samples is as follows: （5） In the formula, These are the normalized values of the concrete composition parameters. These are the original values of the concrete composition parameters. The minimum value of the concrete composition parameters among all test samples. This represents the maximum value of the concrete composition parameters across all test samples.
6. The method for predicting the compressive strength of medium-high strength concrete based on the GA-BP neural network model according to any one of claims 1 to 5, characterized in that, The activation functions of the hidden layer and the output layer are both logarithmic S-functions.
7. The method for predicting the compressive strength of medium- and high-strength concrete based on the GA-BP neural network model according to claim 1, characterized in that, The weights of the input layer and the hidden layer, the weights of the hidden layer and the output layer, the threshold of the hidden layer, and the threshold of the output layer satisfy the following: the weights of the input layer and the hidden layer are a 7-row, 30-column matrix, with the 7 rows corresponding to the input nodes of cement, slag, fly ash, water, water-reducing agent, stone, and sand, respectively; the weights of the hidden layer and the output layer are a 1-row, 30-column matrix; and the threshold of the hidden layer is a 1-row, 30-column matrix.
8. A device for predicting the compressive strength of medium- and high-strength concrete based on a GA-BP neural network model, characterized in that, The device includes: The data acquisition module is configured to acquire multiple sets of medium- and high-strength concrete test samples with different mix proportions. The high-strength concrete test samples include concrete composition parameters and corresponding measured values of compressive strength. The concrete composition parameters include cement content, slag content, fly ash content, water content, water-reducing agent content, stone content, and sand content. The model building module is configured to build a GA-BP neural network model to predict the compressive strength of high-strength concrete, including: Set the number of input layer nodes M, the number of hidden layer nodes q, and the number of output layer nodes L. The number of input layer nodes M is consistent with the number of types of concrete composition parameters. Using the quadratic error criterion function as the objective function, the optimal solutions for weights and thresholds are obtained through the initialization, selection, crossover, mutation, and individual evaluation steps of the genetic algorithm, thus constructing a GA-BP neural network model. After normalizing the high-strength concrete test samples, they were input into the constructed GA-BP neural network model to complete the model training. The composition parameters of the high-strength concrete to be predicted are normalized and then input into the trained GA-BP neural network model, which outputs the predicted value of the concrete compressive strength.
9. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes the computer execution instructions stored in the memory to implement the method for predicting the compressive strength of medium-high strength concrete based on the GA-BP neural network model as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method for predicting the compressive strength of high-strength concrete based on the GA-BP neural network model as described in any one of claims 1-7.