CN115293051A - Concrete rheological property prediction method based on BP neural network - Google Patents

Concrete rheological property prediction method based on BP neural network Download PDF

Info

Publication number
CN115293051A
CN115293051A CN202211022596.6A CN202211022596A CN115293051A CN 115293051 A CN115293051 A CN 115293051A CN 202211022596 A CN202211022596 A CN 202211022596A CN 115293051 A CN115293051 A CN 115293051A
Authority
CN
China
Prior art keywords
concrete
neural network
output
layer
rheological property
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202211022596.6A
Other languages
Chinese (zh)
Inventor
陈嘉奇
周念
吴小敏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Academy Of Sciences Jiangmen Industrial Technology Research Institute Co ltd
Original Assignee
Guangdong Academy Of Sciences Jiangmen Industrial Technology Research Institute Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangdong Academy Of Sciences Jiangmen Industrial Technology Research Institute Co ltd filed Critical Guangdong Academy Of Sciences Jiangmen Industrial Technology Research Institute Co ltd
Priority to CN202211022596.6A priority Critical patent/CN115293051A/en
Publication of CN115293051A publication Critical patent/CN115293051A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a concrete rheological property prediction method based on a BP neural network, which comprises the following steps: according to the influence factors of the concrete rheological property, nine parameters including the air content of the concrete, the needle sheet content of the coarse aggregate, the sand rate, the aggregate distribution of the coarse aggregate, the fineness modulus of sand, the using amount of a cementing material, the water cement ratio, the consistency of cement paste and the admixture doping amount are selected as input of a BP (back propagation) neural network, and a (VDH) 90 value of the corresponding concrete is used as output of the BP neural network, wherein the (VDH) 90 value is V, the volume number of the concrete flowing out of the inverted outflow barrel in 90 seconds, the D is the concrete collapse diameter flowing out of the inverted outflow barrel in 90 seconds, and the H is the concrete overtopping height flowing out of the inverted outflow barrel in 90 seconds. The invention can greatly save production and experiment cost, improve the construction quality of concrete and has wide social and economic benefits.

Description

Concrete rheological property prediction method based on BP neural network
Technical Field
The invention relates to a concrete rheological property prediction method based on a BP neural network.
Background
Currently, in the field of premixed concrete production, the processes from quality control of raw materials to production process control and even quality control of concrete construction are mainly low-efficiency and extensive production management means, which leads to a series of economic and social problems of great waste of raw materials, ineffective investment of production cost, unstable product quality, increasingly prominent environmental protection and the like.
The concrete rheological parameters (mainly refer to yield stress and plastic viscosity) are basic physical parameters for describing the flowing performance of concrete mixture, and for the newly mixed fluid concrete mixture (called newly mixed concrete for short), the rheological parameters reflect the construction performance of the newly mixed concrete, and the numerical value change of the rheological parameters is obviously influenced by the components of the newly mixed concrete.
The coarse aggregate is the most main component of the fresh concrete, and under a certain proportion, the gradation, the particle size, the particle shape, the surface texture, the solid accumulation state and the like of the coarse aggregate are key factors influencing the rheological property of the fresh concrete. The existing method capable of reflecting the shape and the stacking state of the coarse aggregate generally obtains macroscopic geometric index parameters such as the void ratio, the sphericity, the roundness, the needle sheet-shaped coarse aggregate proportion and the like of the coarse aggregate according to a certain design mixing proportion of concrete, and is used for qualitative analysis and regulation of the workability of fresh concrete. The fractal theory has been applied to the performance analysis of materials such as concrete and the like as an advanced research method for reflecting the distribution characteristics of target objects, but the general research method mainly analyzes the distribution rule of the target objects through single fractal, the fractal dimension of the single fractal reflects the complexity of the distribution and the form of the stacking density degree of coarse aggregates from the integral angle, and the characteristics of the actual component stacking structure of each subdivided region cannot be reflected. With the intensive research on the theory of fractal, the method of multi-fractal is introduced into the field of material performance analysis, and the multi-fractal is measured and calculated through a probability density function, so that the heterogeneity of local regions is emphasized, but the overall characteristics are not grasped. For the fluid concrete mixture, because the interaction relationship among the coarse aggregates and between the coarse aggregates and the mortar body is very complex, the performance statistical analysis of multiple characteristics of the aggregates among the mortar bodies in a motion state cannot be effectively obtained by adopting single fractal dimension or adopting fractal spectrum to describe the distribution of the concrete aggregates in a discrete state, and the comprehensive performance, distribution and actual component stacking structure relationship of the coarse aggregates in fresh concrete cannot be accurately described, so that the influence of the coarse aggregate parameters on the rheological property of the mixture is hardly reflected, and the application of the scientific quantitative regulation and control technology of the rheological property of the concrete is restricted.
In addition, in the existing methods, the performance of the concrete is evaluated by using an image analysis method, but the performance analysis is performed by using a single image or by performing slicing reprocessing analysis after the concrete is formed, so that the performance characteristics are obviously not reasonable and complete. For the fresh concrete, because the coarse aggregate and the mortar are three-dimensionally distributed, a single picture has high randomness, the characteristics of the concrete cannot be accurately represented, and the slice analysis result cannot accurately reflect the performance of the fresh concrete in a fluid state.
High-performance concrete is a mainstream product in the current concrete engineering application, the rheological property of high-performance concrete pumping is extremely critical, the working performance of the concrete is evaluated mainly by researching the rheological property of cement paste in recent years, but the batch performance of concrete raw materials is greatly changed, so that the rheological factor of the concrete is influenced, the rheological property of the concrete is predicted by simply measuring the rheological property of the cement paste, the result is far away, and a method for predicting the rheological property of the concrete is urgently needed.
Disclosure of Invention
In view of the above defects in the prior art, the present invention aims to provide a method for predicting rheological property of concrete based on a BP neural network, which can comprehensively and accurately reflect the key control effect of the technical parameters of the concrete raw material in the rheological property of concrete, solve the existing defect that the rheological property of concrete based on the macroscopic geometric parameters of the coarse aggregate part and the rheological property of cement paste can only be qualitatively evaluated or cannot be precisely quantitatively evaluated and regulated in detail, and scientifically and accurately predict the rheological property of concrete by combining with a BP neural network data model.
The technical scheme is as follows: a concrete rheological property prediction method based on a BP neural network comprises the following steps:
the first step is as follows: selecting nine parameters, namely concrete air content, coarse aggregate needle flake content, sand rate, coarse aggregate collection and distribution, sand fineness modulus, cementing material consumption, water-cement ratio, cement paste consistency and admixture mixing amount, as input of a BP (back propagation) neural network, and taking a (VDH) 90 value of corresponding concrete as output of the BP neural network, wherein V refers to the volume number of concrete flowing out of an inverted outflow barrel within 90 seconds, D refers to the concrete collapse diameter flowing out of the inverted outflow barrel within 90 seconds, and H refers to the concrete rise height flowing out of the inverted outflow barrel within 90 seconds;
the second step: collecting sample related input and output parameters on the basis of concrete (VDH) 90 value rheological property test data of a practical samples and test data of nine parameters, namely concrete air content, coarse aggregate needle sheet content, sand rate, coarse aggregate collection and preparation, sand fineness modulus, cementing material consumption, water-cement ratio, cement paste consistency and additive mixing amount, taking the collected data as a training sample set of a BP neural network, and determining the node number, the number of hidden layers, the number of hidden layer nodes, a transfer function, a training function, a weight and a threshold parameter of an input layer and an output layer of the BP neural network;
the third step: training a BP neural network by using a training sample set to construct the BP neural network for predicting the concrete rheological property;
the fourth step: and carrying out simulation test, taking nine parameters of the concrete sample to be tested as the input of the BP neural network, and taking the (VDH) 90 value of the corresponding concrete output by the BP neural network as the prediction result of the concrete rheological property.
The method for determining the BP neural network in the second step of the method is as follows: defining the number of nodes of an input layer of a BP neural network as n, the number of nodes of an output layer as m, knowing that the number of nodes of the input layer and the number of nodes of the output layer of the BP neural network are respectively the dimensions of an input variable and an output variable, defining the number of nodes of a hidden layer as q, and defining the weight between the input layer and the hidden layer as theta ji J =1,2, \ 8230;, q; i =1,2, \ 8230;, n, threshold between input layer and hidden layer is b j J =1,2, \8230;, q, hidden layer and outputThe weight of a layer is w kj K =1,2, \ 8230;, m; j =1,2, \8230q, the threshold values of the hidden layer and the output layer being Z k K =1,2, \ 8230;, m, transfer function of the hidden layer is f 1 (. DEG), the transfer function of the output layer is f 2 (. Cndot.), the hidden layer of the BP neural network is layered as one layer.
The method for constructing the BP neural network model for predicting the concrete rheological property in the third step of the method comprises the following steps: inputting an input variable through an input layer of a BP neural network, then outputting through calculation of a hidden layer and an output layer to obtain a prediction result, taking the prediction result as actual output, defining expected output, reversely transmitting a difference value between the actual output and the expected output to the input layer and the hidden layer, adjusting a weight and a threshold value according to a direction which enables an error function to be minimum through the BP neural network, and training the network.
In the third step, the output of the hidden layer node is:
Figure BDA0003814619850000031
wherein S is the output of the hidden layer node; x is the input variable set of the invention; j. i and n are preset values.
In the third step, the output of the output layer node is:
Figure BDA0003814619850000032
wherein Y is the output of the output layer node; k. j is a preset value.
When the actual output does not coincide with the desired output, the output error E is:
Figure BDA0003814619850000033
where t is the desired output value of the output layer, t = t 1 ，t 2 ，····，t m (ii) a a is the actual number of samples; m and k are preset values.
In order to minimize the error E to meet the requirement, the process of training the BP neural network is to modify the weight coefficients according to the direction of the negative gradient of the error function, that is:
Figure BDA0003814619850000034
Figure BDA0003814619850000035
Figure BDA0003814619850000036
Figure BDA0003814619850000037
in the formulas (4) and (5), j =1,2, \8230;, q; i =1,2, \8230;, n; in formula (6) and formula (7), j =1,2, \8230;, q; k =1,2, \ 8230;, m; the value range of eta in the formulas (4), (5), (6) and (7) is more than or equal to 0 and less than or equal to 1.
The invention has the advantages that:
the invention applies a BP neural network model to the prediction of concrete rheological property, selects nine parameters as the input of the BP neural network according to the influence factors of the concrete rheological property, such as concrete air content, coarse aggregate needle sheet content, sand rate, coarse aggregate collection and preparation, sand fineness modulus, cementing material consumption, water-cement ratio, cement paste consistency and admixture mixing amount, and uses the (VDH) 90 value of corresponding concrete as the output of the BP neural network.
Drawings
FIG. 1 is a diagram of one embodiment of an inverted pour bucket according to one preferred embodiment of the present invention;
FIG. 2 is another embodiment of an inverted outflow bucket according to a preferred embodiment of the present invention;
FIG. 3 is a schematic diagram of a concrete rheological property prediction method based on a BP neural network.
Detailed Description
The following examples are given to illustrate the embodiments of the present invention, and the detailed embodiments and specific procedures are given on the premise of the technical solution of the present invention, but the scope of the present invention is not limited to the following examples.
As shown in fig. 3, the invention provides a concrete rheological property prediction method based on a BP neural network, which comprises the following steps:
the first step is as follows: according to the influence factors of the rheological property of the concrete, nine parameters including the air content of the concrete, the needle sheet content of the coarse aggregate, the sand rate, the aggregate distribution of the coarse aggregate, the fineness modulus of the sand, the using amount of a cementing material, the water-cement ratio, the consistency of cement paste and the admixture mixing amount are selected as the input of a BP (back propagation) neural network, and the (VDH) 90 value of the corresponding concrete is used as the output of the BP neural network, wherein V refers to the volume number of the concrete flowing out of the inverted outflow barrel within 90 seconds, D refers to the concrete collapse diameter flowing out of the inverted outflow barrel within 90 seconds, and H refers to the concrete overtopping height flowing out of the inverted outflow barrel within 90 seconds.
The second step: collecting related input and output parameters of samples on the basis of 5000 concrete (VDH) 90 value rheological property test data of actual samples (the range of the samples is 1000-20000), and nine parameter test data including concrete air content, coarse aggregate needle sheet content, sand rate, coarse aggregate collection and preparation, sand fineness modulus, cementing material consumption, water cement ratio, cement paste consistency and additive mixing amount;
the third step: training a BP neural network by using a training sample set to construct a BP neural network model for predicting the concrete rheological property;
the fourth step: and carrying out simulation test, taking nine parameters of the concrete sample to be tested as the input of the BP neural network, and taking the (VDH) 90 value of the corresponding concrete output by the BP neural network as the prediction result of the concrete rheological property.
The method for determining the BP neural network in the second step of the method is as follows: defining the number of nodes of an input layer of a BP neural network as n, the number of nodes of an output layer as m, knowing that the number of nodes of the input layer and the number of nodes of the output layer of the BP neural network are respectively the dimensionalities of an input variable and an output variable, namely n =9, m =1, defining the number of nodes of a hidden layer as q, and defining the weight value between the input layer and the hidden layer as theta ji J =1,2, \ 8230;, q; i =1,2, \8230n, the threshold between the input layer and the hidden layer is b j J =1,2, \ 8230;, q, the weight of the hidden layer and the output layer is w kj K =1,2, \8230;, m; j =1,2, \ 8230;, q, threshold value of hidden layer and output layer is Z k K =1,2, \8230;, m, transfer function of the hidden layer is f 1 (. O) the transfer function of the output layer is f 2 (. Cndot.), the hidden layer of the BP neural network is layered as one layer.
The method for constructing the BP neural network model for predicting the concrete rheological property in the third step of the method comprises the following steps: inputting an input variable through an input layer of a BP neural network, then outputting through calculation of a hidden layer and an output layer to obtain a prediction result, taking the prediction result as actual output, defining expected output, reversely transmitting a difference value between the actual output and the expected output to the input layer and the hidden layer, adjusting a weight and a threshold value according to a direction which enables an error function to be minimum through the BP neural network, and training the network.
In the third step of the method, the hidden layer nodes are output as follows:
Figure BDA0003814619850000051
wherein S is the output of the hidden layer node; x is the input variable set of the invention; j. and i and n are preset values.
The output layer nodes are output as:
Figure BDA0003814619850000052
wherein Y is the output of the output layer node; k is a preset value.
When the actual output does not coincide with the desired output, the output error E is:
Figure BDA0003814619850000053
where t is the desired output value of the output layer, t = t 1 ，t 2 ，····，t m (ii) a m and k are preset values.
The error E is enabled to be minimum to meet the requirement, and the training process of the BP neural network is to modify weight coefficients according to the direction of the negative gradient of an error function, namely:
Figure BDA0003814619850000054
Figure BDA0003814619850000055
Figure BDA0003814619850000056
Figure BDA0003814619850000057
in the formulas (4) and (5), j =1,2, \8230;, q; i =1,2, \8230;, n; in formula (6) and formula (7), j =1,2, \8230;, q; k =1,2, \ 8230;, m; in the step (7), eta is 0.5;
the foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the prior art according to the concepts of the present invention should be within the scope of protection determined by the claims.

Claims (8)

1. A concrete rheological property prediction method based on a BP neural network is characterized by comprising the following steps:
the first step is as follows: according to the influence factors of the concrete rheological property, nine parameters including the concrete air content, the needle sheet content of the coarse aggregate, the sand rate, the aggregate preparation of the coarse aggregate, the fineness modulus of sand, the using amount of a cementing material, the water cement ratio, the consistency of cement paste and the admixture doping amount are selected as the input of a BP (back propagation) neural network, and a (VDH) 90 value of corresponding concrete is used as the output of the BP neural network, wherein V refers to the volume number of the concrete flowing out of the inverted outflow barrel within 90 seconds, D refers to the concrete collapse diameter flowing out of the inverted outflow barrel within 90 seconds, and H refers to the concrete overthrow height flowing out of the inverted outflow barrel within 90 seconds;
the second step is that: collecting sample related input and output parameters on the basis of concrete (VDH) 90 value rheological property test data of a practical samples and test data of nine parameters, namely concrete air content, coarse aggregate needle sheet content, sand rate, coarse aggregate collection and preparation, sand fineness modulus, cementing material consumption, water-cement ratio, cement paste consistency and additive mixing amount, taking the collected data as a training sample set of a BP neural network, and determining the node number, the number of hidden layers, the number of hidden layer nodes, a transfer function, a training function, a weight and a threshold parameter of an input layer and an output layer of the BP neural network;
the third step: training a BP neural network by using a training sample set to construct the BP neural network for predicting the concrete rheological property;
the fourth step: and carrying out simulation test, taking nine parameters of the concrete sample to be tested as the input of the BP neural network, and taking the (VDH) 90 value of the corresponding concrete output by the BP neural network as the prediction result of the concrete rheological property.
2. The method for predicting rheological behavior of concrete based on BP neural network as claimed in claim 1, wherein the method for determining BP neural network in the second step is as follows: defining the number of nodes of an input layer of a BP neural network as n, the number of nodes of an output layer as m, knowing that the number of nodes of the input layer and the number of nodes of the output layer of the BP neural network are respectively the dimensions of an input variable and an output variable, defining the number of nodes of a hidden layer as q, and defining the weight between the input layer and the hidden layer as theta ji J =1,2, \ 8230;, q; i =1,2, \ 8230;, n, threshold between input layer and hidden layer is b j J =1,2, \ 8230;, q, the weight of the hidden layer and the output layer is w kj K =1,2, \8230;, m; j =1,2, \8230q, the threshold values of the hidden layer and the output layer being Z k K =1,2, \8230;, m, transfer function of the hidden layer is f 1 (. DEG), the transfer function of the output layer is f 2 (. Cndot.), the hidden layer of the BP neural network is layered as one layer.
3. The method for predicting the rheological property of the concrete based on the BP neural network as claimed in claim 2, wherein the method for constructing the BP neural network model for predicting the rheological property of the concrete in the third step comprises the following steps: inputting input variables through an input layer of a BP neural network, outputting through calculation of a hidden layer and an output layer to obtain a prediction result, taking the prediction result as actual output, defining expected output, reversely transmitting the difference value between the actual output and the expected output to the input layer and the hidden layer, adjusting a weight and a threshold value according to the direction which enables an error function to be minimum by the BP neural network, and training the network.
4. The method for predicting the rheological property of the concrete based on the BP neural network as claimed in claim 3, wherein in the third step, the output of the hidden layer node is as follows:
Figure FDA0003814619840000021
wherein S is the output of the hidden layer node; x is the input variable set of the invention; j. and i and n are preset values.
5. The method for predicting the rheological property of the concrete based on the BP neural network as claimed in claim 4, wherein in the third step, the output of the node of the output layer is as follows:
Figure FDA0003814619840000022
wherein Y is the output of the output layer node; k is a preset value.
6. The method of claim 5, wherein when the actual output is inconsistent with the expected output, the output error E is:
Figure FDA0003814619840000023
where t is the desired output value of the output layer, t = t 1 ，t 2 ，…，t m (ii) a a is the actual number of samples; m and k are preset values.
7. The method for predicting the rheological property of the concrete based on the BP neural network as claimed in claim 6, wherein in order to minimize the error E to meet the requirement, the process of training the BP neural network is to modify the weight coefficients according to the negative gradient direction of the error function, namely:
Figure FDA0003814619840000024
Figure FDA0003814619840000025
Figure FDA0003814619840000026
Figure FDA0003814619840000027
wherein eta is a preset value.
8. The method for predicting rheological property of concrete based on BP neural network as claimed in claim 7, wherein in formula (4) and formula (5), j =1,2, \8230;, q; i =1,2, \8230;, n; in formula (6) and formula (7), j =1,2, \8230;, q; k =1,2, \ 8230;, m; the value range of eta in the formula (4), the formula (5), the formula (6) and the formula (7) is more than or equal to 0 and less than or equal to 1.
CN202211022596.6A 2022-08-25 2022-08-25 Concrete rheological property prediction method based on BP neural network Pending CN115293051A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211022596.6A CN115293051A (en) 2022-08-25 2022-08-25 Concrete rheological property prediction method based on BP neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211022596.6A CN115293051A (en) 2022-08-25 2022-08-25 Concrete rheological property prediction method based on BP neural network

Publications (1)

Publication Number Publication Date
CN115293051A true CN115293051A (en) 2022-11-04

Family

ID=83832025

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211022596.6A Pending CN115293051A (en) 2022-08-25 2022-08-25 Concrete rheological property prediction method based on BP neural network

Country Status (1)

Country Link
CN (1) CN115293051A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116090502A (en) * 2022-11-22 2023-05-09 杭州信之威信息技术有限公司 Concrete slump control method and device based on Internet of things
CN116108739A (en) * 2022-12-07 2023-05-12 佛山市顺德区新广利混凝土有限公司 Concrete mortar ready-mix performance prediction method, system and storage medium

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116090502A (en) * 2022-11-22 2023-05-09 杭州信之威信息技术有限公司 Concrete slump control method and device based on Internet of things
CN116090502B (en) * 2022-11-22 2024-01-19 杭州信之威信息技术有限公司 Concrete slump control method and device based on Internet of things
CN116108739A (en) * 2022-12-07 2023-05-12 佛山市顺德区新广利混凝土有限公司 Concrete mortar ready-mix performance prediction method, system and storage medium

Similar Documents

Publication Publication Date Title
CN115293051A (en) Concrete rheological property prediction method based on BP neural network
Reinold et al. Extrusion process simulation and layer shape prediction during 3D-concrete-printing using the particle finite element method
Sonebi et al. Modelling the fresh properties of self-compacting concrete using support vector machine approach
Zeng et al. Accurate prediction of concrete compressive strength based on explainable features using deep learning
Liu et al. Prediction of the durability of high-performance concrete using an integrated RF-LSSVM model
CN110263431B (en) Concrete 28d compressive strength prediction method
CN111650088B (en) Real-time detection method for rheological property of fluid concrete mixture
Da Silva et al. Expert system applied for classifying self-compacting concrete surface finish
CN112185486B (en) Cement-emulsified asphalt mixture shrinkage behavior prediction method based on deep learning
Fasihihour et al. Experimental and numerical model for mechanical properties of concrete containing fly ash: Systematic review
Gogineni et al. Predictive modelling of concrete compressive strength incorporating GGBS and alkali using a machine-learning approach
CN115497574A (en) HPC compressive strength prediction method and system based on model fusion
CN116597919A (en) 3D printing cement-based material design method based on artificial intelligence
Strohle et al. Prospect and barrier of 3D concrete: a systematic review
Akeke et al. Experimental investigation and modelling of the mechanical properties of palm oil fuel ash concrete using Scheffe’s method
Kazemi et al. The properties of cement-mortar at different cement strength classes: Experimental study and multi-objective modeling
Li et al. Effect of air void structure and mechanical properties of high strength fiber reinforced mortar composed with different construction sand species based on X-CT technology
CN103105485A (en) Hardened ordinary cement paste thermal expansion coefficient multiscale predication method
Felfili et al. Performance evaluation of the 3d printing system through fault tree analysis method (FTAM)
Tuvayanond et al. Efficient machine learning for strength prediction of ready-mix concrete production (prolonged mixing)
Vargas et al. Grading material properties in 3D printed concrete structures
Mohammad et al. 2D stationary computational printing of cement-based pastes
Wu et al. Optimal analysis of material ratio for artificial rock by 3D printing technique
Naik et al. Investigating the potential of waste glass in paver block production using RSM
CN103105484B (en) Hardened ordinary cement paste elastic modulus multiscale predication method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination