模糊神经网络在粉煤灰混凝土强度预测中的应用

粉煤灰混凝土强度预测是一个典型的多变量,非线性系统。现在预测的方法准确性较差,难以在实际中被普遍应用。本文将模糊理论与神经网络相结合,综合利用二者的优点,进行粉煤灰混凝土强度的建模与预测。     

自密实混凝土工作性能和强度的BP神经网络预测

如何在进行自密实混凝土配合比设计前对其工作性能和强度进行有效预测,为配合比设计提供指导,是一大难点。本文利用BP神经网络,对自密实混凝土的工作性能(坍落度和扩展度)和28d强度进行预测。结果表明,利用大量试验数据样本训练的BP网络可以预测不同情况下的自密实混凝土的坍落度、扩展度和28d强度,预测精度高。     

Flexural ductility design of high‐strength concrete columns

In the seismic design of a structure, it is necessary to provide not only sufficient strength, but also a minimum level of flexural ductility for reinforced concrete (RC) columns. Eurocode EN1998‐1 directly specifies such minimum flexural ductility, while Chinese code GB50011 limits the normalized design axial force to achieve a nominal minimum flexural ductility. American code ACI 318‐08 uses the tension steel strain at peak resisting moment to control the failure mode. To provide the required flexural ductility, a much lower axial strength reduction factor is assigned to compression‐controlled failure than to tension‐controlled failure. To develop an effective strategy for flexural ductility design of RC columns, it is necessary to identify the essential parameters and control them properly. This is particularly important to those cast of high‐strength concrete that is inherently more brittle. The essential parameters identified include the maximum normalized axial force and maximum normalized neutral axis depth at peak resisting moment, as they help to guarantee various flexural ductility requirements. Their relationship with the flexural ductility is studied using a rigorous full‐range moment‐curvature analysis procedure. Empirical formulae and tables are also developed to facilitate flexural ductility design of RC columns. Copyright © 2010 John Wiley & Sons, Ltd.     

Flexural ductility design of high‐strength concrete beams

In the seismic design of a reinforced concrete (RC) structure, it is necessary to provide not only sufficient strength, but also adequate flexural ductility. This is particularly important to the design of RC beams cast of high‐strength concrete that is inherently more brittle. Eurocode EN1998‐1 directly specifies such minimum flexural ductility. To provide adequate flexural ductility to RC beams, Chinese code GB50011 limits the normalised depth of simplified rectangular stress block at peak resisting moment, whereas American code ACI 318–08 requires that the tension steel strain at peak resisting moment shall not be smaller than 0.004. The essential parameters identified for effective flexural ductility design of RC beams include the maximum difference of tension and compression reinforcement ratios and maximum normalised neutral axis depth at peak resisting moment, as they help to guarantee various flexural ductility requirements. Their relationship with the flexural ductility is studied using a rigorous full‐range moment–curvature analysis procedure. Empirical formulae and tables are also developed to facilitate flexural ductility design of RC beams. A comparison shows that the allowable differences of tension and compression ratios may be smaller than those specified in Eurocode 8 particularly for those cast of high‐strength concrete. Copyright © 2011 John Wiley & Sons, Ltd.     

Experimental tests for improving buildability of construction methods for high‐strength concrete columns in high‐rise buildings

High‐strength concrete columns have the advantage of increasing the amount of usable area in the building because the cross‐section of the columns takes up less space compared with columns using normal strength concrete. However, it is difficult to weld the steel reinforcement and steel members because of the narrow column width due to a decrease in the cross‐section of the column, thereby causing construction delay in many cases. In this paper, five construction methods with different details for high‐strength reinforced concrete columns are tested to improve the buildability of the columns. Five specimens with different construction details were tested and analyzed based on four aspects: (a) the relationship between load and displacements, (b) strain distributions, (c) axial stiffness, and (d) crack patterns. Specimens were constructed using concrete with a compressive strength of 55 MPa, and the design strength of all five specimens were set to about 10,740 kN. From results of the experiment, the specimen with a reduced number of vertical reinforcements from 24 of HD22 (SD400, Fy = 400 MPa) to 16 of UD22 (SD600, Fy = 600 MPa) was the most effective specimen to improve the buildability of the column without deteriorating the structural performance of the reference specimen.     

Experimental behaviour of high‐strength concrete deep beams with web openings

This paper focuses on an experimental study undertaken on high‐strength concrete (HSC) deep beams with various opening sizes and locations on the web. The test covers a wide scope of variables that have not been investigated in previous research. Apart from highlighting the experimental setup, failure loads and typical crack patterns of the test specimens are also reported. Experimental results are then compared with predictions using currently available design methods. The comparison indicates that the predictions using current design methods can overly underestimate or sometimes overestimate the ultimate strength of these HSC deep beams. Further, the reduction of ultimate strengths due to the existence of web openings is not considered adequately in these design methods. To rectify the shortcomings of current design formulae, a new design equation is proposed and compared with the experimental results and those from previous studies on the related topics. The accuracy and reliability of the proposed new equation is subsequently confirmed. With the outcome of this work, more experimental tests with various opening configurations including shape and location of web openings are recommended for future study. Copyright © 2011 John Wiley & Sons, Ltd.     

Seismic performance and cost‐effectiveness of high‐rise buildings with increasing concrete strength

The relationship between the seismic performance and economics of high‐rise buildings when designed to different material strengths is investigated in this paper. To represent the modern high‐rise construction, five 60‐story reinforced concrete buildings with varying concrete strengths, ranging from 45 MPa to 110 MPa, are designed and detailed to fine accuracy keeping almost equal periods of vibration. Detailed fiber‐based simulation models are developed to assess the relative seismic performance of the reference structures using incremental dynamic analyses and fragility functions. It is concluded that a considerable saving in construction cost and gain in useable area are attained with increasing concrete strength. The safety margins of high‐strength concrete in tall structures may exceed those of normal‐strength concrete buildings, particularly at high ground motion intensity levels. The recommendations of this systematic study may help designers to arrive at cost‐effective designs for high‐rise buildings in earthquake‐prone regions without jeopardizing safety at different performance levels. Copyright © 2014 John Wiley & Sons, Ltd.     

混凝土强度预测的混合神经网络模型

针对由于混凝土配方的多样性,直接利用神经网络模型对其进行强度预测存在精度差和收敛差的缺点,采用多种神经网络模型相结合的方法,建立了混凝土强度预测的混合神经网络模型。模型的建立主要包括以下三个过程:(1)建立混凝土强度影响参数样本数据库;(2)利用自组织神经网络对样本数据分类;(3)以混凝土原材料和制作工艺为输入参数,混凝土最终性能为输出参数,训练双并联神经网络模型。通过实例验证了模型的可靠性。     

人工神经网络在掺活化煤矸石粉混凝土强度预测中的应用

人工神经网络技术综合考虑了掺活化煤矸石混凝土强度的各种影响因素,可用于预测混凝土强度.选取了掺活化煤矸石粉混凝土配料中7个主要因素作为输入值,混凝土28d强度作为输出值,建立起混凝土强度预测BP网络模型,进而对掺活化煤矸石配合比强度试验数据进行分析预测,效果良好.结果表明该方法用于掺矿物掺合料混凝土强度预测方面是可行的.     

基于人工神经网络的回弹-拔出综合法检测混凝土强度的研究

在试验研究的基础上,建立了混凝土强度回弹-拔出综合法检测的人工神经网络模型.探索并应用了神经网络的改进算法,其中包括附加冲量法、自适应学习算法及S型函数输出限幅算法等,以保证建立的神经网络的快速有效.与传统的回归算法相比,采用人工神经网络模型推测出的混凝土强度具有更高的精度.     

Artificial neural network model for steel–concrete bond prediction

In this paper, an Artificial Neural Network (ANN) is proposed for modelling the bond between conventional ribbed steel bars and concrete. The purpose is to predict the ultimate pull-out load from the concrete mix constituents (first ANN model) or the compressive strength (second ANN model) and from the steel bar diameter according to the RILEM test configuration [RILEM. Essai portant sur l’adhérence des armatures du béton: essai par traction. Materials and Structures 1970; 3 (3) 175–78]. The ANN models were implemented using an experimental database of 112 pull-out test results performed with ribbed bars 10 mm or 12 mm in diameter and three concrete mixes with different constituent proportions. A Multi-Layer-Perceptron was trained according to a back-propagation algorithm. The first model has six inputs (ANN-6): the diameter of the ribbed bar, the water to cement ratio, the gravel to sand ratio, the crushed to rolled gravel ratio, the type of cement and the concrete maturity. The second model has two inputs (ANN-2): the diameter of the bar and the concrete compressive strength. The ultimate pull-out load was the output data for both models.The results show that the implemented models have good prediction and generalisation capacity with low errors. The ANN-6 model is more accurate, regarding the generalisation capacity, than the ANN-2 model. Concrete mix constituents as input parameter, instead of the compressive strength, are more representative of the local phenomenon at the steel-ribs-to-concrete interface.     

Analytical compressive stress–strain model for concrete confined with high‐strength multiple‐tied‐spiral transverse reinforcement

The behavior of reinforced concrete members subjected to seismic loads is mainly based on the ultimate strength of concrete and its ductility. Based on this, an additional configuration of transverse reinforcement using high‐strength multiple‐tied‐spiral was proposed to improve the strength and ductility of concrete. In this paper, an experimental study of a number of axial loading tests on reinforced concrete columns confined with high‐strength multiple‐tied‐spiral transverse reinforcement is described. The effects of spacing of circular spiral and rectangular hoop, the confined area of circular spiral and concrete strength on axial behavior of confined concrete were investigated. The formulas of confined compressive strength and corresponding axial strain, factor to control the slope of descending branch, and stress in high‐strength circular spiral at confined strength are proposed based on the test results. A stress–strain model is also proposed that is found to give reasonably good prediction of the experimental behavior of reinforced concrete (RC) columns confined by high‐strength multiple‐tied‐spiral transverse reinforcement.     

Predicting the compressive strength and slump of high strength concrete using neural network

High Strength Concrete (HSC) is defined as concrete that meets special combination of performance and uniformity requirements that cannot be achieved routinely using conventional constituents and normal mixing, placing, and curing procedures. HSC is a highly complex material, which makes modelling its behavior very difficult task. This paper aimed to show possible applicability of neural networks (NN) to predict the compressive strength and slump of HSC. A NN model is constructed, trained and tested using the available test data of 187 different concrete mix-designs of HSC gathered from the literature. The data used in NN model are arranged in a format of seven input parameters that cover the water to binder ratio, water content, fine aggregate ratio, fly ash content, air entraining agent, superplasticizer and silica fume replacement. The NN model, which performs in Matlab, predicts the compressive strength and slump values of HSC. The mean absolute percentage error was found to be less then 1,956,208% for compressive strength and 5,782,223% for slump values and R² values to be about 99.93% for compressive strength and 99.34% for slump values for the test set. The results showed that NNs have strong potential as a feasible tool for predicting compressive strength and slump values.     

基于人工神经网络的混凝土抗冻性预测

简要介绍了MATLAB人工神经网络,阐述了人工神经网络预测模型的基本思想和模型特点。随着神经网络技术的不断完善和发展,人工神经网络广泛地用于土木工程的各个领域,可针对混凝土冻融与各种因素之间的相互关系,并且随着数据的不断积累,可在新样本基础上进行自学,形成较完整的评估预测系统。针对混凝土冻融后的相对动弹性模量的变化,应用BP网络建立了评估混凝土冻融后的性能的人工神经元网络模型,较好地解决了多影响因子的非线性映射问题,但真正用于工程实践还有许多问题需要进一步的研究解决。     

Theoretical analysis of post‐peak flexural behaviour of normal‐ and high‐strength concrete beams

A new method of analysing the post‐peak flexural behaviour of reinforced concrete beams has been developed and applied to normal‐ and high‐strength concrete beams. It was revealed that at the post‐peak stage the neutral axis depth keeps on increasing, and at a certain point the strain in the tension reinforcement starts to decrease, even though the curvature is increasing monotonically. Such strain reversal in the tension reinforcement occurs in all concrete beams and has significant effects on the post‐peak behaviour and flexural ductility of concrete beams. Therefore, the stress path dependence of the tension reinforcement needs to be taken into account in the analysis. By means of a parametric study, the variation of ultimate concrete strain with tension steel ratio and the effects of various structural parameters on flexural ductility have been studied. Based on the numerical results, design values of ultimate concrete strain that are independent of tension steel ratio have been recommended and a simple formula for predicting the flexural ductility of reinforced normal‐ and high‐strength concrete beams has been developed. Copyright © 2003 John Wiley & Sons, Ltd.     

Modelling criteria for r/c elements constructed with ultra‐high‐strength concrete and subjected to cyclic loading

High‐strength concrete (HSC) has several benefits in high‐rise concrete buildings; however, its structural use in active seismic regions may be questioned due to the lower ductility of such concretes. In addition, seismic macro‐models being used currently are based on R/C elements constructed with normal‐strength concretes (f ≤ 40 MPa (5.8 ksi)). In this paper, the performance of plastic hinges in flexural elements constructed with a concrete strength up to 175 MPa (25.4 ksi) is investigated. In addition, other variables were studied such as the sectional reinforcement asymmetry, hinge shear strength and hinge shear demand. The seismic performance is presented in terms of stiffness deterioration, strength degradation, pinching phenomenon and displacement ductility. The requirements to implement the use of HSC in current macro‐models are examined. Current hysteretic models may be used to evaluate structural components constructed with higher concrete strength; however, the influence of concrete strength is controlled by the other test variables. Copyright © 2009 John Wiley & Sons, Ltd.     

泡沫混凝土强度的影响因素及质量控制

介绍了泡沫混凝土的制备原理，分析了影响泡沫混凝土强度的因素，提出了提高泡沫混凝土强度的途径，指出泡沫混凝土是一种利废、环保、节能、低廉且轻质高强、不吸水、不燃性的新型建筑节能材料。     

基于RBF神经网络方法的混凝土强度预测

在分析普通混凝土强度各影响因素的基础上,选取6个影响因素组成输入层,以混凝土28 d强度作为输出,建立径向基函数网络,经网络训练和仿真结果对比,表明所建网络结构合理、收敛速度快、精度高,可以满足普通混凝土强度预测要求,具有广泛的应用前景。     

泡沫混凝土的制备及强度影响研究

自行研制了泡沫混凝土的松香皂发泡剂,通过大量试验确定了泡沫混凝的最佳配合比,配制出密度为450kg/m3,强度达到1.3MPa左右的泡沫混凝土。分别对石灰、玻璃纤维、三乙醇胺对增加泡沫混凝土强度影响进行了研究,试验表明三乙醇胺对增加泡沫混凝土强度影响较大。     

Simulation of foamed concrete compressive strength prediction using adaptive neuro-fuzzy inference system optimized by nature-inspired algorithms

Concrete compressive strength prediction is an essential process for material design and sustainability. This study investigates several novel hybrid adaptive neuro-fuzzy inference system (ANFIS) evolutionary models, i.e., ANFIS–particle swarm optimization (PSO), ANFIS–ant colony, ANFIS–differential evolution (DE), and ANFIS–genetic algorithm to predict the foamed concrete compressive strength. Several concrete properties, including cement content (C), oven dry density (O), water-to-binder ratio (W), and foamed volume (F) are used as input variables. A relevant data set is obtained from open-access published experimental investigations and used to build predictive models. The performance of the proposed predictive models is evaluated based on the mean performance (MP), which is the mean value of several statistical error indices. To optimize each predictive model and its input variables, univariate (C, O, W, and F), bivariate (C–O, C–W, C–F, O–W, O–F, and W–F), trivariate (C–O–W, C–W–F, O–W–F), and four-variate (C–O–W–F) combinations of input variables are constructed for each model. The results indicate that the best predictions obtained using the univariate, bivariate, trivariate, and four-variate models are ANFIS–DE– (O) (MP= 0.96), ANFIS–PSO– (C–O) (MP= 0.88), ANFIS–DE– (O–W–F) (MP= 0.94), and ANFIS–PSO– (C–O–W–F) (MP= 0.89), respectively. ANFIS–PSO– (C–O) yielded the best accurate prediction of compressive strength with an MP value of 0.96.