Burst detection using hydraulic data from water distribution systems with artificial neural networks

This paper presents research into the application of artificial neural networks (ANNs) for analysis of data from sensors measuring hydraulic parameters (flow and pressure) of the water flow in treated water distribution systems. Two neural architectures (static and time delay) are applied for time series pattern classification from the perspective of detecting leakage. Results are presented using data from an experimental site in a distribution system of a UK water company in which bursts were simulated by hydrant flushing. Field trials have shown how ANNs can be used effectively for a leakage detection task. Both static and time delay ANNs learned patterns of leaks/bursts. The time delay neural network showed improved performance over the static network. It is concluded that the effectiveness of an ANN in discovering relationships within the data is dependent upon two key factors: availability of sufficient exemplars and data quality.     

Backfilling missing microbial concentrations in a riverine database using artificial neural networks

Predicting peak pathogen loadings can provide a basis for watershed and water treatment plant management decisions that can minimize microbial risk to the public from contact or ingestion. Artificial neural network models (ANN) have been successfully applied to the complex problem of predicting peak pathogen loadings in surface waters. However, these data-driven models require substantial, multiparameter databases upon which to train, and missing input values for pathogen indicators must often be estimated. In this study, ANN models were evaluated for backfilling values for individual observations of indicator bacterial concentrations in a river from 44 other related physical, chemical, and bacteriological data contained in a multi-year database. The ANN modeling approach provided slightly superior predictions of actual microbial concentrations when compared to conventional imputation and multiple linear regression models. The ANN model provided excellent classification of 300 randomly selected, individual data observations into two defined ranges for fecal coliform concentrations with 97% overall accuracy. The application of the relative strength effect (RSE) concept for selection of input variables for ANN modeling and an approach for identifying anomalous data observations utilizing cross validation with ANN model are also presented.     

Modeling construction processes using artificial neural networks

The paper evaluates a neural network approach to modeling the dynamics of construction processes that exhibit both discrete and stochastic behavior, providing an alternative to the more conventional method of discrete-event simulation. The incentive for developing the technique is its potential for (i) facilitating model development in situations where there is limited theory describing the dependence between component processes; and (ii) rapid execution of a simulation through parallel processing. The alternative ways in which neural networks can be used to model construction processes are reviewed and their relative merits are identified. The most promising approach, a recursive method of dynamic modeling, is examined in a series of experiments. These involve the application of the technique to two classes of earthmoving system, the first comprising a push-dozer and a fleet of scrapers, and the second a loader and fleet of haul trucks. The viability of the neural network approach is demonstrated in terms of its ability to model the discrete and stochastic behavior of these classes of construction processes. The paper concludes with an indication of some areas for further development of the technique.     

基于人工神经网络的混凝土抗渗性能预测

在进行了正交试验的基础上,采用人工神经网络方法,建立混凝土的氯离子扩散系数与混凝土配比六个参数之间的非线性映射关系,研究各个参数对混凝土抗渗性能的影响,该研究成果可以减少混凝土试配次数,节约大量的人力、物力和时间,为高性能混凝土的研究发展奠定了基础。     

Prediction of blast-induced ground vibration using artificial neural networks

Blasting is still being considered to be one the most important applicable alternatives for conventional tunneling. Ground vibration generated due to blasting is an undesirable phenomenon which is harmful for the nearby habitants and dwellings and should be prevented. In this paper, an attempt has been made to predict blast-induced ground vibration using artificial neural network (ANN) in the Siahbisheh project, Iran. To construct the model maximum charge per delay, distance from blasting face to the monitoring point, stemming and hole depth are taken as input parameters, whereas, peak particle velocity (PPV) is considered as an output parameter. A database consisting of 182 datasets was collected at different strategic and vulnerable locations in and around the project. From the prepared database, 162 datasets were used for the training and testing of the network, whereas 20 randomly selected datasets were used for the validation of the ANN model. A four layer feed-forward back-propagation neural network with topology 4-10-5-1 was found to be optimum. To compare performance of the ANN model with empirical predictors as well as regression analysis, the same database was applied. Superiority of the proposed ANN model over empirical predictors and statistical model was examined by calculating coefficient of determination for predicted and measured PPV. Sensitivity analysis was also performed to get the influence of each parameter on PPV. It was found that distance from blasting face is the most effective and stemming is the least effective parameter on the PPV.     

Prediction of seismic-induced structural damage using artificial neural networks

Contemporary methods for estimating the extent of seismic-induced damage to structures include the use of nonlinear finite element method (FEM) and seismic vulnerability curves. FEM is applicable when a small number of predetermined structures is to be assessed, but becomes inefficient for larger stocks. Seismic vulnerability curves enable damage estimation for classes of similar structures characterised by a small number of parameters, and typically use only one parameter to describe ground motion. Hence, they are unable to extend damage prognosis to wider classes of structures, e.g. buildings with a different number of storeys and/or bays, or capture the full complexity of the relationship between damage and seismic excitation parameters. Motivated by these shortcomings, this study presents a general method for predicting seismic-induced damage using Artificial Neural Networks (ANNs). The approach was to describe both the structure and ground motion using a large number of structural and ground motion properties. The class of structures analysed were 2D reinforced concrete (RC) frames that varied in topology, stiffness, strength and damping, and were subjected to a suite of ground motions. Dynamic structural responses were simulated using nonlinear FEM analysis and damage indices describing the extent of damage calculated. Using the results of the numerical simulations, a mapping between the structural and ground motion properties and the damage indices was than established using an ANN. The performance of the ANN was assessed using several examples and the ANN was found to be capable of successfully predicting damage.     

Estimation of strength parameters of rock using artificial neural networks

The accurate determination of geomechanical properties such as uniaxial compressive strength and shear strength requires considerable time in collecting appropriate samples, their preparation and laboratory testing. To minimize the time and cost, a number of empirical relations have been reported which are widely used for the estimation of complex rock properties from more easily acquired data. This paper reports the use of an artificial neural network to predict the deformation properties of Coal Measure rocks using dynamic wave velocity, point load index, slake durability index and density. The results confirm the applicability of this method.     

Modelling optimal risk allocation in PPP projects using artificial neural networks

This paper aims to establish, train, validate, and test artificial neural network (ANN) models for modelling risk allocation decision-making process in public-private partnership (PPP) projects, mainly drawing upon transaction cost economics. An industry-wide questionnaire survey was conducted to examine the risk allocation practice in PPP projects and collect the data for training the ANN models. The training and evaluation results, when compared with those of using traditional MLR modelling technique, show that the ANN models are satisfactory for modelling risk allocation decision-making process. The empirical evidence further verifies that it is appropriate to utilize transaction cost economics to interpret risk allocation decision-making process. It is recommended that, in addition to partners' risk management mechanism maturity level, decision-makers, both from public and private sectors, should also seriously consider influential factors including partner's risk management routines, partners' cooperation history, partners' risk management commitment, and risk management environmental uncertainty. All these factors influence the formation of optimal risk allocation strategies, either by their individual or interacting effects.     

Analysis of durability of high performance concrete using artificial neural networks

This study aims to determine the influence of the content of water and cement, water–binder ratio, and the replacement of fly ash and silica fume on the durability of high performance concrete (HPC) by using artificial neural networks (ANNs). To achieve this, an ANNs model is developed to predict the durability of high performance concrete which is expressed in terms of chloride ions permeability in accordance with ASTM C1202-97 or AASHTO T277. The model is developed, trained and tested by using 86 data sets from experiments as well as previous researches. To verify the model, regression equations are carried out and compared with the trained neural network. The results indicate that the developed model is reliable and accurate. Based on the simulating durability model built using trained neural networks, the optimum cement content for designing HPC in terms of durability is in the range of 450–500 kg/m³. The results also revealed that the durability of concrete expressed in terms of total charge passed over a 6-h period can be significantly improved by using at least 20% fly ash to replace cement. Furthermore, it can be concluded that increasing silica fume results in reducing the chloride ions penetrability to a higher degree than fly ash. This study also illustrates how ANNs can be used to beneficially predict durability in terms of chloride ions permeability across a wide range of mix proportion parameters of HPC.     

Mix proportions for HPC incorporating multi-cementitious composites using artificial neural networks

High performance concrete (HPC) is defined in terms of both strength and durability performance under anticipated environmental conditions. HPC can be manufactured involving up to 10 different ingredients whilst having to consider durability properties in addition to strength. The number of ingredients and the number of properties of HPC, which needs to be considered in its design, are more than those for ordinary concrete. Therefore, it is difficult to predict the mix proportions and other properties of this type of concrete using statistical empirical relationship. An alternative approach is to use an artificial neural network (ANN). Based on the experimentally obtained results, ANN has been used to establish its applicability to the prediction and optimization of mix proportioning for HPC. It was demonstrated that mix proportioning for HPC can be predicted using ANN. However, some trial mixes are necessary for better performance and elimination of material variability factors from place to place. ANN procedure provides guidelines to select appropriate material proportions for required strength and rheology of concrete mixes and will reduce the number of trial mixes.     

Hierarchical community classification and assessment of aquatic ecosystems using artificial neural networks

Benthic macroinvertebrate communities in stream ecosystems were assessed hierarchically through two-level classification methods of unsupervised learning. Two artificial neural networks were implemented in combination. Firstly, the self-organizing map (SOM) was used to reduce the dimension of community data, and secondly, the adaptive resonance theory (ART) was subsequently applied to the SOM to further classify the groups in different scales. Hierarchical grouping in community data efficiently reflected the impact of the environmental factors such as topographic conditions, levels of pollution, and sampling location and time across different scales. New community data not included in the training process were used to test the trained network model. The input data were appropriately grouped at different hierarchical levels by the trained networks, and correspondingly revealed the impact of environmental disturbances and temporal dynamics of communities. The hierarchical clusters based on a two-level classification method could be useful for assessing ecosystem quality and community variations caused by environmental disturbances.     

Non-destructive evaluation of concrete physical condition using radar and artificial neural networks

This paper deals with the combination of radar technology and artificial neural networks (ANN) for the non-destructive evaluation of the water and chloride contents of concrete. Two networks were trained and tested to predict these concrete properties. Input data to the statistical models were extracted from time-domain signals of direct and reflected radar waves. ANN training and testing were implemented according to an experimental database of 100 radar measurements performed on concrete slabs having various water and chloride contents. Both networks were multi-layer-perceptrons trained according to back-propagation algorithm.The results of this research highlight the potential of artificial neural networks for solving the inverse problem of concrete physical evaluation using radar measurements. It was found that the optimized statistical models predicted water and chloride contents of concrete laboratory slabs with maximum absolute errors of about 2% and 0.5 kg/m³ of concrete, respectively.     

Predicting the average size of blasted rocks in aggregate quarries using artificial neural networks

The prediction of the average size of fragments in blasted rock piles produced after blasting in aggregate quarries is essential for decresing the cost of crushing and secondary breaking. There are several conventional and advanced processes to estimate the size of blasted rocks. Among these, the empirical prediction of the expected fragmentation in most cases is carried out by Kuznetsov’s equation (Sov Min Sci 9:144–148, 1973), modified by Lilly (1986) and Cunningham (1987). The present research focuses on the effect of the engineering geological factors and blasting process on the blasted fragments using a more powerful, advanced computational tool, an artificial neural network. In particular, the blast database consists of the blastability index of limestone on the pit face, the quantities of the explosives and of the blasted rock pile, assessing the interaction of these parameters on the blasted rocks. The data were collected from two aggregate quarries, Drymos and Tagarades, near Thessaloniki, in the Central Macedonia region of Greece. This approach indicates significant performance stability, providing the fragmentation size with high accuracy.

     

一类大规模系统神经网络自适应控制方案设计

针对一类大规模非线性机械系统,如自动化高速公路汽车空位调节和连续搅拌反应釜等系统,设计了一套基于前向神经网络的分散自适应控制方案以实现对其有效控制．作为直接自适应控制器,神经网络被用于逼近未知函数,所设计的两个鲁棒控制项分别用于消除互连效应和干扰项．系统稳定性得到严格证明,通过仿真进一步验证了方案的有效性．     

Prediction of pressure coefficients on roofs of low buildings using artificial neural networks

This paper describes an artificial neural network (ANN) approach for the prediction of mean and root-mean-square (rms) pressure coefficients on the gable roofs of low buildings. The ANN models, which employ a backpropagation training algorithm, are capable of generalizing the complex, nonlinear functional relationships between the pressure coefficients and eave height, wind direction and spatial location on the roof. The performance of the ANN is demonstrated by the prediction of the pressure coefficients for roof tap locations in a corner bay. The mean bay uplift can be predicted accurately with an average error less than 2% for three cornering wind directions not seen by the ANN during training. The mean-square errors of all of the individual pressure taps in the corner bay were 12% and 9% for the mean and rms coefficients, respectively. This approach could be used to expand aerodynamic databases to a larger variety of geometries and increase its practical feasibility.     

Modeling and optimization of HVAC systems using artificial neural network and genetic algorithm

Intelligent energy management and control system (EMCS) in buildings offers an excellent means of reducing energy consumptions in HVAC systems while maintaining or improving indoor environmental conditions. This can be achieved through the use of computational intelligence and optimization. The paper thus proposes and evaluates a model-based optimization process for HVAC systems using evolutionary algorithm for optimization and artificial neural networks for modeling. The process can be integrated into the EMCS to perform several intelligent functions and achieve optimal whole-system performance. The proposed models and the optimization process are tested using data collected from an existing HVAC system. The testing results show that the models can capture very well the system performance, and the optimization process can reduce cooling energy consumption by about 11% when compared to the traditional operating strategies applied.     

Quantification and localisation of damage in beam-like structures by using artificial neural networks with experimental validation

This paper presents a damage detection algorithm using a combination of global (changes in natural frequencies) and local (curvature mode shapes) vibration-based analysis data as input in artificial neural networks (ANNs) for location and severity prediction of damage in beam-like structures. A finite element analysis tool has been used to obtain the dynamic characteristics of intact and damaged cantilever steel beams for the first three natural modes. Different damage scenarios have been introduced by reducing the local thickness of the selected elements at different locations along finite element model (FEM) of the beam structure. The necessary features for damage detection have been selected by performing sensitivity analyses and different input–output sets have been introduced to various ANNs. In order to check the robustness of the input used in the analysis and to simulate the experimental uncertainties, artificial random noise has been generated numerically and added to noise-free data during the training of the ANNs. In the experimental analysis, two steel beams with eight distributed surface-bonded electrical strain gauges and an accelerometer mounted at the tip have been used to obtain modal parameters such as resonant frequencies and strain mode shapes. Finally, trained feed-forward backpropagation ANNs have been tested using the data obtained from the experimental damage case for quantification and localisation of the damage.     

Prediction of strength properties of some schistose rocks from petrographic properties using artificial neural networks

Petrographic features of a rock are intrinsic properties, which control the mechanical behaviour of the rock mass at the fundamental level. This paper deals with the application of neural networks for the prediction of uniaxial compressive strength, tensile strength and axial point load strength simultaneously from the mineral composition and textural properties. Statistical analysis has also been conducted for prediction of the same strength properties and compared with the predicted values by neural networks to investigate the authenticity of this approach. The network was trained to predict the uniaxial compressive strength, tensile strength and axial point load strength from the mineralogical composition, grain size, aspect ratio, form factor, area weighting and orientation of foliation planes (planes of weakness). A data set having 112 test results of the four schistose rocks were used to train the network with the back-propagation learning algorithm. Another data set of 28 test results of the four schistose rocks were used to validate the generalization and prediction capabilities of the network.