Basis Of Artificial Neural Network Research Articles

Daily Reservoir Runoff Forecasting Method Using Artificial Neural Network Based on Quantum-behaved Particle Swarm Optimization

Accurate daily runoff forecasting is of great significance for the operation control of hydropower station and power grid. Conventional methods including rainfall-runoff models and statistical techniques usually rely on a number of assumptions, leading to some deviation from the exact results. Artificial neural network (ANN) has the advantages of high fault-tolerance, strong nonlinear mapping and learning ability, which provides an effective method for the daily runoff forecasting. However, its training has certain drawbacks such as time-consuming, slow learning speed and easily falling into local optimum, which cannot be ignored in the real world application. In order to overcome the disadvantages of ANN model, the artificial neural network model based on quantum-behaved particle swarm optimization (QPSO), ANN-QPSO for short, is presented for the daily runoff forecasting in this paper, where QPSO was employed to select the synaptic weights and thresholds of ANN, while ANN was used for the prediction. The proposed model can combine the advantages of both QPSO and ANN to enhance the generalization performance of the forecasting model. The methodology is assessed by using the daily runoff data of Hongjiadu reservoir in southeast Guizhou province of China from 2006 to 2014. The results demonstrate that the proposed approach achieves much better forecast accuracy than the basic ANN model, and the QPSO algorithm is an alternative training technique for the ANN parameters selection.

Informational Modeling of the Behavior of a Teacher in the Learning Process Based on Fuzzy Logic

The purpose of this paper is to develop rules, knowledge, and algorithms that will engender a real shift in most of the educational process on the computer environment. On the basis of artificial neural networks, this work is a mechanism that allows taking into account the subjective opinion of experts without requiring changes in the system. With the help of fuzzy logic, this paper offers an original method of monitoring a student's knowledge by closely mimicking the behavior of the teacher in the student survey, which combines the power and brevity, not previously accessible to automated systems.

INTELLECTUAL MODEL FORMATION OF RAILWAY STATION WORK DURING THE TRAIN OPERATION EXECUTION

Purpose . The aim of this research work is to develop an intelligent technology for determination of the optimal route of freight trains administration on the basis of the technical and technological parameters. This will allow receiving the operational informed decisions by the station duty officer regarding to the train operation execution within the railway station. Metodology . The main elements of the research are the technical and technological parameters of the train station during the train operation. The methods of neural networks in order to form the self-teaching automated system were put in the basis of the generated model of train operation execution. Findings . The presented model of train operation execution at the railway station is realized on the basis of artificial neural networks using learning algorithm with a «teacher» in Matlab environment. The Matlab is also used for the immediate implementation of the intelligent automated control system of the train operation designed for the integration into the automated workplace of the duty station officer. The developed system is also useful to integrate on workplace of the traffic controller. This proposal is viable in case of the availability of centralized traffic control on the separate section of railway track. Originality. The model of train station operation during the train operation execution with elements of artificial intelligence was formed. It allows providing informed decisions to the station duty officer concerning a choice of rational and a safe option of reception and non-stop run of the trains with the ability of self-learning and adaptation to changing conditions. This condition is achieved by the principles of the neural network functioning . P ractical valu e. The model of the intelligent system management of the process control for determining the optimal route reception for different categories of trains was formed. I n the operational mode it offers the possibility to the station duty officer or the traffic controller to determine the appropriate park (receiving, sending, transit one ) and efficient reception way o r handling one on condition of train safety control . The cardinal difference of this technology from the existing ones is the possibility to adapt the model to changing conditions. It means that in case of a situation that had not been encountered previously the model will calculate the most efficient way of train operation execution .

Application of Neural Network Models in Modelling Economic Time Series with Non-constant Volatility

Abstract In this paper, we investigate the volatility dynamics of EUR/GBP currency using statistical as well as the neural network approach which is an alternative way for time series modelling and forecasting in economics. The goal of this paper is to provide an alternative and reasonable way in modelling dynamic economic time series. We suggest an alternative approach for forecasting time series with non-constant volatility – we suggest and implement several neural network prediction models; we also use a large number of statistical models as well as different optimization techniques for artificial neural network. After discussing the basics of statistical volatility modelling and the basis of artificial neural networks we perform the experiment on real financial data. We quantify several ARCH and GARCH models; we also implement various RBF neural network prediction models. The comparative analysis of out-of-sample forecasts evaluated using MSE evaluation measures is performed. Finally, we state that suggested neural network models performed almost as good as the standard statistical models and are therefore reasonable and acceptable in economic modelling.

신경망 알고리즘을 이용한 차체용 강판 아크 용접 조건 도출

Famous artificial neural network (ANN) is applied to predict proper process window of arc welding. Target weldment is variously combined lap joint fillet welding of automotive steel plates. ANN's system variable such as number of hidden layers, perceptrons and transfer function are carefully selected through case by case test. Input variables are welding condition and steel plate combination, for example, welding machine type, shield gas composition, current, speed and strength, thickness of base material. The number of each input variable referred in welding experiment is counted and provided to make it possible to presume the qualitative precision and limit of prediction. One of experimental process windows is excluded for predictability estimation and the rest are applied for neural network training. As expected from basic ANN theory, experimental condition composed of frequently referred input variables showed relatively more precise prediction while rarely referred set showed poorer result. As conclusion, application of ANN to arc welding process window derivation showed comparatively practical feasibility while it still needs more training for higher precision.

Extreme learning machine for the displacement prediction of landslide under rainfall and reservoir level

Landslide prediction is always the emphasis of landslide research. Using global positioning system GPS technologies to monitor the superficial displacements of landslide is a very useful and direct method in landslide evolution analysis. In this paper, an EEMD–ELM model [ensemble empirical mode decomposition (EEMD) based extreme learning machine (ELM) ensemble learning paradigm] is proposed to analysis the monitoring data for landslide displacement prediction. The rainfall data and reservoir level fluctuation data are also integrated into the study. The rainfall series, reservoir level fluctuation series and landslide accumulative displacement series are all decomposed into the residual series and a limited number of intrinsic mode functions with different frequencies from high to low using EEMD technique. A novel neural network technique, ELM, is employed to study the interactions of these sub-series at different frequency affecting landslide occurrence. Each sub-series extracted from accumulative displacement of landslide is forecasted respectively by establishing appropriate ELM model. The final prediction result is obtained by summing up the calculated predictive displacement value of each sub. The EEMD–ELM model shows the best accuracy comparing with basic artificial neural network models through forecasting the displacement of Baishuihe landslide in the Three Gorges reservoir area of China.

Intrusion Detection in MANET using Neural Networks and ZSBT

ad-hoc network is a collection of mobile nodes that organize themselves into a network without any predefined infrastructure. The characteristics of MANET are dynamic topology; bandwidth and energy constrained and limited physical security. Due to the dynamic nature of the network, these networks can be easily vulnerable to attacks. Many type of attacks can threat the MANET and the classification of attacks are Black hole, Routing loops, Network partition, Selfishness, Sleep deprivation and Denial of Service. The MANET is mainly applied in military environments, personal area networking and emergency operations. This work provides a technique to improve the security level of MANET. The mechanism of intrusion detection is designed in MANET on the basis of artificial neural networks (ANNs) and Zone- Sampling Based Trace back algorithm (ZSBT) for detecting DOS attacks. Intrusion detection system is a type of security management system for computers and networks. The prevention mechanisms are thwarted by the ability of attackers to forge, or spoof, the source addresses in IP packets. By employing IP spoofing, attackers can evade from detection. But with the help of IDPF, the attacker node can be identified easily and it also has the ability to limit the IP spoofing capability. Artificial neural network and ZSBT modeling uses simulated MANET environments for detecting nodes under DOS attack effectively. But the ZSBT method works well for detecting DoS attacks in the network when compared with ANN methods.

Study on Correlation between Urban Development and Air Pollution Based on Artificial Neural Network

The paper presents a method of researching the impact of urban development on air quality on the basis of artificial neural network (ANN). Statistical data in a monitoring period constitute a sample which contains monitoring values of environmental impact factors and air pollution indicators. Several samples are employed to train the ANN, and the mapping relationship between environmental impact factors and air pollution indicators is established through the trained ANN. The impact degree of each environmental impact factor on each air pollution indicator can be obtained by using the connection weights of the trained ANN. The case study illustrates the feasibility of the method mentioned in the paper which explores a new idea to the study of environmental impact of urban development.

ANNIE—Artificial Neural Network-based Image Encoder

Data and especially image compression is becoming increasingly important for efficient resource utilization. Many digital image file formats therefore include universally usable compression methods. They treat every image separately and do not profit from a larger image data set's similar image contents, which are present in numerous biomedical applications. This situation provided the impetus to develop and implement a technical system that incorporates a priori information on typical image contents in image compression on the basis of artificial neural networks and thus increases compression performance for larger image data sets with frequently recurring image contents.

Application of Artificial Neural Networks (ANNs) in Wine Technology

In recent years, neural networks have turned out as a powerful method for numerous practical applications in a wide variety of disciplines. In more practical terms neural networks are one of nonlinear statistical data modeling tools. They can be used to model complex relationships between inputs and outputs or to find patterns in data. In food technology artificial neural networks (ANNs) are useful for food safety and quality analyses, predicting chemical, functional and sensory properties of various food products during processing and distribution. In wine technology, ANNs have been used for classification and for predicting wine process conditions. This review discusses the basic ANNs technology and its possible applications in wine technology.

Displacement prediction model of landslide based on a modified ensemble empirical mode decomposition and extreme learning machine

In this paper, an M–EEMD–ELM model (modified ensemble empirical mode decomposition (EEMD)-based extreme learning machine (ELM) ensemble learning paradigm) is proposed for landslide displacement prediction. The nonlinear original surface displacement deformation monitoring time series of landslide is first decomposed into a limited number of intrinsic mode functions (IMFs) and one residual series using EEMD technique for a deep insight into the data structure. Then, these sub-series except the high frequency are forecasted, respectively, by establishing appropriate ELM models. At last, the prediction results of the modeled IMFs and residual series are summed to formulate an ensemble forecast for the original landslide displacement series. A case study of Baishuihe landslide in the Three Gorges reservoir area of China is presented to illustrate the capability and merit of our model. Empirical results reveal that the prediction using M–EEMD–ELM model is consistently better than basic artificial neural networks (ANNs) and unmodified EEMD–ELM in terms of the same measurements.

Introducing a new formula based on an artificial neural network for prediction of droplet size in venturi scrubbers

Droplet size is a fundamental parameter for Venturi scrubber performance. For many years, the correlations proposed by Nukiyama and Tanasawa (1938) and Boll et al. (1974) were used for calculating mean droplet size in Venturi scrubbers with limited operating parameters. This study proposes an alternative approach on the basis of artificial neural networks (ANNs) to determine the mean droplet size in Venturi scrubbers, in a wide range of operating parameters. Experimental data were used to design the ANNs. A neural network was trained based on the liquid to gas ratio (L/G) and throat gas velocity (Vgth), as input parameters, and the Sauter mean diameter (D32) as the desired parameter. The back-propagation learning algorithms were used in the network and the best approach was found. A new formula for the prediction of D32 using the weights of the network was then generated. This formula predicts mean droplet size in Venturi scrubbers more accurately than the correlations of Boll et al. (1974) and Nukiyama and Tanasawa (1938). The Average Absolute Percent Deviation (AAPD) of our formula and the Boll et al. and Nukiyama and Tanasawa correlations for the full ranges of experimental data are 26.04%, 40.19% and 32.99%, respectively.

Color-texture image segmentation and recognition through a biologically-inspired architecture

In this paper, we present a multiple scale neural architecture for recognizing colored and textured images. The description and results of the architecture are preceded by an explanation of the Human Visual System, bio-inspired models for image segmentation, and ART-based neural networks used in image recognition. The presented architecture achieved good segmentation and recognition results in comparison to other methods.

Analysis of experimental data sets for local scour depth around bridge abutments using artificial neural networks

The performance of soft computing techniques to analyse and interpret the experimental data of local scour depth around bridge abutment, measured at different laboratory conditions and environment, is presented. The scour around bridge piers and abutments is, in the majority of cases, the main reason for bridge failures. Therefore, many experimental and theoretical studies have been conducted on this topic. This study sought to answer the following questions: Firstly, can data collected by different researchers at different times be combined in one data set? Secondly, can we determine any unquantified effects such as data differences, laboratory conditions and measurement devices? Artificial neural networks (ANN) are used and a basic ANN model is selected to observe the application problems, in order to avoid any misleading conclusion arising due to the model parameters selected and the compilation of different subsets of experimental data into one set. At the first stage, seven experimental data sets are compiled to address the first question and an ANN model is used to discovery any existing discrepancies between available data groups. The importance of selected model parameters for the model’s performance was demonstrated by increasing the number of parameters. Then, each data subset was inspected to expose the importance of the homogeneity of data groups in order to obtain a best-fit ANN model. Finally, a sensitivity analysis was carried out to obtain the dominant parameters of the problem. It was concluded that the use of ‘soft’ computational techniques such as ANN can be beneficial, provided the user is aware of the heterogeneity of the data set and the physical context of the subject or problem being addressed. However, as with other data analysis techniques, elaborate inspection of data and results is required. Keywords: Scour prediction methods, bridge abutment scour, soft computing techniques, ANN

Temperature variation effect compensation in impedance-based structural health monitoring using neural networks

In this article, a new method for temperature compensation on the basis of artificial neural networks (ANNs) in impedance-based structural health monitoring (ISHM) has been introduced. ISHM using piezoelectric wafer active sensors (PWAS) has been extensively developed to provide detection of fault in structure. The principle of this method is based on the electromechanical coupling effect of PWAS materials. Any change in structure leads to changes in mechanical impedance of structure. The electrical impedance of PWAS can sense this change by the electromechanical coupling effect of PWAS. Therefore, the difference in this electrical impedance for undamaged and damaged structures can be considered as a damage index to detect the damage in structure. Since physical and mechanical properties of structure also PWAS materials are temperature dependent, so this electrical impedance of PWAS will be affected by temperature changes. Consequently, the variation in environmental or service temperatures can be detected erroneously as damage in ISHM method. In this article, a new method using ANN based on radial basis function (RBF) has been proposed and developed to compensate the temperature effect on the damage index. A steel plate and gas pipe with bolted joints are considered as two case studies for the performance evaluation of the proposed fault detection methodology. Results confirm that the proposed method using the ANN can be effectively utilized to compensate temperature variation for damage detection in different structures.

Identification of crack-like defects in elastic structural elements on the basis of evolution algorithms

A method for the identification of defects in elastic structural elements (SEs) on the basis of artificial neural network (ANN) tools was proposed. The problems of constructing a simulation model of an object, the optimal arrangement of transducers with the use of a genetic algorithm, and the subsequent identification of a defect were solved. The application of different learning architectures and algorithms to feed-forward neural networks (FFNs) was studied and the influence of errors on the estimation accuracy of the parameters of a defect was analyzed.

Integration of supervised ART-based neural networks with a hybrid genetic algorithm

In this paper, two evolutionary artificial neural network (EANN) models that are based on integration of two supervised adaptive resonance theory (ART)-based artificial neural networks with a hybrid genetic algorithm (HGA) are proposed. The search process of the proposed EANN models is guided by a knowledge base established by ART with respect to the training data samples. The EANN models explore the search space for “coarse” solutions, and such solutions are then refined using the local search process of the HGA. The performances of the proposed EANN models are evaluated and compared with those from other classifiers using more than ten benchmark data sets. The applicability of the EANN models to a real medical classification task is also demonstrated. The results from the experimental studies demonstrate the effectiveness and usefulness of the proposed EANN models in undertaking pattern classification problems.

Assessment of structural sensitivity on the basis of artificial neural networks

AbstractIn engineering practice, the assessment of sensitivity is utilized to detect influential parameters in order to facilitate subsequent numerical simulation techniques. As sensitivity analyses are preprocessing methods for sophisticated numerical simulation techniques, e.g. reliability based optimization procedures, their application is always linked to an increase of the computational expense. In result, it is reasonable to couple sensitivity analysis and artificial neural networks (ANN). Therefore, multi‐faceted global sensitivity measures (GSM) may be formulated, taking advantage of different characteristics of the ANNs. Additionally, to take into account nonlinearities of the response surface, a new approach of sectional global n sensitivity measures is introduced. Generally, the sensitivity can be determined with $S_{i}=\hat{S}_{i}/\sum\limits_{j=1}^{n}\hat{S}_{j}$. Thereby, $S_{i}$ denotes the sensitivity of interest and $\hat{S}_{i}$ a characteristic of the function $f:\mathcal{R}^n\rightarrow\mathcal{R}$ under investigation. This can be either the response $f$ itself or the first partial derivative thereof $\partial_{i}f$. (© 2010 Wiley‐VCH Verlag GmbH & Co. KGaA, Weinheim)

Automatic model decomposition and reuse in an evolutionary cognitive mechanism

This paper addresses the problem of automatically obtaining primitives of the models an evolutionary cognitive mechanism is producing for a robot through its real time interaction with the world. The models are instantiated as Artificial Neural Networks (ANNs) and the objective is to obtain ANNs that cooperate in the process of modelling complex functions. An algorithm where the combination of networks takes place at the phenotypic or functional level is proposed. Thus, a population of networks that are automatically classified into different species depending on the performance of their phenotype is evolved, and individuals from each species cooperate forming a group to obtain a complex output. The components that make up the groups are basic ANNs (primitives) and may be reused in other modelling processes as seeds or combined to generate new solutions. The parameter that reflects the difference between ANNs is their affinity vector, the value which is automatically created and modified for each ANN through a competition based clustering process within the evolutionary process. The main objective of this approach is to explore one path to gradually model complex functions similar to those that arise when obtaining environment or internal models within robotic cognitive systems.

Using artificial neural networks to predict potential complications during trauma patients' hospitalization period.

Complications during treatment of seriously injured trauma patients cause an increase in mortality rates, and increased treatment costs, including bed occupancy. Current methods treat those at risk, and include numbers of false positives. By finding a method to predict those at risk of the three most common recorded Trauma Registry complications, considerable savings in mortality and treatment costs could arise. Artificial Neural Networks (ANN) work well with classification problems using feed-forward/back propagation methodology. Using the National Trauma Data Bank (V6.2) data files, Tiberius Software created the ANN models. Best models were identified by their Gini co-efficient, ability to predict the complication outcome selected, and their Root Mean Squared Error (RMSE) scores. The model ensemble for the three major complications recorded in the registry were determined, variables ranked and model accuracy recorded. The basic ANN is fairly accurate for those likely to contract Acute Respiratory Disease Syndrome (ARDS) though with a high rate of false positives. The ANN ability to predict Ventilator Associated Pneumonia (VAP) is less effective, though better at producing fewer false positives. Predicting Urinary Tract Infections (UTI) cases is not good enough using these input variables. Both VAP and UTI relate to those aged over 55 years, while ARDS related more to those under 16 years. The models need improving.