Abstract: A complex valued neural network is a neural network, which consists of complex valued input and/or weights and/or thresholds and/or activation functions. Complex-valued neural networks have been widening the scope of applications not only in electronics and informatics, but also in social systems. One of the most important applications of the complex valued neural network is in image and vision processing. In Neural networks, radial basis functions are often used for interpolation in multidimensional space. A Radial Basis function is a function, which has built into it a distance criterion with respect to a centre. Radial basis functions have often been applied in the area of neural networks where they may be used as a replacement for the sigmoid hidden layer transfer characteristic in multi-layer perceptron. This paper aims to present exhaustive results of using RBF units in a complex-valued neural network model that uses the back-propagation algorithm (called 'Complex-BP') for learning. Our experiments results demonstrate the effectiveness of a Radial basis function in a complex valued neural network in image recognition over a real valued neural network. We have studied and stated various observations like effect of learning rates, ranges of the initial weights randomly selected, error functions used and number of iterations for the convergence of error on a neural network model with RBF units. Some inherent properties of this complex back propagation algorithm are also studied and discussed.
Abstract: This paper presents a critical study about the
application of Neural Networks to ion-exchange process. Ionexchange
is a complex non-linear process involving many factors
influencing the ions uptake mechanisms from the pregnant solution.
The following step includes the elution. Published data presents
empirical isotherm equations with definite shortcomings resulting in
unreliable predictions. Although Neural Network simulation
technique encounters a number of disadvantages including its “black
box", and a limited ability to explicitly identify possible causal
relationships, it has the advantage to implicitly handle complex
nonlinear relationships between dependent and independent
variables. In the present paper, the Neural Network model based on
the back-propagation algorithm Levenberg-Marquardt was developed
using a three layer approach with a tangent sigmoid transfer function
(tansig) at hidden layer with 11 neurons and linear transfer function
(purelin) at out layer. The above mentioned approach has been used
to test the effectiveness in simulating ion exchange processes. The
modeling results showed that there is an excellent agreement between
the experimental data and the predicted values of copper ions
removed from aqueous solutions.
Abstract: Software estimation accuracy is among the greatest
challenges for software developers. This study aimed at building and
evaluating a neuro-fuzzy model to estimate software projects
development time. The forty-one modules developed from ten
programs were used as dataset. Our proposed approach is compared
with fuzzy logic and neural network model and Results show that the
value of MMRE (Mean of Magnitude of Relative Error) applying
neuro-fuzzy was substantially lower than MMRE applying fuzzy
logic and neural network.
Abstract: Stochastic models of biological networks are well established in systems biology, where the computational treatment of such models is often focused on the solution of the so-called chemical master equation via stochastic simulation algorithms. In contrast to this, the development of storage-efficient model representations that are directly suitable for computer implementation has received significantly less attention. Instead, a model is usually described in terms of a stochastic process or a "higher-level paradigm" with graphical representation such as e.g. a stochastic Petri net. A serious problem then arises due to the exponential growth of the model-s state space which is in fact a main reason for the popularity of stochastic simulation since simulation suffers less from the state space explosion than non-simulative numerical solution techniques. In this paper we present transition class models for the representation of biological network models, a compact mathematical formalism that circumvents state space explosion. Transition class models can also serve as an interface between different higher level modeling paradigms, stochastic processes and the implementation coded in a programming language. Besides, the compact model representation provides the opportunity to apply non-simulative solution techniques thereby preserving the possible use of stochastic simulation. Illustrative examples of transition class representations are given for an enzyme-catalyzed substrate conversion and a part of the bacteriophage λ lysis/lysogeny pathway.
Abstract: A number of studies highlighted problems related to
ERP systems, yet, most of these studies focus on the problems during
the project and implementation stages but not during the postimplementation
use process. Problems encountered in the process of
using ERP would hinder the effective exploitation and the extended
and continued use of ERP systems and their value to organisations.
This paper investigates the different types of problems users
(operational, supervisory and managerial) faced in using ERP and
how 'feral system' is used as the coping mechanism. The paper
adopts a qualitative method and uses data collected from two cases
and 26 interviews, to inductively develop a casual network model of
ERP usage problem and its coping mechanism. This model classified
post ERP usage problems as data quality, system quality, interface
and infrastructure. The model is also categorised the different coping
mechanism through use of 'feral system' inclusive of feral
information system, feral data and feral use of technology.
Abstract: One of the criteria in production scheduling is Make
Span, minimizing this criteria causes more efficiently use of the
resources specially machinery and manpower. By assigning some
budget to some of the operations the operation time of these activities
reduces and affects the total completion time of all the operations
(Make Span). In this paper this issue is practiced in parallel flow
shops. At first we convert parallel flow shop to a network model and
by using a linear programming approach it is identified in order to
minimize make span (the completion time of the network) which
activities (operations) are better to absorb the predetermined and
limited budget. Minimizing the total completion time of all the
activities in the network is equivalent to minimizing make span in
production scheduling.
Abstract: This paper presents two simplified models to
determine nodal voltages in power distribution networks. These
models allow estimating the impact of the installation of reactive
power compensations equipments like fixed or switched capacitor
banks. The procedure used to develop the models is similar to the
procedure used to develop linear power flow models of transmission
lines, which have been widely used in optimization problems of
operation planning and system expansion. The steady state non-linear
load flow equations are approximated by linear equations relating the
voltage amplitude and currents. The approximations of the linear
equations are based on the high relationship between line resistance
and line reactance (ratio R/X), which is valid for power distribution
networks. The performance and accuracy of the models are evaluated
through comparisons with the exact results obtained from the
solution of the load flow using two test networks: a hypothetical
network with 23 nodes and a real network with 217 nodes.
Abstract: In a particular case of behavioural model reduction by ANNs, a validity domain shortening has been found. In mechanics, as in other domains, the notion of validity domain allows the engineer to choose a valid model for a particular analysis or simulation. In the study of mechanical behaviour for a cantilever beam (using linear and non-linear models), Multi-Layer Perceptron (MLP) Backpropagation (BP) networks have been applied as model reduction technique. This reduced model is constructed to be more efficient than the non-reduced model. Within a less extended domain, the ANN reduced model estimates correctly the non-linear response, with a lower computational cost. It has been found that the neural network model is not able to approximate the linear behaviour while it does approximate the non-linear behaviour very well. The details of the case are provided with an example of the cantilever beam behaviour modelling.
Abstract: In this paper, the modelling and design of artificial neural network architecture for load forecasting purposes is investigated. The primary pre-requisite for power system planning is to arrive at realistic estimates of future demand of power, which is known as Load Forecasting. Short Term Load Forecasting (STLF) helps in determining the economic, reliable and secure operating strategies for power system. The dependence of load on several factors makes the load forecasting a very challenging job. An over estimation of the load may cause premature investment and unnecessary blocking of the capital where as under estimation of load may result in shortage of equipment and circuits. It is always better to plan the system for the load slightly higher than expected one so that no exigency may arise. In this paper, a load-forecasting model is proposed using a multilayer neural network with an appropriately modified back propagation learning algorithm. Once the neural network model is designed and trained, it can forecast the load of the power system 24 hours ahead on daily basis and can also forecast the cumulative load on daily basis. The real load data that is used for the Artificial Neural Network training was taken from LDC, Gujarat Electricity Board, Jambuva, Gujarat, India. The results show that the load forecasting of the ANN model follows the actual load pattern more accurately throughout the forecasted period.
Abstract: The Beijing road traffic system, as a typical huge
urban traffic system, provides a platform for analyzing the complex
characteristics and the evolving mechanisms of urban traffic systems.
Based on dynamic network theory, we construct the dynamic model
of the Beijing road traffic system in which the dynamical properties
are described completely. Furthermore, we come into the conclusion
that urban traffic systems can be viewed as static networks, stochastic
networks and complex networks at different system phases by
analyzing the structural randomness. As well as, we demonstrate the
evolving process of the Beijing road traffic network based on real
traffic data, validate the stochastic characteristics and the scale-free
property of the network at different phases
Abstract: The mitigation of crop loss due to damaging freezes requires accurate air temperature prediction models. An improved model for temperature prediction in Georgia was developed by including information on seasonality and modifying parameters of an existing artificial neural network model. Alternative models were compared by instantiating and training multiple networks for each model. The inclusion of up to 24 hours of prior weather information and inputs reflecting the day of year were among improvements that reduced average four-hour prediction error by 0.18°C compared to the prior model. Results strongly suggest model developers should instantiate and train multiple networks with different initial weights to establish appropriate model parameters.
Abstract: Most of the commonly used blind equalization algorithms are based on the minimization of a nonconvex and nonlinear cost function and a neural network gives smaller residual error as compared to a linear structure. The efficacy of complex valued feedforward neural networks for blind equalization of linear and nonlinear communication channels has been confirmed by many studies. In this paper we present two neural network models for blind equalization of time-varying channels, for M-ary QAM and PSK signals. The complex valued activation functions, suitable for these signal constellations in time-varying environment, are introduced and the learning algorithms based on the CMA cost function are derived. The improved performance of the proposed models is confirmed through computer simulations.
Abstract: This paper aims to establish a delayed dynamical relationship between payoffs of players in a zero-sum game. By introducing Markovian chain and time delay in the network model, a delayed game network model with sector bounds and slope bounds restriction nonlinear function is first proposed. As a result, a direct dynamical relationship between payoffs of players in a zero-sum game can be illustrated through a delayed singular system. Combined with Finsler-s Lemma and Lyapunov stable theory, a sufficient condition guaranteeing the unique existence and stability of zero-sum game-s Nash equilibrium is derived. One numerical example is presented to illustrate the validity of the main result.
Abstract: The so-called all-pass filter circuits are commonly
used in the field of signal processing, control and measurement.
Being connected to capacitive loads, these circuits tend to loose their
stability; therefore the elaborate analysis of their dynamic behavior is
necessary. The compensation methods intending to increase the
stability of such circuits are discussed in this paper, including the socalled
lead-lag compensation technique being treated in detail. For
the dynamic modeling, a two-port network model of the all-pass filter
is being derived. The results of the model analysis show, that
effective lead-lag compensation can be achieved, alone by the
optimization of the circuit parameters; therefore the application of
additional electric components are not needed to fulfill the stability
requirement.
Abstract: The paper presents the potential of fuzzy logic (FL-I)
and neural network techniques (ANN-I) for predicting the
compressive strength, for SCC mixtures. Six input parameters that is
contents of cement, sand, coarse aggregate, fly ash, superplasticizer
percentage and water-to-binder ratio and an output parameter i.e. 28-
day compressive strength for ANN-I and FL-I are used for modeling.
The fuzzy logic model showed better performance than neural
network model.
Abstract: A complex valued neural network is a neural network
which consists of complex valued input and/or weights and/or thresholds
and/or activation functions. Complex-valued neural networks
have been widening the scope of applications not only in electronics
and informatics, but also in social systems. One of the most important
applications of the complex valued neural network is in signal
processing. In Neural networks, generalized mean neuron model
(GMN) is often discussed and studied. The GMN includes a new
aggregation function based on the concept of generalized mean of all
the inputs to the neuron. This paper aims to present exhaustive results
of using Generalized Mean Neuron model in a complex-valued neural
network model that uses the back-propagation algorithm (called
-Complex-BP-) for learning. Our experiments results demonstrate the
effectiveness of a Generalized Mean Neuron Model in a complex
plane for signal processing over a real valued neural network. We
have studied and stated various observations like effect of learning
rates, ranges of the initial weights randomly selected, error functions
used and number of iterations for the convergence of error required on
a Generalized Mean neural network model. Some inherent properties
of this complex back propagation algorithm are also studied and
discussed.
Abstract: Natural resources management including water resources requires reliable estimations of time variant environmental parameters. Small improvements in the estimation of environmental parameters would result in grate effects on managing decisions. Noise reduction using wavelet techniques is an effective approach for preprocessing of practical data sets. Predictability enhancement of the river flow time series are assessed using fractal approaches before and after applying wavelet based preprocessing. Time series correlation and persistency, the minimum sufficient length for training the predicting model and the maximum valid length of predictions were also investigated through a fractal assessment.
Abstract: Earth reinforcing techniques have become useful and economical to solve problems related to difficult grounds and provide satisfactory foundation performance. In this context, this paper uses radial basis function neural network (RBFNN) for predicting the bearing pressure of strip footing on reinforced granular bed overlying weak soil. The inputs for the neural network models included plate width, thickness of granular bed and number of layers of reinforcements, settlement ratio, water content, dry density, cohesion and angle of friction. The results indicated that RBFNN model exhibited more than 84 % prediction accuracy, thereby demonstrating its application in a geotechnical problem.
Abstract: The main goal in this paper is to quantify the quality of
different techniques for radiation treatment plans, a back-propagation
artificial neural network (ANN) combined with biomedicine theory
was used to model thirteen dosimetric parameters and to calculate
two dosimetric indices. The correlations between dosimetric indices
and quality of life were extracted as the features and used in the ANN
model to make decisions in the clinic. The simulation results show
that a trained multilayer back-propagation neural network model can
help a doctor accept or reject a plan efficiently. In addition, the
models are flexible and whenever a new treatment technique enters
the market, the feature variables simply need to be imported and the
model re-trained for it to be ready for use.
Abstract: Soil organic carbon (SOC) plays a key role in soil
fertility, hydrology, contaminants control and acts as a sink or source
of terrestrial carbon content that can affect the concentration of
atmospheric CO2. SOC supports the sustainability and quality of
ecosystems, especially in semi-arid region. This study was
conducted to determine relative importance of 13 different
exploratory climatic, soil and geometric factors on the SOC contents
in one of the semiarid watershed zones in Iran. Two methods
canonical discriminate analysis (CDA) and feed-forward back
propagation neural networks were used to predict SOC. Stepwise
regression and sensitivity analysis were performed to identify
relative importance of exploratory variables. Results from sensitivity
analysis showed that 7-2-1 neural networks and 5 inputs in CDA
models output have highest predictive ability that explains %70 and
%65 of SOC variability. Since neural network models outperformed
CDA model, it should be preferred for estimating SOC.