The aim of this paper is to approximate multidimensional functions by using the type of Feedforward neural networks (FFNNs) which is called Greedy radial basis function neural networks (GRBFNNs). Also, we introduce a modification to the greedy algorithm which is used to train the greedy radial basis function neural networks. An error bound are introduced in Sobolev space. Finally, a comparison was made between the three algorithms (modified greedy algorithm, Backpropagation algorithm and the result is published in [16]).
A model using the artificial neural networks and genetic algorithm technique is developed for obtaining optimum dimensions of the foundation length and protections of small hydraulic structures. The procedure involves optimizing an objective function comprising a weighted summation of the state variables. The decision variables considered in the optimization are the upstream and downstream cutoffs lengths and their angles of inclination, the foundation length, and the length of the downstream soil protection. These were obtained for a given maximum difference in head, depth of impervious layer and degree of anisotropy. The optimization carried out is subjected to constraints that ensure a safe structure aga
... Show MoreFinding the shortest route in wireless mesh networks is an important aspect. Many techniques are used to solve this problem like dynamic programming, evolutionary algorithms, weighted-sum techniques, and others. In this paper, we use dynamic programming techniques to find the shortest path in wireless mesh networks due to their generality, reduction of complexity and facilitation of numerical computation, simplicity in incorporating constraints, and their onformity to the stochastic nature of some problems. The routing problem is a multi-objective optimization problem with some constraints such as path capacity and end-to-end delay. Single-constraint routing problems and solutions using Dijkstra, Bellman-Ford, and Floyd-Warshall algorith
... Show MoreIn this research, we dealt with the study of the Non-Homogeneous Poisson process, which is one of the most important statistical issues that have a role in scientific development as it is related to accidents that occur in reality, which are modeled according to Poisson’s operations, because the occurrence of this accident is related to time, whether with the change of time or its stability. In our research, this clarifies the Non-Homogeneous hemispheric process and the use of one of these models of processes, which is an exponentiated - Weibull model that contains three parameters (α, β, σ) as a function to estimate the time rate of occurrence of earthquakes in Erbil Governorate, as the governorate is adjacent to two countr
... Show MoreThis paper deals with the modeling of a preventive maintenance strategy applied to a single-unit system subject to random failures.
According to this policy, the system is subjected to imperfect periodic preventive maintenance restoring it to ‘as good as new’ with probability
p and leaving it at state ‘as bad as old’ with probability q. Imperfect repairs are performed following failures occurring between consecutive
preventive maintenance actions, i.e the times between failures follow a decreasing quasi-renewal process with parameter a. Considering the
average durations of the preventive and corrective maintenance actions a
... Show MoreAbstract In this paper the effect of light exposure duration on Anthracene solution in chloroform is studied. It is found that: the Anthracene solution change its color when it is exposed to light, and that its relative quantum efficiency, Φ, decreases as the light exposure duration, t, increases and this govern by following empirical equation:- Φ = 0.7918-0.0762 In (t)
In this research is estimated the function of reliability dynamic of multi state systems and their compounds and for three types of systems (serial, parallel, 2-out-of-3) and about two states (Failure and repair) depending on calculating the structur function allow to describing the behavior of
In this paper, the maximum likelihood estimates for parameter ( ) of two parameter's Weibull are studied, as well as white estimators and (Bain & Antle) estimators, also Bayes estimator for scale parameter ( ), the simulation procedures are used to find the estimators and comparing between them using MSE. Also the application is done on the data for 20 patients suffering from a headache disease.
Intended for getting good estimates with more accurate results, we must choose the appropriate method of estimation. Most of the equations in classical methods are linear equations and finding analytical solutions to such equations is very difficult. Some estimators are inefficient because of problems in solving these equations. In this paper, we will estimate the survival function of censored data by using one of the most important artificial intelligence algorithms that is called the genetic algorithm to get optimal estimates for parameters Weibull distribution with two parameters. This leads to optimal estimates of the survival function. The genetic algorithm is employed in the method of moment, the least squares method and the weighted
... Show Morehis analysis aims to establish Riemann-Liouville derivation andintegral operators regarding the recently suggested seven-parameter Mittag-Leffler function then investigates the corresponding special cases. In addition,certain notable results associated with those new operators have been dis-cussed
The 3-parameter Weibull distribution is used as a model for failure since this distribution is proper when the failure rate somewhat high in starting operation and these rates will be decreased with increasing time .
In practical side a comparison was made between (Shrinkage and Maximum likelihood) Estimators for parameter and reliability function using simulation , we conclude that the Shrinkage estimators for parameters are better than maximum likelihood estimators but the maximum likelihood estimator for reliability function is the better using statistical measures (MAPE)and (MSE) and for different sample sizes.
Note:- ns : small sample ; nm=median sample
... Show More