SCIENTIA IRANICA
Year:2005 | Volume:12 | Issue:1

Let p be a prime number and let n be a positive integer prime to p. By an Ihara-result, one means the existence of an injection with torsion-free cokernel, from a full lattice, in the space of p-old modular forms, into a full lattice, in the space of all modular forms of level np. In this paper, Ihara-results are proven for genus two Siegel modular forms, Siegel-Jacobi forms and Hilbert modular forms. Ihara did the genus one case of elliptic modular forms[1]. A geometric formulation is proposed for the notion of p-old Siegel modular forms of genus two, using clarifying comments by R. Schmidt[2] and, then, following suggestions in an earlier paper[3] on how to prove Ihara results. The main theorem in[3] is used, where an argument by G. Pappas has been extended to prove the torsion-freeness of certain cokernel, using the density of Hecke-orbits in the moduli space of principally polarized abelian varieties and in the Hilbert-Blumenthal moduli space, which was proved by C. Chai[4].

In this paper, it is proven that a semigroup is regular and RGC_n-commutative if, and only if, it is a spined product of a commutative Clifford semigroup and a right regular band

In this paper, an adaptive random search method, based on continuous action-set learning automata, is studied for solving stochastic optimization problems in which only the noise-corrupted value of a function at any chosen point in the parameter space is available. First, a new continuous action-set learning automaton is introduced and its convergence properties are studied. Then, applications of this new continuous action-set learning automata to the minimization of a penalized Shubert function and pattern classification are presented.

In this paper, optimal production and maintenance planning of a flexible manufacturing system under a time variant demand is considered. There is a preventive maintenance plan to reduce the failure rate of the machine. It is assumed that the failure rate of the machine is a function of its age and its maintenance rate. It is, also, assumed that the demand of the manufacturing product is time dependent and its rate depends on the level of the advertisement on that product. The objective is to maximize the expected discounted total profit of the firm over an infinite time horizon. To solve this optimization problem, first, an optimal control is characterized by a set of Hamilton-Jacobi-Bellman partial differential equations. Then, since this set of equations cannot be solved analytically, this stochastic optimal control model is approximated by a deterministic optimal control problem. By solving this new deterministic problem under practical assumptions, a set of suboptimal controls can be found.

Due to deregulation and restructuring in many countries, it is expected that the amount of small-scale generations connected to the distribution networks will increase. So, it is necessary that the impact of these kinds of generators on Volt/Var control should be investigated. This paper presents a new approach to Volt/Var control in distribution systems with Distributed Generation (DG). It has been shown that DG can improve the entire performance of a network system, by means of better control and decreasing losses. In this approach, the Genetic Algorithm (GA) has been used as the optimization method, where the amount of DG and its controlling parameters, the voltage regulators situation, the status of the load tap changers and, finally, the amount of switched capacitor, have been assumed as state variables. This method is tested on IEEE 34 bus radial distribution test feeders and a rural distribution network. The results are presented and it is shown that in the case of the selection of a correct location for DG, the system losses can be decreased by up to 70\%.

In this paper, the convergence and stability conditions of extended DMC in the control of nonlinear SISO and MIMO systems are investigated. The formulations are based on the ordinary DMC in which, with successive linearization of the nonlinear model and new interpretation of disturbance, the nonlinear extension is deduced. In addition, new convergence and stability criteria are derived for SISO and MIMO systems. These criteria include convergence and stability in the case of longer control (M>1) and prediction (P>1) horizons, as well as the finite and infinite sampling time. Finally, the simulation results for a MIMO (3\times 3) model, based on a power unit nonlinear plant, are presented.

Rebeca is an actor-based language for modeling concurrent and distributed systems, Its Java-like syntax makes it easy-to-use for practitioners and its formal foundation is a basis to make different formal verification approaches applicable. Compositional verification and abstraction techniques are used in formal verification of Rebeca models to overcome state explosion problems, The main contribution of this paper is to show how model checking and deduction are integrated for verifying certain properties of these models. Deduction is used to prove that abstraction techniques preserve a set of behavioral specifications in temporal logic and is also used in applying the compositional verification approach, on the basis of the model checked components.

In this paper, a delay - difference second - order proportionally - fair rate allocation algorithm has been proposed. As conventional proportionally - fair rate allocation algorithms deploy some form of scaled gradient ascent iterative algorithm for converging to user optimal rates, using fast second - order algorithms, such as Jacobi or approximate Newton methods, can be considered as natural and good candidates for increasing the convergence speed of the rate allocation algorithms, Stability analysis, related to scaled gradient ascent algorithms, in the presence of propagation delays, has been performed by some researchers, such as R. Johari et al., in Cambridge, In the current paper, the stability conditions of a second - order Jacobi method in the presence of propagation delays, with the simplifying premise of equality between all the users_ propagation delays, is derived mathematically_ Simulation results show that even in the general case of different propagation delays, stability is maintained.

Active components in amplifiers generate noise. Thus, amplifiers too, produce noise at their output and, therefore, increase the input signal to noise ratio. This is an undesirable feature, more particularly when one is dealing with weak input signals already combined with noise. Thus, reducing amplifiers' noise becomes a necessity for some applications. To overcome this problem, one approach consists of improving electronic device technology. Another approach is the use of advanced techniques in designing low noise amplifiers by taking advantage of pulse shapers. Pulse shapers generally consist of calculated filters made of integrator(s) and differentiator(s) that limit properly the amplifier bandwidth and, thereby, limit noise through the amplifier. In this study, the effect of variations of parameters of a single differentiator dual integrator pulse shaper on the reduction of noise in an amplifier circuit is investigated and their corresponding values are found for an optimal design.

In this paper, a genetic algorithm for solving a class of project scheduling problems, called Resource Investment Problems, is presented. Tardiness of the project is permitted with a defined penalty. The decision variables are the level of resources and the start times of the activities. The objective is to minimize the sum of resources and delay penalty costs, subject to the activities" precedence relations and some other constraints. A revised form of the Akpan heuristic method for this problem is used to find better chromosomes. Elements of the algorithm, such as chromosome structure, unfitness function, crossover, mutation, immigration and local search operations, are explained.The performance of this genetic algorithm is compared with that of other published algorithms for Resource Investment Problems. Also, more than 700 problems are solved using an enumerating algorithm and their optimal solutions are used for the performance tests of the genetic algorithm. The tests results are quite satisfactory.

This paper presents an application of Artificial Neural Networks (ANN) to control the voltage and reactive power in power systems. The technique is based on using a feed - forward artificial neural network with an error back - propagation training algorithm, based on the Liebenberg Marquardt method to train the networks. The training data is obtained by solving several abnormal conditions using Linear Programming (LP) Generator voltages, reactive power sources and transformer taps are considered as control variables and load bus voltages and generator reactive powers as dependent variables. The method presented in this paper has been tested IEEE 14-bus and 30-bus standard systems. The obtained results clearly indicate that the trained neural networks are capable of controlling the voltage and reactive power in power systems with a high level of precision and speed.