Two efficient algorithms for computing the incomplete beta function are considered. They have comparable performance for a wide range of values of parameters and the variable.
Two efficient algorithms for computing the incomplete beta function are considered. They have comparable performance for a wide range of values of parameters and the variable.
The purpose of optimal short-term scheduling is to maximize electrical energy production depending on the natural water inflows of a river, according to a daily load duration curve and the operation of thermal power plants with minimal oscillating load. Thermal power plants have a long start-up time and cannot follow changes in load demand so they are used to cover the constant part of the load, but peaks in load demand are covered by hydro units. A new mathematical model of a hydro plant chain was developed which used only two, instead of three, variables. The optimization algorithm used Newton's method which is applied on the exact penalty function which allows a nonfeasible starting point so no phase simplex or projection method is needed at the beginning. The mathematical model was implemented on a PC as a computer program using the FORTRAN 77 programming language and spreadsheet software. The program was applied to a hydro-chain on the river Drava. Preliminary results show that applying optimization techniques can greatly reduce costs of operating a hydro plant chain.
This paper presents an efficient parallel algorithm for solving nonlinear Partial Differential Equations (PDEs), occurring in heat transfer and fluid flow simulation, on hypercube machines. To evaluate its performance, an expression for the efficiency of the algorithm is derived. The results show that the hypercubes are well suited for solving nonlinear PDEs.
This paper presents an interactive computer package which aids the distributed network designers in choosing the optimum capacity for different links. This package gives the designer a tool to experiment different assignment strategies and compare their results. Four assignment strategies are implemented in this package: Minimum Average Time delay, Min-Max, Equal Capacities and Proportional Capacities Assignment strategies. The designer can also use it to investigate the effect of different routing policies on the values of optimum capacity. Reliability of the network can be examined by choosing routes which avoid broken links. The package is designed to be flexible and extendable so any module can be replaced or new modules can be added without change in other modules of the package.
In this paper a CAD code is presented for the design of pipe networks employed in the irrigation and reclamation of agricultural desert land. The program can be used for both design and simulation purposes, and allows the matching of the pipe network characteristics with the pump's characteristics. It incorporates a highly efficient, specially developed numerical solution procedure to solve the mathematical model equations. The code is employed in the design of a representative network and the results are displayed and discussed.
The spatial coordination of a two-arm robot is achieved by imposing on the robotic system model the proper geometrical and kinematical constraints. The closed loop formation is considered when both end-effectors of the two arms have one point in common or hold the same object from each side. Specified motions resulting from torque generated at the joint links are also considered. The collision avoidance between the two arms in space is illustrated through the use of a minimum distance to be maintained between two different points of the two arms. The governing equations of motion are derived using Kane's equations, and the constraint equations are handled through the generation of orthogonal complement arrays using the Pseudo-Uptriangular Decomposition method (PUTD). The algorithms developed are presented in a computer automated form suitable for computer-aided-desing and dynamic simulation.