Please use this identifier to cite or link to this item: http://localhost:8081/xmlui/handle/123456789/2833
Authors: Rajiv J., Naik
Issue Date: 1987
Abstract: Pipe networks are used to transport liquids and gases from the sources to various locations of their utility. Analysis and design of pipenetworks, in general, involves the solution of a set of nonlinear simulataneous equations. Research efforts are naturally made in order to either improve the efficiency/convergence of existing methods or to reduce the total number of equations in the given set of nonlinear equations, used to simulate the network. There are four types of formulations of pipe network which are described in detail in the literature review alongwith their advantages and disadvan age present work, a new method of formulation, called Mixed me .hod of formulation, is proposed to solve the pipe network problems, Its storage requirements are low and at the same time it is computationally efficient. In essence it has the advantages of both the formulation-.1 and formulation-.3. Efficient software packages have been developed to analyse the pipe networks by using existing methods or by the proposed mixed method of formulation. These packages are also capable of carrying out sensitivity analysis of the network, which may be of help in identifying critical nodes or to plan future expansion of the network. In software packages Newton Raphson method has been used to solve the set of nonlinear equations due to its quadratic convergence. During development of the computer 1:l] programs, it is observed that this method is sensitive to initial guess of values of parameters-. A convergence scheme is proposed to reduce its dependence on initial guess S. Lastly a computer-aided loop selection method has also been developed and programmed, which selects the set of independent loops of pipes in a given network. This set of independent loops is required in the Formulation--4 and any error in the selection of this set of independent loops, which is likely during manual inspection of the large networks, results in the divergence of the solution.
Other Identifiers: M.Tech
Research Supervisor/ Guide: Jain, S. C.
Kumar, Surendra
metadata.dc.type: M.Tech Dessertation
Appears in Collections:MASTERS' THESES (MIED)

Files in This Item:
File Description SizeFormat 
MIED179307.pdf3.56 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.