| dc.contributor.author | Hartati, Rukmi Sari. | en_US |
| dc.date.accessioned | 2014-10-21T12:38:05Z | |
| dc.date.available | 2002 | |
| dc.date.issued | 2002 | en_US |
| dc.identifier.other | AAINQ77593 | en_US |
| dc.identifier.uri | http://hdl.handle.net/10222/55883 | |
| dc.description | A Hopfield neural network is modified to handle inequality constraints by introducing an exterior penalty function. The use of a penalty function converts constrained optimization problems into unconstrained problems. | en_US |
| dc.description | The optimal power flow is a general non-linear programming problem with a non-linear objective function and non-linear functional equality and inequality constraints. Security Constrained Dispatch is defined as an Optimal Power Flow problem, in which the objective function is the total cost of generations and the security constraints are placed on the bus voltage magnitudes, phase angles and the generated active powers. | en_US |
| dc.description | This thesis presents an alternative method for solving optimal active power flow and active security-constrained dispatch using a modified Hopfield neural network. The objective function of security-constrained dispatch is the incremental generation cost function in quadratic form which is expanded in a second-order Taylor series. The equality and inequality constraints are modelled using a linearized network and appended to the objective function using suitable penalty functions to form an augmented cost function. | en_US |
| dc.description | The goal of this research is to model and study the applicability of the modified Hopfield Neural Network for solving optimal active power flow and security-constrained dispatch problem. In addition, this thesis aims to discover the advantages and disadvantages of using this technique instead of the methods that currently exist. | en_US |
| dc.description | The Hopfield Neural Network was simulated on a digital computer for four standard IEEE test systems varying in size from a 5-bus system to a 57-bus system. The optimal solution obtained using this approach is consistent with the solution obtained using the conventional method. | en_US |
| dc.description | The advantage of this method is in the ease of formalization of the problem. It is simple, straightforward, and easy to apply. The method requires modest memory resources and is efficient in computation time. This representation is applicable to many problems other than the economic load-dispatching problem. | en_US |
| dc.description | Thesis (Ph.D.)--DalTech - Dalhousie University (Canada), 2002. | en_US |
| dc.language | eng | en_US |
| dc.publisher | Dalhousie University | en_US |
| dc.publisher | | en_US |
| dc.subject | Engineering, Electronics and Electrical. | en_US |
| dc.subject | Computer Science. | en_US |
| dc.title | Active security-constrained optimal power flow using modified Hopfield neural network. | en_US |
| dc.type | text | en_US |
| dc.contributor.degree | Ph.D. | en_US |
