University of Twente Student Theses

Login

Internship Report on benchmarking of Python based turbine model of cost of energy model; and generation of load coefficients to be used in the turbine model

Adhikari, S. (2014) Internship Report on benchmarking of Python based turbine model of cost of energy model; and generation of load coefficients to be used in the turbine model.

[img] PDF
1MB
Abstract:This report has been prepared on the basis of internship work done during August 4, 2014 – November 4, 2014. The report has been divided into two sections, each section based on a different task assigned during the internship. Each section discusses about the objectives, methodology, results, conclusion and recommendation for the respective task. The first section explains the benchmarking of the python based turbine model which has been developed at DNV-GL; against the industrial data and the old excel based Turbine model. The Turbine model is a part of Cost of Energy model. For the benchmarking task internal turbines (i.e. wind turbines designed at DNV-GL) and external turbines (i.e. wind turbines which are not designed by DNV-GL) were considered. Results of the benchmarking done against internal turbines were more useful as more turbine data were available for them. The data for the external turbines were limited. The benchmarking task was an iterative process. Results of the benchmarking were useful in identifying the bugs in the python based Turbine Model. This report only discusses about the latest benchmarking done using the internal turbines data, which was completed on October 3, 2014. The second section describes about various methods for obtaining the power law fit coefficient for load components by using the load database available at DNV-GL. In addition, power law fit coefficient for calculating the expected power is also obtained. The power law fit coefficients obtained are exported to a CSV file, which will be used in the Turbine Model for calculating the loads. These coefficients for various load components were obtained considering various components of load data. Scripts were written in Python to access the required data from the load database and to perform the calculations. Results from the various methods to obtain Power law fit coefficient are compared and the better approach among these methods is identified. Future works needed for improving the results has been recommended.
Item Type:Internship Report (Master)
Clients:
DNV GL (voorheen KEMA), the Netherlands
Faculty:ET: Engineering Technology
Subject:52 mechanical engineering
Programme:Sustainable Energy Technology MSc (60443)
Keywords:Turbine model, Cost of Energy model, Load component coefficent, Power Law fit
Link to this item:https://purl.utwente.nl/essays/70268
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page