Optimal Computer Control of a Group of Interacting Hydro Electric Power Stations

Clink, James (1986) Optimal Computer Control of a Group of Interacting Hydro Electric Power Stations. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 10995587.pdf] PDF
Download (12MB)

Abstract

The North of Scotland Hydro Electric Board (NSHEB) are currently upgrading the control equipment for hydro generating plant. An important part of this modernisation is the introduction of optimisation methods to the control system. Highly constrained plant poses an unique optimisation problem. This project reports the application of optimisation techniques to one such highly constrained group, namely, the Tummel Valley Scheme. The Tummel Valley scheme consists of nine main reservoirs, coupled to each other by the flow through eight hydro power stations. Each individual station of the scheme has a number of unique constraints which must be placed upon its operation. The problem of optimising the scheme has been simplified by decomposing the problem in time. The operation of the group was studied in three time scales : long term, or strategic, operational planning of the large reservoirs, daily decoupling of the interactions between the valley's reservoirs, and, the hourly scheduling of the available water. The method used in the long term planning of the large reservoir, Loch Ericht, was a two stage process based upon stochastic dynamic programming. The method was fast, and enabled annual planning of a reservoir's water resources. The optimisation results were compared to the actual annual strategy used by the NSHEB for Loch Ericht. For smaller hydro stations, the water available for generation from day to day is much more dependent upon the natural inflow from the catchment area. The prediction of this natural inflow allows optimal use to be made of the water available in small stations. Methods have been applied to the catchment at Gaur power station which allow a prediction of the inflow to be made from limited data which can be easily collected on site. The daily decoupling procedure reconciles the daily demand for power with the optimal water available for generation from each station. The coupling constraints between the reservoirs and other operational constraints are considered. The procedure uses non-linear programming to produce daily outflows from each station, which effectively decouple the reservoirs of the Tummel Valley scheme. The daily outflow from each station is allocated hourly according to a hourly demand curve to give an hourly schedule for each station. The method of dynamic programming with successive approximations is used to find a solution to the close coupling between Clunie and Pitlochry stations. A set point controller has been tested for dispatching the hourly schedule produced by the optimisation. Models have been developed of existing Temporary Droop governed hydro turbines. The set point controller has been fully tested against this simulation and real plant for both fast and controlled loading.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: Electrical engineering, Computer engineering, Civil engineering
Date of Award: 1986
Depositing User: Enlighten Team
Unique ID: glathesis:1986-77539
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 14 Jan 2020 11:53
Last Modified: 14 Jan 2020 11:53
URI: https://theses.gla.ac.uk/id/eprint/77539

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year