Objective
The objective of TOPGEN is to apply the results of two previous RTD projects to the on-line optimisation of fuel costs in a Fossil Power Plant, thus reducing both the economical and environmental costs of Power Generation. Medium term objectives are the demonstration and development of the technologies involved, mainly in (but not restricted to) Power Generations Plants, as a product that can be easily adapted to different plants by using a specific methodology. This methodology will be progressively improved starting from the first application, reducing future costs of analysis, adaptation, and testing of further applications of the product.
One of the related projects is called CORAGE, by UITESA. It comes from the PASO programme, project code PC095, partially supported by the Spanish Government and the European Community. It is a software product designed to help the operator of a complex process, by means of operation recommendations, in order to achieve performance improvements in the process. It uses machine learning techniques to automatically create and tune sets of fuzzy control rules which are chosen to help drive a given high level process variable closer to a given goal (e.g. to minimise the costs or to drive a given variable towards a given setpoint).
The other EC project is VPS, Virtual Plant for Industrial and Process Control, project code for Northern Ireland IRTU TDP4, by Queen's University of Belfast (QUB). VPS is a Plant Simulator for Power Plants. It will be used to test and tune CORAGE in an early stage, thus enabling the later application of CORAGE to the real Power Plant and reducing implementation risks.
The idea is to apply CORAGE to recommend on-line adjustments to the operators of Ballylumford Power Plant (N. Ireland), property of Premier Power Plc (PREMIER), in order to achieve performance improvements during steady-state and in fault-free conditions, given a set for exploitation constraints. At the same time, continuous performance monitoring will be provided. Thus, the expected result is a software system, fully installed at Ballylumford Power Plant, that by means of operation recommendations and performance monitoring, will help improve the performance of the plant, reducing the amount of fuel necessary to obtain the electrical power required. This system will work on-line, continuously receiving data from the process.
The chosen approach is to face the problem in two main:
- Firstly, to test and tune CORAGE by means of on-line connection with the VPS. This allows to approximate the target application in a simpler, noisy-free but still representative test. Noise and non-controlled biases can mask the effects of an optimisation system when tested directly on the process, making it difficult to tune it efficiently.
- Once tuned, CORAGE will be applied to the process itself, in this case Ballylumford Power Plant. There, starting from the point of the previous system tuning, it will pass through a period of on-line connection and automatic learning of the control rules to fit the real process. After such period, the system will be tested and finally used in the plant's operation.
Increasing competitive pressures, efforts to reduce emissions, and stringent regulatory measured, are forcing power companies to improve energy conversion efficiency of generating units. In a thermal power plant, fuel constitutes about 85% of the total running cost. As a typical example, if a 500 MW generating unit is continuously operated at an efficiency of half a percentage point below optimum, the extra fuel bill would be about 1 MECCA a year. Thus it is worth going to considerable trouble and some expense to reduce the losses to a practical minimum. Electric Power Research Institute (EPRI) studies have shown that typically a 3% to 5% improvement in heat rate of existing fossil power plant units is feasible.
The results of the project will be exploited internally by the partners and related companies, and externally, by means of a marketing strategy based on integration of the products within existing control systems, using the partner's marketing structure and establishing strategic alliances with companies involved in the fields of control, power generation and systems suppliers.
Fields of science (EuroSciVoc)
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectrical engineeringelectric energy
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringcontrol systems
- natural sciencescomputer and information sciencessoftwaresoftware applicationssystem software
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
- engineering and technologyenvironmental engineeringenergy and fuelsenergy conversion
You need to log in or register to use this function
Topic(s)
Call for proposal
Data not availableFunding Scheme
CSC - Cost-sharing contractsCoordinator
28006 Madrid
Spain