A district cooling system (DeS) is a.scheme for centralised cooling energy distribution which takes advantage of economies of scale and load diversity. . A cooling medium (chilled water) is generated at a central refrigeration plant and then supplied to a district area, comprising multiple buildings, through a closed-loop piping circuit. Because of the substantial capital investment involved, an optimal design of the distribution piping . configuration is one of the crucial factors for successful implementation of a district 1'. cooling scheme. Since there. exists an enormous number of different combinations of the piping configuration, it is not feasible to evaluate each individual case using an exhaustive approach. This thesis exammes the problem of determining an optimal distribution piping configuration using a genetic algorithm (GA). In order to estimate the spatial and temporal distribution of cooling loads; the climatic conditions of Hong Kong were investigated and a weather database in the form of a typical meteorological year (TMY) was developed. Detailed thermal modelling of a number of prototypical buildings was carried out to determine benchmark cooling loads. A novel Local Search/Looped Local Search algorithm was developed for finding optimal/near-optimal distribution piping configurations. By means of computational . experiments, it was demonstrated that there is a promising improvement to GA performance by including the Local Search/Looped Local Search algorithm, in terms of both solution quality and computational efficiency. The effects on the search performance of a number of parameters were systematically investigated to establish the most effective settings. In order to illustrate the effectiveness of the Local Search/Looped Local Search algorithm, a benchmark problem - the optimal communication,spanning tree (OCST) was used for comparison. The results showed that the Looped Local Search method developed in this work was an effective tool for optimal network design of the distribution piping system in DCS, as well as for optimising the OCST problem.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:485621 |
| Date | January 2008 |
| Creators | Chan, Lok Shun Apple |
| Publisher | De Montfort University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/2086/4109 |
Page generated in 0.0105 seconds