Back to results list
Please use this identifier to cite or link to this item:
|Title:||Heuristic-engineering-statistical approach for chiller optimization||Authors:||Mak, Cheuk Wai||Keywords:||Air conditioning -- Equipment and supplies -- Energy consumption.
Air conditioning -- Control.
Hong Kong Polytechnic University -- Dissertations
|Issue Date:||2014||Publisher:||The Hong Kong Polytechnic University||Abstract:||Since the energy crisis occurred in 1973, the issue of energy use became the first priority for engineers and researchers to tackle this problem. In the recent two decades, ozone depletion and global warming have become two critical issues that affecting our global environment. Starting from 1990s, the Hong Kong government has been concerning about the energy use in air-conditioning systems within buildings. The government launched code of practice and guidelines for energy efficiency of air conditioning installations as well as guidelines on energy audit. Nonetheless, they were voluntary base which did not incentive to the building owners to follow. Until 2012, the government turned the codes and guidelines with some modifications into Building Energy Efficiency Ordinance under Chapter 610 that all building services engineers and building owners shall follow the regulation. Nevertheless, some statements regarding the issue of assessing the energy performance of the centralized air-conditioning system within buildings and recommendations on optimizing the control for chiller plant are still vague. In addition, when assessing the energy performance of the chiller plant, inadequate and invalid building information and operating data were discovered in most of buildings in Hong Kong. Notwithstanding, building energy performance or chiller performances are traditionally investigated by engineering modeling especially in the striking of the optimum point, this approach may be very successful in research level, whether it is laboratory based or simulation based. All through the past decades, there exist no successful optimization protocols in real building. This thesis presents a protocol for assessing and optimizing individual chillers within building with supplementing the deficiency of the codes addressed by the Hong Kong government. This research starts with the concept of data mining with two approaches: (a) preliminary building energy performance assessment; and (b) detailed chiller performance assessment. For preliminary building energy performance assessment, the energy signature for building performance analysis based on statistical model is discussed. Using this technique, the applications of the energy signature for 20 surveyed buildings are studied. Through this analysis, the effects of the outdoor environment conditions on the building energy consumption especially the air-conditioning system for different types of functional use can be investigated. Furthermore, the survey on the availability of essential building information for establishing the building energy signature is also presented. Apart from using statistical modeling approach, a dissection of building energy consumption using engineering mode ling approach in order to simulate the building energy performance for a selected commercial building is also highlighted. The results are the annual building electricity consumption breakdown, the effects of the climatic variable (To2 and ToWo) on the building electricity consumption together with the degradations of the equipment of the centralized chiller plant within building.
For detailed chiller performance assessment, a modified refrigerant model for simulating the thermal-dynamic characteristic of R134a is firstly presented. The accuracy of this model is significantly high when comparing the laboratory data. Secondly, a data management protocol with 5 steps (including data acquisition, data synchronization, data conditioning, range validity conditioning and uncertainty analysis) in order to deal with the typical problems (i.e. due to data acquisition or sensor errors etc.) occurred in BMS is introduced. This protocol is a prior procedure that facilitates the next step for individual chiller performance assessment and optimization. Two chiller plants are selected for the study. Lastly, the Heuristic-Engineering-Statistical (HES) approach is developed in this study It innovates this practice in two ways. On one hand, this study tries to identify low optimization operation zones (operating conditions) of chillers instead of highly optimized operating point or zones where the operating conditions are usually cannot be matched in real life operation, and converts these low optimized conditions into more optimized operating conditions. On the other hand, this study takes every site collected performance conditions as the operating settings in an experiment. At the completion of one annual cycle, the chiller can be thought of going through all most likely operating conditions in the building. This approach is then more realistic and applicable in the building. The HES model is then self-validated by the true operation of the chiller. An integrated solution for individual chiller optimization involved the comprehensive working procedures. Therefore, a chiller server (electronic web-based integrated analysis scheme) for individual chiller performance optimization is established and demonstrated in this research project which is comprised of the data logic management protocol and the HES approach for individual chiller performance assessment and optimization. This chiller server is welcome by the industry and is expected to innovate the method for individual chiller performance analysis and optimization.
|Description:||Iv, 383, 238 pages : color illustrations ; 30 cm
PolyU Library Call No.: [THS] LG51 .H577P BSE 2014 Mak
|URI:||http://hdl.handle.net/10397/7396||Rights:||All rights reserved.|
|Appears in Collections:||Thesis|
Show full item record
Files in This Item:
|b27629053_link.htm||For PolyU Users||203 B||HTML||View/Open|
|b27629053_ir.pdf||For All Users (Non-printable)||22.8 MB||Adobe PDF||View/Open|
Citations as of Jun 18, 2018
Citations as of Jun 18, 2018
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.