First System Identification Competition
Organised end Edited by J.J. Bloem
INTRODUCTION
A wide range of new techniques is now being applied to the analysis problems involved with the estimation of thermal characteristic parameters of buildings and building components. Similar problems arise in most observational disciplines, including physics, biology and economics. New commercially available software tools promise to provide results that were unobtainable just a decade ago. Unfortunately, the realisation and evaluation of this promise has been hampered by the difficulty of making rigorous comparisons between competing techniques, particularly ones that come from different disciplines.
The objective of the competition is to set up a comparison between alternative techniques and to clarify particular problems of system identification applied to the thermal performance of buildings.
Research on energy saving in buildings can be divided into three major areas:
- building components
- test cells and unoccupied buildings in real climate and
- occupied buildings.
Three competitions are planned along this line, of which the first competition concerned with building components is described in the book: SYSTEM IDENTIFICATION COMPETITION, edited by J.J. Bloem. Parts of the book can be downloaded via this site. For ordering the book, see at the bottom of this page.
The competition has been organised to help clarify the conflicting claims among many researchers who use and analyse building energy data and to foster contact among these persons and their institutions. The intent is not necessarily only to declare winners, but rather to set up a format in which rigorous evaluations of techniques can be made. Because there are natural measures of performance, a rank-ordering has been given. In all cases, however, the goal is to collect and analyse quantitative results in order to understand similarities and differences among the approaches.
There is also a second System Identification Competition.
To order the book:
‘System Identification Competition’ by J.J. Bloem
Ref.: EUR 16359 EN (1996)
ISBN 92-827-6348-X
OFFICE FOR OFFICIAL PUBLICATIONS OF THE EUROPEAN UNION
L-2985 Luxembourg
For more information contact J.J. Bloem
DESCRIPTION OF THE PROBLEM CASES AND THE RULES FOR THE COMPETITION
Competition rules and instructions (pdf file; 27kB)
This chapter describes the competition as it was announced in 1994. To give complete information about the competition and to keep the data available for future use by other interested people, it is included in the book. It gives the information how the data can be obtained and which rules are to be applied for participating in the competition.
Description of the cases (pdf file; 32 kB)
In this chapter definitions are given and the five cases are presented as they were described in the instructions of the system identification competition. Furthermore the authors have added some additional comments with every case, to provide some background information why the particular cases have been selected for the competition. The statistics of each data set are given at the end of the chapter.
Generation of the data (pdf file; 89 kB)
This chapter is concerned with the generation of data for the System Identification Competition. A simulation model based on the physical properties of a wall has been developed in order to simulate a wall in practical situations. Various data series were generated to simulate internal and external surface temperatures of the concerning wall which are the inputs for the simulation model. The generation of these temperatures as well as the heat flow is described in detail. Furthermore different existing identification are applied on the generated data series in order to check the quality of the data.
EVALUATION OF THE RESULTS SUBMITTED BY THE PARTICIPANTS.
The statistical evaluation (pdf file; 92 kB)
This chapter describes the statistical evaluation of the submittals. Definitions are given about the applied statistics and the results of the evaluation can be found separately for each case. This chapter concludes with the remark that the applied identification technique can be known, but that the applicant needs a certain level of skill to apply the method successfully in order to perform well. Furthermore the competition has shown that the estimation of the uncertainty of the parameter estimates is a difficult one.
INVITED PAPERS ON THE FIVE CASES & DESCRIPTION OF THE PHYSICAL CONTEXT OF THE PROBLEM CASES
An introduction to building physics (pdf file; 48 kB)
This chapter places the System Identification Competition in a broader context of evaluating the thermal performances of building components. It gives a global context for identification techniques in relation to the thermal performances of building components. Basic information is provided about the various processes involved in heat transfer in building components as well as the commonly used formulas to express these heat flow processes. The key parameters in which the building sector is interested are given with an explanation for these priorities. Moreover, thermal testing in general and dynamic testing specifically is situated in this overall context and finally the role and importance of the System Identification Competition is explained.
Application of the average method to all cases (pdf file; 31 kB)
This chapter describes the application of the classical steady state method, also referred to as the average method. The method is a straightforward analysis technique which provides some quantitative and qualitative information about the measured data and is therefore very useful as a first step in the analysis process. However in general the disadvantage of the application of the average method is the long test duration to obtain a relative accurate result. Furthermore it does not give information about the dynamics of the physical system.
DOWNLOAD DATA
Download data of the System Identification Competition (zip file; 266 kB)