Model-based Diagnosis of HVAC Systems Peter Struss Tech. Univ. of - - PowerPoint PPT Presentation

▶

Oct 29, 2023 208 likes •302 views

Model-based Diagnosis of HVAC Systems Peter Struss Tech. Univ. of Munich University College Cork Infokaffee 0712 HVAC diagnosis 1 Model-Based Systems & Qualitative Reasoning Group of the Technical University of Munich Air Handling

SLIDE 1

Model-Based Systems & Qualitative Reasoning Group of the Technical University of Munich

Infokaffee 0712 HVAC diagnosis

Model-based Diagnosis of HVAC Systems

Peter Struss

Tech. Univ. of Munich

University College Cork

1

SLIDE 2

Model-Based Systems & Qualitative Reasoning Group of the Technical University of Munich

Infokaffee 0712 HVAC diagnosis

Air Handling Unit (AHU) within HVAC System

2

SLIDE 3

Model-Based Systems & Qualitative Reasoning Group of the Technical University of Munich

Infokaffee 0712 HVAC diagnosis

Demonstrator: AHU 09

3

SLIDE 4

Model-Based Systems & Qualitative Reasoning Group of the Technical University of Munich

Infokaffee 0712 HVAC diagnosis

School of Music

4

SLIDE 5

Model-Based Systems & Qualitative Reasoning Group of the Technical University of Munich

Infokaffee 0712 HVAC diagnosis

AHU 09 – Contd

5

Return Duct Supply Duct

SLIDE 6

Model-Based Systems & Qualitative Reasoning Group of the Technical University of Munich

Infokaffee 0712 HVAC diagnosis

AHU 09

6

Preheat Coil Cooling Coil Return Fan

SLIDE 7

Model-Based Systems & Qualitative Reasoning Group of the Technical University of Munich

Infokaffee 0712 HVAC diagnosis

BMS

7

SLIDE 8

Model-Based Systems & Qualitative Reasoning Group of the Technical University of Munich

Infokaffee 0712 HVAC diagnosis

EMWINS - Challenge

HVAC systems: Very robust Monitored But hardly diagnosed and maintained Waste of energy Through undetected faults E.g. passing valve heating coil Compensated for by cooling coil Adaptation of solution: Fairly standard components, but Different configurations Different parameters

8

SLIDE 9

Model-Based Systems & Qualitative Reasoning Group of the Technical University of Munich

Infokaffee 0712 HVAC diagnosis

EMWINS - Solution

! Model-based solution Model library of HVAC components Configurable model Adaptation to plant parameters Numerical model (Automated) abstraction to diagostic model Consistency-based diagnosis Goals: Covering 70% of the faults Significant cost reduction Demonstrator AHU 09: real and simulated data Comparison with commercial rule-based system