4. EINSTEIN Project
• EINSTEIN (FP7, 2014-2017)
Simulation Enhanced Integrated Systems for
Model-based Intelligent Control(s)
– Development of automated prediction,
optimization and fault detection algorithms
for integration with the IES software for
building performance optimization
• Funding: 4-Year EU-funded Marie Curie IAPP
– Jan. 2014 – Dec. 2017
• Partners: IES and TCD
• Topics
– Fault Detection
– Prediction
– Optimisation
– Overall system integration
5. IES-VE & SCAN / ERGON
SCAN – Cloud-based data acquisition,
analysis and visualisation for Buildings.
ERGON - Import, manage and interrogate
real building data / schedules and use
them inVE simulations.
IES-VE – Integrated suite of simulation
software for the built environment.
• Dynamic thermal models;
• Detailed HVAC and Control systems;
• Ratings and accreditation (LEED,
BREEAM etc.)
6. Building Information Silos
Building Information Model (BIM)
• Geometry,
• Materials,
• Constructions etc.
Building Management
System (BMS)
• Temperature,
• CO2
• Set points,
• Schedules,
• Energy & Water;
Occupants
• Connected devices
(WiFi, Cellular);
• Feedback;
Rooms
• Booking schedules;
• Lecture timetables;
Weather Station
• Wind speed;
• Temperature;
• Humidity;
• Rainfall.
Technical
documentation
• Drawings
• Op & Maintenance
manuals for
equipment etc.
8. Model Based FDD
Knowledge / Rule-based:
uses expert user-defined
rules (e.g. APAR) which
govern system behaviour;
Data-driven: uses historical
building data, Statistical
Methods, Empirical Data,
Machine Learning;
Model-based; uses a
calibrated detailed system
model
9. Prediction / Optimisation
• Multi-objective
control optimisation;
– Comfort;
– Economics;
– Carbon, Etc.;
• Integrates
– Historic/predicted
weather conditions;
– Building & system
thermal response;
– Occupant schedules
& feedback;
– Economics (i.e.
electricity / gas
tariff);
11. Operational WBS Models
• Increasing prevalence of BIM and energy modelling at
building design phase;
• Compliance and rating systems (e.g. LEED) starting to
recognise operational performance, rather than traditional
design performance;
• Advantage of operation models for FDD / MPC;
– Can be used to effectively monitor and diagnose discrepancies
between design intent & operational performance (i.e.
performance gap);
– Adaptable to changes in building or system operation
(compared to solely data-driven approaches);
– Capable of simulating different control scenarios, recognising
actual system response;
– Allows optimisation of control strategies using real performance
feedback
Today I’m going to talk about the topic of smart cities and buildings, and the IES R&D vision to make them a reality;
Holistic view from Smart Cities down to Smart Building Control;
At a City Scale, IES are focussed on MASTERPLANNING: This will help city architects and urban planners to design their cities and subsystems in the most sustainable way possible.
At a New Building DESIGN level, and for RETROFIT projects, IES have been offering Building Simulation solutions for over 20 years.
REAL-TIME CONTROL is required to intelligently exploit the controls of the building to automatically adjust them according to weather and other conditions.This can all link to SMART CITIES: Through enabling existing cities to make better use of their existing resources, and by clustering the energy needs of buildings together.
Intelligent and Model Based Control
Using Simulation for Fault Detection and Analysis
Real-Time Optimisation and Control of Buildings
Real-Time Prediction of Building Use
Direct Connections between Simulation Environment and BMS / Building Sensors
ERGON is an IES Cloud service that allows you to import, manage and interrogate real building profile or schedule data for use within your VE simulations. You can utilise measured data from the actual building you’re investigating to create profiles that enhance model calibration. Or use normalised benchmark data from other buildings of the same type. As ERGON is based on the cloud it enables you to manipulate vast amount of data and create profiles that go down as low as 1 minute time steps. Therefore you can not only deal with 8760 profiles, but also 525600 profiles!
Such profiles can be used to:
Investigate the impact of retrofit options using real building data
Undertake Post Occupancy Evaluations
Improve operational models for performance contracting
Aid in delivering Soft Landings
Undertake Monitoring Based Commissioning for LEED V4
Undertake LEED Measurement and Verification
Help close the performance gap by simulating designs closer to reality
Help Close the Performance Gap and simulate closer to reality
Monitoring-based commissioning for LEED V4;
Undertake Post Occupancy Evaluations
Improve operational models for retrofit & performance contracting
Within ERGON users can easily import .csv files, graphically interrogate the completeness of data, undertake some initial analysis using in-build analytics, and export as a Free Form Profile Data file (.ffd) into ApPro for use in VE Apache simulations. Users can also generate from scratch bespoke profiles at deep granularity (consisting of up to 10 daily profiles which can be assigned hourly across a year).
Traditionally, approaches to FDD can be split into three areas.
One issue is the ability to confidently predict and diagnose system faults, and minimise false positives.
EINSTEIN looking at how we can combine these approaches to deliver a more robust FDD system.
Building data
- Ontologies (Haystack), mapping
- Hardware
- Communications protocols
SCAN
- Visualisation tools
- Web dashboards
- Statistical analysis
Modelling / Prediction
FDD and Optimisation
Returned improved control logic to Building