Digital 210 King Project Feed

Updated Digital 210 King Files (BIM, IFC, and XYZ) (Blog Entry)

Tue, 07 Feb 2012 13:01:44 -0700

These are complete versions of the Digital 210 King model for Revit Architecture and MEP.

We have also included a fully integrated IFC file that contains all the architectural and MEP components including connectivity information associated with electrical circuits and HVAC equipment. To create an integrated IFC file, we merged all the individual IFC files using an open source BIM server.

Finally, we provide three decimated versions of the point cloud produced by the laser scans of the 5th floor, lobby, and building exterior.

Digital 210 King - Revit Architecture Model
ZIP - 53.332MB

Digital 210 King - Revit Electrical Model
ZIP - 12.550MB

Digital 210 King - Revit Mechanical Model
ZIP - 18.923MB

Digital 210 King - Merged IFC Model
ZIP - 30.321MB

Digital 210 King - XYZ Point Cloud (Low-Res)
ZIP - 5.655MB

Digital 210 King - XYZ Point Cloud (Med-Res)
ZIP - 18.973MB

Digital 210 King - XYZ Point Cloud (High-Res)
ZIP - 61.969MB

Electrical Model of 210 King (Blog Entry)

Mon, 28 Feb 2011 14:21:20 -0700

Figure 1: Single line diagram of electrical room on the 6th floor of 210 King building.

In this blog entry we briefly highlight some important aspects of creating our BIM electrical model for the 210 King building. In general, our electrical model includes the power, system, and lighting layouts that are based on our existing set of electrical drawings. Using these drawings we were able to quickly place components such as transformers, electrical panels, receptacles, and lighting fixtures throughout the building. Placing these elements was only a small step in building a BIM model and additional information was required to define circuits and logical connections. It is also important to note that we were not able to incorporate power switches in this model since our existing drawings of light switches do not correspond to the actual built condition. Instead we focused on proper circuiting of all the light fixtures and power receptacles.

Our process started by understanding the single line diagram (Figure 1 & 2) that illustrate the overall electrical flow and distribution of electrical panels throughout the 210 King building. This drawing was greatly helpful in modeling our electrical room layout while defining the distribution systems and voltage definitions in Revit (Figure 3).

Figure 2: Single line diagram of electrical distribution in 210 King building.

Figure 3: (Top) Distribution system. (Bottom) Voltage definition.

The next step was to model the electrical rooms by placing all the electrical panels, transformers, and disconnect switches. In case of the 210 King building, there are two electrical rooms per typical floor covering buildings 1 & 2 and building 3. There are also some additional electrical closets per floor as part of building 4. Figure 4 displays a sample electrical room on the 5th floor of 210 King building.

Figure 4: A typical electrical room in 210 King building.

Similar to any BIM object we need to provide our electrical panels with certain semantic information before we can fully define a circuiting system. Figure 5 highlights panel name, Max #1 Pole breakers, and distribution system as three main semantics that had to be defined for each electrical panelboard in our building.

Figure 5: Semantic information for a typical electrical panelboard.

Having finished all the electrical room layouts and placed all the electrical fixtures, we were now ready to define circuits pertaining to receptacles and light fixtures. We also had to make sure that all our electrical components had electrical connectors, otherwise we would not be able to include them in a defined system. As mentioned we relied on our existing drawings for circuiting. Our drawings had a tag for each electrical device representing their corresponding circuit and electrical panel. For instance, Figure 6 shows a sample label for a receptacle. According to the label, this receptacle connected to panel 5B and circuit 5. Following this information, Figure 7 shows all the lighting fixtures on a selected floor after being circuited and labeled in Revit.

Figure 6: A typical label showing the corresponding electrical panel and the circuit.

Figure 7: Illustrating all the light fixtures on a selected floor after being circuited and labeled in Revit.

When circuited properly, we can see how by selecting one light fixture causes all fixtures that belong to the same circuit to be highlighted (Figure 8). Creating a circuit schedule is also an important part of this process. Panel schedules display information about the panel, the circuits connected to the panel, and their corresponding loads.

Figure 8: A series of lights that belong to the same circuit.

Our current electrical model represents a very comprehensive documentation of the existing condition. Our Revit electrical families were also carefully chosen to be as close as possible to the existing scenario. However, in the future we hope to have a larger selection of manufactured products as part of a BIM library. A complete catalogue of all electrical fixtures can be implemented within the Revit model for asset management and other related facility management activities. For the time being, our model presents a complete lighting layout for the entire 210 King building in addition to full circuiting of the three floors utilized by Autodesk.

Placing Thermostats in the Revit Model (Blog Entry)

Fri, 14 Jan 2011 12:47:40 -0700

Figure 1: (Left) Two sample thermostats on the 5th floor. (Right) The same thermostats are placed in the Revit model.

In our previous blog we provided a brief overview of how to place custom sensors in a Revit model. Our goal was to create semantic links between physical sensors and their corresponding BIM sensors. In addition to our custom sensor network, our 210 King building is also equipped with thermostats that are distributed throughout the building as a means to report thermal values from various locations (Figure 1). Our building operator can remotely monitor these thermostats through our Building Control System (BCS) interface (Figure 2). The distribution scheme for these thermostats however is still based on the original design assumptions that consisted of large open areas. To meet the ongoing business needs, our building has gone through many retrofits and the existing distribution of thermostats do not provide a high resolution sample of thermal values within each environment. When combined with our custom sensor network we can achieve a rich environment for sampling various environmental data.

Figure 2: BCS interface for monitoring thermostat values.

Similar to our custom sensor network, we need to create a semantic link between existing thermostats in our Revit model and data collected from our BCS. Each thermostat already has a unique ID within our BCS therefore we had to spend some time to acquire corresponding IDs from our BCS system. The following are some sample IDs from various data points in 210 King:

Tracer_Summit_BCU-1-Analog_Input-152-VAV-2-2-123
Tracer_Summit_BCU-1-Analog_Input-2-VAV-1-1-67
Tracer_Summit_BCU-1-Analog_Input-56-VAV-3-2-66

Before adding the semantic information, we placed all our existing thermostats inside our Revit model according to our HVAC drawings (Figure 3). We also tested selected sensors to make sure they correspond to data points found in BCS. To do this, we simply held a heating device against the selected thermostat while monitoring its value through our system (Figure 4).

Figure 3: VAV system as built drawing indicting each thermostat and its name.

Figure 4: Testing a selected thermostat against values monitored through our BCS system.

For modeling purposes, we used a default thermostat family grouped under Electrical Fixtures. Moreover, to meet our future needs we had to incorporate some custom parameters to make sure thermostats will be recognized as IFC sensor objects when exported out of Revit (Figure 5).

Figure 5: Adding custom IFC parameters to a default thermostat family.

Having placed all the thermostats in the Revit model, we cross referenced thermostats IDs collected from our BCS to our VAV system as built drawings to physically locate and properly name each individual thermostat. Figure 6 illustrates a sample thermostat family with an appropriate mark and label that correspond to data collected from the BCS.

Figure 6: Semantic link between a BIM thermostat and BCS data point.

Our building consists of many data points. Therefore, we have created a thermostat schedule to quickly identify thermostats according to their label or associated level (Figure 7).

Figure 7: Thermostat schedule.

In parallel to our HVAC, we have also been developing an electrical and plumbing Revit model based on the drawings provided by our facility manager. In our future blogs will provide an overview of our electrical and plumbing Revit model.

Placing Sensors in the Revit Model (Blog Entry)

Fri, 05 Nov 2010 13:40:59 -0700

Figure 1: (Left) Physical sensors embedded in each cubicle. (Right) Digital sensors placed in the Revit model.

Following our blog on Implementing a Sensor Network at 210 King, in this blog we describe how to place sensors within a Revit model, so we can make semantic links between physical sensors and their corresponding BIM sensors (Figure 1). In general, sensors are just like any other families found in a typical Revit file, although Revit currently does not include a native sensor family by default. Therefore, we created a custom sensor family for the Digital 210 King project. Our sensor family is called "Networked Sensor" and it appears by default under the Specialty Equipment category in Revit Project Browser. Our Networked Sensor family includes the following types of sensor:

Camera
CO
CO₂
Current
Humidity
Irradiance
Light
Moisture
Movement
Occupancy
Pressure
Smoke
Sound
Temperature
Wind Direction
Wind Speed

When creating our sensor family, we had to create a number of custom parameters to address the following requirements for our project:

IFC Parameter: A key consideration in creating a sensor family is to make sure that sensor is actually recognized as a sensor object when exported through international file standards such as Industry Foundation Class (IFC). Figure 2 shows the IFC parameter associated with a sensor, which also includes sensor type i.e. LIGHTSENSOR.

Figure 2: IFC parameter to sensor family.

Linking sensors to a cubicle Another important consideration is how to create a parent-child relationship between the embedded sensors and their hosting cubicle. In doing so we can link sensor data such as Motion to occupancy information associated with a cubicle. Within our Revit model each cubicle is defined as a station (Figure 3). This is a custom parameter we added in order to distinguish among all the instances of our cubicle family. As for the sensors, first we have to semantically indicate at what level this sensor is collecting data (e.g. station), and second, we have to identify a specific hosting cubicle by referencing numbers provided from our facility manager. Therefore, we added two custom parameters to accurately link each sensor to a particular cubicle or station (Figure 4).

Figure 3: Adding a custom parameter to each cubicle to distinguish among all the instances of our cubicle family distributed throughout our office floor.

Figure 4: Sensor parameter to define its scope for data collection, and its specific hosting cubicle.

Linking physical sensors to sensors in Revit: Collected data from each physical sensor is stored in our web server under a specific identification number found on each Phidget sensor. To correlate a Revit sensor to an actual sensor we have to individually select each sensor in the Revit model to write down their specific ID in the field highlighted below (Figure5).

Figure 5: Linking Revit sensors to actual sensor data by sharing the same ID.

We can now collect live and archival building data that is correctly associated with the building information model for visualization and analysis.

Revit Model Sensors - Networked Sensor Family
ZIP - 0.175MB

Mechanical, Electrical and Plumbing (MEP) Modeling at 210 King - Version 1 (Blog Entry)

Mon, 04 Oct 2010 10:59:16 -0700

Figure 1: HVAC model at 210 King St. East.

Since our last blog post describing the 210 King Revit model we have made significant progress in incorporating mechanical, electrical, and plumping (MEP) information into our BIM model. The integration of MEP information with our existing architectural model introduced an additional level of complexity in managing the massive amount information found in a typical MEP scheme. Compared to the long tradition of 3D architectural visualization, production of MEP drawings in a 3D environment is rather new to the AEC industry. Typically MEP drawings are drawn in 2D and then passed on to the architect who will be responsible for coordination among all the consultants. A large aspect of this coordination involves checking for clashes among architectural, structural, and MEP components. This is a particularly challenging task when dealing with large projects. Given the ability of BIM to incorporate information pertaining to all disciplines in a singular 3D environment, we can expect a better workflow and clearer communication among all the disciplines describing a building. Figure 2 illustrates a few screenshots where our 210 King BIM model was directly exported from Revit to Autodesk Navisworks for inspection. Using this method we were able to quickly identify errors and verify the accuracy of our MEP model.

Figure 2: (Top to bottom). Models from Revit Architecture and MEP are combined in Navisworks for quick inspection and annotation. A tagged-view marking a clash between a duct and building envelope.

As mentioned above, MEP modeling has a very short history and existing workflows utilize only a small part of BIMs' capacity in visualizing, analyzing, and simulating a building. Furthermore, lack of appropriate guidelines in BIM for existing buildings presents a challenge in defining an appropriate level of detail (LOD) when dealing with complex MEP systems.

Our HVAC model is mainly comprised of:

Ducts
VAV Units
Zoning Scope
Air Terminals
Roof Top Units

The combination of these components together defines a system that can be described on a topological level. The topological description essentially refers to physical configuration and connections among the components. For instance, our building consists of five HVAC zones supplied through five Roof Top Units (RTU). By selecting a RTU we can follow an entire branch of connected ducts for a selected zone (Figure 3).

Figure 3: Looking at connected ducts supplying fed through a selected RTU.

In our first attempt to model the HVAC system we had to rely on old drawings we retrieved from MEP consultant involved in the 1997 renovation. However, we quickly noticed discrepancies between the drawings and actual layout of mechanical units on our rooftop. Since the time of renovation some equipment, such as a chiller and many A/C units (dedicated to server rooms), have been added to the roof. Figure 4 illustrates an on-site sketch of our current roof top configuration. This diagram is a simplified version that does not include the gas line and our boiler room, but it is being used for a quickly laying out the existing pieces at their approximate location on the roof.

Additionally, finding an appropriate BIM object for all our existing mechanical equipment has been rather challenging. BIM objects are digital models of both the physical and mechanical characteristics for mechanical equipment. For the time being, we have incorporated simpler representation of BIM objects into our Revit model. However, we have been documenting the existing equipments and we hope to work with existing manufacturers to produce higher quality BIM objects. It is worth to note that we can refine the level of detail on our model over time. Therefore, we can replace our existing BIM objects with BIM objects provided by manufacturer of MEP equipment.

In our next blog we will provide some additional details about our MEP model and some of the keys points we have learned during our modeling process.

Figure 4: On-site documentation of mechanical layout on the roof.

Installing a Weather Station to Collect Microclimate Information (Blog Entry)

Thu, 16 Sep 2010 15:06:29 -0700

Figure 1: Weather Station is installed on the roof.

A key aspect of collecting data pertaining to building performance is to accurately measure both the indoor and outdoor environment. The correlation of these data may reveal behavioral patterns of interaction between the indoor and outdoor conditions. Therefore, in addition to our indoor sensor-network implementation we have installed a weather station to collect microclimate information. While many analyses in energy modeling methods use weather files acquired through the nearest airport as input, we are interested in real-time sensing of weather data that can be directly fed into our system. Furthermore, we believe that collecting microclimate information will give us a finer resolution of site specific data.

Our weather station is called WeatherHawk and can measure the following:

Atmospheric pressure
Atmospheric temperature and relative humidity
Evapotranspiration
Irradiance caused by solar radiation
Rainfall
Wind Speed and direction

The data collected is transmitted through a radio modem to a local computer that subsequently transfers the data to our server over HTTP. The following figures illustrate sensor charts of various factors for a selected timeline.

Figure 2: Sensor Charts collected from 210 King weather station.
(Top to bottom) Temperature, solar radiation, and wind speed.

Useful links
Weather Hawk

Implementing a Sensor Network at 210 King - Version 2 (Blog Entry)

Wed, 08 Sep 2010 14:31:31 -0700

Figure1: Interface boards are prepared to be installed at each cubicle.

As part of our focus on collecting, mapping and analyzing building performance data within our office building, we have reached our first milestone by instrumenting a total of 24 cubicles in addition to an office space all located in the research area on the 5th Floor (Figure 2). We are essentially treating each cubicle as a cell for sampling data within the floor space (Figure 2). The advantage of this approach is twofold: First, each cubicle essentially defines a meaningful local boundary for evaluating comfort and energy usage per occupant. Second, the cumulative effect of data collection at such a high resolution per square foot could itself result in the understanding of new qualities about the larger space and more detailed information for the building control system (BCS).

Figure 2: Instrumented cubicles and office space shown as green dots.

Figure 3: A typical instrumented cubicle.

Our implementation consists of pre-manufactured, commercially available sensors. These sensors include temperature, humidity, light, CO2, motion and an alternating current sensor per outlet (Figure 4). Since there are currently no guidelines on the overall arrangement of sensors, our sensor configuration is a result of informal tests we ran to determine areas of high importance within each cubicle.

Figure 4: Sensors (Left to right). Current, Light, Temperature and Humidity, Motion, Co2.

Sensors output measured data as (analog) electronic signals. In order to convert the analog sensor outputs into digital signals, all sensors are attached to interface boards that had to be assembled on site. (Figure 1, 3, 5) In turn, one or more interface boards are connected to a Wi-Fi enabled embedded computer using a standard Universal Serial Bus (USB) interface (Figure 6).

Figure 5: Making of interface boards.

Figure 6: Custom embedded computer mounted at a cubicle partition.

Collected data is sent to a web application that monitors the rate and frequency of sensor updates, as well as storage of the data. In addition to sensor data, this database also manages BCS and weather station data. The web application implements an interface for querying the data.

Useful Links

Phidgets Products

SenseAir

Sensor Network Data Collection, Flow, and Monitoring (Blog Entry)

Thu, 26 Nov 2009 15:17:48 -0700

Left to right: wireless outlet reporters are installed at selected cubicles to transmit power-usage data. Data is collected through a receiver connected to a local PC for processing. Once processed, data is sent to a web database, which in turn can be viewed online by many clients.

As illustrated in the image above, our sensor network data collection system uses wireless outlet reporters to transmit power-usage data from a few selected cubicles. A local PC is set up to receive the data through an XBee receiver and FTD1 cable. Collected-data is processed and organized using a custom application. Finally, this data is sent to a web database where it can be accessed in real-time by any client running the appropriate application on their machine.

In addition to collecting power-usage data, we are also experimenting with how collected-data should be structured and represented for power usage analysis. Essentially, this is closely related to testing the flexibility of semantics and relationships that we can embed in our Building Information Model. The image below shows an idealized version of our approach towards creating a hierarchical Tree View, allowing for multi-level readings of power consumption in the 210 King building. The overall power consumption of 210 King can be broken down into floors, groups (i.e., the Research Group), a specific cubicle, or a specific outlet associated with a cubicle.

A Tree View for representing a hierarchical display of power-usage data.

Alternatively, we can look at the power-usage data outside of a hierarchical relationship. This is particularly important if we are interested in evaluating and comparing a particular group of semantics such as all the outlets in the building. This can be achieved in a Table View fashion as displayed in the image below.

A Table View for representing a categorical display of power-usage data based on a particular semantic, such as building outlets.

We will continue to explore different methods of aggregating and displaying the data collected from our sensor network. As well, as we further integrate the sensor data with the Building Information Model of the office building, we hope to better leverage the existing semantics available to us.

Data Set Research Paper as part of SimAUD: Symposium on Simulation for Architecture and Urban Design (Blog Entry)

Mon, 23 Nov 2009 18:45:25 -0700

To better support the sharing of data amongst researchers working in the area of Building Simulation, the new Symposium on Simulation for Architecture and Urban Design (SimAUD) has added a Data Set Track in addition to the typical full and short paper submission tracks. As the concept of a Data Set submission is new, the Digital 210 King project will be presented at the symposium in April 2010 as an invited Data Set paper, but is available now to show the intended content of this novel type of research contribution. The invited data set contains the detailed 3D building model, the corresponding gbXML file, etc. The PDF outlines all the details of the contributed data files.

The paper Abstract is as follows:

210 King Street: A Dataset for Integrated Performance Assessment
This paper presents a Building Information Modeling (BIM) re-creation of a designated heritage building located in Toronto, Canada. By taking advantage of BIM as a centralized database, which describes both geometric and semantic aspects of a building, this model can be leveraged as a source of input for many forms of analysis. In addition to the BIM model, we present a comprehensive point cloud dataset gathered using terrestrial laser scanning technology. Based on an existing and a living building, this model is an ideal candidate for simulations that can be cross referenced with information gathered on-site.

Implementing a Sensor Network at 210 King (Blog Entry)

Tue, 17 Nov 2009 16:41:46 -0700

The next focus of the Digital 210 King project is to collect, map and analyze the energy performance of the office building. These data are crucial for gaining insights towards validating our future simulations and we are currently experimenting on how to implement and manage the real-time collection of building performance data.

To start, we implemented a simple sensor network to collect power-usage data. Due to the scale and complexity of our building we decided to start this new phase by experimenting at a smaller scale, which involves collecting power consumption at three different cubicles co-located on the 5th floor. In this blog we provide a brief overview of basic steps in preparing and implementing our wireless power-monitoring system.

Our system basically consists of a number of power meters, purchased from P3 INTERATIONAL, in addition to a number of wireless transmitters that would transmit the data from a power meter to our server.

Kill-A-Watt power meter (left) and wireless transmitter (right)

To allow for the power meter to report power usage directly to our server, we had to incorporate wireless transmitters on each Kill-A-Watt™ power meter. This enables the transmitter to listen to data coming from the Kill-A-Watt measurement sensor.

Incorporating the transmitter into the power meter

Once the wireless outlet reporter has been set up, this device can be installed into any plug at a cubicle to read the power consumption. Data is sent through the transmitter and our receiver will receive this data and store it in the server.

Wireless outlet reporter (left), installation at a cubicle outlet (center), and the wireless receiver (right)

We have already observed interesting power usage profiles from several devices in the office environment. In future posts we will present in situ visualization strategies for dealing with multiple types of sensor readings.

Useful links
Kill-A-Watt power meter
Wireless transmitter
Build a wireless home-power monitoring system

210 King St East Revit Model – Version 2 (Blog Entry)

Wed, 23 Sep 2009 11:15:07 -0700

Creating an accurate model of the 210 King office will be an ongoing, iterative process. Our first attempt at modeling the existing 210 King office building in Revit was a tremendous exercise in filtering through the mountain of information we collected from laser scans, hand drawn sketches, AutoCAD drawings, and onsite inspections. We used the AutoCAD drawings of the building from the time of the renovation as a basis for the model and integrated measurements from the other sources to fill in missing information, such as wall thickness measurements and composition of brick walls.

Our next step was to look for ways to further improve and refine our model of the office building. The first version model acted as a great environment where we could compare and reconcile the complexities of the many existing datasets. However, we agreed that our approach to the problem of modeling an existing building needed to be reviewed. We approached the first model by considering the existing office building as a whole from the very start. Many of the issues we highlighted stemmed from this approach. In our second attempt, we instead thought of the task as though we were designing four new buildings that would later interact. We started the model by building the major party walls defining each of the four buildings, thus embedding a clear conceptual separation among the four existing structures. With each building still being considered separately, we added internal structural components: the pillars, the floors, and major interior walls. As a final step, we created openings in the walls where the buildings now connect to one another.

Sample wall section sketch

By working with wall sections and onsite sketches, we tried to refine the relationship between the existing structures while taking advantage of Revit's parametric capabilities. By establishing the entire building's structural grid and each individual building's structural wall as a datum, we were able to create an overall anatomical setup for interior modeling. Furthermore, we paid more attention to the visual fidelity of our model by creating drawings and rendering outputs throughout the process. In particular, the laser scan data proved invaluable for refining measurements of difficult to reach areas, such as external wall details.

Completed envelopes of the 4 buildings

It is interesting to note that the role of BIM in this project is not fundamentally different than when modeling a new design. We are relying on BIM as a platform to help us speculate about the existing building's construction and rationalize certain assembly configurations. By allowing us to easily experiment, we hope to achieve a higher level of conceptual clarity which will be crucial to accurately simulating the overall performance of the building.

Detail of the structure and floor

We are currently working on adding more detail to the interior of the office, including defining office spaces, work spaces, meeting rooms, and common areas. Stay tuned for further updates.

210 King St East Revit Model - Version 2 High Res Images
ZIP - 4.541MB

Green Building XML (gbXML) (Blog Entry)

Fri, 26 Jun 2009 11:08:58 -0700

With recent emphasis on energy efficiency in buildings, energy and environmental analysis are becoming increasingly important aspects of design consideration. Within this context, gbXML is becoming a defacto industry standard for defining thermal models, with support from various industry standard CAD tools. Once the model has been prepared in a CAD tool, i.e. Revit, then gbXML can be used to export an analytical model to analysis software. For example, this analytical model can be used either in conjunction with Autodesk Green Building Studio to evaluate the energy profiles and carbon footprints of design early in the design cycle or it can be imported to Autodesk Ecotect to perform a full range of environmental analysis and simulation. Green Building Studio is for whole building energy analysis, forecasting how a building will consume resources and providing estimates in tabular report form. Ecotect tools measure how the environment may impact the building performance over time and provides graphical displays of information that allows architects and designers to interact with the data in real time. The following chart provides a high-level summary of each product's analysis capabilities:

	Autodesk Green Building Studio Measures how the building consumes resources	Autodesk Ecotect Measures how the environment impacts building performance
Life cycle costs	+
CO2 Emissions	+
LEED Credit Analysis (Lighting)	+
Water use	+
Energy use and cost	+
Airflow/Ventilation	+	+
Thermal Loads	+	+
(Day) Lighting (LEED & Energy Savings) Luminosity Fluxes	+	+
Shading		+
Solar		+
Acoustic		+
Climate	+	+
Energy code compliance		+

Green Building Extensible Markup Language (gbXML) is a schema that was originally developed by GeoPraxis for use in Green Building Studio, an online energy analysis advising tool. This schema is geared towards converting a CAD model into input for the DOE-2 energy analysis tool and EnergyPlus.

gbXML consists of a very specific set of definitions and data requirements focusing only on energy analysis. Its geometric requirements deal only with spatial volumes and thermal zones with relatively simple polygonal boundary surfaces. Therefore, any CAD tool that supports gbXML must export its complex model information in a simplified form.

Prior to exporting the model into gbXML format, we need to define zones within our model to create a series of distinct homogenous volumes of air. Using software such as Revit Architecture or Revit MEP, a designer can add room objects, spaces and HVAC zone objects which will be imported into Ecotect or Green Building Studio through the gbXML format for further analysis.

This simple Revit example consists of some very basic elements such a window, skylight, door, floor, roof and four walls (above, left). Using Revit Architecture we define a room object (above, right) and then we will export this model into gbXML format in order to create an analytical model.

In this view, we are looking at a room object as it appears in the gbXML export window within Revit Architecture (above, left). We can also view the analytical surfaces and their hierarchy prior to exporting into gbXML (above, right). A list of elements are presented in as a tree structure on the right side of the window.

We can import the gbXML file into Autodesk Ecotect for further analysis. The above image shows the import window displaying the content summary of the gbXML file.

In Ecotect we can perform analysis ranging from solar studies to acoustics analysis, thermal analysis, ventilation and air flow. The following image shows a simple configuration of a shadow study based on the time of the day and location on Earth.

We can also import the same gbXML model into Autodesk Green Building Studio. Using Green Building Studio we can run whole building analysis and assess energy goal viability of an intended design. The image above displays an early analysis of our simple example outlining the estimated energy and cost summary. Furthermore, Green Building Studio also provides guidance on the LEED daylighting 8.1 credit and LEED water efficiency credits 1, 2 and 3.

Green Building XML (gbXML) Sample Files
ZIP - 6.428MB

210 King St East Revit Model (Blog Entry)

Wed, 17 Jun 2009 11:06:45 -0700

The development of a BIM model for 210 King required a different approach than what could be expected from the BIM process for new construction. Critical elements needed to be discovered and identified rather than drawn and detailed, structural ambiguities had to be resolved empirically rather than inferred from construction drawings, hidden wall sections had to be assumed rather than be assembled. In the case of 210 King which involves the interaction of 4 buildings built at different times, the interaction between the buildings themselves had to be understood beyond simple connecting blocks. But above and beyond this, the nature of the approach to understanding this heritage building in terms of BIM was very unique.

The process was hampered by unknown wall compositions, surveying inconsistencies, various sets of blueprints and plans that did not match, and ambiguous interactions between elements which could not be discovered without actually ripping holes in the walls. The use of BIM 'families' had to transcend their usual purpose as reusable building elements; instead, 'families' were often used to connect the standard elements to the adhoc maintenance and repairs which occurred over the life of the building. Whether or not this extended use of families will cause an issue in the future simulations is unknown at this time.

The main difference between BIM for existing construction and BIM for new construction is that existing architecture has a life story – it has experience and exists in time. All of this is etched in its architecture, its maintenance, and its inhabitation. By going through this process, some discoveries were made that helped the building communicate its story. For 210 King, the process of creating the BIM dataset had the added effect of revealing the story embedded in the architecture.

Read the complete case study, which details the approach and process used to develop the BIM model, as well as some of the difficulties encountered along the way.

The preliminary version of the Revit model is available for download below. The model includes the exterior walls, foundation walls, roof, and roof deck; interior walls and floors; circulation including elevators and emergency stairs; entrances and exits; windows, doors, and other openings; structural system (beams and columns); and 'room' boundaries and 'space' boundaries for sustainability simulations.

We will be refining and adding to the model in the near future, so check back for updates.

210 King St East Revit Model Case Study
PDF - 0.295MB

210 King St East Revit Model - Revit Architecture
ZIP - 26.811MB

210 King St East Revit Model - Revit MEP
ZIP - 28.735MB

210 King St East Revit Model - IFC
ZIP - 5.663MB

210 King St East Revit Model - gbXML
ZIP - 0.710MB

Building Information Modeling (BIM) (Blog Entry)

Mon, 08 Jun 2009 15:47:27 -0700

Many types of geometry can be used to represent the shape of a building. We have previously shown how simple point cloud data can be collected from a laser scanner to display the locations of points on surfaces and objects in the building. A more common way to represent three-dimensional geometry in a digital file is to connect a few points together to create simple polygons like triangles and rectangles. In this way, a large surface can be represented by just 3 or 4 points. Typically, the geometry is tagged by a user to further describe its properties such as the color and transparency level of the surface. To properly describe a model of a building, additional properties are needed to describe the real physical materials that will be used when the building will be constructed. For example, a rectangle may be tagged as being a "wall" or as a "door". In this way, the building model can be made of known objects with known properties. When a model contains these higher-level building-specific tags, it is called a Building Information Model (BIM).

To create a BIM for a specific building project, advanced software is needed to help the user author sensible models. For example, the software will know that a "door" exists within a "wall" and that it can open in a certain direction. These applications are sometimes called "parametric" because the entire model can be easily changed by modifying a few parameters. In the same manner that a spreadsheet program can recalculate all the totals when a single value is changed, BIM software can update the whole building model when a parameter, like a wall type for example, is changed. Some BIM applications are also connected to vast databases of building materials catalogs. This supports the advanced feature that when the user adds a "door" to the model, for example, the program can present a catalog where the user can choose a real door from a specific supplier and the correct shape and size will be used in the building model. Also, an ongoing parts list can be maintained automatically by the software so that, at any time, the user can see how many sheets of drywall are needed or how many doors to order and how much they will cost.

Beyond tags for materials and parts, BIM can include structural and mechanical data, information about the location and orientation of the building, or construction staging and scheduling. Essentially, BIM is a digital representation of the physical and functional characteristics of a design. In contrast to most common 3D applications that present a building through a set of simple geometrical components, BIM facilitates the assembly of actual building components by incorporating intelligent and contextual semantics such as floors, beams, pipes, roof, etc. Software applications, like Autodesk Revit, are specifically designed to support BIM and the collaborative BIM process. The process of creating a BIM dataset typically involves a number of experts in different disciplines to create various parts of the building model. BIM is essential for large construction projects. Without BIM, small changes could mean hundreds of floor plans must be redrawn. Also, as mentioned above, coordinating many simultaneous users would be impossible. However, when using BIM software and methodology, updates from multiple users can propagate through the model instantly. Furthermore, using BIM, another major aspect of modern building design becomes feasible: simulation.

Simulation software can import a BIM database and it will have all the information it needs to analyze lighting, heating, water usage, and many other emergent properties to which a certain building design may lead. Simulation can be used to help optimize the use of resources to create a more sustainable building. From Revit, the BIM can be exported to the Green Building XML (gbXML) format. Specific simulation products, like Autodesk Green Building Studio and Autodesk Ecotect, can then import the gbXML file and begin simulation to measure and analyze the sustainability of the design. Linking these products to Revit gives architects quick feedback on design alternatives early in the design process.

Energy analysis requires spatial information – it is essentially a simulation of energy movement in, out, and through the rooms and volumes within a building. Which surfaces are exposed to the outside? How many are exposed to sunlight? What are the number, size, and orientation of openings in each space? How much heat is generated by internal lighting and equipment? In the past, this information was manually calculated from 2D drawings. An engineer would use building plans, elevations, and details to collate spaces (type, area, volume), surfaces (including adjacency and thermal properties), and shading. All this information is latent in a Revit model, and in a form that is much easier to interpret than 2D drawings. And, if the project is consistently structured, software such as Autodesk Ecotect Analysis can be used repeatedly right from the beginning of the project. This is a very important contribution to the design process at a stage when change is still possible. [excerpt from Ecotect User Guide]

To effectively use the simulation tools, similar spaces in the building should be blocked out as "zones". These building volumes are individual units, or thermal zones, in the simulation. Perimeter rooms facing the same direction should be grouped together. Core zones with little or no exterior exposure should be grouped together. Similarly, unconditioned support spaces, such as restrooms, stairwells, elevator shafts, and storage spaces, should be grouped together. If need be, a mechanical engineer can provide guidance on the zone layout.

Building Information Model (BIM) - Sample Revit File
ZIP - 6.237MB

Validating measurements (Blog Entry)

Tue, 02 Jun 2009 13:00:00 -0700

Although elevation plans exist for the building (see Backgrounder entry), the accuracy of those plans is unknown. Normally, a number of physical measurements can be made. However, there are many situations where it can be very helpful to have point cloud data for a site, especially for very large sites. For example, the 210 King lobby is two storeys in height and would be difficult to measure with traditional methods. Also, non-flat objects, such as curved walls or cylindrical ventilation ducts can be challenging to measure and place within spaces. In some cases, contractors may discover additional measurements are needed once a project has begun requiring additional site visits. However, with a detailed 3D building scan, enormous numbers of measurements can be taken without ever revisiting the physical site.

To build confidence that the point cloud accurately reflects the real dimensions, we went through a validation phase of comparing a number of physical measurements against virtual distances measured from the point cloud data. We ensured that comparison measurements were made in all three dimensions both within a single scan and, more importantly, across combined scans.

Shown here in color are measurements from the physical site together with distances queried from the point cloud data, shown in grayscale. Although measuring modern machined elements (like drywall or metal posts) was a trivial task, we found it to be challenging to exactly measure physical dimensions for brick or wood materials, like interior walls and beams. The same challenge existed in taking virtual measurements where the width of a wooden beam could be measured from any number of (non-planar) samples, for example. Given these issues, we still found a stable average difference to be .06 inches which is equivalent to a precision value of 1.524 millimeters.

We found Autodesk Navisworks (building information modeling [BIM] project review software for construction simulation, scheduling, and coordination), to be a good tool for 3D navigation around the point cloud data. Using the Walk tool in the Steering Wheels (shown below), we could more easily move to the correct location in the data set. Also, the measurement tool in Navisworks was easy to use and we could confirm distances quickly.

Overall, this process gave us great confidence in the point cloud scan data set. Finally, the accuracy of the elevation plans could now be seen to be quite low. The real physical building plan was significantly different from the original drawings we had so this data set was very helpful in the accurate creation of our 3D BIM model of the 210 King building.

Validating Measurements High-Res Images
ZIP - 12.110MB

Working with point cloud data (Blog Entry)

Tue, 26 May 2009 12:00:00 -0700

Point cloud data is conceptually quite simple. There is no structure to the points; they all exist independently of each other in an unordered list. Each point is specified by three numbers indicating the distance from the Origin point which is simply written as (0, 0, 0). Each number indicates the distance from the origin along the x-axis, y-axis, and z-axis, written as (x, y, z). One of the axes (often z) is interpreted as the height so when this value is zero, the point is said to lie on the ground plane. The x and y values are typically used to represent the distance from the origin along the "length" and "width" on the ground plane. Each point can also have a color value associated with it, expressed as a combination of red, green, and blue values (that range from 0 to 1), written as (r, g, b). In our data set, the scanner returned only a reflectance value so this is stored as a grey value, or brightness. A grey color will have equal values for r, g, and b.

The point list format is six values (x, y, z, r, g, b) with one point on each line in a file. For example, if a point was ten units above the origin and was a mid-tone grey, the point data would be (0, 0, 10.0, 0.5, 0.5, 0.5). Above is some code is shown to read in the point list from a file. Using the QT user-interface framework, we have made a small program to read in the point list from a file (with a sample point cloud file from the 210 King lobby, as part of the downloads). The source code is provided, but there is also a Microsoft Windows sample application pre-built that can be run directly.

CloudTool Binary and Source
ZIP - 4.950MB

Lobby Scan Ascii Point Cloud
ZIP - 8.359MB

Some sample datasets (Blog Entry)

Mon, 11 May 2009 15:28:37 -0700

We are in the process of anonymizing the complete dataset, removing identifiable objects, license plates, etc. The full dataset is coming soon. In the interim, we are releasing a subset of the scans.

In this image, one can see a 2D representation of the 3D point cloud scan data collected in a single scan from one position. The scanner was placed at the bottom of the stairway to the 6th floor that appears prominently on the right-hand side of the image. The distortion in the image is similar to the distortion seen on a map of the Earth. For example, all the pixels along the bottom of the image form a small circle at the bottom of the 3D data set. When stretched out to form a 2D image, the small circle becomes a long line. This is similar to the way that Antarctica is shown as a long thin land mass at the bottom of a map, even though it is really a (mostly) circular land mass.

Note that the brightness of a particular pixel does not represent the distance from the scanner but rather represents its reflectance to the laser. That is, in addition to generating a 3D value for each point captured, an associated brightness value is stored. On the images and scans shown here, we have made no changes or transformations to the data file we received from the scanner. In the center of this scan image, taken in the doorway of an office, the metal doorframe is quite reflective and shows a white "hot spot" in the middle where the laser was reflected significantly. However, if one examines the 3D data along the door frame, it is completely flat and contains the correct distance values.

Another noteworthy relationship is that objects closer to the scanner cast a larger "shadow". This means that, for example, the desk in the office hides quite a bit of the floor behind it. When looking at the 3D scan data from another angle, it becomes quite clear that no data is collected "behind" the desk and the resulting empty space in the data set looks like a shadow even though no lighting is present in the 3D point cloud data set. This is why it is helpful to have overlaps in multiple scans to better capture the surfaces that may be obscured from other angles.

In the case of this meeting room, combining two scans, taken close to each entrance to the room, gives much better coverage of the room than could be achieved from a single scan in the center of the room. The Faro software can align multiple scans (as described in a previous article), so that a single point cloud data set can be created of complex spaces with multiple rooms and areas, from which geometric surfaces and other features can be created.

Finally, the Autodesk lobby at 210 King is shown here also combined from multiple scans. Again, the use of multiple scans increases coverage around obstructions like the posts in the middle of the space and the corner wall in the middle of the L-shaped lobby.

The image content also contains some interesting artifacts. For example, at the top of the post near the chair, there is a flag that was swaying to one side from some internal air flow. However, the flag is distorted in a way similar to the distortion achieved when moving a document around on the scanning glass of a photocopier. This distortion is created by the different positions and shapes of the flag as it passes through the vertical scan trace at the moment that the scanner was pointing in that direction. These artifacts reveal the scanning technique and to some degree, the speed of the rotation of the scanner head.

Completing a scan of a building provides a great deal of data that could be used for many purposes. Our primary use is to provide a way to perform multitudes of measurements that could perhaps be done using traditional methods but would be incredibly time consuming and resource consuming. As the accuracy of our floor plans is unknown, we can now use the data set to properly measure, in all three dimensions, the size and shape of 210 King.

Download the full resolution scans below. To view the file, you will need to download Faro Scout LT viewer, a free download available from Faro.

Open a faro workspace file (.fws) in Faro Scout. Once loading the dataset(s) is complete, you can view the 3D point data by right-clicking on the folder that contains the scan files and selecting New View and then 3D View. To view the Planar projections as in the images above, you can right-click on the individual scans and select New View and then Planar View.

Navigation tools can be found in the toolbar at the top:

The Examine tool allows you to orbit around the data.
The Pan 3D tool lets you move from side to side and up and down.
The scroll wheel lets you zoom in and out.

There are also preset views:

Canonical views along the primary axes can be access using the Top View, Right View, and Back View buttons .
A fit-to-view button moves the camera far enough away so that all data points can be seen.

The right-click context menu in a 3D View provides more options under "Visibility Settings...", where the camera projection can be changed from perspective to orthographic, the background color can be changed, as well as the display size of the data points.

Sample Scan: Fifth Floor Landing
ZIP - 61.195MB

Sample Scan: Office Space
ZIP - 60.969MB

Sample Scan: Meeting Room
ZIP - 91.343MB

Sample Scan: Lobby
ZIP - 114.867MB

Combining the scans (Blog Entry)

Thu, 07 May 2009 15:06:13 -0700

We spent several days setting up for and capturing the scan data for the fifth floor and exterior of 210 King East. Our next task was to combine all of these into a single data set. As mentioned in a previous entry, we posted registration markers around the office. The purpose of these was to provide a common point of reference between the scans that could be aligned in 3D space. This is why we tried to capture at least 3 registrations markers in each scan, which is enough to define a plane in 3D space.

We used Faro Scene to combine the scan data to map the registration markers to each other. This process is interactive: each marker must be highlighted by the user in each individual scan and then manually tagged with the appropriate ID number. The software then automatically calculates the relative 3D offset of each scan to every other. In addition, using the position and elevation information from the survey team, it positions the scans from street level and lobby to those on the fifth floor, effectively positioning all the scans absolutely in 3D world-space.

The software is also able to automatically match unique features within a scan. However, this is most accurate when the features in a scan are very large and distinct. We use this method to align the scans of the exterior of the building.

This map shows where the markers were located in and around the office, as well as the location of the scanner during each scan. Higher resolution versions are available in the download file below.

Here we see the complete dataset of all 53 scans. Since the exterior of the building was scanned from street level, we have no scan of the building's rooftop, so the fifth floor is clearly visible. In the lower image, each individual scan is uniquely color-coded to show how the data from different scans overlap.

Combining The Scans High-Res Images
ZIP - 8.187MB

The 3D scanning process (Blog Entry)

Wed, 29 Apr 2009 18:38:30 -0700

The first digital dataset we want to provide for 210 King Street East is a 3D laser scan of the office. As the name suggests, laser scanning employs a laser to determine the distance from the scanning device to surfaces around it. This real-world spatial data is collected as points in 3D space relative to the position of the scanner. Carrying out a full scan of the entire building would have been a huge task, so we limited the scope of the scanning to the entire fifth floor and the rooftop terrace, as well as the lobby and the exterior of the building. In total, the dataset contains over 1.3 billion data points.

In support of this project, Faro Technologies graciously provided us with a Faro Photon 80 laser scanner. This self-leveling, tripod-mounted unit is capable of scanning a radius of 0.6 to 76 meters, with a precision of 2mm. Since a laser scanner can only capture data in its line of sight, it is necessary to take multiple scans of a given room to eliminate the effect of obstructing features, such as pillars, doors, and desks.

To facilitate merging the data from these individual scans, we need to identify unique features in each of the scans that relate to each other. We accomplished this through the use of position registration markers. We spent the first morning setting up these markers around the office. When positioning these, we had to ensure that any given scan would have line of sight to at least 3 of these registration markers. A total of 61 registration markers were positioned.

The next step was to scan the office. Using a floor plan, we figured out the best place to position the scanner to minimize the number of scans required, which also reduced the amount of redundant data captured. Each scan took approximately 5 minutes. Over the next day, we carried out 53 individual scans of the office.

Simultaneous to our scanning efforts, surveyors from Bennett Young were determining the terrestrial space position of the registration markers in terms of coordinates and elevation above sea level. This will allow us to know the precise location of any 3D point in our dataset in terms of the real-world. The surveyors measured the registration points on both the interior and exterior of the office, correlating the interior measurements to exterior street measurements from the deck.

We have just started post processing of the data. Here is a sample of the point data from one scan – approximately 27 million points – the spherical data is shown as a flattened image. Registering each scan to the other is an interactive process whereby the ID numbers of the registration markers need to be manually confirmed. Stay tuned; more data will be posted shortly.

The 3D Scanning Process High-Res Images
ZIP - 22.173MB

Backgrounder: 210 King Street East (Blog Entry)

Wed, 22 Apr 2009 16:08:56 -0700

Situated in downtown Toronto, 210 King Street East is the current home of Autodesk's Toronto office. This eclectic workspace spans four historic Toronto warehouses, built between the 1930s and 1960s, with a total 145,000 square feet of office space. Toronto architecture firm Kuwabara Payne McKenna Blumberg was commissioned to carry out the integration and renovation of the warehouses, which was completed in November 1997. At the time, the company was called Alias|wavefront, later renamed to Alias Systems in 2004, and subsequently acquired by Autodesk in January 2006.

Three of the warehouses, including 204 and 214 King Street East, are designated as heritage buildings, requiring that original features of the buildings be preserved. This greatly limited the extent of possible renovations. 210 King Street East, on the other hand, suffered damage from a fire and did not receive the same designation, making it an ideal candidate for the main entrance to the office. This space was renewed through the creation of an expansive two-storey lobby and striking steel staircase, the signature feature of the office.

The open-concept office layout balances both public and private spaces. Designed with socialization and collaboration in mind, employees can congregate in the central kitchen and cafe of each floor, as well as on a rooftop terrace atop 214 King Street East. Quiet work environments are found around the periphery of the office, offering secluded spaces to focus on work. The landmarks of each floor, including the kitchen, cafe, main foyers are mirrored on each floor.

The renovation incorporates exposed brick walls and original hardwood floors with industrial-inspired steel and concrete accents. Adding to the loft-like atmosphere, ceilings span 12 to 14 feet in height and wooden beams, cable routing, and ventilation systems are exposed. In total, 600 miles of cable was laid throughout the building. Diamond plate steel ramps mitigate the differences in floor heights of the three westerly buildings. The floors in 214 King Street East, which lay almost half a floor below the rest of the office, create a split-level effect.

It was not without difficulty that both the goals of preserving historic buildings and creating a unique high-tech workspace were achieved. However, 210 King Street East is a prime example of how these two values can be blended into a functional, yet appealing and interesting work environment.

In the past, stories about this renovation have been featured in the following publications:

Small, Medium, Large
Azure
November/December 1998
Vol. 14, no. 120, pp. 34-39.

Case Study 3: The Funky Fort of Nimble Teams
Eciffo
Spring 2000
Vol. 36, pp. 36-43.

Backgrounder: 210 King Street East High-Res Images
ZIP - 33.707MB

Azam Khan keynote at SpringSim 2009 (Blog Entry)

Thu, 02 Apr 2009 14:01:12 -0700

At the ACM SIGSIM SpringSim 2009 conference in San Diego, California on March 23, 2009, Azam Khan, head of the Environment & Ergonomics Research Group of Autodesk Research, gave a keynote talk called Systems Architecture. The focus of the talk was the great complexity involved in architecture and urban design which has led to the problem that buildings are the largest Green House Gas emissions problem. This complexity results in many counterintuitive aspects of sustainability that are only clear from modeling and simulation outcomes. The talk also announces the Digital 210 King project to the simulation research community and announces the new SpringSim 2010 Symposium on Simulation for Architecture and Urban Design.

Abstract
Specialization has helped us make progress in all aspects of academia, industry, and life. The benefits of modern technology, from indoor plumbing to the internet, are the product of this basic "scientific" reductionist approach. However, we now find ourselves faced with many strange counter-intuitive real world problems that reveal the true complexity of human activity and our relationship to the environment. The intuition of groups in several sciences is that "what worked before won't work now." From specialization, we turn to integration. For example, systems biology is a field that now works toward a holistic model of life as a single complex system to help describe emergent properties that would be difficult or impossible to model directly. We borrow this integrative approach applying it to the fields of architecture and urban design. These fields have been using simulation in a variety of specific areas for some time and we now propose an integrative methodological and technological framework for collaborative investigation to help take a bite out of the world's largest ecological problem: buildings.

SpringSim 2009 Keynote Presentation
PDF - 5.960MB

Symposium on Simulation for Architecture and Urban Design (Blog Entry)

Mon, 30 Mar 2009 15:54:51 -0700

We are happy to announce the Symposium on Simulation for Architecture and Urban Design! This is a new event designed to bring together the architecture research and simulation research communities that will be part of the ACM SIGSIM SpringSim 2010 multi-conference in April 2010 in Orlando, Florida. The Call for Papers will be available soon and we are working hard on the 210 King BIM model to support new research in this area.

We hope the symposium will bring together currently disparate efforts in design, modeling, data capture, sensor networks, simulation, visualization, and validation of architecture and urban design research. Please see the symposium website for details and please let your colleagues know about this exciting new research venue!

Symposium on Simulation for Architecture and Urban Design

Website launch (Blog Entry)

Mon, 23 Mar 2009 15:53:40 -0700

Welcome to the launch of our new site documenting the process of creating an advanced Building Information Model (BIM) of our building and offices of Autodesk Research in Toronto, Canada.

The purpose of creating this data set, and sharing it with the world, is to provide a solid foundation to the architecture research and to the simulation research communities with the hope of accelerating innovation in sustainable design. Buildings are the primary cause of global climate change attributable to human activity creating 48% of all Green House Gas (GHG) emissions, primarily from the use of electricity for heating and air conditioning. Here at Autodesk Research, we feel we can contribute to the solution by developing new technologies to help architects and engineers to make informed design decisions.

Our offices in downtown Toronto are at 210 King Street East near Sherbourne Street and this is where we begin our journey. We are planning to create several data sets and will share all of the data on this site. The advantage of working with a digital data set of an existing structure is that design variations and simulations can be validated against real data. Also, real usage data can be collected over a long period of time.

The quest for sustainable design will involve a diverse set of partners and expertise so please contact us and help us to redesign design.