< Back to Blog

February 2020

,

Aremi MacDonald

Winds of Change – Or How We Can Change the Wind

 

“Wind is not necessarily an uncontrollable force of nature. The wind can be quite complicated. We are not helpless before it, but we have to be much more proactive and plan for it.”

– Isla Tanaka, Winter City Planner in Edmonton, Canada/The Winter City Podcast.

 

The way we design our cities has a huge impact on the surrounding microclimate and pedestrian comfort, which is why it's so important to design with care. Susan Holdsworth, Strategic Planner for the City of Edmonton and host of The Winter City Podcast recently interviewed our co-founder, Sebastian Desand alongside Gwyn Richards, the Assistant Director & Head of Design for the City of London.

The candid discussions dig deeper into the importance of implementing wind studies at early stages of the design process, and how accurately predicting wind behaviour helps mitigate potential risks associated with the aerodynamics of our urban skylines. These efforts can help build an inclusive and sustainable city for all. 

 

Sebastian: My name is Sebastian Desand. I'm the CEO and co-founder of Ingrid Cloud, a spin-out from research at KTH in Stockholm, Sweden.

Susan: Can you tell me a bit about Ingrid Cloud and what it is?

Sebastian: Ingrid Cloud offers software as a service (SaaS) that allows users to run an accurate and high fidelity wind simulation without having any prior expertise within the area of cloud simulation or CFD as it's called in computational fluid dynamics.

We have a system that takes care of the entire workflow for that end, and that means that the user can upload a model 3D geometry of an urban area or a building in order to simulate how the wind behaves, and how the building affects the environment- the wind microclimate in that area with the purpose of understanding how to design the area in as good of a way as possible to ensure good pedestrian comfort and safety.

Susan: How is it better than other software or other ways of measuring wind impacts?

Sebastian: Ingrid Cloud is fully automated. As I mentioned, Ingrid Cloud is a spin-out from research, and that research has developed quite unique technology where we use intelligent algorithms to take care of the simulation, which is a quite complex workflow or process to actually do.

Traditionally, when you do a simulation today using traditional software you need to be an expert and it takes quite a long time to run a simulation, especially if it's a high fidelity simulation. You need to set a lot of parameters and try it in iterations and do a lot of manual work with that.

With Ingrid Cloud that is not necessary. Our system and our algorithms take care of that and replace all of the manual work with our automated flow. We use supercomputers to run the simulation on hundreds or thousands of cores, and that enables us to deliver a result within a few hours after you initiate the simulation. It's reliable results that are not dependent on the knowledge or expertise of the user. It takes away that human factor or possible human error as well.

Susan: What would you say the human factor is for other ways of measuring wind? What is the margin for human error?

Sebastian: First of all, I would say that there are two different areas where you can make mistakes. The first one is the actual technical knowledge of performing the simulation and the other part is; analysing the results, understanding the results and also post-processing and creating good visualisations so that you actually draw all the right conclusions from the simulation.

I heard a joke from a researcher once that said if you ask three CFD engineers to do a simulation, you will get at least four different results, where every one of them will be different and they will be claimed that they are correct. Every one of them. There is a very broad area where these things can go wrong. And actually, that is also the case with wind tunnel experiments, so it's not only limited to CFD simulations.

There have been tests where the same model was running in wind tunnels and there was a 20 percent difference in the results of the three different wind tunnel vendors or facilities. Even a wind tunnel is a model of reality. There is a big possibility of actually getting the wrong results. That’s why we think that by automating it and really making sure that we have quality-assured software and numerics, we can ensure that the results of the simulation are accurate and reliable.

One way of doing that is to only run high fidelity simulations in contrast to low fidelity. To explain that let's say that you look at a waterfall. You see the water going down at high speed and you want to capture that. A low fidelity version would be to have a camera and then have long exposure on that- it makes it a blur. It's an average picture of a waterfall, right?

A high fidelity simulation would be to actually either take a lot of pictures or to record it with a video camera so you can actually see the turbulence that is going on. In our experience, that reduces the risk of not getting the solution right as well.

Susan: That's a great analogy to explain the difference between the way wind analysis is normally done and what your technology offers. Can you tell me how accurate Ingrid Cloud is? How fast is it, and how expensive is it?

Sebastian: With regards to the accuracy, it depends on what you want to have from it. You can either do a four-wind-direction simulation which is not as expensive and more affordable, but it won't be as refined or high resolution. Then you could run an eight-wind- direction simulation or 16, or even a 36-wind-direction simulation.

36 wind directions are, for example, what the City of London recommends when you do wind studies. But it could be fine to run a four-wind-direction or eight-wind-direction [simulation] when you're early in the design process, when you may have different design options, you want to try different things and the resolution level is not super important.

You just want to be able to compare different options. There are cases where they have done wind studies or even wind tunnel experiments to verify that design. But it still led to pretty much a disaster. The reason for that is that they didn't do the wind simulations early in the design process. And when you do revocation, it's very late. It's when you typically want to have a building permit, or you want to get a green light. In those cases, it's very expensive to go back and change the design.

The property developer doesn't want to do it, maybe the municipality perhaps doesn't even want to change things and the architects and designers certainly don't want that. No one really wants to change. It's easy to sort of, "Yeah, well, you know, we think it's okay. People are used to wind." So they sort of talk themselves into the, "Yeah, this is fine. It's not perfect, but it's OK," because they don't want to make these expensive changes. That is one of the big risks in the design process by not doing it early.

By doing it early, it's not only about mitigating risk. It could actually create a really nice environment and ensure that property value won't be affected, or that there won't be a safety issue and so on. You need to take that holistic view of always looking at different aspects like lighting and wind and temperature and so on.

Susan: It sounds to me like the point you made about being able to do early wind studies at the very early concept stage is important also because it's before the proposed development is shared with the community- and a whole bunch of community consultation is done before they've already negotiated changes with the community. It's really too late at that point to change the urban forms significantly, so it sounds like it's the biggest opportunity to have an impact on reducing the wind impacts of a proposed development of a building.

How many people are using your software at that early stage? Of very early concept stages of development?

Sebastian: Perhaps 70 or 80 percent of our customers/customer base is using or considering using it early as well- and not only for verification purposes. And on that topic, we are right now developing another application for the early-stage design process where architects, urban planners, and designers can upload a number of different design options and then we rank them. For that purpose, it's sort of a comparative study where you want to see which three designs are the best out of these 10, for example. At that point, you're not interested in looking at specific simulation results. You're just interested in comparing them, and that is one way of introducing simulations early.

Susan: That sounds like a really exciting thing you're developing there with the various design options to evaluate at that early stage. The City of London has just released its new guidelines around wind, and I think they're one of the strictest ones out there. Does your software meet their guidelines?

Sebastian: First of all, we think it's great that they initiated that work and that they launched these new guidelines.

It's good to have those guidelines there. That really helps designers know when they should do wind studies, and what kind of wind studies they should do to begin with. We are right now actually launching and implementing some changes in the software that makes sure we are compliant with those guidelines.

Susan: We've been getting pretty granular about your software and so I want to move into more higher-level thoughts. Why is it important to understand wind?

Sebastian: If you're designing and planning and building in an area where the municipality has knowledge about this, you could actually risk not getting a building permit for one. Otherwise, I mean, you don't want to develop an area that could have such bad wind conditions that it will affect the property value. If you're a real estate developer, that affects the investment.

"As a responsible or sustainable company, and as a city or municipality, there is the issue of creating a city or area that is uncomfortable or potentially dangerous for pedestrians."

 

I would like to also state that something that many times is overlooked when it comes to wind studies, is that you look at how the wind is affecting pedestrians. Then you're making any difference between pedestrians because not all pedestrians are young, strong and healthy. Some pedestrians are old or have physical challenges or have some illness for example. They are affected by wind in a totally different way compared to a young and healthy person. You need to take that into account as well, and that is actually part of the U.N.'s Global Goals, target 11.3 which is inclusive and sustainable urbanisation. You can't design a city for some. You need to design a city for all.

Susan: Yes. Completely agree, and as we know, we're all only temporarily abled. Are municipalities and developers interested? Do they get how important this is?

Sebastian: More and more. Many still do not have knowledge about this, but it's really fun when you meet municipalities for example, or real estate developers who are architects and talk about this. You really can see that they understand and they say they know that it's important. Then they start thinking about areas where they know there are problems.

I think sometimes people look at wind like it's a force of nature. It's something you can't control or predict, or it just happens, and you just have to live with it. But no, you actually can predict it and it's affected by how we design. And that's something that many or most people actually aren't aware of.”

 

That knowledge is getting more and more spread I think. That helps when cities have guidelines because they also not only demand certain things, but it helps spread the word and spread the knowledge.

Susan: Okay. Well, thank you very much.

Sebastian: Thank you for having me.

 

Listen to the full podcast here: