This project considers how to combine urban farming into the plan of new Xiong’An using neural style transfer.

STORY: A DAY OF AN ORDINARY RESIDENT

8:50 AM rushing to work | in the community garden

.

.

.

.

reference: https://youtu.be/OLh74U_zVLk

12:00 PM dining with client | at the restaurant

.

.

.

reference: https://youtu.be/ea0cndjPjPw

3:30 PM taking a break | at the balcony of the office

.

.

.

.

reference: https://youtu.be/OLh74U_zVLk

3:45 PM ordering vegetables for dinner | at the balcony of the office

7:30 PM taking a walk by the lake

.

.

BACKGROUND
Most investment strategies are based on the principle of diversification— to reduce risk, create a portfolio that includes multiple investments. That said, our current food system does just the opposite. Its success depends on concentrated, large-scale industrial agriculture in only a few places. In Beijing, only 10% of the vegetable are from local farms. With limited additional capacity available to produce more food for a growing global population, Beijing also suffers from periodic, severe droughts that can impact food availability and pricing. As it stands today, our food system is at risk of failure. To feed a growing population and to reduce the energy, we need to grow our food closer to where we live—in cities.

location of Xiong’An

Located 100km southeast of Beijing, Xiong’An aims to allocate the non-core function of Beijing, the capital. Unlike other big cities in China where land resources are insufficient, Xiong’An is the ideal context for developing urban farming. Even more significant is the Chinese diet, which consists of approximately 60% leafy greens including spinach, lettuce, and kale. Leafy greens are an excellent choice for hydroponic and aquaponics growing systems. They thrive in the simplest of setups and don’t need a lot of extra attention. They grow quickly and weigh little, both of which make them an economical and efficient option.

URBAN FARMING SYSTEM

the farming system

The urban food system consists of 3 layers. The first layer is the foodservice points scattered in the city. They help to gather vegetables grew in the city landscape as well as grow vegetables indoors. If someone in the community makes an order for the vegetable, they’ll pick the fresh vegetable and have it delivered to the client’s home in a couple of hours.
The second layer is the landscape system which is the combination of small farm landscape, landscape architecture, vertical farms and rooftop farms. They are not merely as sources of food, but also serve as educational landscapes where urban children learn about food and nature. They are part of our green infrastructure as well, providing ecosystem services rather than exporting pollutants. They bring communities together around the elemental joy of the pastoral landscape.
The third layer is the shared space in the community which can be used for farming as well. But since the landscape system and the service points have produced enough vegetables for the resident, farming is not necessary. Residents are free to use the space to do what they want.

NEURAL STYLE TRANSFER

details of Maya fields

In this project, the first step is to create a set of Maya fields, including the urban infrastructure patterns and the residential patterns, then use these patterns as content pictures in neural style transfer.

After careful consideration and experiments, the satellite image of Barcelona and Venice are used as style pictures because the former one has a very clear urban structure and the latter one has recognizable red roofs and wonderful landscape.

satellite image of Xiong’An (6km*6km)
Maya field

The second step is to create the landscape. This time the satellite image of Xiong’An is used as the content picture and the fields created previously are used as style pictures. The next step is to map the service points in the city. Then combining the landscape and the architecture together.

urban plan
Xiong’An urban plan (6km*6km)
Berilin

nolli map (6km*6km)

satellite image of Xiong’An (6km*6km)
online grocery service points in Shenzhen

Xiong’An map is the content picture again, and the map of online grocery service points in Nanshan district, Shenzhen is used as style. The last step is to create a clear urban structure from the result of the previous steps, the nolli map of Berlin is used as a style picture and the result of the previous steps is the content picture.

3D MODELS

scenarios
CNC model

CONCLUSION

From the perspective of technique, neural style transfer allow more possibilities in design. It helps to generate patterns that beyond our imagination and inspire designers.

patterns generated with different style pictures and parameters

From the perspective of the research topic, the urban farming system actively supports a more sustainable local food network and improve the quality of life in the city.

the farm city (6km*6km)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s