AI vs. Engineering: Designing a Keel Cooler for an Aluminum Catamaran

Leveraging AI for Complex Engineering Problems

• 💡 The build of the custom aluminum catamaran "Delos 2.0" involves extensive research, prompting the use of AI to work smarter, not harder.
• 🧠 AI has been initially used for simple calculations, such as determining the volume between two tubes, demonstrating its utility in streamlining engineering tasks.

The Challenge of Electric Motor Cooling

• ⚡ Electric motors are not 100% efficient, meaning a portion of their energy is lost as heat that needs to be dissipated.
• 🌡️ Two cooling methods are considered: a closed coolant loop with a heat exchanger, or a keel cooler that exchanges heat directly with the surrounding saltwater.
• ⛵ For an aluminum sailboat, the concept is to use the hull itself as a heat sink, conducting heat from a closed-loop system to the outside ocean water.

AI Agents Tackle Keel Cooler Design

• 🔍 The primary engineering challenge is determining the required surface area of the keel cooler to dissipate the motor's heat.
• 🤖 Three AI agents—DeepSeek, ChatGPT, and Google Gemini—were tasked with calculating the heat dissipation needs for a 25kW motor at 90% efficiency, requiring the removal of 2.5kW of heat.
• 📊 DeepSeek estimated 0.22 sq meters, Gemini suggested 0.33 sq meters, while ChatGPT provided a significantly larger estimate of 1.25 sq meters.

Analyzing AI-Generated Estimates

• ⚠️ The significant variation in AI outputs highlights the importance of verifying AI assumptions, such as heat transfer coefficients and temperature differences.
• 🧐 The AI's estimates provide a useful starting point, but human engineering judgment and verification are crucial before implementation.
• 🤝 The video seeks input from marine and mechanical engineers on dissipating 2,500 watts through an aluminum hull, considering a maximum water temperature of 35°C and a 10-15°C temperature variance.