Authors:

Darren Lie

Date Created:

Not specified

Course Title:

Professor:

Not specified

About Paper:

Oceans compose over 70% of the earth's surface area. Such a large body of water is capable of absorbing and storing an incredibly large amount of useful energy, or exergy, in the form of heat, which mainly comes from sunlight. Exergy from a specific location in the ocean can be calculated easily, however, there does not exist a global mapping of available exergy around the globe. The goal of this project is to generate a global map of exergy stored around the globe. The exergy stored at a specific location in the ocean can be quantified by analyzing the region in the ocean where temperature changes most rapidly, known as the thermocline, and using thermodynamic equations to calculate exergy based on temperature and fluid property difference. However, ocean thermoclines are generally difficult to predict accurately since they depend on factors such as location and time of year, which is where the role of machine learning comes in. A neural network was developed to predict ocean temperature based on geographical location and time, capable of predicting within 2 °C of the actual recorded temperature. Results from the model are then used to predict the thermocline at each location and subsequently, the available exergy, which averages about 2.5 x 10^8 J/m^2 around the equator and 0 J/m^2 near the poles. The global exergy map created from this study helps visualize the distribution of exergy around the globe and can be useful in determining the most viable locations to build an ocean energy plant.

Source:

Purdue University / 2023

Topics:

No topics listed

Co-authors:

Darren Lie