Date of Award


Document Type



Santa Clara : Santa Clara University


Computer Engineering

First Advisor

Rani Mikkilineni


ELEN 288 / COEN 282, Energy Management Systems, is a graduate course o ered at Santa Clara University concerning di erent methods and procedures energy companies utilize. Many algorithms taught in this class for predicting and measuring energy usage are dependent on the weather. For example, to predict energy usage for an upcoming day, if the day is predicted to have a max temperature of eighty degrees and a minimum temperature of fty- ve degrees, students can look in historical databases of weather for days that have similar weather, and infer that the energy usage might be similar as well. Energy usage prediction methods such as this one that involve numerous database lookups are fairly accurate, but extremely time consuming to calculate. Currently, if a student enrolled in the class is solving a problem that requires this \similar day" algorithm, he/she has to obtain a historical weather database, and compare the upcoming day they are trying to calculate the energy usage for with the rst day in the database. Then he/she compares it with the second day, etc. This process is extremely tedious, and provides almost no educational utility to the student. If the student wants to check his/her work, he/she would have to do the calculations all over again. These large repetitive tasks are extremely ine ciently done by humans, but can be done extremely rapidly by a computer. At the moment, students in the class use historical databases that contain one week of data, because of the amount of time it takes to perform lookups and comparisons. However, for a computer, it would be trivial to perform hundreds or even thousands of lookups and comparisons in a matter of seconds. The solution is to create a web application that has access to all of the required historical databases, and performs all of the comparisons automatically. The application will take as input an upcoming day and predicted weather parameters, and search through its databases for the most similar day, and return the predicted energy usage. This will alleviate the tedium for the students, allowing them to focus on the actual ideas behind the methods used and why they work. The logic behind the algorithm will be extremely portable, and could be easily adapted to other methods that are used in the ELEN 288 / COEN 282, or even other classes in the Sustainable Energy Program, such as predicting power generation from solar panels or wind farms in ELEN 282 or ELEN 286. The web application will be simple and user friendly, and will reduce stress, tedium, and human error that arise from having to perform these calculations manually.