Smart Grid

Team: 75

School: Los Alamos Mid

Area of Science: Engineering and Computing Science

Interim: Problem Definition:
When a normal electrical grid reaches peak demand the grid becomes very inefficient and more expensive inefficient generators are required to support this demand. These generators called “peakers” are required to support the demand of power during peak times of the day. A smart grid is able to reduce the peak load by more evenly distributing the power usage across the course of the day. As the amount of energy being used in peak load time increases the resistance of the electrical wires also increases, and thus the efficiency of the grid is greatly decreased.

For example, all day long your fridge is keeping your food cold by constantly using energy. The idea of a smart grid is to create a grid that can communicate with appliances in your house like a fridge. If the fridge was able to decide whether or not to use energy during peak load we would be able to drastically improve the amount of energy used during peak times of the day. The fridge would also be able to decrease the temperature inside the fridge before peak times of the day, so the fridge doesn’t have to use as much energy during the times of peak load.

Also these devices can choose to use more power when renewable energy sources, such as sunlight or wind, are available.


Problem Solution:
To solve this problem computationally we have come up with around ten possible agents in each house that consume electricity. The computer will model how much energy is used normally by the appliances during peak load time relying on existing end-use surveys reported by DOE. The Smart grid has instrumentation and controls necessary to alert end-users of impending peak conditions and automatically turn off home appliances. On the other hand, abruptly and simultaneously shutting off too many household appliances would cause major problems with the grid voltage and stability. Similarly, injection of too much electricity into the grid will surge electric voltage and could cause unsafe operating conditions.

We will use agent-based modeling to examine how to sequence shutting down appliances without damaging the grid. The computer program then will model how much energy appliances use when connected to the smart grid during the peak time. It will compare the results of the energy used by the appliances in each of the simulations. The difference in the amount of energy used is the amount of energy we have saved during the peak load time by using smart grid. This decrease in the peak load, as we stated above, will increase the efficiency of the electrical wire greatly. Because of limitation of number of agents that can be run on each computer, we expect one of the problems will be how to simulate a town as large as Los Alamos. If possible we will use multiple computers to solve this problem.

Progress to Date:
We have thoroughly researched the smart grid in many ways. Three people have volunteered to mentor us. These people, Jim Redman, Andy Erickson, and Venkat Dasari, have provided us with three different Los Alamos National Laboratory and Sandia National Laboratory reports about the smart grid. The New Mexico Green Grid report, the NEDO/NM Green Grid Collaboration report and The Smart Grid: An Introduction report, are our main sources.

Second we have come up with a possible list of agents for our simulation. These agents are household appliances that do not have to be in use during peak load times in the day. Also on this list are possible renewable energy sources that we will incorporate in our simulation. An example of these agents are, wind, solar, and possibly geothermal. We will use our energy-use survey results and control algorithms to develop and demonstrate an effective control strategy for peak management.

Expected Results:
One of the things we expect from our study is a strategy for reducing energy usage during peak load times. Also we expect that the strategy would also evenly distribute the amount of energy used over the day. With the smart grid in effect we will not have to rely on the inefficient and polluting “peaker” generators. Finally, we will examine methods for integrating more renewable energy sources into the smart grid system.

Team Members:

  Colin Redman
  Sudeep Dasari
  Michael Erickson

Sponsoring Teacher: Clara Vigil

Mail the entire Team