## Energy Informatics## MIST4550 & MIST 6550 |
Revised: November 11, 2017 |

- Read the data for solar radiation (a timestamp and solar radiation in watts/m2/sec) and electricity prices (a timestamp and cost in cents per kWh) for a city in the South East of the US. The files measure data for different time periods, one every 2 minutes and the other hourly. Assume the 'on the hour' measure of solar radiation is a good estimate for the 30 minutes either side. Merge the two files.

Compute the correlation between solar radiation and electricity price. What do you conclude?

- Using the solar radiation data, compute the annual average and monthly averages of solar radiation. Compare the data with solar radation for Athens, GA, which measures solar radiation in kWh/m2/day, so you will need to make a conversion.

Using the merged file created in the previous assignment, do the following

- Graph the relationship between solar radation and electricity price.
- Graph the relationship between solar radiation and electricity price during August.
- Create a single graph showing the relationship between solar radiation and electricity price differentiating by color when the solar radiation is above or below 500 watts/m2/sec. (Hint: Try recoding).
- Create a bar graph of average monthly solar radiation in kWh/m2/day. Hint: group_by(month = month(TimeStamp)) creates a column name, month, that you can reference in ggvis.

- Read the solar radiation and electricity price data. For each of the data sets, convert the data to a time series and use dygraph to visualize it. You might find it helpful to first convert the radiation data to a daily value such as mean or total. Don't bother with setting start or frequency for either time series.
- A spreadsheet <http://people.terry.uga.edu/rwatson/data/electricity_sales_revenue.xls> contains extensive data on electricity sales and prices in the US. Use the function read_excel in the package readxl (readxl::read_excel) to read the spreadsheet. Create a time series for the average retail prices for residential customers `Average Retail Price Residential (c/kWh)` for Georgia. Decompose the time series. What are your conclusions?

- Solve the following transportation problem.
- Determine the shipment plan that minimizes transportation costs and report it in matrix format.
- What is the minimum transportation cost?
- Where should the company increase production capacity to reduce transportation costs?

Factory | Supply |
---|---|

Alpha | 20 |

Beta | 6 |

Gamma | 9 |

Delta | 11 |

Distribution center | Demand |
---|---|

Phi | 10 |

Chi | 15 |

Psi | 12 |

Omega | 9 |

Shipping cost per unit | ||||
---|---|---|---|---|

From/To |
Phi | Chi | Psi | Omega |

Alpha | 10 | 30 | 25 | 15 |

Beta | 20 | 15 | 20 | 10 |

Gamma | 10 | 30 | 20 | 20 |

Delta | 30 | 40 | 35 | 45 |

b. A pizza chain in Athens has three locations, and in the last five minutes has received orders for pizzas to be delivered to five addresses in Athens. Assuming ordering is centrally managed, which store should assigned the delivery of the five orders to minimize travel distance. What is the shortest route?

Delivery locations

- 170 River Rd
- 897 S. Milledge Ave.
- 558 West Broad St
- 125 Greek Park Circle
- 634 Prince Ave.

Store locations

- 350 E. Broad
- 1591 S. Lumpkin
- 145 Baxter St

Using the 'Storms' dataset that is part of trajectories package, (1) plot the path of Sandy showing the wind speed and (2) showing the category. Once the trajectories package is loaded, you can access the file by its name (e.g., summary(Storms)).