Wednesday, October 24, 2012

Federal Government Electricity Use


While studying the Federal Energy Management Program (FEMP) I came across an interesting dataset published by the Data.Gov (https://explore.data.gov/) initiative (Reference #1). I did a quick re-organization of the published raw data, analyzed it and created the following charts. I was also interested to see if there is any correlation with factors external to the dataset. One of the factors I looked at was the size of the agency in term of number of employees. While I looked for the historical data on the number of employee by agency, I could only find data on the cumulative personnel size for the Federal government (Reference #2). I have listed the names of the analyzed agencies under Reference #3.

Following are the charts from my analysis on the annual consumption of electrical energy (in MWh).

Top 5 Electrical Energy Intensive Agencies



Chart 1

The Data.Gov has published energy consumption data from 38 Federal Agencies. I focused on the top 5 electrical energy intensive agencies which constituted 86% of the overall consumption, remaining 33 agencies only contributed 14% of the total, which I combined them under “Others” category (the segmentation is a classic 80:20 rule example; i.e. more than 80% of the energy intensity came from less than 20% of the agencies). From Chart 1 it can be seen that the DOD is by far the biggest electricity consumer, which I assumed would be the case before I had even looked at the data, but had not imagined this big a size of the pie for them. Secondly, seeing the VA and the GSA among the top 5 was a surprise. I admit that I am not familiar with operations of various federal agencies, but I was surprised to see the VA and GSA’s consumption is even more than some of the other prominent technology intensive agencies such as NASA. I guess VA’s many medical centers are contributing to this large energy consumption percentage.

Yearly Trend of Electrical Energy Consumption of the Top 5 Agencies


Chart 2

Chart 2 shows the trend of electricity consumption from 1975 through 2007. The immediate take-a-way from this chart is the big difference between the DOD energy profile and the rest of the immediate other five profiles. While the DOD has undertaken several programs to improve energy efficiency, I was more interested in the profiles of the next four agencies. Chart 3 is the zoomed-in view of these four agencies.

Chart 3

From Chart 3, while the DOE and GSA have pretty much maintained flat lines over the years, it was interesting to note the continued upward trends of USPS and VA consumption of which the former is more significant. Given the economic conditions and competitive situation the USPS is in and looking at the data, I believe they would gain critical financial benefits by implementing energy efficiency measures. (Note: Looking at the downward trend of energy consumption starting at year 2005 point, USPS may already be on track to improve energy efficiency, but without the data beyond 2007 it is difficult to tell).

Per Capita Electrical Energy Footprint for the Federal Government

Finally, I was interested to see if there is a connection between the energy consumption and any other factors associated with the agency size.  One of the factors I studied was the number of employees in the Federal Government.  Chart 4 shows this relationship. (Note: This analysis was for all the agencies combined.)

Chart 4

From Chart 4, it is interesting to note that while the number of federal employees dropped over the years from 5 million to 4 million (blue line), the electrical energy consumption on per employee basis actually increased at a high rate (I calculated this value by dividing the total federal energy consumption by the number of employees). I am not sure the reason for this, but I have the following hypothesis:
  • Agencies are beginning to report energy consumption data more accurately and comprehensively compared to earlier years
  • Increasing use of energy consuming electronic devices such as computers and IT systems is  driving the increase in energy consumption rate
  • Agencies are becoming more geographically spread-out. Maybe the increase in number of facilities (roof-tops) is increasing the per-capita energy consumption as shown in Chart 4.


Conclusion


Thanks to the Data.gov initiative for making this dataset available for curious analysts like me to study the data and share some of the insights. I hope to update these charts when new dataset is published by Data.gov that includes data beyond year 2007.

References

  1. Annual Federal Government Energy Use and Costs by Agency, 1975 – 2007
    • URL: https://explore.data.gov/d/pib3-yjd4
  2. Historical Federal Workforce Tables
    • URL: http://www.opm.gov/feddata/historicaltables/totalgovernmentsince1962.asp
  3. Agency Acronyms: 
    • DOD – Department of Defense; DOE – Department of Energy; USPS – United States Postal Service; GSA – General Services Administration; VA – The Department of Veterans Affairs.

Wednesday, August 1, 2012

LMP - Locational Marginal Pricing


The Energy Market operates much like a stock exchange, with market participants establishing a price for electricity by matching supply and demand. Some deregulated markets, most notably in the PJM ISO, ERCOT, New York ISO, and New England ISO markets in the USA use locational marginal pricing that reflects the value of the energy at the specific location and time it is delivered.

Ideally, if the lowest-priced electricity can reach all locations, prices are the same across the grid. However, when there is transmission congestion, energy cannot flow freely to certain locations. In that case, more expensive electricity is ordered to meet that demand. As a result, the locational marginal price (LMP) is higher in those locations.

Locational Marginal Price (LMP) is defined as the marginal price for energy at the location where the energy is delivered or received. For accounting purposes, LMP is expressed in dollars per megawatt-hour ($/MWh). LMP is a pricing approach that addresses Transmission System congestion and loss costs, as well as energy costs. Therefore, each spot market energy customer pays an energy price that includes the full marginal cost of delivering an increment of energy to the purchaser‘s location.

In simple terms (the actual computation of LMP is more complicated):

LMP = Generation Marginal Cost + Transmission Congestion Cost + Cost of Losses

References:

  • http://en.wikipedia.org/wiki/Electricity_market
  • PJM Manual 11: Energy & Ancillary Services Market Operations Revision: 50; 2012

Dynamics of LMP

The following are a couple of  interesting charts that I created from weighted-average day-ahead LMP data in PJM's operating area. The purpose of the chart is to visualize the variation of  representative electricity wholesale market price by hour of the day (daily peaks) and by the month (seasonal pricing), to get better insights into wholesale energy market dynamics. (The horizontal axis for the charts represents the time in 24-hour format.)

Chart 1: LMP movement by hour, by month

Note: Please use the scroll bar at the bottom of the chart to see how the LMP varies by hour and then month, starting with June 2011 up to June 2012 (I have deliberately used a 13 month cycle to show the seasonal peaking that happens in July every year). If you hover the mouse over the bar graphs, you would be able to see the data values associated with those bars.


Chart 2: Seasonal Variation of LMP  

The following chart shows the seasonal (by month) variation of LMP in a single view. As you see from the chart, the variation is very drastic during peak power usage (from approximately 2PM to  approximately 8PM). Similar to the observation from Chart 1: the highest LMP is during the Summer months and the lowest LMP is around the Spring.

Note: If you hover your mouse over the bars on the chart you would be able to see the variation of LMP over the months as part of the seasonal variation.

Data Source: http://www.pjm.com/markets-and-operations/energy/real-time/monthlylmp.aspx

Tuesday, July 10, 2012

US Electricity Production by State

I was curious to see what the latest data from the EIA (Energy Information Agency) is telling us. So, I did some quick visualization using Tableau Public. The following chart shows the electricity production by State for year 2011 based on the preliminary data (please footnote for EIA's clarification on this data). The chart shows some interesting characteristics of how different States generated electricity last year.



The chart is interactive. You can use Zoom-out button (-) on the top left corner (it will appear when you hover your mouse over the graph) to see information for Alaska and Hawaii.


If you move the mouse over each State you will be able to see the number of generators and the amount of generation for that State (The darkness of redness in each State is proportional to the amount of electricity generated).  


Additionally, you can change the fuel type used for generation by selecting different filter setting on the right to see the predominant type of fuel used by the State.

Legend: Aggregated Fuel Type
COL Coal 
DFO Distillate Petroleum
GEO Geothermal
HPS Hydroelectric Pumped Storage
HYC Hydroelectric Conventional
MLG Biogenic Municipal Solid Waste and Landfill Gas 
NG Natural Gas
NUC Nuclear
OOG Other Gases
ORW Other Renewables 
OTH Other (including nonbiogenic MSW)
PC Petroleum Coke
RFO Residual Petroleum
SUN Solar PV and thermal
WND Wind
WOC Waste Coal
WOO Waste Oil
WWW Wood and Wood Waste

source: preliminary data published by Energy Information Agency for 2011

Which US States are generating electricity using IC (Internal Combustion) engines running on NG (Natural Gas)?

CHPs (Combined Heat Power) being one of my current areas of interest, I also wanted to get a quick answer to the above question from EIA's data. 


The X-axis shows States with NG, IC Engine CHPs. Y-axis shows the number of CHPs in each state.

If you move the mouse over to each of the plant represented on the bar, it will also show the Plant Name and the total electricity produced by this plant in 2011. You can explore other scenarios by changing the filter settings on the right. 


With current record-low NG prices, I believe the number of generators producing electricity from this fuel is likely to increase significantly.


Prime Mover Code
BA Energy Storage, Battery
BT Turbines used in a Binary Cycle (geothermal)
CA Combined Cycle – Steam Part
CE Energy Storage, Compressed Air
CP Energy Storage, Concentrated Solar Power
CS Combined Cycle Single Shaft (combustion turbine and steam turbine share a single generator)
CT Combined Cycle Combustion – Turbine Part
FC Fuel Cell
GT Combustion (Gas) Turbine (includes jet engine design)
HA  Hydrokinetic, Axial Flow Turbine
HY Hydraulic Turbine (includes turbines associated with delivery of water by pipeline)
IC Internal Combustion (diesel, piston) Engine
OT Other
PS Hydraulic Turbine – Reversible (pumped storage)
PV Photovoltaic
ST Steam Turbine, including nuclear, geothermal and solar steam (does not include combined cycle)
WT Wind Turbine


source: preliminary data published by Energy Information Agency for 2011

Footnote:

Clarification on the preliminary data by EIA: This is an early release of the final EIA-923 data for calendar year 2011.  The early release is provided for the express purpose of providing immediate access to individual plant and generator data for analysts who use this type of information.  The data has not been fully edited and is inappropriate for aggregation, such as to state or national totals.  Also, in some cases, data for a certain number of plants and generators has been excluded from this early release pending further data validation. Final, complete, and fully-edited data will be released by EIA later in 2012.