Distributed Generation: Transmission Panacea or
Utility Pain Point?Michael GoldmanSenior Research AnalystNortheast Utilities Westwood, MAMichael.Goldman@nu.com
On June 7th, 2013, President Obama issued a Presidential Memorandum instructing several Cabinet-level departments to work together in order to expedite the siting, permitting, and review process for electric transmission infrastructure. The memo states that a reliable transmission grid is necessary for economic growth and national security. Additionally, the memo explicitly lists reduced congestion as a goal of a modernized grid. This renewed emphasis on transmission infrastructure raises the question of whether distributed generation (DG) opportunities can be targeted on specific geographic areas (“geo-targeted”) in order to help alleviate transmission congestion in those areas and help improve the overall reliability of the bulk electricity system. Looking at recent developments and trends from across the country, many of the economic and operational costs and benefits associated with geo-targeting DG can be highlighted.
If it was possible to successfully geo-target DG opportunities and reduce transmission congestion, it would result in significant monetary savings from fewer electrical outages, less line loss, less variability in wholesale electricity prices, and savings from deferred/cancelled T&D projects. But in order to realize the reliability and monetary benefits that DG can provide, many challenges first have to be addressed. This is especially true when trying to target specific geographic areas and in a time frame when the DG resource might be able to offset the need for new transmission or reduce the load on existing infrastructure. Some of these issues include system and interconnection compatibility matters, timing, and whether it is feasible to install enough DG to materially affect energy demand.
Benefits of Geo-targeted DG
The potential benefits of geo-targeted DG could be quite large, both economically and in terms of reliability. According to the DOE, the number of customers who have experienced an electric outage increased over the period 2009-2012 and peaked in 2011 with over 25 million households impacted (see Figure 1).
Figure 1. National Customer Outages by Outage Cause
Source: Author, adapted from DOE
The majority of outages were due to weather related issues such as wind, snow or trees damaging components in the transmission/distribution system. In addition to weather outages,approximately 2.6 million outages can be attributed to transmission-specific problems such as equipment failure, system interruption or the loss of a substation. A report released from the Executive Office of the President estimated that the annual cost to the economy of power outages due to weather is between $18 Billion and $30 Billion (EOP 2013). With DG resources located closer to load and a higher diversity of resources, it is likely that equipment failure or weather impacting transmission lines would affect fewer customers.
Locating generating capacity near demand also has the additional benefit of lower transmission line losses. Electricity in the line is lost when there is increased heating of line wires by the electric current. This effect is more pronounced in longer transmission lines. Reducing line losses can provide a substantial economic impact. In all planning activities, ISO-NE assumes an 8% line loss, with national estimates at 7% (EIA 2012). While there is a tremendous amount of variability in terms of the average hourly day ahead hub/zone locational marginal pricing (LMP), the average $/MWh for 2013 was $56.80 in the region that ISO-NE covers. In 2013, there was 127,247,000 MWh of non-PTF demand in ISO-NE (ISO-NE 2014 a), which equals $7.2 Billion in costs. Using the assumed 8% line loss factor, there could have been approximately $578 Million worth of line losses in 2013. This is just one example from a relatively small ISO, and only in one year. Nationally, economic loss from line loss would be considerably larger. While DG certainly couldn’t eliminate line losses altogether, it could start to chip away at this economic deadweight loss.
Besides reducing line losses, DG may help mitigate variations in wholesale electricity prices caused by transmission congestion. Electricity prices may spike in certain areas due to a host of reasons, but if electricity was able to easily flow to high demand load centers, you would expect to see a moderation in overall prices. Instead, if there is transmission congestion, it may be necessary to utilize inefficient, expensive generation or import high cost electricity. LMPs represent electricity prices at specific points within a geographic area and variations can be reflective of transmission congestion. Figure 2 is an example from the Midwest ISO that shows how widely LMPs can vary within a given geographic area at a given time point.
Figure 2. Real time LMPs from Midwest ISO, 2/17/2014
Source: Midwest ISO (MISO) LMP Contour Map and Table
Figure 3 shows LMPs in the Midwest ISO on an average cold weather day. In Figure 3, prices are relatively similar across the whole region.
Figure 3. Real time LMPs from Midwest ISO 2/18/2014
Source: Midwest ISO (MISO) LMP Contour Map and Table
By reducing congestion, geo-targeted DG may be able to keep LMPs lower across a region by reducing the need to send electricity across large areas but retaining the ability to do so in an emergency situation. This has the effect of keeping overall pricing down because as ISO NE describes on their website, in some instances higher LMP “costs are absorbed by all load, or demand, across the [New England] system, regardless of their areas’ contribution to the transmission constraint” (ISO-NE 2014 b).
If new DG resources reduce peak load demand, then it would be possible to build less transmission infrastructure and maintain a high degree of reliability. However, if equipment or energy efficiency programs only serve to shift load but the peaks remain unchanged, then the same level of infrastructure would be necessary to serve that peak demand, and there would be no savings. Recent estimates of transmission capital costs show that a 500 kV single circuit line costs approximately $1.85 million per mile, a 500 kV double circuit line costs approximately $2.96 million per mile, a 345 kV single circuit line costs approximately $1.3 million per mile and a 345 kV double circuit line costs approximately $2 million per mile (B&V 2012). In addition to transmission capital costs, ancillary costs such as siting, right of way acquisition, and certification can represent up to 20% of a project’s cost (PJM 2006). In 2006, AEP estimated that its AEP Interstate Project, a 550-mile, 765 kV line would cost $3 billion. That means that siting, right of way acquisition, and certification costs alone would be approximately $600 Million. If geo-targeting DG in certain areas can help avoid the need to build transmission lines, especially higher kV lines, there will be substantial cost savings. These would ultimately benefit the ratepayer either through lower bills or by re-allocation of money to increase reliability in other parts of the system.
In order to realize the benefits described above, it is necessary to consider the degree of difficulty of geo-targeting certain types of DG. DG is commonly solar, wind, or combined heat and power (CHP). Estimates of nation-wide planned DG capacity are above 30 gigawatts (GW) cumulatively by 2040. Figure 4 shows EIA’s forecast (in GW) that the majority of future DG capacity will be from solar power.
Figure 4. EIA’s 2013 Annual Energy Outlook Reference Case DG Forecast
Solar Photovoltaic (PV) systems, however, are usually installed in areas with beneficial sun and solar resources, which may not be the same areas where system congestion is occurring. Therefore, the most eligible technology to geo-target areas experiencing transmission congestion may be natural gas CHP units as they are not dependent on co-location with solar or wind resources.
Within CHP, there is an important difference in the interconnection review process if the generator is synchronous, induction, or inverter based technology. Synchronous generators are popular as they allow for black start generation but as such require extensive safety systems and review by utility protection engineers, as it is possible that these generators could electrify a portion of the grid without the utility’s knowledge or approval. Induction generators do not have black start capabilities and require reactive power from the grid, which can create an amperage burden. Inverter technology is becoming more popular as it combines beneficial aspects of both synchronous and inverter generators. These different configurations of DG systems are important as it can affect the review process and application process time.
A common issue with integrating DG into the grid is the complexity and length of time it takes for interconnection requests to be processed and reviewed. Depending on the size of the DG unit, where it is to be interconnected, if it requires system modifications, and whether it needs utility and/or ISO approval, processing times can fluctuate substantially. If the DG will feed into the transmission system, then ISO approval is also necessary. In Massachusetts, the average application processing time for DG interconnection that required no system modifications was 60.5 days but that number jumps to 239.5 days if system modifications are required. The size of the proposed DG project is also critical in application processing time, with those projects over 2.5MW averaging 373 days and those less than 2.5 MW averaging only 108 days (DOER 2014). This can tie up capital and represents an opportunity cost as that money could be invested elsewhere.
If system modifications are necessary for a DG project, not only is the application process time longer, there are additional costs that are borne by the applicant. In addition to system modification costs, other costs associated with the interconnection process can include cost of utility review, interconnection studies, interconnection equipment costs, application fee, and a reservation fee. Some of these costs are outlined in the Standards for Interconnection of Distributed Generation tariff (NSTAR 2013) with the application fee at $4.50/kW, up to $7,500, and supplemental reviews at $150 per engineering hour, up to $4,500. The real costs, however, are embedded in system modifications and facility upgrades.
Once a congested area is identified, the type of existing infrastructure that is at the site of the interconnection needs to be considered when trying to geo-target DG resources. The dichotomy between radial and networked systems delineates whether or not DG might be an economical solution to local congestion. Radial systems are designed for power to flow in one direction while a networked system can permit multiple supplies to enter the system. Networked systems may make better candidates to integrate DG but the interconnection process can be more complicated and expensive. Even if certain DG resources could not easily interconnect with the T&D system, it does not necessarily mean that installing DG would have no effect on congestion. In fact, quite the opposite is true. DG could still help alleviate congestion by taking current and future load off the grid. For example, a factory that installs a CHP unit will no longer need to derive as much energy off of the grid, lowering the demand on the T&D system that serves that facility. While a single site might not be large enough to affect transmission congestion, the aggregation of all sites in an area might.
Finding enough customers who are interested in installing larger DG systems may be challenging. Unless the utility is willing to develop, own, and operate the resource, it will be necessary for non-utility entities to become involved in the process. Certain utilities that do not own or operate any generation resources can encourage the development of DG by providing financial subsidies either through energy efficiency incentives or other methods. Subsidized projects normally need to meet a set of criteria that include certain levels of costs and savings. This is one way in which utilities can help spur the development of new DG. In addition to incentives from energy efficiency programs, there may be other municipal or federal incentive programs.
One additional issue to consider is the potential of future transmission congestion due to lack of visibility into existing DG resources behind a customer’s meter. During the transmission planning process, a utility may assume a certain level of load in a given area, not knowing whether or not that level of load includes DG resources. If there is a large amount of DG in that area and it fails, it will unmask load leading to a large draw on the grid, potentially causing congestion issues. Without knowledge of where existing DG resources are located, utilities may not be able to properly locate where future congestion issues could occur and would be unable to target those areas.
Can Geo-targeting DG Really Help?
At the end of the day, two real questions emerge. The first question is whether enough DG can be installed to displace the need for transmission lines in a given area. The second question is whether it is more cost effective to install DG rather than transmission lines. There are other reasons besides economics to construct transmission lines such as achieving public policy objectives like adding more renewable power to the grid or to increase reliability for N-1 events (loss of any one single system element) and N-1-1 (loss of a second system element 15 minutes after loss of a first element) contingencies. Reliability is usually the main reason for undertaking transmission projects but for illustrative purposes the following examples are over simplified and focus only on costs.
According to AEP, with the correct terminal systems over a 250 mile span, a single 765 kV transmission line has a practical loading limit of 2770 MW (PJM 2006). The costs associated with a 500 mile, 765 kV line with the 2770 MW loading capacity is estimated at $3 billion or $1,083,032 per MW of loading capacity. Table 1 shows how the installed capital costs of several types of DG compare with a 765 kV transmission line and the number of average size units that would be required to displace the electrical capacity of a line of that size.
Table 1. DG Size and Cost Comparisons
Source: Author, adapted from ICF and EIA
To highlight an example from this table, the average size of a residential PV system is 5 kW (SEIA 2012) and it would take over half a million of these systems to displace the need for the 765 kV transmission line. Additionally, the EIA estimates the installed capital costs for a residential solar PV system in 2015 will be $4,965/kWDC (EIA 2013 b), which equates to $4,965,000 per MW. The capital costs in $/MW for a gas fired combined cycle plant is $917,000 (EIA 2013 a) and the cost per MW loaded on a 500 mile 765 kV transmission line is $1,083,032.49. This results in a combined cost of just over $2 million per MW, or 40% of the cost for residential solar PV. This is an overly simplistic example, as it does not take into account fuel costs, O&M costs, and the capacity factors of the various energy sources. But this simplified example does illustrate some of the cost and sizing issues associated with this topic.
Part of the difficulty in addressing this problem is understanding where in the T&D system that geo-targeting DG could be most helpful. It is easiest to geo-target DG to displace transmission lines as transmission lines serve a wide geographic area so the aggregate savings from DG anywhere in that area could help alleviate transmission congestion. But transmission lines, especially the larger kV lines, can transmit such a large amount of electricity that it would take massive amounts of DG installations to actually negate the need for these lines, as Table 1 above demonstrates. DG may be more effective at increasing reliability at the distribution or circuit level because of the lower loads involved but the complication here is that circuits serve a very narrow geographic area so the installation of DG would have to be in very specific locations, which can be difficult to achieve.
Ultimately, geo-targeting DG likely is not a panacea for transmission congestion but it can be part of a wider solution for tackling this problem. While DG alone cannot displace new transmission projects, the aggregate effect of geo-targeting DG in certain areas, coupled with other activities that decrease load such as energy efficiency, can start to alleviate transmission and distribution congestion and unlock the wealth of benefits described in this paper. By working towards overcoming the challenges associated with geo-targeting DG and implementing a variety of solutions aimed at reducing load, utility companies can be a key partner in achieving the goal of reducing transmission congestion.
[B&V] Black & Veatch, (2012). Capital Costs for Transmission and Substations,
Recommendations for WECC, Transmission Expansion Planning.
[DOER] Massachusetts Department of Energy Resources, (2014). Distributed
Generation and Interconnection in Massachusetts Application Process Time
[EIA 2012] U.S. Energy Information Administration, (2012). How much electricity
is lost in transmission and distribution in the United States?
[EIA 2013 a] U.S. Energy Information Administration, (2013). Updated Capital Cost
Estimates for Utility Scale Electricity Generating Plants
[EIA 2013 b] U.S. Energy Information Administration, (2013). Distributed Generation
System Characteristics and Costs in the Buildings Sector
[EOP] Executive Office of the President, (2013). Economic Benefits of Increasing
Electric Grid Resilience to Weather Outages.
[ICF] ICF International on behalf of the California Energy Commission, (2012).
Combined Heat and Power: Policy Analysis and 2011-2030 Market Assessment.
[ISO-NE 2014 a] ISO New England Staff. (2014). Monthly Summary of Hourly Data,
Monthly Data by Load Zone.
[ISO-NE 2014 b] ISO New England Staff. (2014). Locational Marginal Pricing, available at http://www.iso-ne.com/nwsiss/grid_mkts/how_mkts_wrk/lmp/index-p2.html
[NSTAR] NSTAR Electric Company, (2013). Standards for Interconnection of
[PJM] PJM Media Website, (2006). AEP Interstate Project: Why 765KV
[SEIA] Solar Energy Industries Association, (2012). Photovoltaic Solar Technology.
 Adapted from DOE OE-417 annual forms
 Costs and sizes for Microturbines, Fuel Cells, Gas Turbines, and Reciprocating Engines come from (ICF 2012) and costs and sizes for Solar PV and Small Wind come from (EIA 2013 b)