As all of us in the distributed-resources community are aware, actual markets have lagged the early optimistic projections. This should not be unexpected in an emerging industry attempting to compete head-to-head with a successful, 100-year-old way of doing business.

Fortunately distributed resources do not have to compete based on the same conventional decision criteria used for electricity-investment decision-making. In fact, I believe that the distributed-resources industry often fails to use the right metrics (coins of the realm) in evaluating its products and projects. We understand cost, and, to a limited extent, benefits, but we rarely address risk. I also will offer an example of a way to emphasize and monetize unique features of distributed resources, by taking full advantage of their modularity and portability. Using a more inclusive coin of the realm will lead to better recognition of the most advantageous opportunities for distributed resouces, higher revenues, and faster market penetration.

Three Purchasing Metrics

When decision-makers think through the purchase of a distributed resouce (i.e., distributed generation, distributed storage, and targeted demand management), the first consideration is obviously cost. Capital cost usually is the first hurdle, and normally variable operating costs also are factored into buying decisions. Of course, the distributed-resouces community constantly is striving to reduce costs.

Close behind cost as an important metric is benefit. Even at the beginning of the modern distributed-power era, it was recognized that distributed resources would have a very difficult time competing with central power purely on a cost basis. Distributed-resource markets are based instead on a value proposition: The value of the benefits of distributed resources exceeds their costs. Thankfully the full benefits (e.g., utility generation, transmission and distribution deferrals, combined heat and power, improved reliability, power quality, and reduced demand charges) of distributed resources extend well beyond that of providing wholesale/commodity energy, making distributed resources cost-effective in many instances.

Benefits are much more complicated to estimate than cost is, and their perceived value often is highly dependent on the eye of the beholder. As John Jimison notes in his January/February 2004 Guest Editorial, benefits are multiple, diverse, and complicated. In fact, one person's benefit might even be another's detriment, as in the case of reduced utility payments. Further complicating benefits is the immaturity of the distributed-resource marketplace and its - temporary, I hope - inability to quantify, allocate, and monetize the benefits of distributed resources (Iannucci et al., 2003). Smart purveyors of distributed resources emphasize the extensive benefits of their projects. Informed buying decisions usually are based on comparing costs to the subset of easily monetizable benefits. But let me offer a pivotal yet even less-understood and -discussed metric that affects the attractiveness of distributed resources: risk. Obvious sources of risk for anyone wanting to install a distributed resource are fuel availability and price volatility; the distributed system's reliability, performance, and life; and environmental and siting delays or changes that will impact the ability to dispatch the resources.

For energy end user installations of distributed resources, poor knowledge and/or uncertainty about interconnection costs and delays, future standby charges, and evolving-rate provisions must be added to the list of uncertainties that affect risk.

Even utilities that would install distributed resources must face load-growth uncertainty on a specific circuit (plus or minus), the permanence of customers' distributed generation, the persistence of demand-side management/conservation measures, the reliability of customer-installed distributed resources, the regulatory uncertainty of rate-basing the distributed-energy resource assets, and even potentially stranding existing distribution assets.

It will be the continuing job of the distributed-resources community to manage these three objectives:

• Minimize costs
• Maximize benefits
• Mitigate risk - technically and contractually

Managing the Risks

Performance (i.e., the odds of the unit operating properly and reliably for its design life) and fuel (i.e., its availability and price) probably are the two most obvious aspects of risk. Successful field experience is the best measure of performance risk. When this is augmented by solid performance guarantees, product warranties, and possibly even distributed-resources unit redundancy (N+1 units), little risk remains for the hardware. Fuel-price and availability contracts probably can mitigate fuel contingencies, but these might be more expensive for small projects than for large ones.

Smart, holistic, proactive project design and planning are required to minimize delays that result from environmental issues, to anticipate interconnection costs, to mitigate the regulatory uncertainty (especially regarding utility rate-basing of utility-owned distributed-energy resources, so utility stockholders are made whole), and to estimate standby charges.

Some uncertainties are beyond the realm of planning: New utility rates might be instituted, and there always will be uncertainty about load growth and the impacts of distributed resources used by customers.

Modularity and Portability

Finally, let me propose a way that distributed resources actually can take advantage of risk.

Utilities are very accustomed to entering all capital costs into a rate base and passing through all expenses, such as fuel for power plants, to ratepayers. This leads to indifference about using distributed resources when they actually might be the clear choice. If cost is not the only criterion used to evaluate the relative merits of distributed resources versus a conventional "wires" upgrade, there must be a different coin of the realm that includes explicit consideration of benefits and exposure to risk for each potential solution being considered.

A key point is that risk exists whether distributed resources are under consideration or not. A distribution planner or operator always is managing risk associated with wires failing or loads unexpectedly exceeding wires ratings. The risk on any one line is not perceptible to the regulators but immediately is visible to the customers on the line and to the utility. A limited number of such problems are expected every year and are by and large unavoidable.

Years ago, when utilities were less competitive, a debate broke out about whether there was an incentive for utilities to "gold-plate" their transmission and distribution systems to increase revenue while minimizing risk. The actual answer is always somewhere in between minimal design standards and gold-plating. A smart utility will try to minimize outages and constrained operations to below the level set by the regulators or below the radar entirely without asking for rate increases for dramatically more wire investments each year. Tree-trimming usually is the first transmission and distribution expense to be sacrificed to budget constraints or the need to pay for sudden catastrophic problems.

Thus, a hidden risk - that distribution capacity will not be adequate - is not taken into account when distributed resources are being considered. First, the feeders that don't quite make it to the top of the upgrade list in the current year indeed might fail. An even greater risk is that the upgraded feeder will never grow to the capacity to which it is increased. In both cases, this risk can be reduced by using modular and perhaps portable distributed resources.

By its very nature, distribution - and even transmission - upgrades are relatively lumpy investments. For example, it would not be unusual to add 4 MW to a 10-MW feeder. Conversely load growth in many locations is often about 2%/yr. In this case, a 2% load growth really would require only a 200-kW capacity increase to balance load growth versus the 4,000-kW upgrade. Grossly oversimplifying the argument, this gives a 20:1 advantage to the modular distributed-resource option (4,000 kW/200 kW=20), in the case where growth is deterministic. Thus, even if the distributed-resource cost is 10 times more per kilowatt than the upgrade cost is, it still would be half as expensive the first year as the permanent wires upgrade.

But, of course, load growth is not deterministic. There is a chance that an upgrade will be made before it is needed, and there is even a chance that a permanent upgrade will be made to serve load growth that does not materialize. This risk manifests itself as (1) an underused distribution asset and (2) an opportunity cost related to not solving the next-most-important distribution-capacity needs and problems. Despite this risk, under cost-based rate-making used in most states, utilities receive the authorized rate of return no matter which investment is made, obscuring the risk.

Consider the distributed-resource alternative to investing the same capital with the same utility return on equity (this makes risk the only important coin of the realm). If modular and - even better - portabledistributed resources are used temporarily (for a few years of load growth), the distribution planner with a finite budget retains maximum flexibility. Perhaps 20 circuits could be nominated for upgrade this year but there is only enough budget for the top 10 feeders to be upgraded permanently. A viable alternative might be to make permanent upgrades to the five most problematic circuits and use modular and portable - actually relocatableis good enough - distributed resources to solve the other 15 problems. Keeping a few portable distributed resources in reserve to move to hot spots developing during the year and/or to back up distributed resources in service adds even more flexibility to distribution planning and operations.

Using the previous example, once the feeder load grows to some point (say an additional 600 kW, after three years), it is easy to relocate the temporary distributed resources to more optimal locations and then to make permanent upgrades to the distribution system. In subsequent years, the load on several of these circuits will have grown to the point where permanent upgrades are the best option. Conversely several locations will no longer be at risk. The fleet of portable distributed resources should be augmented annually and adjusted as needed, at no higher capital cost than the wires upgrades.

Bringing the risk metric explicitly into the distribution planning process allows the value of distributed resources and two of their key features - modularity and portability - to be fully recognized. In this manner, distributed-resource users can take advantage of these unique attributes. Such creative applications will open the way to these and other markets.

In conclusion, based on experience and common sense, it is no surprise that distributed resources cannot compete head-to-head with conventional utility options using the existing utility coin of the realm. Distributed resources willcompete in a growing number of circumstances as benefit and risk become important elements of decision-making.

Reference

Iannucci, Joseph J., Lloyd Cibulka, James M. Eyer, Roger L. Pupp. DER Benefits Studies: Final Report. Report to the National Renewable Energy Laboratory. www.nrel.gov/docs/fy03osti/34636.pdf. 2003.

JOSEPH IANNUCCI is principal of Distributed Utility Associates in Livermore, CA, and is a member of DISTRIBUTED ENERGY's Editorial Advisory Board.

DE - March/April 2004