Distributed Photovoltaics: Utility Integration Issues and Opportunities
- March 18, 2009
- 678 views
PV technology has proven to be an effective power generation option in a variety of on-grid and off-grid applications. Steady advances over the past 25 years, combined with favorable policies, have fueled rapid expansion in deployment, particularly in the last decade when the annual growth rate has ranged from 25 to 40 percent. Globally, PV sales have surpassed $10 billion annually, with many leading multinational energy and electronics manufacturing companies involved. Total installed PV capacity--just over 1 gigawatt (GW) in 1999--was projected to exceed 10 GW in 2008 and 25 GW by 2011.
Grid-connected PV systems are expected to account for more than 90 percent of capacity additions in 2011, up from about 50 percent in 2000 and from less than 10 percent a decade earlier. The accelerating worldwide growth in grid-tied PV will be driven by continuing technology performance and cost improvements, strong deployment incentives, and growing consumer interests, as well as renewable portfolio standard (RPS), climate, and other policy mandates.
The number of relatively large PV projects feeding power directly to the grid will increase, but most systems will be deployed in behind-the-meter applications, where the technology competes with the retail rate of delivered electricity rather than the wholesale cost of energy supplied by central-station generating plants. Germany and Japan currently lead the world in PV deployment, due in large part to favorable policies, generous incentives, and high retail rates. To date, the United States has lagged behind these countries. For the future, the issue facing the U.S. electric sector is not whether PV will play an increasingly important role, but where, when, and how it will influence consumer decision-making, retail and wholesale markets, and grid operations and planning, as well as the balance sheets of utilities, distribution companies, and other industry participants.1
Forecasting U.S. PV deployment has proven extremely difficult. Past projections, thought to be optimistic at the time, have understated the current size of the PV market, which is expanding dramatically in California and other states with strong incentives and other favorable conditions. Further, retail rates are 20 cents/kWh or higher in some U.S. regions, levels at or near parity with the cost of energy from larger-scale distributed PV systems in resource-rich locations before accounting for policies that increase the technology's competitiveness.
By 2015, total U.S. PV capacity is projected to range from a low of 7.5 GW to a high of 24 GW, depending on the policy environment.2 Under even the most optimistic scenario, PV is expected to make marginal contributions to the overall U.S. electricity supply mix across the next decade or so, but large quantities of distributed PV capacity are expected to be added in areas with advantageous solar resources, policy conditions, and market circumstances.
Conventional grid-tied and off-grid PV will be used on rooftops, in ground-mounted arrays, and as components in end-use devices. Building-integrated PV (BIPV) systems and novel grid-interactive and grid-independent devices will be enabled by advanced PV, inverter, storage, communications, and end-use technologies. According to previous EPRI analysis, distributed PV likely will begin driving significant changes in the electricity enterprise within the next decade in many regions of the country.1
On the one hand, widespread PV deployment will impose new demands on the grid and may compete for markets served by conventional infrastructure. Successful integration of substantial quantities of distributed PV will require grid modernization to maintain performance and reliability and to enable the communications, metering, and control functions needed for effective energy management, market access, service delivery, and grid operation. On the other hand, distributed PV will create opportunities for industry participants to serve consumers in different ways, develop new revenue streams, and meet societal objectives in a political, regulatory, and business climate likely to place an increasing premium on an intelligent grid capable of delivering clean energy and precision power in a reliable and secure manner.
Evolving U.S. Markets
Historically, the U.S. PV market has been dominated by consumers purchasing and installing distributed PV systems and owning and operating them based on end-use wants and needs. Grid-connected, behind-the-meter systems have been deployed by consumers interested primarily in PV's environmental, energy security, and other attributes rather than concerned about payback periods, which depend most critically on up-front costs and on the value of system output. By contrast, off-grid PV systems have proven to be a cost-effective and dependable solution for delivering energy services in locations remote from the grid, relative to the alternatives of extending a distribution line or relying on other power supply options. Most on-grid and grid-independent PV projects have ranged in capacity from a handful up to a few dozen kilowatts.
Utilities and other industry participants generally have not engaged in PV installations except when required to offer incentives or to facilitate interconnection or when motivated by rationales other than return on investment. All these trends are changing rapidly due to a confluence of technology, market, and policy developments. Most importantly, PV economics are improving. The cost of PV modules has declined from about $10,000/kW in 1990 to under half that today. Within the next few years, module prices are projected to fall under $2,000/kW as new manufacturing capacity comes on line and advanced PV technologies mature. In addition, "balance of system" and labor costs are declining as modules of all types become more efficient, inverter and other components improve, and the system size required to produce a given amount of energy decreases. As a result, installed costs for residential rooftop systems are expected to fall to about $4,000/kW in the relatively near term, about half the current costs, and to about $2,000/kW by 2020. Economies of scale continue to yield progressively lower costs for progressively larger systems.
Rebates, tax credits, and additional incentives offer further reductions in the up-front costs of grid-tied systems--up to 25 percent or more in some states. Meanwhile, the value of PV system output, the other critical determinant of payback, is increasing due to rising energy costs as well as net metering and RPS requirements. For net-metered systems, consumers are compensated at the full retail rate, rather than the wholesale supply rate, for solar energy exported to the grid. RPS mandates establish a market for renewable energy certificates (RECs) associated with solar generation, creating a revenue stream for PV system owners.
These developments are making small-scale PV projects more attractive across a broad region and spurring deployment of multi-megawatt distributed PV installations on large buildings and in ground-mounted arrays. Going forward, they will expand the number of economically feasible PV applications. Already, the number of cordless PV-powered products--from backpacks to trash compactors--is expanding rapidly and encompassing end uses traditionally powered by the grid. Complementing the improvement in PV economics and the expansion in PV applications are significant changes in PV system ownership and utility involvement. Driven largely by the RPS mandates and climate policies adopted at the state level and anticipated at the federal level, utilities and other businesses are becoming increasingly active in the deployment of distributed PV systems and in the sale and purchase of solar energy and RECs. Decoupling policies that put load reduction on an equal footing with supply procurement are expected to have complementary effects.
In the past two years alone, large PV systems have been installed under varying partnerships involving consumers, independent power producers (IPPs), and/or utilities; third-party providers have emerged pursuing business models designed to eliminate the "first cost" barrier constraining widespread PV application; and utilities have announced or implemented novel approaches for accelerating PV deployment consistent with policy objectives. Notably, in several states with restructured markets, distribution companies required to divest their generation assets within the past decade or so have recently been granted regulatory approval to own PV systems. Utilities are highly involved, either owning the majority of added capacity or purchasing its output via PPAs. In the second, third-party providers become established and capture a major share of the market. These scenarios, which are illustrative rather than predictive, differ in another important way: total PV capacity additions may be slightly higher when utilities are actively involved because they are well-positioned to accelerate deployment and they have the incentive to do so; and increased levels of utility engagement--up to and including system ownership--provide greater control over PV integration, improved grid support capabilities, and expanded business opportunities.3
PV Configurations & Attributes
Challenges and opportunities relating to distributed PV integration will be strongly influenced by the current and future attributes of PV and balance-of-system technologies. Conventional grid-connected systems include PV panels or modules wired together to create an array. The panels convert incident solar radiation into DC electricity. DC output from the series-connected array is fed to a power conditioning device and converted to AC power. Solar output serves on-site loads or is fed directly to the grid.
The PV array connects to a standardized UL 1741/IEEE 1547-compliant DC-to-AC inverter, which serves loads through the building's electric panel. A customer-owned meter between the inverter and service panel tracks PV output, while a utility-accessible switch allows for manual disconnection. If on-site loads exceed solar generation or the sun is not shining, the grid supplies supplemental power or backup service. During a grid outage, the line-interactive inverter shuts the PV system down to prevent it from operating in islanded mode and to protect utility personnel and equipment. A two-way meter on the utility side of the service panel tracks electricity from the grid, net of the amount of AC power exported when solar generation exceeds on-site demand. PV systems supplying power directly to the grid are generally larger but otherwise incorporate similar components and/or functionalities. Off-grid PV arrays usually connect to a battery charge controller and then battery bank. The battery output is fed directly to DC devices, or it is run through a DC-to-AC inverter before serving AC devices directly or through the electric panel. In many cases, an on-site generator is employed to supply supplemental power or backup service at night and during extended periods of cloudy weather. As noted above, PV technology has advanced substantially over the years, and further progress is expected. Similarly, the performance of line-interactive inverters has improved markedly over the past decade. Higher DC-to-AC conversion efficiencies allow more energy to be produced from the same size PV array, and increased robustness to failure reduces the lifetime cost of PV installations. While inverters today cost on the order of $500 to $1500/kW of rated capacity (depending on their features and size), production economies and design improvements are expected to shave their costs to $100 to $500/kW while further improving overall performance.
An alternative approach to the conventional grid-tied system configuration--where DC output from series-connected PV modules feeds a single inverter--is currently being explored. Here, the inverter is integrated within the module. The end result is a "plug-and-play" PV system that could be connected to a standard AC electrical outlet or an array of modules that could be connected in parallel to supply AC power to a building's electric panel or directly to the grid. Potential advantages of the "AC module" design are inverter production economies, simplified installation, and reduced DC wiring, as well as increased robustness: no longer will an inverter failure take the entire system off line. Disadvantages include reduced economies of scale for the inverter, plus the need for grid interaction with multiple inverters. Distributed grid-connected PV can do more than simply serve as an alternative to conventional bulk power generation and delivery. It can supply clean energy, consistent with RPS mandates and other policy requirements. Complementing energy efficiency and demand response, it can help shave peak loads and defer or avoid distribution system upgrades and generating capacity additions. It can provide ancillary power quality and grid reliability functions. It also can feed DC power directly to motors, vehicle charging systems, and other DC loads, as well as microgrids.
Several of these functions require PV systems with advanced communications, metering, power electronics, and control capabilities. Smart grid technology combined with intelligent interface devices will deliver the core functionalities: two-way communications between the PV system and the utility, advanced metering infrastructure for monitoring system output and on-site loads, and advanced controls enabling automated response to grid conditions, pricing signals, and other factors.4 Some commercially available inverters enable PV systems to provide backup power services, but advances in inverter technology promise additional premium power and grid support functions. In off-grid applications, PV typically has been deployed to supply electricity to homes, farms, and facilities not served by conventional infrastructure and to power communications installations, security systems, environmental monitoring stations, signs, and other equipment in remote locations. Innovations in PV, storage, and end-use technologies and changing market factors promise novel off-grid applications that could yield slow but steady erosion in demand in some load categories located at or near the distribution system that were formerly clear candidates for grid connection.
Part II of this article will consider conventional grid-connected applications.