Southern California Edison Anticipates a Significant Amount of New Wind Generation on its transmission system, more than 4000 MW within the next several years. This will add to the industry-leading 18% renewable energy content that SCE (Rosemead, California, U.S.) already delivers to its customers and will be a key component to meeting its renewable portfolio standard (RPS) target of 20% renewable energy by 2010, and the just recently state-mandated RPS target of 33% renewable energy by 2020.

While wind is a key renewable resource, its fundamental intermittent and nondispatchable characteristics pose grid-integration and operating challenges. However, SCE's recent technical activity and research on this topic give the utility confidence that energy storage is an available technical solution that can be an essential element in meeting the industry's wind-integration challenges.

KEY ROLE OF STORAGE

The concept behind SCE's Distributed Energy Resource (DER) program is that there is a role for energy storage to aid wind-generation integration. This expectation is consistent with technical conclusions from the California Independent System Operator's (CAISO) renewable integration study that assessed a 20% renewable-resource penetration scenario. Released in November 2007, a report on the CAISO study concludes: “Additional storage capability would be of considerable benefit with the integration of large amounts of renewables, especially intermittent renewables.”

Relevant to even more aggressive long-term California renewable penetration targets, a California Energy Commission (CEC)-sponsored report has stated that the growth in electricity produced by renewables is expected to account for 20% to 30% of electrical energy supplied. Specific to storage, the CEC report states: “We suggest further and more comprehensive assessments of multiple energy-storage technologies … to gain a better sense of the actual value propositions for these technologies in the California energy system.”

Prior to the referenced CEC and CAISO reports, SCE built and operated the 10-MW Chino Battery, a large-scale battery energy-storage system that successfully demonstrated and implemented several battery-system capabilities. The success of the Chino Battery project, which was commissioned in 1988, is particularly relevant to wind integration, specifically the dynamic injection and absorption of power to support grid stability and energy shifting.

POWER CONDITIONING SYSTEM

General Electric (GE; Fairfield, Connecticut, U.S.) developed and constructed a power conditioning system (PCS) for the Chino Battery system, with several additional grid-friendly capabilities, including frequency regulation via dynamic power output modulation. Demonstrating this particular capability, at a multimegawatt grid-scale, was very prescient and may now have a role in guiding the industry to a storage-system capability that will enable high penetration of otherwise nondispatchable renewables onto the grid.

In 1994, GE built an Energy Source Power System Stabilizer (ESPSS) that was added to the Chino Battery system. With the ESPSS functionality, the battery could detect and actively counter grid-frequency deviation and grid power oscillations with fast-acting, closed-loop-controlled battery output. Within two days of going into service, dynamic grid-supportive response by the battery was captured and validated by project metering.

The increased presence of wind-turbine generators by themselves lowers overall system-damping capability (a lower inertia within the total system) and increases exposure to system-frequency excursions from sudden loss of wind output. The deployment of large batteries with real power-stability support capability can provide counter measures for important grid-protective functions.

SIMULATING ENERGY-STORAGE SUPPORT

More recently, several energy-storage vendors and researchers have pointed out that a fast-acting subsecond time-frame grid interface can allow an energy-storage system to address wind-integration challenges; SCE research supports this. However, SCE also believes that fast-acting dynamic capabilities from energy-storage systems cannot be characterized nor quantified through a simple accounting of steady-state loads, resources and delivery-system capacities. Additional analyses are still needed to understand and define system dynamic stability impacts and solutions to properly scope and facilitate future successful deployments. To that end, SCE DER performed a simple illustrative system simulation of dynamic power stability with and without energy storage.

SCE's illustrative electrical system dynamic simulations were conducted with GE electrical system simulation software, using a system model based on a CAISO 2015 peak-summer scenario. The system model contained equivalents for nearly 3000 wind-turbine generators, representing a combined total of more than 4000 MW of new wind generation in the Tehachapi Mountain region of Southern California, as well as the transmission-system upgrades that are necessary to implement the planned generation additions.

Using this system model, a set of dynamic simulation runs were executed for a N-1 transmission-line outage scenario that isolates and drops 700 MW of wind generation. A range of energy-storage post-outage output capacities were modeled: 0 MW, 250 MW, 500 MW and 750 MW. The modeling of the energy storage used available model components to simulate performance of an advanced energy-storage system with two operational characteristics:

• Battery-power megawatt output is modeled via a constant real-power load whose magnitude is user defined and switched in at a specified megawatt capacity and time (within five cycles after fault-clearing line relay).

• Battery reactive mega-volt-ampere-reactive (MVAR) output is modeled using a synchronous condenser with ±50-MVAR reactive limits and nominal 0-MW real-power output. The reactive capability of the battery is already on-line pre- and post-fault.

This provided a simplistic but informative demonstration of system-level stability support by fast-acting storage following loss of significant wind generation. Major 230-kV buses across SCE's modeled system were monitored for frequency through the simulated 10-second window.