No abstract

This paper presents control algorithms and sizing strategies for using energy storage to manage energy imbalance for variable generation resources. The control objective is to minimize the hourly generation imbalance between the actual and the scheduled generation of wind farms. Three control algorithms are compared: 1) tracking minute-by-minute power imbalance; 2) postcompensation; and 3) precompensation. Measured data from a wind farm are used in the study. The results show that tracking minute-by-minute power imbalance achieves the best performance by keeping hourly energy imbalance zero. However, the energy storage system (ESS) will be significantly oversized. Postcompensation reduces the power rating of the ESS but the hourly energy imbalance may not be reduced to zero when a large and long-lasting power imbalance occurs. A linear regression forecasting algorithm is developed for a two-stage precompensation algorithm to precharge or predischarge the ESS based on the predicted energy imbalance. An equivalent charge cycle estimation method is proposed to evaluate the effect of providing the energy balancing service on battery life. The performance comparison shows that the precompensation method reduces the size of the ESS by 30% with satisfactory performance.

No abstract

No abstract

PHASE I WECC PHASE II EIC + ERCOT v 2. What are the most cost-effective technology options for providing additional balancing requirements today and in 2020 assuming technological progress? Our analysis includes the following technologies: i. Combustion turbine as a base case technology ii. Na-S (Sodium Sulfur) batteries iii. Li-ion (Lithium-ion batteries) iv. Flywheel v. CAES (Compressed Air Energy Storage) vi. Redox Flow batteries vii. Pumped Hydroelectric (PHES) Storage viii. Demand Response ix. Hybrid energy storage systems (configurations of various above mentioned storage technologies) 3. What is the market size (quantified in MW and MWh) for energy storage and its respective cost targets (expressed in $/kWh) for balancing and energy arbitrage applications by regions? Key Outcomes Pacific Northwest National Laboratory (PNNL) analyzed a hypothetical 2020 grid scenario in which additional wind power is assumed to be built to meet WECC's 20% renewable energy portfolio standard target. Several models were used to address the three questions, including a stochastic model for estimating the balancing requirements using current and future wind statistics and the statistics of forecasting errors. A detailed engineering model was used to analyze the dispatch of energy storage and fast-ramping generation devices for estimating capacity requirements of energy storage and generation that meet the new balancing requirements. Financial models estimated the life-cycle cost (LCC) of storage and generation systems and included optimal sizing of energy storage and generation to minimize LCC. Finally, a complex utility-grade production cost model was used to perform security constrained unit commitment and optimal power flow for the WECC. Outcome 1: Total Balancing Market in the WECC is Estimated to be 6.32 GW Assuming about 24 GW of Installed Wind Capacity in 2020 The total amount of power capacity for a 20% RPS scenario in the WECC for 2020 would require a total intra-hour balancing capacity of approximately ~6.32 GW. The total market size was estimated for the WECC by NERC sub-regions based on the potential for energy storage in the high-value balancing market. The energy capacity, if provided by energy storage, would be approximately 2.0 GWh, or a storage that could provide power at rated capacity for about 20 minutes. The additional intra-hour balancing capacity that is required to accommodate the variability due to capacity addition in wind technology and load growth from 2011-2020 was estimated to be 1.53 GW. If these additional balancing services were provided by new energy storage technology, the energy capacity would be about 0.58 GWh, or a storage capable of providing electricity at the rated power capacity for about 20 minutes. vi The regional distribution of balancing requirements within the WECC is driven by load forecasting wind prediction errors. Because of the non-homogeneous distribution of the loads and wind across the WECC region, the balancing requirements increase with load and wind capacity. NWPP and Ca...