Requirements for Defining Utility Drive Cycles: An Exploratory Analysis of Grid Frequency Regulation Data for Establishing Battery Performance Testing Standards [report]

Ryan P. Hafen, Vilanyur V. Vishwanathan, Krishnappa Subbarao, Michael CW Kintner-Meyer
2011 unpublished
Battery testing procedures are important for understanding battery performance, including degradation over the life of the battery. Standards are important to provide clear rules and uniformity to an industry. The work described in this report addresses the need for standard battery testing procedures that reflect real-world applications of energy storage systems to provide regulation services to grid operators. This work was motivated by the need to develop Vehicle-to-Grid (V2G) testing
more » ... V2G) testing procedures, or V2G drive cycles. Likewise, the stationary energy storage community is equally interested in standardized testing protocols that reflect real-world grid applications for providing regulation services. As the first of several steps toward standardizing battery testing cycles, this work focused on a statistical analysis of frequency regulation signals from the Pennsylvania-New Jersey-Maryland Interconnect with the goal to identify patterns in the regulation signal that would be representative of the entire signal as a typical regulation data set. Results from an extensive time-series analysis are discussed, and the results are explained from both the statistical and the battery-testing perspectives. The results then are interpreted in the context of defining a small set of V2G drive cycles for standardization, offering some recommendations for the next steps toward standardizing testing protocols. v that, unless otherwise compensated, on average there would be a resulting net charge of a battery plugged in during the night and, conversely, a net discharge during the day. A net charge during the night for plug-in electric vehicles can be accommodated easily without any compensation, but a net discharge during the day will need to be compensated so the vehicle battery can be recharged for the next drive.  Amplitude of regulation cycling: The amplitude swing (i.e., the minimum-maximum spread) of the cycles in the regulation signal has a propensity to be higher at night and lower during the day. In fact, we found that a significant relationship exists between the system load level and the amplitude of the regulation signal; that is, under low-load conditions, the amplitude of regulation grows exponentially compared regulation under high-load conditions. This means that, during the day, either fewer vehicle resources are necessary to meet the regulation requirements, or if the resource availability constant is kept constant during the 24-hour period (i.e., the same number of vehicles offering the same service for 24 hours), the individual contribution by each vehicle is less during the day than during the night. As a consequence, the amplitude for defining a performance test or drive-cycle could potentially be smaller during the day than at night. Seasonal (month-of-the-year) differences observations Exploring the range of low-frequency (1-hour cycles) to high-frequency (7.5-minute cycle) signals across all seasons indicates that the lower frequency component in the regulation signal increases in significance during winter to summer months. The mid-range frequency (30-minute cycle) remains constant over the months studied, while the higher-frequency components (7-minute cycle) slightly decrease their contribution to the overall signal from March through August. Low-frequency cycles require more energy (kWh) capacity than do high-frequency cycles. Low-frequency cycles with higher energy requirements cause large cycles across the state-of-charge (SOC) scale for any given battery size. For transportation batteries, this would place a higher burden on the remaining life of the battery, as it may encounter deeper depths of discharge over any given period of time than would occur if the battery was exposed to higher frequency regulation signals. This also could affect thermal management of batteries at low SOCs as the internal resistance is higher at low SOCs. Hence, in addition to energy considerations, the ability of the batteries to provide the required power at low SOCs needs to be assured as part of the test profiles being developed.
doi:10.2172/1028571 fatcat:2gmeme3oj5aeviecimrjxjifny