Arthur K. Barnes

Placement of energy storage coordinated with smart PV inverters

Energy storage (ES) is increasing used in electrical transmission and distribution systems because it can perform many functions. These include peak shaving, voltage regulation, frequency regulation, spinning reserve, and aiding integration of renewable generation by mitigating the effects of intermittency. This work focuses on the usage of energy storage for peak shaving and voltage regulation on a distribution system having a high penetration of photovoltaic (PV) generation. The PV stations considered make use of smart PV inverters as proposed by the Electric Power Research Institute (EPRI). These inverters assist the energy storage with voltage regulation. Additionally, the proposed method includes support for varying energy storage unit (ESU) sizes, non-radial distribution systems, and reverse power flow, both real and reactive. The method is applied to the worst-case voltage regulation scenario. The impact of the placement and voltage regulation on the profitability of energy storage is assessed. This is accomplished by adding voltage regulation as a constraint to the problem scheduling energy storage in order to maximize profit. Applying the method shows that the best place to put an ESU is near the end of a feeder. Validation of the method shows that it does not impact the ability of ES to be scheduled in order to maximize economic benefits with time-of-use pricing.

Smart grid applications of selected energy storage technologies

Large-scale energy storage has recently been discussed as part of the future of the smart grid because of the many opportunities for improvement in the reliability and quality of the electric grid that can stem from their use. Information exchange from utility to consumer and vice versa makes it possible to send real-time signals regarding electricity prices and consumption. This enables many applications such as arbitrage, electric reserves, and load following to be served immediately by energy storage systems that are on the grid. For this reason, large-scale energy storage technologies must be implemented that have the ability to serve these applications. This paper examines three energy storage technologies that appear to be well suited for large-scale implementation: sodium-sulfur, vanadium-redox flow batteries, and lithium-ion batteries. These technologies were examined along with many other current technologies and chosen due to the potential to operate at grid-scales. Also, several of their potential applications are discussed.

Optimal battery chemistry, capacity selection, charge/discharge schedule, and lifetime of energy storage under time-of-use pricing

Energy storage units (ESU) can reduce the cost of purchased electricity under time-of-use (TOU) pricing. To maximize the cost reduction, the chemistries, capacities, and charge/discharge schedules of the batteries used in the ESU must be selected appropriately. The batteries must have sufficient capacities to supply the energy demanded by the charge/discharge profiles and to meet the project lifetime. The ESU responds to a TOU price structure. The ESU output power is limited by the rating of the power electronic interface. The cost of the ESU is assumed to increase linearly with battery capacity. A method using linear optimization is developed that determines the battery chemistries, capacities, and charge/discharge schedules simultaneously. The method shows that the Li-Ion battery chemistry is the most cost effective technology due to its high efficiency and that an 11-year project lifetime is most profitable.

Selection of converter topologies for distributed energy resources

Distributed energy resources (DER) are becoming increasingly common on the electrical grid. Depending on the operating conditions of the DER, which depend on the application, different topologies need to be selected in order to achieve the maximum efficiency of each DER. Complicating the selection is the fact that operating conditions vary over time. For example, the voltage and current drawn from a PV panel varies over the course of a day. To calculate the overall efficiency, the efficiency of a topology at each operating point and the amount of time spent at that operating point must be considered. This work extends existing analytical methods for loss calculations by taking this into account. The specific DER applications considered are a three-phase ultracapacitor energy storage unit (UC-ESU), battery energy storage unit (B-ESU), and photovoltaic array (PV). This work determines for each application if an inverter-only (single-stage) or an inverter plus boost converter (double-stage) topology is more efficient. The results show that a single-stage topology is better for the B-ESU and PV, while the double-stage topology is better for the UC-ESU. The method is applicable to other DER types, including wind turbines, micro-hydro generators, variable-speed gensets, and microturbines.

A Semi-Markov Model for Control of Energy Storage in Utility Grids and Microgrids With PV Generation

Photovoltaic (PV) penetration levels in the power grid have significantly increased during the last years. However, issues such as cloud-induced intermittency in PV generation forces equipment on the electrical grid to cycle excessively, preventing PV generation from being considered as a reliable or dispatchable source of power, particularly by utilities. A semi-Markov process model is proposed to model PV power. Unlike existing models of PV power, the proposed model has a wide range of applicability across both small and large timescales. These applications include simulating PV power, short-term forecasting of PV power, design of rule-based controllers for energy storage units (ESU), and stochastic scheduling of ESU in conjunction with other resources. The model is applied to study two cases of coordinating ESU with PV generation. In the first case, the model serves to design a coordination scheme for a hybrid battery-ultracapacitor (UC) ESU where the UC serves to extend the lifetime of a lead-acid battery. In the second case, the model allows probabilistic scheduling in a standalone PV/diesel/battery-ESU microgrid. In both cases, the model improves performance in terms of either battery lifetime or fuel consumption.

A local voltage regulator that improves energy savings under Advanced Volt-Var Control

Advanced Volt Var Control (AVVC) systems offer distributed, granular, and dynamic visibility and control to distribution system engineers working to comply with ANSI C84.1 standards while achieving broader goals, such as reducing energy consumption or integrating Distributed Energy Resources (DERs). Local voltage regulation provided by an AVVC system is a means for mitigating the small percentage of loads in a distribution feeder that limit the effectiveness of conservation voltage reduction (CVR) initiatives, for example. An idealized local voltage regulator (LVR) enables complete decoupling from the feeder voltage while providing precise voltage to connected loads and RMS current reduction to the upstream feeder. Fast-acting distributed power electronics acting on the edge of the grid are a natural solution for achieving this type of local voltage control. In this paper, we describe the implementation of a particular type of power electronics based LVR device, including basic theory of operation and functionality and test data. We use this new LVR in distribution system models to estimate the energy savings impact on CVR for representative feeder classes.

Modelling PV Clouding Effects Using a Semi-Markov Process with Application to Energy Storage

Abstract Cloud-induced intermittency of photovoltaic (PV) generation forces equipment on the electrical grid to cycle excessively preventing PV from being considered as a reliable or dispatchable source of power. Energy storage units (ESU) are proposed to turn PV power dispatchable. In order to use an ESU most effectively, it must be controlled appropriately by considering cloud-induced effects. To this end, the cloud structure is modeled as a random sequence inferred from clouding data. The proposed model is valid for centralized PV installations and serves to develop not only a control methodology to coordinate an ESU with existing grid equipment but also as a sizing criterion for an ESU. The above methodology is demonstrated on both clouding data collected from a rooftop PV installation that includes a pyranometer.

Implementation of a three-phase multilevel boosting inverter using switched-capacitor converter cells

Manufacturers of power electronic converters seek to increase not only power and voltage ratings but also power density. Inductorless converters help achieve the latter because capacitive energy storage has higher energy density than inductive energy storage. Multilevel topologies, which include some inductorless topologies, enable converters to operate at higher voltage levels by allowing voltage sharing between devices. This paper presents a novel multilevel inductorless boosting three-phase inverter that is constructed using a series configuration of switched-capacitor converter cells. This allows for the inverter to be used without an additional boost converter or output transformer when powered from a low-voltage dc source, such as the battery of a grid-connected energy storage system or an electric vehicle.

Placement of distributed energy storage via multidimensional scaling and clustering

Energy storage has long been proposed at the distribution level, where it can provide additional benefits via ancillary services. This work studies how to place energy storage units (ESU) on a distribution feeder in the most cost-effective manner while still meeting voltage regulation requirements. The feeder also has photovoltaic (PV) generation, and the PV ability to supply reactive power is considered. The placement of the ESU is performed via a fast heuristic, in which multidimensional scaling (MDS) is used to transform the combinatorial placement problem into a continuous-valued problem by mapping buses to points in a space. In the new space, clustering algorithms can be applied to determine the ESU locations from a set of candidate locations. The method reduces computation time by an order of magnitude, allowing for various distribution feeder configurations to be quickly compared.

Improvement of conservation voltage reduction energy savings via local voltage regulation

Conservation voltage reduction (CVR) offers an inexpensive method to reduce the power consumption of distribution feeders with minimal capital investment. However, the ability to apply CVR on a feeder is limited by a small number of “bottleneck” loads, those with poor voltage regulation. Recent advances in power semiconductor devices are making distributed power and voltage control at the edges of the grid a reality. One example of this is a power electronic local voltage regulator (LVR), which provides a targeted solution to mitigate this problem of “bottlenecks.” The targeted nature of the devices means that with a small number of approximations the authors were able to develop a computationally efficient method to economically place the LVRs. This paper illustrates how the placement method is applied to quantify the economic savings gained by improving CVR through the deployment of LVRs. Deploying a small number of LVRs can yield significant savings versus conventional approaches, while also improving power quality to end user customers.