Jouni Peppanen

Leveraging AMI Data for Distribution System Model Calibration and Situational Awareness

The many new distributed energy resources being installed at the distribution system level require increased visibility into system operations that will be enabled by distribution system state estimation (DSSE) and situational awareness applications. Reliable and accurate DSSE requires both robust methods for managing the big data provided by smart meters and quality distribution system models. This paper presents intelligent methods for detecting and dealing with missing or inaccurate smart meter data, as well as the ways to process the data for different applications. It also presents an efficient and flexible parameter estimation method based on the voltage drop equation and regression analysis to enhance distribution system model accuracy. Finally, it presents a 3-D graphical user interface for advanced visualization of the system state and events. We demonstrate this paper for a university distribution network with the state-of-the-art real-time and historical smart meter data infrastructure.

Increasing distribution system model accuracy with extensive deployment of smart meters

The advent of smart metering infrastructure has paved the way to enhance the accuracy of conventional distribution system models. However, the size and complexity of the smart meter data and information systems present new issues, including the cumbersome process of extracting and analyzing data from various data sources. The focus of this paper is to create a highly accurate distribution system model using Big Data from smart meters. It presents details on how traditional modeling can be improved by smart meter data as well as practical methods to make the data functional for analysis.

Distribution system low-voltage circuit topology estimation using smart metering data

Operating distribution systems with a growing number of distributed energy resources requires accurate feeder models down to the point of interconnection. Many of the new resources are located in the secondary low-voltage distribution circuits that typically are not modeled or modeled with low level of detail. This paper presents a practical and computational efficient approach for estimating the secondary circuit topologies from historical voltage and power measurement data provided by smart meters and distributed energy resource sensors. The accuracy of the algorithm is demonstrated on a 66- node test circuit utilizing real AMI data. The algorithm is also utilized to estimate the secondary circuit topologies of the Georgia Tech distribution system. Challenges and practical implementation approaches of the algorithm are discussed. The paper demonstrates the computational infeasibility of exhaustive secondary circuit topology estimation approaches and presents an efficient algorithm for verifying whether two radial secondary circuits have identical topologies.

Neighborhood electric vehicle charging scheduling using particle swarm optimization

Chargeable electric vehicles are projected to gain increasing market share becoming a significant load in distribution systems. An un-controlled charging of a large number of electric vehicles can potentially lead to problems in distribution circuits including low voltage levels and component overloads. These problems can be avoided by implementing a vehicle charging control scheme. This paper proposes a particle-swarm optimization-based method to centrally control vehicle charging on a neighborhood level. Vehicle charging is scheduled day-ahead for a given distribution system area while minimizing the total charging cost subject to grid and vehicle constraints. The proposed computationally efficient algorithm reduces the charging cost while enforcing voltage or line flow limits applying linear sensitivities. We demonstrate the method in a model of a real meshed 121-bus, 57-vehicle European low voltage distribution system.

Handling bad or missing smart meter data through advanced data imputation

Smart meters and other the modern distribution measurement devices provide new and more data, but usually they are subject to longer delays and lower reliability than transmission system SCADA. Accurate and robust use of the modern distribution system measurements will be a cornerstone of the future advanced distribution management systems. This paper presents a novel and computationally efficient data processing method for imputing bad and missing load power measurements to create full power consumption data sets. The imputed data periods have a continuous profile with respect to the adjacent available measurements, which is a highly desirable feature for time-series (power flow) analyses. The method is shown to be superior in accuracy to a utility best practice approach. Our simulations use actual AMI data collected from 128 smart meters on the Georgia Tech campus.

Secondary circuit model generation using limited PV measurements and parameter estimation

This paper presents an approach for generating simplified secondary circuit models with limited SCADA and PV micro-inverter measurement data. The proposed method is computationally efficient and can be utilized with typically available measurement data. The method is applied to models of three real U.S. utility feeders with PV micro-inverter measurements. The proposed simplified secondary circuit modeling approach decreases the PV voltage simulation errors in all the three feeders compared to using generic secondary circuit models. This paper also presents approaches for improving the feeder voltage regulating device model set points by utilizing the PV voltage measurements.

Using distribution LMP and time-of-delivery pricing to promote optimal placement and increased profitability of residential PV systems

In order to stimulate continued expansion of residential PV in the future, PV should be reimbursed for the benefits it provides to the grid with local distributed generation. This includes market mechanisms to compensate for the time-of-delivery (TOD) of the solar energy and the location of the injection using distribution locational marginal prices (DLMPs). The benefits of using TOD pricing is analyzed for three LMP nodes in NYISO for the real-time and the day-ahead markets for 2011. A local distribution system market is demonstrated with varying DLMP prices that depend on distribution system losses. The DLMPs change with local generation, substation price curves, time of day, line losses, and feeder load. Such a market allows the PV owner to be reimbursed appropriately for a reduction in system losses and allows PV to naturally be installed in the optimal locations due to the pricing mechanism.

Secondary circuit model creation and validation with AMI and transformer measurements

Accurate distribution secondary circuit models are needed to effectively monitor and coordinate the distributed energy resources located in the secondary circuits and to enhance overall distribution system operations and planning. Accurate secondary models are also needed to fully leverage the measurement data received from smart meters and distributed energy resources at the customer premises. This paper discusses approaches for creating distribution system secondary low-voltage circuit models utilizing smart meter measurements. This paper also discusses methods to model secondary circuits when the loads and distributed energy resources are only partially metered. The presented methods are demonstrated on a real distribution secondary circuit with smart meter measurements and transformer low voltage measurements. Practical challenges related to real measurement data are discussed.

Thermal Energy Storage for Air Conditioning as an Enabler of Residential Demand Response

Residential Demand Response (DR) has been associated with many benefits. In the residential sector, air conditioning (AC) currently has the largest peak demand reduction potential, but it is limited by the comfort bounds set by the user. This paper studies the limitations of AC load shifting and the attractiveness of using thermal energy storage (TES) to increase residential demand response potential. A general building thermodynamic model is developed and is used to evaluate AC load shifting potential with different AC control principles and electricity rate structures. The viability of optimal AC operation and TES is demonstrated with estimates of achievable over-the-lifetime electricity cost savings.

Tools for Enhanced Grid Operation and Optimized PV Penetration Utilizing Highly Distributed Sensor Data.

Increasing number s of PV on distribution systems are creating more grid impacts , but it also provides more opportunities for measurement, sensing, and control of the grid in a distributed fashion. This report demonstrates three software tools for characterizing and controlling distribution feeders by utilizing large numbers of highly distributed current, voltage , and irradiance sensors. Instructions and a user manual is presented for each tool. First, the tool for distribution system secondary circuit parameter estimation is presented. This tool allows studying distribution system parameter estimation accuracy with user-selected active power, reactive power, and voltage measurements and measurement error levels. Second, the tool for multi-objective inverter control is shown. Various PV inverter control strategies can be selected to objectively compare their impact on the feeder. Third, the tool for energy storage for PV ramp rate smoothing is presented. The tool allows the user to select different storage characteristics (power and energy ratings) and control types (local vs. centralized) to study the tradeoffs between state-of-charge (SOC) management and the amount of ramp rate smoothing.