EPIC projects use innovative technology to enhance areas tied to PG&E's core values of providing safe, reliable and affordable energy for our customers. The reports below document all of our completed EPIC projects and provide a summary of the objectives, scope of work, results, technology transfer plan and alignment to the EPIC principles and metrics. Additional reports are added as they are completed, and information about the progress of each active project can be found in the latest PG&E EPIC Annual Report.
EPIC 1.01 - Energy Storage End Uses
This project successfully utilized PG&E's Vaca-Dixon and Yerba Buena Battery Energy Storage Systems (BESSs) to gain experience and data by participating in CAISO's Non-Generator Resource (NGR) market model. PG&E developed and deployed an automated communications and control solution to fully utilize and evaluate BESS fast-response functionalities.
EPIC 1.02 - Demonstrate Use of Distributed Energy Storage for Transmission and Distribution Cost Reduction
This project demonstrated the ability of a utility-owned and controlled energy storage resource to deliver autonomous distribution peak shaving functionality. Energy storage resources hold significant promise to help California address a variety of grid planning and operations challenges, both today and in the future, and can be used to provide more reliable and clean power to customers for lower overall costs. The learnings from this project can help inform utility procurement and operation of future energy storage resources, both utility-owned and utility-contracted, through compliance with the IOU energy procurement targets as set forth in CPUC D. 10-03-040 and beyond.
EPIC 1.05 - Demonstrate New Resource Forecast Methods to Better Predict Variable Resource Output
This project successfully developed and demonstrated a new mesoscale meteorological model to provide more granular and accurate weather forecasting input to PG&E's storm damage prediction model and to other PG&E forecasting applications, such as catastrophic wildfire risk, large storms and photovoltaic (PV) generation. This model has improved the accuracy of forecasting for large storms, allowing for increased efficiencies in storm preparation, as well as enhanced the accuracy of identifying fire risks, helping enable improved reliability and safety. Finally, leveraging granular solar irradiance data in a new framework has improved PG&E's ability to understand the impacts of PV generation for grid management.
EPIC 1.08 - Distribution System Safety and Reliability through New Data Analytics Techniques
This project demonstrates a visualization and decision support system to support PG&E's risk management efforts to enhance public and system safety, as well as improve asset management strategies and investment plans.
EPIC 1.09A - Test New Remote Monitoring and Control Systems for Existing Transmission & Distribution Assets: Close Proximity Switching
This project focused on increasing system reliability and improving the safe operation of three-phase Load Break Oil Rotary switches, which are used for making or breaking the path in an electrical circuit. In both a lab and field setting, this project successfully demonstrated and evaluated various robotics that would allow PG&E workers to more safely operate certain subsurface or underground (UG) oil switches.
EPIC 1.09B/10B - Test New Remote Monitoring and Control Systems for T&D Assets / Demonstrate New Strategies and Technologies to Improve the Efficacy of Existing Maintenance and Replacement Programs
This project successfully demonstrated methods of evaluating and potentially extending the longevity, resiliency and data integrity of Supervisory Control and Data Acquisition (SCADA) condition-monitoring components over time. The overall strength of the monitoring and communication systems currently installed across the distribution network was confirmed and methods for improving the life and data integrity of its components were demonstrated. Real-time condition monitoring of this system provides a key input to support proactive mitigation of equipment-related issues.
EPIC 1.09C - Test New Remote Monitoring and Control Systems for T&D Assets
This project successfully demonstrated a new technology deployed directly onto transmission conductors to detect potential overloads and increase line impedance to shift this load to parallel facilities. These devices can potentially enable optimization of line flows, mitigation of overloads, and delay of costly new transmission line or reconductoring projects.
EPIC 1.14 – Next Generation SmartMeter™ Telecom Network Functionalities
This project evaluated the radio mesh telecommunications network that connects SmartMeter™ devices across PG&E's territory, including demonstration of new potential use cases for that network. The project created a methodology to determine available bandwidth, tested a variety of smart grid devices to demonstrate their potential to leverage the network for communications, and demonstrated potential enhancements to the existing outage reporting capabilities of SmartMeter™ devices.
EPIC 1.15 - Grid Operations Situational Intelligence (GOSI)
This project demonstrated a technology platform to visualize grid operations data to improve both real-time and short-term operational decisions, such as outage anticipation, construction planning, circuit loading research, and emergency operations. The project developed key data, system, and user experience learnings through integrating more than 20 data sources into a single visualization tool allowing users to view complex data sources in ways that were not possible through current solutions. This project formed the foundational learnings which will allow PG&E to potentially explore other complex situational awareness tools and applications to allow users to target information to help manage changes on the grid.
EPIC 1.16 - Demonstrate Electric Vehicle as a Resource to Improve Grid Power Quality and Reduce Customer Outages
This project successfully developed and demonstrated a new Vehicle On-Site Grid Support System (VOGSS), for utility-grade power export from Plug-in Hybrid Electric Vehicle (PHEV) fleet trucks. This new technology enables a source of mobile power that can connect directly to distribution circuits, minimizing the impact of an outage for common preventative maintenance tasks such as transformer replacements. Additionally, VOGSS can provide power to facilities in emergency events, maintaining or quickly restoring service to customers.
EPIC 1.18 - Demonstrate SmartMeter™-Enabled Data Analytics to Provide Customers with Appliance-Level Energy Use Information
This project conducted a demonstration to understand and compare disaggregation vendors' ability to itemize monthly appliance-level usage for residential customers, as well as their current analytical capability and accuracy of their energy disaggregation software. Additionally, this project surveyed customers to understand their perception of the end-use energy presentations and the value of the disaggregated data.
EPIC 1.19 - Enhanced Data Techniques and Capabilities via the SmartMeter™ Platform
This project successfully demonstrated new ways to leverage the SmartMeter™ platform to provide greater visibility and granularity to additional SmartMeter™ data. The project proved the ability to collect power quality data and potentially enable a proactive response to address customer satisfaction concerns on voltage issues. The project also connected difficult to reach meters to the AMI network to potentially reduce manual meter reading operation and maintenance costs. Finally, the project improved the ability to identify ‘Line Side Tap' scenarios to improve the efficiency and effectiveness of investigating energy diversion cases and to mitigate safety hazards for customers, the public or PG&E.
EPIC 1.21 - Auto Identification of Photovoltaic (PV) Resources
This project focused on developing and demonstrating technology to identify existence of PV systems using SmartMeter™ and other data not otherwise recorded in PG&E's interconnection database. Additionally, the project explored the ability of detecting underperforming or malfunctioning PV systems. The project was able to develop key inputs necessary to identify a PV system, filter for those identified systems with unauthorized interconnections, support high quality interconnection records by validating the size of PV systems, understand the limitations in the ability to detect if a PV system is underperforming or not functioning, and established a process to engage with solar customers to provide appropriate notice.
EPIC 1.22 - Demonstrate Subtractive Billing With Submetering for EVs to Increase Customer Billing Flexibility
This project was part of a California Statewide effort to demonstrate and evaluate the use of Electric Vehicle (EV) submetering to provide EV owners access to electricity at a less expensive electric rate—without having to install an additional utility meter to an existing service. This project also assessed EV customer demand for submetering and the customer experience with submetering.
EPIC 1.23 - Photovoltaic (PV) Submetering
This project focused on developing, testing, and validating a way of collecting or estimating solar generation output data and enabling a subset of customers to view their estimated solar generation data through integration with PG&E's YourAccount website (previously known as MyEnergy). Upon determining that using estimated PV generation data would be a viable option, the project also assessed the accuracy of the algorithm used by a third-party vendor. The project determined that additional data is necessary to develop a scalable PV generation estimate, including shading impacts, PV system tilt and azimuth, as well as weather data like fog and marine layer.
EPIC 1.24 - Demonstrate Demand-Side Management (DSM) for Transmission and Distribution (T&D) Cost Reduction
This project successfully provided and tested the performance of a near real-time window of PG&E's Air Conditioning (AC) Direct Load Control (DLC) system, which utilizes one-way switch control devices. This allowed us to improve our ability to estimate AC DLC impacts at the distribution system level to better understand the localized impact of AC direct load control devices on meeting distribution feeder level reliability concerns. It also enabled near real-time visibility of AC direct load control installations to support Transmission and Distribution (T&D) Operations and provided Demand Response (DR) program administrators with near real-time feedback on any problems with direct load control devices before, during or after an event is called, which supports T&D operational improvements.
EPIC 1.25 - Direct Current Fast Charging (DCFC) Mapping
DCFC charging stations provide the ability for DCFC-ready EVs to recharge to 80% in 30 minutes or less. This project addressed Electric Vehicle (EV) adoption barriers by identifying optimal locations within PG&E's territory for the placement of DCFCs based on factors such as cost, available service transformer capacity, traffic patterns, as well as site host and driver preference. PG&E worked with industry experts to identify the 300 locations of highest unmet public charging need, forecasted out to 2025. Using a variety of inputs, including publicly-available business listing data, PG&E's distribution network to assess available distribution capacity, results from expert interviews, and PlugShare's database on existing public charging locations, the team then identified over 14,000 individual potential charger host sites, such as businesses, parking lots, and public places. The results of the project were developed into an interactive online map that visualizes the 300 optimal DCFC locations. The publicly-available map is accompanied by guidelines in the final report surrounding best practices for siting DCFCs developed to further encourage EV adoption by drivers, site hosts, and developers.
EPIC 2.02 – Distributed Energy Resource Management System
This project provided an opportunity for PG&E to define and deploy a DERMS and supporting technology to uncover barriers and specify requirements to prepare for the increasing challenges and opportunities of DERs at scale. The DERMS Demo was a ground-breaking field demonstration of optimal control of a portfolio of 3rd party aggregated behind-the-meter (BTM) solar and energy storage and utility front-of-the-meter (FTM) energy storage to provide distribution capacity and voltage support services while also allowing for participation of these same DERs in the CAISO wholesale market.
EPIC 2.03A Smart Inverters
EPIC 2.03B - Test Smart Inverter Enhanced Capabilities – Vehicle to Home
This project assessed the technical feasibility and potential benefits to individual customers and to ratepayers of vehicle to home (V2H) technology which can be utilized for resiliency and reliability. V2H is technically capable of islanding and supporting household load in outage and demand response events and customers reported high initial interest. However, the technology is not yet commercially available and vehicle warranties must be modified to allow for discharge, the cost to customers exceed their perceived benefits, and the net benefits to the utility and ratepayers are likely not sufficient to surmount the low cost-effectiveness for customers. The V2H market is nascent and requires further investigation ahead of PG&E commercialization activities.
EPIC 2.04 - Distributed Generation Monitoring and Voltage Tracking
This project demonstrated an algorithmic process to analyze new data sources (including SmartMeter™ devices and databases of solar irradiance) to predict the likelihood that a Rule 2 voltage violation was caused by distributed solar generation. Solar energy is by nature intermittent, and ebbs and surges of generation can change the voltage for neighboring, downstream customers. As solar adoption continues to grow, there is an increased likelihood of such voltage violations. This functionality, if integrated into a larger grid analytics platform, might improve decision making for Power Quality Engineers responding to customer issues, and Distribution Planners as they work to support safe and reliable solar installation across PG&E's service territory.
EPIC 2.05 - Inertia Response Emulation for DG Impact Improvement
This project explored the capabilities of inverter-based energy resources to provide a set of functions related to system inertia which support the electric system. The project demonstrated via transmission system modeling and Power-Hardware-In-Loop testing that advanced inverter control methods can provide active power support that improves the system’s frequency response in the face of reduced conventional inertia from synchronous machine generators. Inverter control methods were explored including inertia-like response (derivative control) and grid-forming (voltage source) modes for respective benefits in bulk system and isolated distribution system use cases.
EPIC 2.07 - Real Time Loading Data for Distribution Operations and Planning
This project developed analytical methods for generating near real-time load forecast information. The project successfully built and demonstrated a platform to ingest and process SmartMeter™, Supervisory Control and Data Acquisition (SCADA), photovoltaic system (PV) generation, Geographic Information System (GIS) and weather data for two of the eight Areas of Responsibility (AOR) within PG&E’s service territory.
EPIC 2.10 - Emergency Preparedness Modeling
The project developed and demonstrated a decision support system that successfully recommends restoration strategies for PG&E electric assets after a disruptive event occurs. To accomplish this, the following high level key business requirements were achieved:
EPIC 2.14 - Automatically Map Phasing Information
This project successfully developed and demonstrated automated analytical methods for determining meter phasing and meter-to-transformer connectivity using SmartMeter™, Supervisory Control and Data Acquisition (SCADA) and Geographic Information System (GIS) data. The distribution network model is central to multiple existing control systems, system analyses, and work processes. As the load characteristics of the distribution network evolve, such as with the growth of Distributed Energy Resources (DER), it is becoming more important to have accurate and up-to-date network model information to be able to actively manage the distribution system. Automated approaches for obtaining this information can offer a more efficient alternative to the conventional boots-on-the-ground approach.
EPIC 2.15 - Synchrophasor Applications for Generator Dynamic Model Validation
This project installed Phasor Measurement Units (PMUs) on the three generators at PG&E's Colusa Generation Station, developed station generator models using commercial software, and used actual disturbance data collected online (in lieu of offline test data) to test new synchrophasor applications for generator model validation. The integration of PMUs on generators for dynamic model validation is a new technology and the project did not result in a tool that is production ready. As applications evolve, installation of PMUs at generating stations could potentially allow utilities to enhance their generator model validation processes.
EPIC 2.19 - Enable Distributed Demand-Side Strategies & Technologies
This project evaluated the performance and efficacy of using customer-sited behind-the-meter storage for grid and reliability services. The project utilized both residential and commercial assets via two vendor platforms. BTM Energy storage is technically feasible for the use cases evaluated, but before a full program is pursued there are opportunities for improvement.
EPIC 2.21 - Home Area Network (HAN) for Commercial Customers
This project demonstrated the viability and usefulness of access to real-time energy use data for commercial customers. This technology demonstration accomplished three set objectives: 1) verified Zigbee enabled SmartMeters™ for Large Commercial and Industrial customers have the same ability as residential meters to provide real-time usage information via the HAN radio; 2) Identified and assessed LC&I customers' needs and meaningful use cases (i.e. opportunities) for real-time data; 3) Identified the barriers to adoption, integration, and utilization of HAN devices at scale for LC&I customers.
EPIC 2.22 - Demand Reduction through Targeted Data Analytics
This project developed a tool that leverages customer level data along with grid information and forecasts to create a robust optimization engine for identification of the lowest cost solution capable of deferring or mitigating the need for an asset upgrade due to capacity limitations. The tool considers both traditional wires solutions and DER portfolios and allows Distribution Planners to complete advanced scenario analysis.
EPIC 2.23 - Demand Side Utility Planning
This project successfully developed and demonstrated the integration of a broader range of customer-side technologies and Distributed Energy Resources (DER) approaches into the utility planning process. The project served as a necessary and enabling precursor to the fulfillment of Assembly Bill (AB) 327/ Section 769, which requires transparent, consistent and more accurate methods to cost-effectively integrate DERs into the distribution planning process. This project delivered new load shape profiles, enhanced load forecasting tool and overall analytical process that allows PG&E to more accurately and consistently integrate DER impact to the distribution system load profile. With these enhancements, PG&E can evaluate if DER growth could defer or even in some instances eliminate the need for future network upgrades. Leveraging any of the SmartMeter™ data, PG&E created more accurate and granular load shapes that allowed distribution planners to more precisely capture DER impact on the load growth forecast.
EPIC 2.26 - Customer and Distribution Automation Open Architecture Devices
PG&E's AMI Network is one of the largest private Internet Protocol Version 6 (IPv6) networks in the Unites States, with more than 5 million AMI devices connected across its electric network. This project investigated the use of the AMI network for purposes beyond the collection of electricity usage data. The project successfully demonstrated the ability of a Client-Server architecture consisting on an IoT router to establish communication, monitoring, command, and control of various third-party and utility end devices such as smart inverters, sensors, SCADA devices, RFID readers and distributed generation controls over the AMI network using the IEEE 2030.5 protocol.
EPIC 2.27 - Next Generation Integration
This project demonstrated a new AMI Network management system (a "manager of managers") to holistically and more effectively monitor, control, and evolve the existing AMI network and infrastructure. Currently, PG&E leverages multiple AMI networks with separate operational systems. Leveraging disparate systems limits the ability to optimally manage workflow and prioritize and schedule data processes (for instance, ensuring remote connect/disconnect is prioritized over tenant application queries).
EPIC 2.28 - Smart Grid Communications Path Monitoring
This project sought to 1) Conduct an initial noise assessment to establish a baseline of radio frequency interference (RFI) in the AMI Networks, 2) Analyze a continuous flow of data to identify potential locations and sources of RFI, and 3) Develop an end-to-end process/tool from monitoring to mitigation of interference. PG&E identified through a sample of radio frequency (RF) data that there are potential channel congestion issues that can lead to RFI conflicts in the AMI networks, however no specific RF tools existed to identify RFI signal(s) in PG&E’s local Neighborhood Area Network (NAN). Given the RF dataset availability and access limitations, there was no feasible path to demonstrate a successful algorithm-based application for proactive automated interference detection. The preliminary work completed on this project could be leveraged in the development and/or use of future tools and in formulation of strategies around broader prevention of PG&E’s network RFI.
EPIC 2.29 - Mobile Meter Applications
This project designed, built, and tested the Next Generation Meter (NGM). This electricity meter was demonstrated to be the first revenue grade, high resolution real time power meter that fully met national standards for metering including ANSI C12.1 and ANSI C12.20 (accuracy), ANSI C12.19 (meter data table format) and C12.22 (cellular communication protocol format). The NGM was developed with a compact, modular design that takes advantage of a host of new technologies including faster microprocessors, expanded memory, and multiple communications pathways—all contained in a hardware package that is the size of a credit card. The NGM has the ability to: 1) Be installed in a wider range of locations beyond traditional customer premises, 2) Reduce meter maintenance and replacements costs, 3) Improve the grid operator’s situational awareness during outages, and 4) Provide additional services and applications as grid-edge technology evolves
EPIC 2.34 Predictive Risk Identification with Radio Frequency (RF) Added to Line Sensors
This project investigated the use of radio frequency-based Distribution Reliability Line Monitor (DRLM) and Early Fault Detection (EFD) technologies and compared their performance with Distribution Fault Anticipation (DFA) technology for predictive maintenance and risk reduction on electric distribution circuits. The demonstration successfully detected, located and addressed multiple examples of conductor damage, vegetative encroachment, internal transformer discharge, fault induced conductor slap, and insulator and clamp issues. The project concluded that effective grid asset health and performance monitoring can be achieved through an ensemble approach and further work is necessary to improve and integrate sensor technologies into an analytics platform or Distribution Management System (DMS).
EPIC 2.36 Dynamic Rate Design Tool
This project demonstrated a dynamic rate design tool approach built on a cloud platform for modeling customer bill impacts. The project leveraged advanced technologies to experiment with high-level rate designs, new billing determinants, and enabled a more robust, powerful and rapid bill impact analysis process than that which is used by current models. In its current state, the tool can design high-level experimental tiered, time-of-use, and tiered time-of-use rates as well as rates with a maximum demand charge. The dynamic rate design tool can be leveraged and further developed to significantly improve other production-grade tools for rate and bill analysis.