As part of the Energy Department’s commitment to a reliable and resilient power grid, I am pleased to announce that the Office of Electricity Delivery and Energy Reliability (OE) is investing nearly $900,000 in early stage research to address the risk and uncertainty of the power system. This support will allow academic institutions in Pennsylvania, Utah, and Virginia to perform research in wholesale market operations, transmission system design, and demand-side participation. All three areas are transforming electricity markets.
Today’s announcement is another important step in helping to ensure the reliable, resilient, efficient, and secure delivery of electricity to America’s businesses and consumers. Support of cutting edge research and collaboration with U.S. colleges and universities to educate future scientists and engineers are vital to continued improvement of the reliability of America’s power grid. In August 2016, OE announced funding of nearly $1.8 million for five academic institutions in California, Iowa, New York, and Texas under the same Funding Opportunity.
This additional investment is part of the Energy Department’s Grid Modernization Initiative (GMI), a comprehensive effort to help shape the future of our Nation’s grid and solve the challenges of integrating conventional and renewable sources with energy storage and smart buildings, while ensuring that the grid is resilient and secure to withstand growing challenges such as the cybersecurity threat.
The three projects selected for awards are outlined below. Final award amounts are subject to negotiation.
Recipient: The Pennsylvania State University
Title: A Multistage Stochastic Transmission Expansion Algorithm for Wide-area Planning under Uncertainty
Location: State College, Pennsylvania
The Pennsylvania State University will develop and demonstrate a novel computational method for solving the optimal transmission plan under uncertainty in future generation with multiple decision periods applied to a large power system network. The algorithm will take advantage of high-performance computing networks, and will consist of importance sampling-based approximate dynamic programming.
During Phase I, researchers will develop the algorithm and software tools for implementation and validation on small and moderate-sized test networks. Phase II will focus on demonstration of the results of the method for a larger system (e.g., the Western Interconnection and/or PJM Interconnection) to extend the approach to solve for three or more decision stages, and co-optimize generation and transmission over several decision stages. Phase III of the project will involve extending the algorithm to use AC optimal power flow to evaluate the proposed lines in each stage, and compare the results of the new method with other approaches in use, including robust optimization methods.
The project is expected to result in a dramatic reduction in the computation time required to solve for the best high-voltage transmission lines to add to the existing grid in the near-term when it is not known where future generation – in particular, renewable energy sources – will be located. Improved transmission planning would facilitate reduction in the cost of electricity and the integration of greater capacity of renewable generation. The method to be developed, which will be placed in the public domain and shared freely, will allow Independent System Operators (ISOs), Regional Transmission Operators (RTOs), and other regional organizations to plan for changes in the coming decades to the power system more effectively, and consider a broader range of alternatives investments and possible future scenarios than currently possible.
DOE Funds: $174,061
Cost Share: $54,258
Total Project Value: $228,319
Recipient: University of Utah
Title: Stochastic Continuous-time Flexibility Scheduling and Pricing in Wholesale Electricity Markets
Location: Salt Lake City, Utah
The University of Utah’s project will develop the mathematical foundation, implementation practices, and models for performing stochastic continuous-time unit commitment (UC) in wholesale markets operation. The proposed UC will accurately model the continuous-time variations of load and Renewable Energy Sources (RES), and efficiently deploy the ramping capability of flexible resources to compensate the sources of variability and uncertainty in markets – all while respecting the continuous-time flow constraints of transmission network.
Researchers will first formulate the UC problem as a stochastic continuous-time optimal control problem, then discretize the continuous-time UC decisions and forms a decision space with countable dimensions. They will then recast the stochastic continuous-time UC problem into a Mixed-Integer Linear Programming (MILP) optimization problem, where the coefficients of projecting the continuous-time decisions in the finite-dimensional function space represent the decision variables of the discrete-time optimization problem. The resulting formulation will be a two-stage stochastic optimization model, where the first stage decisions model the day-ahead commitment and continuous-time scheduling decisions, and the second stage schedules the continuous-time reserve capacity to compensate the uncertainty of RES and load forecast in real-time operation.
This approach will model the uncertainty of RES and load forecast by continuous-time power trajectory realization scenarios. Thus, the scheduled reserve capacity will not only compensate the real-time energy imbalance, but will also cover the resulting extra ramping requirements, effectively minimizing the risk of ramping scarcity events in the North American power grid, enabling competitive integration of energy storage devices and demand response in wholesale markets operation, and allowing for a tradeoff between model accuracy and computational complexity.
DOE Funds: $357,339
Cost Share: $92,552
Total Project Value: $449,891
Recipient: Virginia Polytechnic Institute and State University (Virginia Tech)
Project Title: A Probability-based Model for Cost-effective Integration of Renewables into the Electricity Grid
Location: Blacksburg, Virginia
Virginia Tech researchers will develop a production costing tool that can take into account the variable nature of renewable energy sources (both solar PV and wind farms), and treat them as generation candidates in a power system expansion plan, unlike the current practice of using them as negative loads. The objective is to develop a probability-based model for Effective Load Carrying Capability (ELCC) calculations for Resource Adequacy (RA) studies and production costing functions for cash flow analyses to help with cost-effective integration of renewables into the power grid.
The fundamental contribution of this project will be developing a probability-based optimization tool that can reflect the probabilistic nature of solar/wind energy outputs. The tool will allow ISOs and RTOs to consider renewable energy as an option for their expansion plans and help them calculate the capacity credit for each solar/wind farm using probabilistic forecasts. The resulting information can serve as a guideline for computing operating reserve requirements to cope with uncertainty and variability of solar/wind output committed into the market. The tool will also offer production costing analysis using uncertain variable generation data, which will give power marketers more powerful capabilities for their cash flow analysis. It will also produce estimates for wind/solar output curtailments. Overall, the tool will address issues of production costing, resource adequacy, and ELCC calculations in a generalized way with applications in the wholesale electric markets in the North American grid.
DOE Funds: $359,691
Cost Share: $89,958
Total Project Value: $449,649