Review of photovoltaic power forecasting

High-impact weather events, such as severe thunderstorms, tornadoes, and hurricanes, cause significant disruptions to infrastructure, property loss, and even fatalities. High-impact events can also positively impact society, such as the impact on savings through renewable energy. Prediction of these events has improved substantially with greater observational capabilities, increased computing power, and better model physics, but there is still significant room for improvement. Artificial intelligence (AI) and data science technologies, specifically machine learning and data mining, bridge the gap between numerical model prediction and real-time guidance by improving accuracy. AI techniques also extract otherwise unavailable information from forecast models by fusing model output with observations to provide additional decision support for forecasters and users. In this work, we demonstrate that applying AI techniques along with a physical understanding of the environment can significantly improve ...

/pdf/using-artificial-intelligence-to-improve-real-time-decision-1dx1ehcrjy.pdf

Using Artificial Intelligence to Improve Real-Time Decision-Making for High-Impact Weather

The electric grid is a highly complex, interconnected machine, and changing one part of the grid can have consequences elsewhere. Adding wind and solar affects the operation of the other power plants and adding high penetrations can induce cycling of fossil-fueled generators. Cycling leads to wear-and-tear costs and changes in emissions. Phase 2 of the Western Wind and Solar Integration Study (WWSIS-2) evaluated these costs and emissions and simulated grid operations for a year to investigate the detailed impact of wind and solar on the fossil-fueled fleet. This built on Phase 1, one of the largest wind and solar integration studies ever conducted, which examined operational impacts of high wind and solar penetrations in the West.

/pdf/the-western-wind-and-solar-integration-study-phase-2-8cgjc8qca7.pdf

The Western Wind and Solar Integration Study Phase 2

This paper proposes a day-ahead stochastic scheduling model in electricity markets. The model considers hourly forecast errors of system loads and variable renewable sources as well as random outages of power system components. A chance-constrained stochastic programming formulation with economic and reliability metrics is presented for the day-ahead scheduling. Reserve requirements and line flow limits are formulated as chance constraints in which power system reliability requirements are to be satisfied with a presumed level of high probability. The chance-constrained stochastic programming formulation is converted into a linear deterministic problem and a decomposition-based method is utilized to solve the day-ahead scheduling problem. Numerical tests are performed and the results are analyzed for a modified 31-bus system and an IEEE 118-bus system. The results show the viability of the proposed formulation for the day-ahead stochastic scheduling. Comparative evaluations of the proposed chance-constrained method and the Monte Carlo simulation (MCS) method are presented in the paper.

Chance-Constrained Day-Ahead Scheduling in Stochastic Power System Operation

The ever-increasing penetration of variable wind energy in power systems affects the hourly dispatch of thermal power generation in electricity markets. The variability of wind energy, which makes the wind energy non-dispatchable and difficult to control, could bring significant challenges to power system operators. The hourly coordination of wind power generation units with pumped-storage hydro (PS) generation could relieve the variability of wind energy and increase its dispatchability. The wind-PS coordination is based on the application of stochastic security-constrained unit commitment (stochastic SCUC). In this study, the coordinated hourly bus-level scheduling of wind-PS is compared with the coordinated system-level operation strategies in the day-ahead scheduling of power systems.

Enhancing the Dispatchability of Variable Wind Generation by Coordination With Pumped-Storage Hydro Units in Stochastic Power Systems

The rapid deployment of wind and solar energy generation systems has resulted in a need to better understand, predict, and manage variable generation. The uncertainty around wind and solar power forecasts is still viewed by the power industry as being quite high, and many barriers to forecast adoption by power system operators still remain. In response, the U.S. Department of Energy has sponsored, in partnership with the National Oceanic and Atmospheric Administration, public, private, and academic organizations, two projects to advance wind and solar power forecasts. Additionally, several utilities and grid operators have recognized the value of adopting variable generation forecasting and have taken great strides to enhance their usage of forecasting. In parallel, power system markets and operations are evolving to integrate greater amounts of variable generation. This paper will discuss the recent trends in wind and solar power forecasting technologies in the U.S., the role of forecasting in an evolving power system framework, and the benefits to intended forecast users.

/pdf/recent-trends-in-variable-generation-forecasting-and-its-3i5mllitym.pdf

Recent Trends in Variable Generation Forecasting and Its Value to the Power System

Wind and solar power are playing an increasing role in the electrical grid, but their inherent power variability can augment uncertainties in the operation of power systems. One solution to help mitigate the impacts and provide more flexibility is enhanced wind and solar power forecasting; however, its relative utility is also uncertain. Within the variability of solar and wind power, repercussions from large ramping events are of primary concern. At the same time, there is no clear definition of what constitutes a ramping event, with various criteria used in different operational areas. Here, the swinging door algorithm, originally used for data compression in trend logging, is applied to identify variable generation ramping events from historic operational data. The identification of ramps in a simple and automated fashion is a critical task that feeds into a larger work of 1) defining novel metrics for wind and solar power forecasting that attempt to capture the true impact of forecast errors on system operations and economics, and 2) informing various power system models in a data-driven manner for superior exploratory simulation research. Both allow inference on sensitivities and meaningful correlations, as well as quantify the value of probabilistic approaches for future use in practice.

/pdf/identifying-wind-and-solar-ramping-events-3yfir0yav3.pdf

Identifying Wind and Solar Ramping Events

The introduction of large amounts of variable and uncertain power sources, such as wind power, into the electricity grid presents a number of challenges for system operations. One issue involves the uncertainty associated with scheduling power that wind will supply in future timeframes. However, this is not an entirely new challenge; load is also variable and uncertain, and is strongly influenced by weather patterns. In this work we make a comparison between the day-ahead forecasting errors encountered in wind power forecasting and load forecasting. The study examines the distribution of errors from operational forecasting systems in two different Independent System Operator (ISO) regions for both wind power and load forecasts at the day-ahead timeframe. The day-ahead timescale is critical in power system operations because it serves the unit commitment function for slowstarting conventional generators.

A Comparison of Wind Power and Load Forecasting Error Distributions

Wind integration studies are conducted routinely to examine the operational impacts of wind on the power system. Wind plant power outputs and forecasts are needed as inputs to these studies and this data is often synthesized by a variety of methods. This paper examines the methodologies used to create these datasets, the pitfalls that may be encountered, and the tradeoffs between different methodological approaches.

/pdf/wind-data-inputs-for-regional-wind-integration-studies-3gnfogncsb.pdf

Wind data inputs for regional wind integration studies

/pdf/solar-data-inputs-for-integration-and-transmission-planning-29zwa5vdb1.pdf

Kirsten Orwig

Papers

Recent Trends in Variable Generation Forecasting and Its Value to the Power System

Identifying Wind and Solar Ramping Events

A Comparison of Wind Power and Load Forecasting Error Distributions

Wind data inputs for regional wind integration studies

Solar Data Inputs for Integration and Transmission Planning Studies