Bonneville Power Administration Research Articles

Wind Power Deviation Charge Reduction using Machine Learning

High penetration of wind power plants in power systems resulted in various challenges such as frequent system imbalances due to highly uncertain and variable wind generation. Additional spinning reserves and specific balancing products such as flexible ramp products are used to handle such frequent imbalances. Incorporation of these ancillary services leads to increased total operational costs. Increased operational costs should be transferred to wind power producers as it is caused by wind power plants. This leads to penalizing the wind power producers for the deviation of power generation from forecasts, called deviation charges. These deviation charges can be reduced by improving the forecasting accuracy. Existing forecasting models show performance in terms of error matrices. Such error matrices do not indicate the financial loss associated with it. This can be overcome by expressing forecasting performance in terms of deviation charge and it will directly encourage wind power producers to improve forecasting accuracy or arrange reserves to accommodate the error. This paper proposes a backpropagation-based artificial neural network model for reducing deviation charges in this context. An analysis is conducted on the data collected from the Bonneville Power Administration (BPA) Balancing Area. Seasonal analysis (Spring, Summer, Fall, and Winter) is conducted to show the performance of the proposed model throughout the year. The proposed model performance is compared with linear regression and ARIMA models. The comparison shows that the proposed ANN model gives the least deviation charges in the Spring, Summer, and Winter seasons and deviation charges in the Fall season are higher than the ARIMA model.

River definitions reciprocally define us

AbstractThis article explores different definitions of river, and in doing so, considers how our efforts to define rivers illuminate aspects of the human condition. Through an exploration of our definitions of river, through the human imagination, in poetry, music, and human's actions, this article considers how our nature is reflected back by rivers—how we in turn are defined by rivers—such as their role in determining borders, movement, escape, the locations of cities, trade, communication, and in the shaping of cultures and societies. This article specifically highlights three definitions of river using examples predominately from North America. River as boundary, exploring how in the reaches of some river basins, rivers act not only as a border, or boundary between two nations or cultures, but also as a line of demarcation between the known and unknown landscapes and the tamed and untamed lands. The definition of river as pathway includes exploration of how the Mississippi River served as a natural highway for people escaping slavery to travel North and reach Underground Railroad routes toward freedom, while also forcibly moving hundreds of thousands of enslaved people in the United States from the Upper South to the Lower South as they were “sold down the river.” Third, this article examines how rivers have been defined as a resource through examples, such as the US government and Bonneville Power Administration's (BPA) efforts, to encourage support of large dam construction and water development projects by hiring Woody Guthrie in 1941 to write songs that would prompt public support for the water projects. Through this discussion of three river definitions, the article considers how rivers have been conceptualized over time and how those different knowledges inform our understanding and treatment of rivers. The article argues that it is vital to include knowledge from across scales, time, boundaries, and ways of knowing in order to improve the protection of rivers and their ecosystems. If we do not, we will continue to lose freshwater biodiversity, wetland and riverine habitat, and their immeasurable qualities that inspire life and imagination at a staggering rate.

The altitude correction term of the corona loss on alternating-current transmission lines

The increase in mobility of charged particles causes high-voltage alternating-current (AC) transmission lines in high-altitude areas to be more likely to undergo severe corona discharge and associated losses, to the detriment of the economic efficiency of power transmission. Evaluating the corona loss on transmission-line conductors in high-altitude areas is conducive to conductor selection, parameter optimization, transmission-line tower arrangement, and optimization of the phase sequence arrangement. The altitude correction factor K of 3.33 dB/km recommended by the Bonneville Power Administration (BPA) in the United States is commonly used in engineering design, however, measured data indicate that the correction method may be subject to significant error, so the altitude correction factor is discussed under different weather conditions. K is suggested to be 4.5 and 1.5 dB/km separately under dry and rainy conditions, the better to evaluate the coronal loss.

Improving Abundance Estimates of Spring–Summer Snake River Chinook Salmon for Fisheries Management

AbstractThe Columbia River basin is home to a run of spring–summer Chinook Salmon Oncorhynchus tshawytscha that returns to the Snake River drainage of Idaho, Oregon, and Washington in the Pacific Northwest. Historically, the run was one of the more productive throughout the Columbia River basin. However, Snake River spring–summer Chinook Salmon have experienced declines in abundance due to overfishing, habitat degradation, and dams. Several stocks are listed as threatened under the U.S. Endangered Species Act and are supported by mitigation hatcheries funded by Idaho Power Company, the Lower Snake River Compensation Plan, and the Bonneville Power Administration. To maximize tribal and state harvest of returning hatchery adults, minimize impacts on wild fish, and ensure that enough hatchery fish return to meet broodstock needs, careful fisheries management is required. Since 2008, managers have used hatchery adults, PIT‐tagged as juveniles and detected at Lower Granite Dam, to generate adult abundance estimates. In season, these estimates inform state and tribal harvest shares and ensure that broodstock needs are met. Postseason, they provide smolt‐to‐adult survival and return rates. Since 2012, parentage‐based tagging (PBT) has provided an alternative method to estimate stock‐ and age‐specific returns at Lower Granite Dam, since returning hatchery adults sampled at Lower Granite Dam can be assigned to their parents. We compared stock‐specific abundance estimates between PIT‐ and PBT‐derived methodologies for return years 2016–2019. Across all years, PIT tag estimates accounted for 65% of the PBT‐based estimates at Lower Granite Dam across all age‐groups and release sites combined. This underrepresentation across all groups equated to 49,833 fish that were not accounted for in PIT tag abundance estimates. It is clear that PBT‐based estimates should aide in‐season harvest management and postseason run reconstruction to avoid the known bias of estimates from PIT tags, especially during years of low returns when increased accuracy is critical.

Assessing the Bonneville Power Administration’s Financial Vulnerability to Hydrologic Variability

Hydrologic variability can cause large swings in hydropower generation, inducing significant volatility in power sales. Dry years often result in low revenues that can threaten a hydropower supplier’s ability to meet its fixed costs, leading to budget shortfalls, lower credit ratings, higher interest rates, and, ultimately, higher rates. This is particularly true for suppliers in hydropower-dominated regions, such as the Bonneville power administration (BPA). The BPA strategy for managing its hydrologic financial risk is multilayered, involving cash reserves, a line of credit, and tariff adjustments. Yet, compared to its long-term energy contracts and debt service, BPA’s risk assessment is conducted on a short-term basis, thereby neglecting medium- and long-term temporal dynamics impacting their financial risk. This paper focuses on (1) evaluating BPA’s hydrologic financial risk; and (2) testing the effectiveness of BPA’s existing risk management strategy. Results suggest that BPA’s financial risk will grow substantially over the next 20 years as its risk management tools become increasingly inadequate, providing cautionary lessons for organizations in similarly hydrodominated systems despite the current use of common risk management tools.

Akihiro ametani: paving the way to power grid expansion through innovation in power system transient analysis [history

Akihiro Ametani was born in Nagasaki, Japan, on 14 February 1944. He received the B.Sc. and M.Sc. degrees from Doshisha University, Kyoto, Japan, in 1966 and 1968, respectively, and the Ph.D. degree from the University of Manchester, England, in 1973. He was with the Bonneville Power Administration (BPA), Portland, Oregon, from 1976 to 1981 and 1991 to 1992, where he contributed to the development of the Electromagnetic Transient Program (EMTP). He was a professor at Doshisha University from 1974 to 2014.

Valuation of operational flexibility: A case study of Bonneville power administration

Real options theory is applied to quantify the value of operational flexibility intrinsic to the hydropower generation capacity from the perspective of US federal power marketing administrations. The option value in this model arises from the tension between the increase in current sales revenue and the increase in the exposure of future power shortages. Using Bonneville Power Administration as an example, we apply the proposed valuation framework, analyze the optimal management policy for Bonneville Power Administration and investigate the potential impacts of Wind Vision 2050.

Synchrophasors Point the Way: The Emergence of a New Technology

I was a test engineer in the laboratory Division at the Bonneville Power Administration (BPA) in 1987. They needed someone to design, implement, and conduct a field test of a new computer-based device that measured the phasor equivalent of voltage and current sine waves. Knowing little about the technology and method, I went to a project meeting where Prof. Arun Phadke from Virginia Tech (VT), the principal investigator and primary creator of the concept, was leading the planning.

Scenario Reduction for Stochastic Day-Ahead Scheduling: A Mixed Autoencoder Based Time-Series Clustering Approach

Scenario based stochastic scheduling has drawn a tremendous amount of interests worldwide in tackling the uncertainty of renewable energy and accounting for risks. It is important to generate representative time-series scenarios of renewable energy, while keeping the dimensionality of the scenario set tractable. This article presents a mixed autoencoder based clustering approach to select a reduced scenario set from high-dimensional time series. In contrast to other techniques targeting on minimizing different probability distances, the proposed architecture accounts for the pattern recognition within a large set of scenarios. The effectiveness of the model is verified in the case studies, where the data sets from the Bonneville Power Administration and Elia are used. The numerical results show that the model outperforms the state of the art, in terms of statistical metrics and through empirical analysis.

Regulation as a force for hybrid organization: evidence from the Bonneville Power Administration (1980–2012)

PurposeThis case study highlights state-logic influence on hybrid organizations and institutionally complex environments through acts of regulation (and deregulation).Design/methodology/approachThis study presents a 30-year narrative case focused on the significant social achievements of the Bonneville Power Administration within the Northwest United States. It combines the analysis of historical documentation, annual reports issued by the organization and interviews with firm management to observe the wax and wane of regulatory influence through time.FindingsThe presented case suggests two ways regulation projects state-logic influence onto hybrid organizations. First, it imposes a “floor” level of baseline social activity that must be met despite pressure from market logic stakeholders. Second, it imposes formal administrative procedures that require interaction with, and often approval from, key social stakeholders. Administrative procedures provide a series of public forums used to promote additional social resource allocation in excess of baseline regulatory mandates.Research limitations/implicationsA narrative case covering a 30-year period will by necessity have to prioritize breadth of analysis over depth. This is a limitation of the analysis presented, but it also provides an opportunity to observe the oscillating impact of state and market-logic influence through time.Originality/valueThe study findings have several implications for the growing accounting literature on institutional complexity and hybrid organization. First, the authors highlight the ways regulation shapes institutionally complex spaces and, as a result, the hybrid organizations formed within those environments. Second, the exogenous nature of regulatory mandates indicate hybrid firms could emerge as both a voluntary and an involuntary adaptation to institutionally complex environments. Finally, this study highlights opportunities to further one’s understanding how state logics influence hybrid organizations through the study of state-owned enterprises (SOEs).

The Bonneville Power Administration Film Archives: Ecologies of Infrastructural Media from 1939 to the Present

The Bonneville Power Administration Film Archives: Ecologies of Infrastructural Media from 1939 to the Present

Grey System Theory Based Net Load Forecasting for High Renewable Penetrated Power Systems

Net-load is the imbalance between aggregated load and renewable generation. System operations like economic dispatch necessitate accurate forecasting of net-load. There has been significant progress in load and renewable generation forecasting, however, little research focus has been there on Net-Load Forecasting (NLF). Therefore, this paper proposes three Direct Grey index net-load forecasting models (DGM (1, 1), DGM (1, 2) and DGM (1, 3)) using Grey System Theory (GST). GST based models are suitable for accurate and fast very short-term (five minutes ahead) forecasting due to their momentum transfer behavior. Proposed NLF models are implemented for the Bonneville Power Administration (BPA) balancing area. Forecasts obtained from proposed models are compared with actual net load data and forecasts obtained from reference Artificial Neural Network (ANN) model. Comparison with actual net load shows that proposed models have strong potential for very short-term NLF. At very short time frames, net load shows a very high correlation with previous time steps data. Proposed models utilize such characteristics for forecasting, compared to continuous error reduction procedure in ANN. Continuous error reduction in a very short time frame can lead to under/overestimation of ANN weights and that lowers the forecasting accuracy. Proposed NLF models, especially DGM (1, 3) show superior performance over ANN.

Pursuing “Fit for Purpose”: An Industry Practioner's Experience With Roadmapping

Since 2006, the Bonneville Power Administration (BPA) Technology Innovation (TI) Office has applied roadmapping processes to help articulate the federal wholesale electric utility's technology research agenda. The TI Office team has evolved its tools, templates, and process steps considerably over the years in response to implementation challenges, shortcomings in deliverables derived from the process, and maturation opportunities enabled by data management and visualization software. Practioners within the TI Office have sustained their commitment to roadmapping during an era of significant evolution of the practice within industry and academia. The practice came out of product manufacturing environments in the 1960s, 1970s, and 1980s. Increasingly since the early 2000s, roadmapping has been applied outside this “traditional” realm, requiring trial, iteration, and customization of processes, tools, and deliverables to meet different requirements. The TI Office's use of roadmapping is one “nontraditional” variant because the agency is not in the business of generating profits by creating products in response to market demands. Instead, the TI Office has used roadmapping to communicate the agency's technology research, development, and demonstration agenda to the broader community—national laboratories, universities, manufacturers, etc.—so these external organizations can respond with proposals delivering products with the desired technology characteristics. These products, in turn, enable BPA to continue to fulfill its public service mission as an electric utility and government entity in providing safe, cost-effective, and reliable electricity throughout the Pacific Northwest region of the United States.

Electromagnetic Transients Program: History and Future

This paper describes the background of Electromagnetic Transients Program (EMTP) and reviews original EMTP development at Bonneville Power Administration of the US Department of Energy starting from 1966 to 1984. Then, the paper briefly explains the development and present situation of four EMTPs, that is, ATP‐EMTP, EMTP‐RV (now EMTP®), EMTDC/PSCAD and XTAP. In addition, the line/cable parameters and the modeling methods prepared in EMTPs are described, and related problems are explained. Finally, the recent trend and future of electromagnetic transient simulations are discussed. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

Multi-stage fault warning for large electric grids using anomaly detection and machine learning

In the monitoring of a complex electric grid, it is of paramount importance to provide operators with early warnings of anomalies detected on the network, along with a precise classification and diagnosis of the specific fault type. In this paper, we propose a novel multi-stage early warning system prototype for electric grid fault detection, classification, subgroup discovery, and visualization. In the first stage, a computationally efficient anomaly detection method based on quartiles detects the presence of a fault in real time. In the second stage, the fault is classified into one of nine pre-defined disaster scenarios. The time series data are first mapped to highly discriminative features by applying dimensionality reduction based on temporal autocorrelation. The features are then mapped through one of three classification techniques: support vector machine, random forest, and artificial neural network. Finally in the third stage, intra-class clustering based on dynamic time warping is used to characterize the fault with further granularity. Results on the Bonneville Power Administration electric grid data show that i) the proposed anomaly detector is both fast and accurate; ii) dimensionality reduction leads to dramatic improvement in classification accuracy and speed; iii) the random forest method offers the most accurate, consistent, and robust fault classification; and iv) time series within a given class naturally separate into five distinct clusters which correspond closely to the geographical distribution of electric grid buses.

Data-Driven Distributionally Robust Stochastic Control of Energy Storage for Wind Power Ramp Management Using the Wasserstein Metric

The integration of wind energy into the power grid is challenging because of its variability, which causes high ramp events that may threaten the reliability and efficiency of power systems. In this paper, we propose a novel distributionally robust solution to wind power ramp management using energy storage. The proposed storage operation strategy minimizes the expected ramp penalty under the worst-case wind power ramp distribution in the Wasserstein ambiguity set, a statistical ball centered at an empirical distribution obtained from historical data. Thus, the resulting distributionally robust control policy presents a robust ramp management performance even when the future wind power ramp distribution deviates from the empirical distribution, unlike the standard stochastic optimal control method. For a tractable numerical solution, a duality-based dynamic programming algorithm is designed with a piecewise linear approximation of the optimal value function. The performance and utility of the proposed method are demonstrated and analyzed through case studies using the wind power data in the Bonneville Power Administration area for the year 2018.

Barriers to energy trading on the Pacific Intertie

This paper explains artificial barriers to inter-regional trade on the Pacific Intertie erected by the Bonneville Power Administration (BPA) in markets for both long-term and short-term transactions, intended to protect incumbent transmission contract customers of BPA from potential competitors and increase the profits earned by those incumbents, by increasing price spreads between spot markets in the Pacific Northwest and California and interfering with market efficiency and market integration on the West Coast. Both market participants and policy-makers should work to dismantle these barriers.

Sequence Generative Adversarial Networks for Wind Power Scenario Generation

With the rapid increase in distributed wind generation, considerable efforts have been devoted to the microgrid day-ahead scheduling. The effectiveness of those methods will highly depend on the selection of the uncertainty sets. We propose a distribution-free approach for wind power scenario generation, using sequence generative adversarial networks. To capture the temporal correlation, the model adopts the long short-term memory architecture and uses generative adversarial networks coupled with reinforcement learning, which, in contrast to the existing methods, avoids manual labeling and captures the complex dynamics of the weather. We conduct case studies based on the data from the Bonneville Power Administration and the National Renewable Energy Laboratory, and show that the generated scenarios can better characterize the variability of wind power and reduce the risk of uncertainties, compared with those produced by Gaussian distribution, vanilla long short-term memory, and multivariate kernel density estimation. Moreover, the proposed method achieves better performance when applied to the day-ahead scheduling of microgrids.

A Systems Approach to Project Stakeholder Management: Fuzzy Cognitive Map Modeling

Projects that make effective use of project stakeholder management (PSM) tend to run smoothly and be successful because stakeholders understand and agree with the project approaches and outcomes. Projects with ineffective stakeholder management, on the other hand, frequently experience delays and cost overruns or may even be terminated. To date, project teams have limited methodological support for PSM: Existing methods are dominantly static and internally focused, making it difficult to manage so-called external stakeholders, who are not under the authority of the project manager. This work aims to improve PSM practice by closing the methodological gap. We developed a novel decision-support methodology, based on Fuzzy Cognitive Map (FCM) modeling that leverages stakeholders’ public comments to anticipate the project’s impacts on them and to make conflicts between stakeholder interests and project objectives transparent. A demonstration of the method is provided using a single case—namely, a longitudinal case study at Bonneville Power Administration (BPA), a federal agency that provides power to the Pacific Northwest.

Economic and environmental implications of different approaches to hedge against wind production uncertainty in two-settlement electricity markets: A PJM case study

This paper examines the economic and environmental outcomes of four two-settlement electricity market clearing designs. The first design corresponds to a Deterministic Market Clearing (DMC) similar to the mechanism currently used in organized wholesale electricity markets in the United States. The other three designs account for the day-ahead (DA) wind power production uncertainty into the DA market mechanisms either implicitly or explicitly. An Augmented Deterministic Market Clearing (ADMC) design introduces DA ramp-capability products. These products ensure adequate and ramp-feasible electricity generation capacity commitments in the DA stage to cope with the real-time realization of wind power generation. A Hybrid Deterministic Market clearing (HDMC) design augments ADMC by explicitly integrating a characterization of wind power production uncertainty into the residual unit commitment (RUC) process, which is run after the DA market is closed, using stochastic programming. The last design, referred to as stochastic market clearing (SMC), uses stochastic optimization to explicitly account for wind power production uncertainty in the DA market clearing mechanisms (i.e. DA unit commitment and economic dispatch).The four market clearing designs are assessed by simulating the electricity market operations of a test system and comparing their results in terms of operating costs, prices, costs and revenues of different types of producers, consumer's payments, integration of wind power, and air emissions. The test system has 12% of the capacity of PJM's fossil-fired power generation fleet, and uses data on coincident demand and wind power production from the Bonneville Power Administration (BPA) system during years 2010–2014. The simulations are performed hourly for a whole year.Results show that SMC is superior as its costs reductions are more than two times the improvements attained by ADMC and HDMC. Also, SMC results in electricity prices that are better aligned with operation costs, cuts the spread between the day-ahead and real-time prices by >40%, reduces out-of-market short-term revenue sufficiency payments by 58%, reduces CO2 emissions by 3.52%, and decreases power plants' cycling. HDMC is a distant second-best market design. Relative to DMC, it achieves a reduction in total costs that is less than half the reduction achieved by SMC, a reduction in out-of-market payments that is 80% of the reduction attained by SMC, and an increase in wind power integration that is <10% the improvement obtained under SMC.