California Independent System Operator Research Articles

Electricity-market price and nuclear power plant shutdown: Evidence from California

Japan׳s Fukushima nuclear disaster, triggered by the March 11, 2011 earthquake, has led to calls for shutting down existing nuclear plants. To maintain resource adequacy for a grid׳s reliable operation, one option is to expand conventional generation, whose marginal unit is typically fueled by natural-gas. Two timely and relevant questions thus arise for a deregulated wholesale electricity market: (1) what is the likely price increase due to a nuclear plant shutdown? and (2) what can be done to mitigate the price increase? To answer these questions, we perform a regression analysis of a large sample of hourly real-time electricity-market price data from the California Independent System Operator (CAISO) for the 33-month sample period of April 2010–December 2012. Our analysis indicates that the 2013 shutdown of the state׳s San Onofre plant raised the CAISO real-time hourly market prices by $6/MWH to $9/MWH, and that the price increases could have been offset by a combination of demand reduction, increasing solar generation, and increasing wind generation.

Coordinating plug-in electric vehicle charging with electric grid: Valley filling and target load following

Plug-in electric vehicles (PEVs) shift energy consumption from petroleum to electricity for the personal transportation sector. This work proposes a decentralized charging protocol for PEVs with grid operators updating the cost signal. Each PEV calculates its own optimal charging profile only once based on the cost signal, after it is plugged in, and sends the result back to the grid operators. Grid operators only need to aggregate charging profiles and update the load and cost. The existing PEV characteristics, national household travel survey (NHTS), California Independent System Operator (CAISO) demand, and estimates for future renewable generation in California are used to simulate PEV operation, PEV charging profiles, grid demand, and grid net load (demand minus renewable). Results show the proposed protocol has good performance for overnight net load valley filling if the costs to be minimized are proportional to the net load. Annual results are shown in terms of overnight load variation and comparisons are made with grid level valley filling results. Further, a target load can be approached in the same manner by using the gap between current load and the target load as the cost. The communication effort involved is quite modest.

On-line voltage security assessment and control

SUMMARY Voltage stability analysis and control are key applications for maintaining and enhancing the voltage security of bulk power systems. In today's control center, as power systems become more stressed and the penetration of renewable energies increases, system operators need to analyze voltage security of the systems based on actual operating conditions, contingency and power transactions of the system. In this paper, the model used to represent a power system losing stability is described. It is shown that the voltage collapse point is reached at saddle-node bifurcation or a structure-induced bifurcation. Computation of P–V curves showing system load margin to voltage collapse, system load margin to voltage-limit violation and system load margin to thermal-limit violation is presented. The architecture and example results of the Voltage Stability Analysis and Enhancement system at the California Independent System Operator implementing this research are shown. The continuation power flow engine at the core of this work is presented in detail, and study results are shown. Other key functions, contingency selection (insecure or critical) and ranking, preventive control to increase the load margin of insecure contingencies such that it is secure and enhancement control to increase the load margin of critical contingencies, are presented. Finally, a novel method to enhance the continuation power flow in handling uncertainty of renewable energy is shown. Copyright © 2014 John Wiley & Sons, Ltd.

Ensemble re-forecasting methods for enhanced power load prediction

Electric load forecasting is a key element for management and operation of the electric grid. In this study we introduce ensemble re-forecast methods that take an initial forecast and produce a better prediction by extracting information from the structured errors. The models in the ensemble rely upon the real-time information obtained from load measurements and estimates over a state-wide domain. The weights in the ensemble are optimized in three different ways based on global, hourly, and weekly performance of the models. The proposed methodology is applied to predict hour-ahead market (HAM) and day-ahead market (DAM) load for California Independent System Operator (CAISO) and Electric Reliability Council of Texas (ERCOT) respectively. Proposed models showed consistent performance enhancements for all the cases. HAM predictions show an improvement of 47% and 36% in terms of Mean Absolute Percentage Error over the forecasts provided by CAISO and ERCOT. For DAM, the improvements are 34% for CAISO and 47% for ERCOT. Temporal analysis comparing the internal forecast produced by the ISOs and re-forecasts shows significant improvement during off-peak hours and small improvement for on-peak hours. Results validate the potential of the proposed methodology to enhance the forecast accuracy, independent of load profile or forecast horizon.

Consolidated Dynamic Pricing of Power System Regulation

The cost and quality of regulation have been brought to the forefront of power system operations by renewable variability. We propose a regulation pricing methodology based on an idea of Berger and Schweppe: in the same manner that locational marginal prices are derived from the dual of economic dispatch, let regulation prices be the dual multipliers or costate of an optimal control problem. By specializing to the linear quadratic regulator, we formulate a regulation pricing policy, which in turn allows us to derive statistically correct payments such as imbalance fees. We then construct a Vickrey-Clarke-Groves mechanism to induce selfish agents to honestly report their private valuations and costs of regulation. Semidefinite programming descriptions of the formulation can be embedded within standard convex economic dispatch constraint sets, enabling cooptimization of base load and regulation pricing. The approach mechanistically produces prices for any dynamic scenario, and can therefore be used to consolidate diverse regulation services. We demonstrate this feature by combining traditional frequency regulation, area control error, and the California Independent System Operator's Flexible Ramping Product in an example.

A comparative study among fossil fuel power plants in PJM and California ISO by DEA environmental assessment

This study compares among fossil fuel power plants in PJM and California ISO by their unified (operational and environmental) performance. DEA (Data Envelopment Analysis) is used as a methodology. For comparative analysis, DEA incorporates strategic concepts such as natural and managerial disposability into the computational process. This study explores both how to measure Returns to Scale (RTS) under natural disposability and how to measure Damages to Scale (DTS) under managerial disposability. This empirical study obtains two implications on US energy policy. One of the two policy implications is that California ISO outperforms PJM in terms of the three unified efficiency measures. The result implies that strict regulation on undesirable outputs, as found in California, is important in enhancing the performance of US fossil fuel power plants. Thus, it is necessary for federal and local governments to regulate the fossil fuel power plants under the strict implementation of environmental protection. Under such a policy direction, it is possible for US fossil fuel power plants to attain economic prosperity (by enhancing their operational efficiencies) and to satisfy environmental regulation (by enhancing their environmental efficiencies). The other policy implication is that coal-fired and gas-fired power plants in PJM and California ISO need to reduce their operational sizes or introduce technology innovation on desirable and undesirable outputs and/or new management for environmental protection within their operations. Meanwhile, oil-fired power plants may increase their operational sizes if they can introduce technology innovation and new management on undesirable outputs.

Envelope Modeling of Renewable Resource Variability and Capacity

One of the greatest challenges to integrating renewable energy resources into power operations is the variability in electricity output from these resources. The electricity system works essentially as a supply chain with a nearly zero tolerance for delay and storage resources that remain very limited despite rapid growth. Yet, because no consistent methodology exists to quantify the capacity contribution and system impact of variable generating resources such as wind and solar, these resources have yet to be effectively evaluated, managed, or compensated. To enable analysis of both renewable and traditional energy sources within the system, we adapt an envelope-based modeling method that is inspired by Network Calculus (NetCal) for deterministic queuing systems from the field of telecommunications engineering. The variability of electricity supply and demand can be described by upper and lower envelopes and their corresponding Legendre conjugates, which not only quantify variability across different time scales, but also characterize the tradeoff between any assigned capacity value and its corresponding Quality-of-Service (QoS) measures of performance. In particular, the QoS measure leads to an intuitive interpretation through storage resources. Envelope modeling leads to the definition of two QoS-based capacity metrics -- Guaranteed Capacity and Best-Effort Capacity -- whose conceptual and numerical properties we analyze and compare against existing capacity metrics. For illustration, the proposed methods are applied to data from the California Independent System Operator (CAISO). We also explicitly quantify the capacity contribution (via the notion of Best-Effort Capacity) of wind during peak hours and its negative system impact at night. The same envelope characterization further demonstrates the capacity value of storage resources.

Energy Imbalance Market Benefits in the West: A Case Study of PacifiCorp and CAISO

An energy imbalance market between PacifiCorp and the California Independent System Operator (CAISO) would bring benefits of $21 million to $129 million for the year 2017, an analysis suggests. Preliminary cost estimates of setting up the EIM range from $3 million to $6 million, with an estimated annual cost of $2 million to $5 million. This suggests that a two-party EIM provides a low-cost, low-risk means of achieving operational savings and enabling greater penetration of variable energy resources.

Using ICT-Controlled Plug-in Electric Vehicles to Supply Grid Regulation in California at Different Renewable Integration Levels

The purpose of this paper is to quantify the potential for plug-in electric vehicles (PEVs) to meet operating reserve requirements associated with increased deployment of wind and solar generation. The paper advances prior PEV estimates in three key ways. First, we employ easily implementable scheduling strategies with very low centralized computing requirements. Second, we estimate PEV availability based on data sampled from the National Household Travel Survey (NHTS). Third, we predict regulation demand on a per minute basis using published models from the California ISO for 20% and 33% renewable electricity supply. Our key results are as follows: First, the amount of regulation up and regulation down energy delivered by PEVs can be balanced by using a hybrid of two scheduling strategies. Second, the percentage of regulation energy that can be delivered with PEVs is always significantly higher than the percentage of conventional regulation power capacity that is deferred by PEVs. Third, regulation up is harder to satisfy with PEVs than regulation down. Fourth, the scheduling strategies we employ here have a limited impact on load following requirements. Our results indicate that 3 million PEVs could satisfy a significant portion-but not all-of the regulation energy and capacity requirements that are anticipated in California in 2020.

Mixed integer non-linear programming and Artificial Neural Network based approach to ancillary services dispatch in competitive electricity markets

Ancillary services represent a good business opportunity that must be considered by market players. This paper presents a new methodology for ancillary services market dispatch. The method considers the bids submitted to the market and includes a market clearing mechanism based on deterministic optimization.An Artificial Neural Network is used for day-ahead prediction of Regulation Down, regulation-up, Spin Reserve and Non-Spin Reserve requirements.Two test cases based on California Independent System Operator data concerning dispatch of Regulation Down, Regulation Up, Spin Reserve and Non-Spin Reserve services are included in this paper to illustrate the application of the proposed method: (1) dispatch considering simple bids; (2) dispatch considering complex bids.

Incorporating Demand Response with Load Shifting into Stochastic Unit Commitment

High costs for fossil fuels and increasing installations of intermittent energy sources are imposing major challenges on power grid management. Uncertainty in generation and demand for electric energy require flexible generation capacity and stochastic optimization of generation schedules. Emerging smart grid technology is one mean to increase efficiency in power generation and to mitigate effects of increasing uncertainty. In this paper, we analyze the potential of demand side resources (DSRs) that can be dispatched to reduce load at peak times. We present a stochastic dynamic programming model for the unit commitment problem in a day ahead market and include dispatch decisions for DSRs. Unlike previous research, we model the stochastic load shifting effect to previous and subsequent periods that must be taken into account when making dispatch decisions. We also present a solution algorithm that combines an approximate dynamic programming algorithm with stochastic progressive hedging to solve the problem. We prove convergence results for our algorithms and derive a lower bound on the optimal solution. Using data from the California Independent System Operator region, we show that our approach can solve real world instances in reasonable time. We conduct an extensive numerical study using real world data to show that substantial savings can be achieved by utilizing DSRs.

The Resilient Organization

Both academics and practitioners have recently discovered resilience as a core topic of interest. Resilience is widely viewed as a potential solution to the challenges posed by crises and disasters. The promise of resilience is an organization or society that absorbs shocks and ‘bounces back’ after a disturbance. While the idea of resilience is increasingly popular, empirical research on resilient organizations is actually quite rare. This article explores whether a relation exists between organizational characteristics, processes and resilience. Building on the insights of high reliability theory and crisis research, it probes this relation in two organizations that experienced deep crises: the California Independent System Operator (CAISO) and National Aeronautics and Space Agency (NASA).

Characterization and Impact of Extreme Forecast Errors on Power Systems

Extreme events in the electrical power system, caused by the load and wind forecasting errors, can impact the power system infrastructure via two main avenues. The first avenue is a sudden and significant power unbalance exceeding reasonable operating reserve capacity. The second is a sudden increase of power flows on the system critical paths causing transmission violations. The challenge in managing these system unbalances is more significant for a stand-alone balancing area operation. The consolidation of balancing authorities into a single balancing area can offset the operating reserve problem, but this strategy enhances incremental power flows on the transmission interfaces, potentially leading to more unpredictable transmission congestion. This article evaluates the expectation of occurrence of extreme events due to forecast error extremes using California Independent System Operator and Bonneville Power Administration data. Having this type of information, independent system operators and operating utilities could be better prepared to address the extreme events by exploring alternative reserve options, such as wide-area control coordination, new operating procedures, and remedial actions.

Optimal Charge Control of Plug-In Hybrid Electric Vehicles in Deregulated Electricity Markets

Plug-in hybrid electric vehicles are a midterm solution to reduce the transportation sector's dependency on oil. However, if implemented in a large scale without control, peak load increases significantly and the grid may be overloaded. Two algorithms to address this problem are proposed and analyzed. Both are based on a forecast of future electricity prices and use dynamic programming to find the economically optimal solution for the vehicle owner. The first optimizes the charging time and energy flows. It reduces daily electricity cost substantially without increasing battery degradation. The latter also takes into account vehicle to grid support as a means of generating additional profits by participating in ancillary service markets. Constraints caused by vehicle utilization as well as technical limitations are taken into account. An analysis, based on data of the California independent system operator, indicates that smart charge timing reduces daily electricity costs for driving from $0.43 to $0.2. Provision of regulating power substantially improves plug-in hybrid electric vehicle economics and the daily profits amount to $1.71, including the cost of driving.

Contributors

Contributors

A Monte Carlo approach to generator portfolio planning and carbon emissions assessments of systems with large penetrations of variable renewables

A new generator portfolio planning model is described that is capable of quantifying the carbon emissions associated with systems that include very high penetrations of variable renewables. The model combines a deterministic renewable portfolio planning module with a Monte Carlo simulation of system operation that determines the expected least-cost dispatch from each technology, the necessary reserve capacity, and the expected carbon emissions at each hour. Each system is designed to meet a maximum loss of load expectation requirement of 1 day in 10 years. The present study includes wind, centralized solar thermal, and rooftop photovoltaics, as well as hydroelectric, geothermal, and natural gas plants. The portfolios produced by the model take advantage of the aggregation of variable generators at multiple geographically disperse sites and the incorporation of meteorological and load forecasts. Results are presented from a model run of the continuous two-year period, 2005–2006 in the California ISO operating area. A low-carbon portfolio is produced for this system that is capable of achieving an 80% reduction in electric power sector carbon emissions from 2005 levels and supplying over 99% of the annual delivered load with non-carbon sources. A portfolio is also built for a projected 2050 system, which is capable of providing 96% of the delivered electricity from non-carbon sources, despite a projected doubling of the 2005 system peak load. The results suggest that further reductions in carbon emissions may be achieved with emerging technologies that can reliably provide large capacities without necessarily providing positive net annual energy generation. These technologies may include demand response, vehicle-to-grid systems, and large-scale energy storage.

Decision making in dynamic stochastic Cournot games

In this paper, the Cournot competition is modeled as a stochastic dynamic game. In the proposed model, a stochastic market price function and stochastic dynamic decision functions of the rivals are considered. Since the optimal decision of a player needs the estimation of the unknown parameters of the market and rivals’ decisions, a combined estimation–optimization algorithm for decision making is proposed. The history of the rivals’ output quantities (supplies) and the market clearing price (MCP) are the only available information to the players. The convergence of the algorithm (for both estimation and decision making processes) is discussed. In addition, the stability conditions of the equilibrium points are analyzed using the converse Lyapunov theorem. Through the case studies, which are performed based on the California Independent System Operator (CA-ISO) historical public data, the theoretical results and the applicability of the proposed method are verified. Moreover, a comparative study among the agents using the proposed method, naïve expectation and adaptive expectation in the market is performed to show the effectiveness and applicability of the proposed method.

Application of adaptive wavelet neural network to forecast operating reserve requirements in forward ancillary services market

Operating reserve (OR) is a major portion of ancillary services (AS) in a competitive electricity market and need to be procured by independent system operator (ISO) , to achieve a high degree of power system reliability and security, following the major generation and transmission contingencies. Several ISOs have adopted deterministic methods to assess the OR requirements, however, such methods do not explicitly consider the unforeseen load swings and the probability of equipment outages. This paper proposes an adaptive wavelet neural network (AWNN) based two-stage approach to forecast OR requirements for both day-ahead and hour-ahead AS market in the California ISO (CAISO) controlled grid. The AWNN is a new class of feed-forward neural network with continuous wavelet function as the hidden layer node's activation function. The forecasting results for winter and summer seasons of the year 2007 are presented and compared with those obtained by feed-forward multi-layer perceptron neural network (MLPNN). It is found that AWNN based proposed method outperforms the MLPNN model.

Modeling environmental policies in probabilistic generation system planning reserve margin assessment

This paper provides an overview of the California Independent System Operator (CAISO)’s efforts to incorporate environmental policies regarding electric power industry into its probabilistic planning reserve margin studies. Currently, load serving entities such as Investor Own Utilities (IOUs) in the State of California, with its retail load under the jurisdiction of the California Public Utility Commission (CPUC), are required to procure reserve level at 15–17%. This requirement is deterministic in nature and does not include the potential impact of additional reserve requirement due to environmental policies. The CAISO, which controls 80% of the state’s electrical load, is investigating the use of probabilistic planning for planning reserve margin (PRM) of the load serving entities under its operational control. The paper provides an overview of CAISO study results for PRM under various load and resource scenarios, increase in generators’ forced outage rate, and impacts due to California’s environmental policies regarding electric supply sector, such as its once-through cooling (OTC) draft policy and Renewable Portfolio Standards (RPS) targets.

Forecasting of Short-Term Electric Load Using Application of Wavelets with Feed-Forward Neural Networks

An accurate and efficient Short-Term Load Forecasting (STLF) plays a vital role for economic operational planning of both regulated and restructured power systems. This paper presents the STLF models by introducing wavelet transforms, in different ways, with feed-forward Neural Networks (NNs). First, a wavelet-based NN is modeled, where the forecasting has been accomplished in three-stages and the wavelet technique is employed to decompose/reconstruct the original signals and non-linearity of the decomposed signals. Second, an Adaptive Wavelet Neural Network (AWNN) is modeled, which is a new class of NN with continuous wavelet function as the hidden layer node's activation function. Unlike the first model, AWNN does not externally decompose/ reconstruct the original signals and, therefore, this model deals with the problem related to loss of high frequency information that might occur in the wavelet-based NN model. AWNN continuously updates the wavelet parameters (translation and dilation) and layer weights through a back-propagation training algorithm as in classical NNs. The performances of these two models are compared with Multi-Layer Perceptron NN (MLPNN) with the application of day-ahead and hour-ahead load forecasting in the California electricity market. The results are also compared with California Independent System Operator (CAISO)'s forecasted system loads. It is found that due to faster and accurate training capability, AWNN outperforms the MLPNN, wavelet-based NN and CAISO load forecasts.