Generation Assets Research Articles

New Design of a Supervised Energy Disaggregation Model Based on the Deep Neural Network for a Smart Grid

Energy management technology of demand-side is a key process of the smart grid that helps achieve a more efficient use of generation assets by reducing the energy demand of users during peak loads. In the context of a smart grid and smart metering, this paper proposes a hybrid model of energy disaggregation through deep feature learning for non-intrusive load monitoring to classify home appliances based on the information of main meters. In addition, a deep neural model of supervised energy disaggregation with a high accuracy for giving awareness to end users and generating detailed feedback from demand-side with no need for expensive smart outlet sensors was introduced. A new functional API model of deep learning (DL) based on energy disaggregation was designed by combining a one-dimensional convolutional neural network and recurrent neural network (1D CNN-RNN). The proposed model was trained on Google Colab’s Tesla graphics processing unit (GPU) using Keras. The residential energy disaggregation dataset was used for real households and was implemented in Tensorflow backend. Three different disaggregation methods were compared, namely the convolutional neural network, 1D CNN-RNN, and long short-term memory. The results showed that energy can be disaggregated from the metrics very accurately using the proposed 1D CNN-RNN model. Finally, as a work in progress, we introduced the DL on the Edge for Fog Computing non-intrusive load monitoring (NILM) on a low-cost embedded board using a state-of-the-art inference library called uTensor that can support any Mbed enabled board with no need for the DL API of web services and internet connectivity.

Governance structures and efficiency in the U.S. electricity sector after the market restructuring and deregulation

In some U.S. states that have undergone electricity market restructuring and deregulation, investor-owned utilities (IOUs) were required to divest the majority of their generation assets in order to purchase power from merchant generators, independent power producers, and power marketers competing in the new wholesale market. This paper examines the effect of the governance structures used to purchase power in the wholesale market, the bilateral forward contracts and market transactions, on the technical efficiency of IOUs during the post-divestiture period. Using a two-stage empirical strategy (the non-parametric data envelopment analysis and the difference-in-differences regression approach), I analyze the performance of 152 distribution utilities in the U.S. from 1994 to 2015. The results show that while the use of contracts has no significant effect, the use of market transactions has a significant negative effect on IOU technical efficiency. Trading arrangements in the restructured wholesale markets that rely on the concept of competition generate transaction costs that make it more costly for the utility to use market transactions rather than other alternatives. Thus, market transactions fall short to promote the efficiency purpose of IOUs as an economic organization in this policy-induced market and the adverse effect persists until 20 years after the divestiture.

Finding Order in Chaos: Grappling with the Policy and Politics of Sustainable Energy Transition

Finding Order in Chaos: Grappling with the Policy and Politics of Sustainable Energy Transition

System state model based multi‐period robust generation, transmission, and demand side resource co‐optimisation planning

This study discusses a multi-period co-optimised generation and transmission expansion planning (GTEP) problem while considering a proliferation of demand side resources (DSR). Uncertain renewable energy variations and load fluctuations in the long-term planning horizon are addressed, and a system state model derived via k-means clustering algorithm is adopted to capture temporal operation features. The problem is formulated as a two-stage robust optimisation model with mixed-integer recourse, in which annual investment decisions of generation, transmission, and DSR assets are determined in the first stage and short-term operation decisions of individual system states are made in the second stage. In recognising that considering DSR deployment and the system state model brings significant computational complexity, an extended column-and-constraint-generation algorithm is adopted to effectively solve the proposed planning problem. Numerical studies show that integrating DSRs into multi-period GTEP could effectively postpone or even avoid expensive generation/transmission investment in the planning stage, and improve economic efficiency in the operation stage.

Business model analysis for the interaction between smart grid and mobile network operators

This paper explores the interaction between mobile network operators (MNOs) and smart grid operators from an economic and strategic point of view. We first present the emerging new actors in the evolving smart grid value chain. We then focus on the role of MNOs and study the different roles they could play in this value chain, not only as connectivity providers for connecting consumers to distributed energy storage and production facilities and for ensuring demand-response, but also with regards to their expertise as Internet of Things (IoT) players and their energy generation and storage assets. Based on this ecosystem analysis, we build three business models canvasses for the MNO as a Demand Response (DR) aggregator, as a Virtual Power Plant (VPP) and as an enhanced VPP where it also acts as a prosumer with generation/storage facilities. We show that the first two services fit very well with the European context, while the third is well suited for sub-Saharan African countries.

Predictive Risk Analytics for Weather-Resilient Operation of Electric Power Systems

Day-to-day operation of the electricity grid generation, transmission, and distribution is environmentally driven and closely dependent on evolving weather patterns. This paper introduces several new weather-driven analytics for accurate spatial–temporal electricity generation forecasts, asset health and reliability assessment, probabilistic load forecasts, and electricity market simulations. A new risk metric is suggested, which accounts for the weather hazards, grid vulnerability, and financial consequences in the face of changing weather patterns and associated meteorological predictions over time. New mitigation formulations for power system topology control through transmission line switching for fast and timely recovery of the weather-caused electricity outages are suggested. The proposed decision support tool enables the operators to predictively evaluate the high-risk weather threats and consequently plan on how to safeguard the grid when exposed to forecasted weather-driven incidents. The efficiency of the proposed toolset is illustrated by application to a part of the IEEE 73-Bus test system.

Regulatory Framework for Long-Term Demand Forecasting and Power Procurement Planning

Electricity supply chain is a capital intense activity requiring a long gestation period for setting up generation, transmission and distribution assets. A robust and stable regulatory and policy framework provides adequate incentive for investment in such capacities. Further, availability of adequate generation and transmission capacities in future needs to be ensured for reliable operation of the power system. A reliable electricity demand projection can not only help the distribution utilities to plan power procurement in advance, but also assist in creation of transmission infrastructure to ensure reliable electricity supply to consumers. Cost of power procurement is the most important element of the Aggregate Revenue Requirement (ARR) of distribution utilities and accounts for about three-fourth of the final consumer tariff. Power procurement plans and contracts typically have a long-term horizon and, therefore, need to be worked out well in advance, based on a reliable and dependable forecast. A national-level exercise for long-term electricity demand estimation is carried out by Central Electricity Authority (CEA), which projects state-wise long-term peak demand for electricity. The Electric Power Survey (EPS), published every five years by CEA, projects electricity demand using partial end use method, which is now supplemented with econometric estimates. Our analysis reveals that there is a deviation of up to 25 percent between the electricity demand projections in the EPS Reports and the actual demand of electricity. This mismatch can be attributed to quality of data as well as methodological shortcomings. Application of a standardised approach across states which vary in terms of economic development, consumption profile, agroclimatic conditions, etc. may also explain the range of deviations in demand projections across states. At the state level, the respective distribution (DISCOMs) and transmission licensees (TRANSCOs) also undertake such an exercise. The objectives, forecast horizon and methodology adopted for such forecasts vary widely. The objectives of some forecasts include transmission capacity planning, power procurement planning and/or regulatory compliance for tariff filings. While licensees in a few states revise their estimates annually, others undertake this activity every five years or more. CEA’s projections are adopted for transmission as well as generation planning in some states. A holistic sectoral approach combining long-term electricity demand forecast with power procurement would not only help utilities economise their overall cost of power procurement but also alleviate supply risks, especially with the growing share of renewable energy sources. We analysed historical demand projections and relevant regulations of selected states – Andhra Pradesh, Assam, Delhi, Gujarat, Madhya Pradesh, Maharashtra, Odisha, Punjab and Uttar Pradesh. Whereas some of the State Electricity Regulatory Commissions (SERCs) have issued specific regulations for demand forecasting and power procurement planning, others have covered only short-term aspects through tariff regulations. The prevailing regulatory framework of the sector across many states does not address the matter of long-term demand forecasting and power procurement planning in an integrated manner. Based on our analysis of different states and some international practices, we identified key aspects that should be considered in formulation of regulations for long-term demand forecasting and power procurement planning. SERCs can consider these key aspects, with appropriate modifications if necessary, while framing model regulations for long-term demand forecasting and power procurement planning.

Integrated System of Predictive Maintenance and Operation of Eletronorte Based on Expert System

Maintenance based on condition and diagnosis technology has a relevant function in Power Electrical System's reliability. This paper proposes the methodological development and computational implementation of a system capable of failure diagnosis on electrical generation assets in order to aid the maintenance crew and the operation at the Coaracy Nunes hydropower plant, located in the Amapa state, in the decision making related with executions of operation and maintenance actions to minimize the incipient issues before they actually occur. The proposed methodology is based in the OSA-CBM (Open System Architecture for Condition-Based Maintenance) model to structure the system’s knowledge base, in addition to implement fuzzy production rules to diagnosis of more complex failures. A case study was realized at the referred plant, showing in the results that the proposed system has a big potential on failure detection. Predictive maintenance and pattern recognition are technological enablers of Industry 4.0.

Turn Down for What? The Economic Value of Operational Flexibility in Electricity Markets

This paper explores how enhanced flexibility through lower minimum load levels (also called “turndown” limits) for coal- and natural-gas-fired power plants can impact operations and profitability. Using soft-linked capacity planning and unit commitment models, this research examines the economic value of this flexibility and variation across different regions, generation asset types, and market environments (e.g., level of renewable deployment). Model results suggest that lower minimum loads are most valuable for mid-merit units with intermediate capacity factors, policy conditions that rearrange the merit order (e.g., carbon pricing), regions and scenarios without seasonal layup of assets, and units that are more flexible than generators with otherwise similar characteristics. Lower minimum loads can increase operating hours, increase net revenues, and decrease startups and long-run maintenance costs. The scenarios in this analysis show that plant flexibility affects operations more than annual profitability, which may have a higher system value for integrating renewables than a private value for the flexible unit. For many plants, the modest changes in net revenues with lower minimums mean that the focus will likely be on making operational modifications to increase flexibility rather than on capital investments. Additionally, this analysis quantifies how revenues from ancillary services markets can boost the value of lower minimum loads, especially for efficient natural gas units, but total market size and depth are uncertain.

The impact of concentrated solar power in electric power systems: A Chilean case study

This paper presents a study about the impacts of the integration of concentrated solar power (CSP) with thermal energy storage (TES) in electric power systems. The main tool for this study is a comprehensive long-term power system capacity expansion planning model that integrates a specific module to represent the operation of CSP-TES power plants. The model determines the optimal investments on generation and transmission assets over a twenty-year planning horizon, ranging from 2018 until 2037, and employs projections for the various parameters involved (e.g. load growth, capital costs for the different generation technologies, fuels costs). One of the main features of the model is its ability to capture the hourly operational dynamics of the system through the consideration of multiple representative days for each of its investment periods. This feature allows a better understanding of the role of CSP-TES as a significant provider of flexibility to support a high penetration of variable renewable energy sources, as compared with traditional planning models based on load blocks. The model is applied to a case study for the Chilean electricity system. In order to study the impacts of CSP-TES, various scenarios of future capital costs and carbon tax levels are defined and analyzed for two market dominant CSP-TES technologies. The results show that for low CSP-TES capital costs, or high carbon taxes, the integration of CSP-TES in the system is significant towards year 2037, potentially reaching about one third of the total dispatched energy in the Chilean electric power system, yielding important operational, economic, and environmental benefits.

Fuzzy Portfolio Optimization for Power Generation Assets

Fuzzy sets theory is proposed as an alternative to the probabilistic approach for assessing portfolios of power plants, in order to capture the complex reality of decision-making processes. This paper presents different fuzzy portfolio selection models, where the rate of returns as well as the investor's aspiration levels of portfolio return and risk are regarded as fuzzy variables. Furthermore, portfolio risk is defined as a downside risk, which is why a semi-mean-absolute deviation portfolio selection model is introduced. Finally, as an illustration, the models presented are applied to a selection of power generation mixes. The efficient portfolio results show that the fuzzy portfolio selection models with different definitions of membership functions as well as the semi-mean-absolute deviation model perform better than the standard mean-variance approach. Moreover, introducing membership functions for the description of investors' aspiration levels for the expected return and risk shows how the knowledge of experts, and investors' subjective opinions, can be better integrated in the decision-making process than with probabilistic approaches.

Fuzzy Portfolio Optimization of Power Generation Assets

Fuzzy theory is proposed as an alternative to the probabilistic approach for assessing portfolios of power plants, in order to capture the complex reality of decision-making processes. This paper presents different fuzzy portfolio selection models, where the rate of returns as well as the investor’s aspiration levels of portfolio return and risk are regarded as fuzzy variables. Furthermore, portfolio risk is defined as a downside risk, which is why a semi-mean-absolute deviation portfolio selection model is introduced. Finally, as an illustration, the models presented are applied to a selection of power generation mixes. The efficient portfolio results show that the fuzzy portfolio selection models with different definitions of membership functions as well as the semi-mean-absolute deviation model perform better than the standard mean-variance approach. Moreover, introducing membership functions for the description of investors’ aspiration levels for the expected return and risk shows how the knowledge of experts, and investors’ subjective opinions, can be better integrated in the decision-making process than with probabilistic approaches.

Enhancing Electric Grid, Critical Infrastructure, and Societal Resilience with Resilient Nuclear Power Plants (rNPPs)

This paper, the fourth in a series, presents the results of a study conducted to explore the role current U.S. commercial nuclear power plants (NPPs) play and the role a new type of NPP—a resilient nuclear power plant (rNPP)—could play in enhancing U.S. electric Grid, Critical Infrastructure, and societal resilience. A rNPP is a NPP intentionally designed, sited, interfaced, and operated in a manner to enhance Grid resilience. Four specific rNPP applications are discussed: (1) rNPPs as “flexible operations” electricity generation assets, (2) rNPPs as anchors of nuclear hybrid energy systems, (3) rNPPs as Grid Black Start Resources, and (4) rNPPs as anchors of Resilient Critical Infrastructure Islands. These four applications, individually and collectively, could enhance U.S. Grid, Critical Infrastructure, and societal resilience during normal conditions and in the wake of major national calamities stemming from natural hazards and/or malevolent human actions. rNPPs would be both tactical and strategic resilience assets, thereby extending the value proposition of nuclear energy well beyond that associated with nuclear power’s traditional baseload electricity generation. These are important topics as society grows increasingly dependent on electricity, and the natural hazard and malevolent human threat portfolio to the Grid continues to evolve.

Enabling Resiliency Operations Across Multiple Microgrids With Grid Friendly Appliance Controllers

Changes in economic, technological, and environmental policies are resulting in a re-evaluation of the dependence on large central generation facilities and their associated transmission networks. Emerging concepts of smart communities/cities are examining the potential to leverage cleaner sources of generation, as well as integrating electricity generation with other municipal functions. When grid connected, these generation assets can supplement the existing interconnections with the bulk transmission system, and in the event of an extreme event, they can provide power via a collection of microgrids. To achieve the highest level of resiliency, it may be necessary to conduct switching operations to interconnect individual microgrids. While the interconnection of multiple microgrids can increase the resiliency of the system, the associated switching operations can cause large transients in low inertia microgrids. The combination of low system inertia and IEEE 1547 and 1547a-compliant inverters can prevent multiple microgrids from being interconnected during extreme weather events. This paper will present a method of using end-use loads equipped with grid friendly appliance controllers to facilitate the switching operations between multiple microgrids; operations that are necessary for optimal operations when islanded for resiliency.

Cybersecurity for shared infrastructure substation networks with IEC 61850 GOOSE and Sampled Values

This study explains an example where select digital technologies are included to face the challenges related to cybersecurity in a windfarm solution. The network is planned, designed and commissioned by one operator, and will be maintained by two operators – one for generation assets for the windfarm, the other for the grid connection power transmission assets. The topology deployed is shown at a high level, describing how the windfarm offshore and onshore networks will be divided between the operators and the specific requirements of the solution. A more detailed view of the network is shown depicting Ethernet switches, routers and Intelligent Electronic Devices (IEDs); with the services and protocols used are explained and aligned with the challenges that each one represents in terms of cybersecurity, for example interconnections that use Parallel Redundancy Protocol for redundancy exchanging GOOSE interlocking. Each challenge is addressed one at the time and the topology will evolve as the solutions are added, going from a unified network under the control of one operator into two that can be easily split between operators.

Estimating the value of demand-side management in low-cost, solar micro-grids

Demand-side management has the potential to reduce the cost of solar based community micro-grids and solar home systems for electricity access. This paper presents a methodology for optimal least-cost sizing of generation assets while meeting explicit reliability constraints in micro-grids that are capable of active demand management. The battery management model considers kinetic constraints on battery operation and represents dispatch in the field to regulate the depth of discharge. The model allows consideration of the trade-off between depth of discharge, cycle life, and calendar lifetime in lead-acid batteries. Separate reliability targets for disaggregated, residential load profiles at hourly timesteps are considered to evaluate the performance and cost reduction potential of demand-side management capabilities — with economic results and sensitivity analyses around key input assumptions subsequently presented. We find that demand-side management can reduce the number and cost of requisite solar panels and batteries with the integration of real-time management and controls – a key result for justifying next generation micro-grids for electricity access.

Role of energy storage in ensuring transmission system adequacy and security

The main purpose of a Transmission System Operator is to ensure stabile, reliable and efficient operation of its power system. Large-scale integration of renewable energy sources has introduced additional challenges to active control of transmission power systems. Traditionally, generation adequacy has been achieved through investments in generating units and transmission adequacy through investments in transmission lines. However, energy storage can be regarded as both the generation asset, as it reduces peak load and acts as a generator when injecting electricity into the network, and transmission asset, as it can move electricity in time thus reducing congestion and curtailment of renewable energy sources. This paper examines the role of energy storage in increasing power system adequacy and security. A method is proposed to define the charging/discharging schedule of energy storage after a contingency in order to preserve the system within the operating limits and to provide the system operator enough time to redispatch the system and relieve the overloaded lines. The method is applied to an actual part of the Croatian power system using scenarios that describe representative network states. The simulations are performed in a transmission operations and planning software using actual operating data. The results are analysed in details and conclusions on the role of the energy storage in providing transmission system adequacy and security is assessed.

Integration of distributed energy resources in isolated microgrids: the Colombian paradigm

The electrification of rural or isolated areas coupled with increasing environmental concerns have promoted the emergence of Distributed Energy Resources (DER) and the operation by isolated microgrids. However, the integration of such resources involves technical issues related to the reliability and continuity of the electricity supply. Indeed, the uncertainty of renewable generation sources and the reduced inertia of isolated microgrids are challenges for the operation of these distribution systems. One way to address them is by providing ancillary services through all the resources involved in the system’s operation (generation assets, demand share, and storage systems). Accordingly, this paper first presents a literature review of the challenges and potential benefits of integrating DERs into the operation of a distribution system. It also includes some common strategies to mitigate the vulnerability of the introduction of these technologies in microgrids. Afterwards, the current state of each type of resource in Colombia is assessed. Finally, some basic strategies that enhance the benefits of DER integration are outlined along with the overcoming of challenges of microgrid operation in said country. To that end, we consider isolated Colombian regions to be natural laboratories where the effects of DER integration and the requirements for the operation by local production units can be analyzed.

Modeling Dependent Outages of Electric Power Plants

In the article “Modeling Dependent Outages of Electric Power Plants,” Malladi, Mendoza-Arriaga, and Tompaidis study simultaneous outages of many electric power plants, a problem that can have severe economic consequences. Using real data, they separate power plants into possibly overlapping groups. Each group is affected by a single cause—input fuel, geographic proximity, or location relative to the transmission network, for example. Using the mathematical technique of time subordination, the authors build dependence in outages among power plants in each group. With outage data from NERC, they determine these groups for Texas and the western United States and calculate the impact on system reliability and on the addition of new generation assets. Looking ahead, the methodology in the article can help account for dependence in generation from renewable sources as their share of generation capacity increases. The article will be of interest to owners of generating assets, regulators, and energy and reliability researchers.

Co-Optimization of Energy and Ancillary Services for Hydrothermal Operation Planning Under a General Security Criterion

One of the most used methods for long-term hydrothermal operation planning is the stochastic dual dynamic programming (SDDP). Using this method, the immediate and future water opportunity cost can be balanced and an economic-dispatch policy can be defined for multiple reservoirs under inflow uncertainty. In this framework, equipment outages and reserve deliverability are generally disregarded, despite their strong impact on the operative plan. However, recent advances in robust optimization have shown how to endogenously account for security criteria in scheduling models with reduced computational burden. Within this framework, reserve deliverability is ensured across the network via the co-optimization of energy and reserves (ancillary services). In this paper, we propose a new multistage model for planning hydrothermal coordination that co-optimizes the nominal energy dispatch and individual up and down reserve allocations. The main goal of this paper is to address a general $n-K$ security criterion, such that, for each inflow scenario, the system is capable of withstanding the loss of up to $K$ components, i.e., generation or transmission assets. The proposed methodology uses the column-and-constraint generation algorithm to efficiently incorporate a compound umbrella set of contingencies in the SDDP algorithm. Results for the Brazilian power system data corroborate the effectiveness of the proposed model.