Showing papers in "IEEE Transactions on Power Systems in 2020"

Microgrid Stability Definitions, Analysis, and Examples

Abstract: This document is a summary of a report prepared by the IEEE PES Task Force (TF) on Microgrid Stability Definitions, Analysis, and Modeling, IEEE Power and Energy Society, Piscataway, NJ, USA, Tech. Rep. PES-TR66, Apr. 2018, which defines concepts and identifies relevant issues related to stability in microgrids. In this paper, definitions and classification of microgrid stability are presented and discussed, considering pertinent microgrid features such as voltage-frequency dependence, unbalancing, low inertia, and generation intermittency. A few examples are also presented, highlighting some of the stability classes defined in this paper. Further examples, along with discussions on microgrid components modeling and stability analysis tools can be found in the TF report.

Deep-Reinforcement-Learning-Based Autonomous Voltage Control for Power Grid Operations

Abstract: In this letter, a novel autonomous control framework “Grid Mind” is proposed for the secure operation of power grids based on cutting-edge artificial intelligence (AI) technologies. The proposed platform provides a data-driven, model-free and closed-loop control agent trained using deep reinforcement learning (DRL) algorithms by interacting with massive simulations and/or real environment of a power grid. The proposed agent learns from scratch to master the power grid voltage control problem purely from data. It can make autonomous voltage control (AVC) strategies to support grid operators in making effective and timely control actions, according to the current system conditions detected by real-time measurements from supervisory control and data acquisition (SCADA) or phasor measurement units (PMUs). Two state-of-the-art DRL algorithms, namely deep Q-network (DQN) and deep deterministic policy gradient (DDPG), are proposed to formulate the AVC problem with performance compared. Case studies on a realistic 200-bus test system demonstrate the effectiveness and promising performance of the proposed framework.

Ultra-Short-Term Industrial Power Demand Forecasting Using LSTM Based Hybrid Ensemble Learning

Abstract: Power demand forecasting with high accuracy is a guarantee to keep the balance between power supply and demand. Due to strong volatility of industrial power load, ultra-short-term power demand is difficult to forecast accurately and robustly. To solve this problem, this article proposes a Long Short-Term Memory (LSTM) network based hybrid ensemble learning forecasting model. A hybrid ensemble strategy—which consists of Bagging, Random Subspace, and Boosting with ensemble pruning—is designed to extract the deep features from multivariate data, and a new loss function that integrates peak demand forecasting error is proposed according to bias-variance tradeoff. Experimental results on open dataset and practical dataset show that the proposed model outperforms several state-of-the-art time series forecasting models, and obtains higher accuracy and robustness to forecast peak demand.

Using Bayesian Deep Learning to Capture Uncertainty for Residential Net Load Forecasting

Abstract: Decarbonization of electricity systems drives significant and continued investments in distributed energy sources to support the cost-effective transition to low-carbon energy systems. However, the rapid integration of distributed photovoltaic (PV) generation presents great challenges in obtaining reliable and secure grid operations because of its limited visibility and intermittent nature. Under this reality, net load forecasting is facing unprecedented difficulty in answering the following question: How can we accurately predict the net load while capturing the massive uncertainties arising from distributed PV generation and load, especially in the context of high PV penetration? This paper proposes a novel probabilistic day-ahead net load forecasting method to capture both epistemic uncertainty and aleatoric uncertainty using Bayesian deep learning, which is a new field that combines Bayesian probability theory and deep learning. The proposed methodological framework employs clustering in subprofiles and considers residential rooftop PV outputs as input features to enhance the performance of aggregated net load forecasting. Numerical experiments have been carried out based on fine-grained smart meter data from the Australian grid with separately recorded measurements of rooftop PV generation and loads. The results demonstrate the superior performance of the proposed scheme compared with a series of state-of-the-art methods and indicate the importance and effectiveness of subprofile clustering and high PV visibility.

Hybrid Deep Neural Networks for Detection of Non-Technical Losses in Electricity Smart Meters

Abstract: Non-technical losses (NTL) in electricity utilities are responsible for major revenue losses. In this paper, we propose a novel end-to-end solution to self-learn the features for detecting anomalies and frauds in smart meters using a hybrid deep neural network. The network is fed with simple raw data, removing the need of handcrafted feature engineering. The proposed architecture consists of a long short-term memory network and a multi-layer perceptrons network. The first network analyses the raw daily energy consumption history whilst the second one integrates non-sequential data such as its contracted power or geographical information. The results show that the hybrid neural network significantly outperforms state-of-the-art classifiers as well as previous deep learning models used in NTL detection. The model has been trained and tested with real smart meter data of Endesa, the largest electricity utility in Spain.

The IEEE Reliability Test System: A Proposed 2019 Update

Abstract: The evolving nature of electricity production, transmission, and consumption necessitates an update to the IEEE's Reliability Test System (RTS), which was last modernized in 1996. The update presented here introduces a generation mix more representative of modern power systems, with the removal of several nuclear and oil-generating units and the addition of natural gas, wind, solar photovoltaics, concentrating solar power, and energy storage. The update includes assigning the test system a geographic location in the southwestern United States to enable the integration of spatio-temporally consistent wind, solar, and load data with forecasts. Additional updates include common RTS transmission modifications in published literature, definitions for reserve product requirements, and market simulation descriptions to enable benchmarking of multi-period power system scheduling problems. The final section presents example results from a production cost modeling simulation on the updated RTS system data.

A P2P-Dominant Distribution System Architecture

Abstract: Peer-to-peer interactions between small-scale energy resources exploit distribution network infrastructure as an electricity carrier, but remain financially unaccountable to electric power utilities. This status-quo raises multiple challenges. First, peer-to-peer energy trading reduces the portion of electricity supplied to end-customers by utilities and their revenue streams. Second, utilities must ensure that peer-to-peer transactions comply with distribution network limits. This article proposes a peer-to-peer energy trading architecture, in two configurations, that couples peer-to-peer interactions and distribution network operations. The first configuration assumes that these interactions are settled by the utility in a centralized manner, while the second one is peer-centric and does not involve the utility. Both configurations use distribution locational marginal prices to compute network usage charges that peers must pay to the utility for using the distribution network.

A Multi-Agent Deep Reinforcement Learning Method for Cooperative Load Frequency Control of a Multi-Area Power System

Abstract: This paper proposes a data-driven cooperative method for load frequency control (LFC) of the multi-area power system based on multi-agent deep reinforcement learning (MA-DRL) in continuous action domain. The proposed method can nonlinearly and adaptively derive the optimal coordinated control strategies for multiple LFC controllers through centralized learning and decentralized implementation. The centralized learning is achieved by MA-DRL based on a global action-value function to quantify overall LFC performance of the power system. To solve the MA-DRL problem, multi-agent deep deterministic policy gradient (DDPG) is derived to adjust control agents’ parameters considering the nonlinear generator behaviors. For implementation, each individual controller only needs local information in its control area to deliver optimal control signals. Numerical simulations on a three-area power system and the fully-modeled New-England 39-bus system demonstrate that the proposed method can effectively minimize control errors against stochastic frequency variations caused by load and renewable power fluctuations.

Wind in Weak Grids: Low-Frequency Oscillations, Subsynchronous Oscillations, and Torsional Interactions

Abstract: Oscillations have been observed in wind farms with weak grid interconnections. While Texas observes 4 Hz low-frequency oscillations, the west region in China observes subsynchronous oscillations at 30 Hz. Furthermore, this oscillation mode caused torsional interactions with a remote synchronous generator and led to shutdown of the power plant. Inspired by those real-world events, this paper aims to present an analytical model of type-4 wind in weak grids that can demonstrate both low-frequency and subsynchronous frequency oscillations. Critical factors, e.g., the parameters of the phase-locked loop (PLL), are examined using small-signal analysis. The analysis results are validated using a testbed built in MATLAB/SimPowerSystems. Except power electronic switching sequences, the testbed has controls (e.g., wind turbine pitch control, maximum power point control, converter controls), machine dynamics and power system dynamics modeled. This testbed includes a 100 MW Type-4 wind farm, a 600 MW synchronous generator, a long transmission line and a grid. The testbed successfully demonstrates two types of dominant oscillations under different PLLs. In addition, the testbed demonstrates torsional interactions due to the proximity of the subsynchronous mode and one of the torsional modes.

A Multi-Agent Deep Reinforcement Learning Based Voltage Regulation Using Coordinated PV Inverters

Abstract: This paper proposes a multi-agent deep reinforcement learning-based approach for distribution system voltage regulation with high penetration of photovoltaics (PVs). The designed agents can learn the coordinated control strategies from historical data through the counter-training of local policy networks and centric critic networks. The learned strategies allow us to perform online coordinated control. Comparative results with other methods show the enhanced control capability of the proposed method under various conditions.

Reliability Evaluation for Integrated Power-Gas Systems With Power-to-Gas and Gas Storages

Abstract: The reliability evaluation of integrated power-gas systems (IPGS) becomes critical due to the high dependency of the two energy systems. Once a contingent incident happens in one system, the other system will accordingly be affected. In this paper, a detailed model for the IPGS is presented with the consideration of the power-to-gas devices and gas storages. Furthermore, a sequential Monte Carlo (SMC) simulation is utilized to evaluate the reliability of the IPGS. In particular, the gas storage life reliability model is considered to characterize the charging and discharging performance. Moreover, an optimal load shedding model is used to coordinate the load shedding of the IPGS. What's more, new reliability indices are given to display the reliability of the IPGS. Finally, the proposed model is tested on an integrated IEEE 24-bus power system and 20-node gas system and an integrated IEEE RTS 96 power system and 40-node gas system. The results show the effectiveness of the proposed model.

A Data-Driven Multi-Agent Autonomous Voltage Control Framework Using Deep Reinforcement Learning

Abstract: The complexity of modern power grids keeps increasing due to the expansion of renewable energy resources and the requirement of fast demand responses, which results in a great challenge for conventional power grid control systems. Existing autonomous control approaches for the power grid requires an accurate system model and a powerful computational platform, which is difficult to scale up for the large-scale energy system with more control options and operating conditions. Facing these challenges, this article proposes a data-driven multi-agent power grid control scheme using a deep reinforcement learning (DRL) method. Specifically, the classic autonomous voltage control (AVC) problem is taken as an example and formulated as a Markov Game with a heuristic method to partition agents. Then, a multi-agent AVC (MA-AVC) algorithm based on a multi-agent deep deterministic policy gradient (MADDPG) method that features centralized training and decentralized execution is developed to solve the AVC problem. The proposed method can learn from scratch and gradually master the system operation rules by input and output data. In order to demonstrate the effectiveness of the proposed MA-AVC algorithm, comprehensive case studies are conducted on an Illinois 200-Bus system considering load/generation changes, N-1 contingencies, and weak centralized communication environment.

Direct Energy Trading of Microgrids in Distribution Energy Market

Abstract: Recent advancement of distributed renewable generation has motivated microgrids to trade energy directly with one another, as well as with the utility, in order to minimize their operational costs. Energy trading among microgrids, however, confronts challenges such as reaching a fair trading price, maximizing participants’ profit, and satisfying power network constraints. In this paper, we formulate the direct energy trading among multiple microgrids as a generalized Nash bargaining (GNB) problem that involves the distribution network's operational constraints (e.g., power balance equations and voltage limits). We prove that solving the GNB problem maximizes the social welfare and also fairly distributes the revenue among the microgrids based on their market power. To address the nonconvexity of the GNB problem, we propose a two-phase approach. The first phase involves solving the optimal power flow problem in a distributed fashion using the alternative direction method of multipliers to determine the amount of energy trading. The second phase determines the market clearing price and mutual payments of the microgrids. Simulation results on an IEEE 33-bus system with four microgrids show that the proposed framework substantially reduces total network cost by 37.2%. Our results suggest direct trading need be enforced by regulators to maximize the social welfare.

Battery Energy Storage Systems in Energy and Reserve Markets

Abstract: Recent Federal Energy Regulatory Commission (FERC) Order 841 requires that Independent System Operators (ISOs) facilitate the participation of energy storage systems (ESSs) in energy, ancillary services, and capacity markets, by including ESS bidding parameters that represent the physical and operational characteristics. However, in the existing market frameworks that allow Battery Energy Storage Systems (BESSs) to participate, the bids and offers do not explicitly represent the physical and operational characteristics such as the state of charge (SOC), discharge rate, degradation, etc. This paper proposes a novel BESS operational cost model considering degradation cost, based on depth of discharge and discharge rate. The model is developed considering Lithium-ion batteries, and the approach can be applied to other conventional electrochemical batteries, but not flow batteries. A detailed bid/offer structure based on the proposed BESS operational cost functions is formulated. Thereafter, a new framework and mathematical model for BESS participation in an LMP based, co-optimized, energy and spinning reserve market, are developed. Three case studies are presented to investigate the impact of BESS participation on system operation and market settlement. The proposed model is validated on the IEEE Reliability Test System (RTS) to demonstrate its functionalities.

Modeling Frequency Dynamics in Unit Commitment With a High Share of Renewable Energy

Abstract: The power system inertia is gradually decreasing with the growing share of variable renewable energy (VRE). This may jeopardize the frequency dynamics and challenges the secure operation of power systems. In this paper, the concept of frequency security margin is proposed to quantify the system frequency regulation ability under contingency. It is defined as the maximum power imbalance that the system can tolerate while keeping frequency within the tolerable frequency range. A frequency constrained unit commitment (FCUC) model considering frequency security margin is proposed. Firstly, the analytical formulation of system frequency nadir is derived while considering both the frequency regulation characteristics of the thermal generators and the frequency support from VRE plants. Then, the frequency security margin is analytically formulated and piecewise linearized. A novel FCUC model is proposed by incorporating linear frequency security constraints into the traditional unit commitment model. Case studies on a modified IEEE RTS-79 system and HRP-38 system are provided to verify the effectiveness of the proposed FCUC model. The impacts of VRE penetration on system frequency security are analyzed using frequency security margin.

Optimal Distributed Feedback Voltage Control Under Limited Reactive Power

Abstract: In this paper, we propose a distributed voltage control in power distribution networks through reactive power compensation. The proposed control can operate in a distributed fashion where each bus makes its decision based on local voltage measurements and communication with neighboring buses, always satisfy the reactive power capacity constraint, drive the voltage magnitude into an acceptable range, and minimize an operational cost. We also perform various numerical case studies to demonstrate the effectiveness and robustness of the controller using the nonlinear power flow model.

On the Interaction Between Aggregators, Electricity Markets and Residential Demand Response Providers

Abstract: To decarbonize the heating sector, residential consumers may install heat pumps. Coupled with heating loads with high thermal inertia, these thermostatically controlled loads may provide a significant source of demand side flexibility. Since the capacity of residential consumers is typically insufficient to take part in the day-ahead electricity market (DAM), aggregators act as mediators that monetize the flexibility of these loads through demand response (DR). In this paper, we study the strategic interactions between an aggregator, its consumers and the DAM using a bilevel optimization framework. The aggregator-consumer interaction is captured either as a Stackelberg or a Nash Bargaining Game, leveraging chance-constrained programming to model limited controllability of residential DR loads. The aggregator takes strategic positions in the DAM, considering the uncertainty on the market outcome, represented as a stochastic Stackelberg Game. Results show that the DR provider-aggregator cooperation may yield significant monetary benefits. The aggregator cost-effectively manages the uncertainty on the DAM outcome and the limited controllability of its consumers. The presented methodology may be used to assess the value of DR in a deregulated power system or may be directly integrated in the daily routine of DR aggregators.

Optimal Tap Setting of Voltage Regulation Transformers Using Batch Reinforcement Learning

Abstract: In this paper, we address the problem of setting the tap positions of load tap changers (LTCs) for voltage regulation in power distribution systems. The objective is to find a policy that maps measurements of voltage magnitudes and topology information to LTC tap ratio changes so as to minimize the voltage deviation across the system. We formulate this problem as a Markov decision process (MDP), and propose a data and computationally efficient batch reinforcement learning (RL) algorithm to solve it. To circumvent the “curse of dimensionality” resulting from the large state and action spaces, we propose a sequential learning algorithm to learn an action-value function for each LTC, based on which the optimal tap positions can be directly determined. By taking advantage of a linearized power flow model, we propose an algorithm to estimate the voltage magnitudes under different tap settings, which allows the RL algorithm to explore the state and action spaces freely offline without impacting the system operation. The effectiveness of the proposed algorithm is validated via numerical simulations on the IEEE 13-bus and 123-bus distribution test feeders.

Continuous Under-Frequency Load Shedding Scheme for Power System Adaptive Frequency Control

Abstract: Frequency drop due to loss of massive generation is a threat to power system frequency stability. Under-frequency load shedding (UFLS) is the principal measure to prevent successive frequency declination and blackouts. Based on traditional stage-by-stage UFLS scheme, a new continuous UFLS scheme is proposed in this paper to shed loads proportional to frequency deviation. The characteristic of the proposed scheme is analyzed with a closed-form solution of frequency dynamics. Frequency threshold and time delay are added to make the proposed scheme practical. A line-by-line scheme based on precise load control is introduced to implement the continuous scheme for systems without enough continuously controllable loads. The load shedding scale factor of the proposed scheme is tuned with an analytical method to achieve adaptability to different operating conditions. The adaptability of the proposed scheme is validated with 39-bus New England model and simplified Shandong Power Grid of China.

Carbon-Oriented Operational Planning in Coupled Electricity and Emission Trading Markets

Abstract: Carbon financing policies such as emission trading have been used to assist in emission mitigation worldwide. As energy end-users/consumers are the underlying driver of emissions, it would be difficult to effectively mitigate carbon emissions by creating an emission trading market without active end-users’ involvement. In electricity markets, demand side management (DSM) in the smart grid can manage demands in response to power supply conditions and influence end-users to contribute to improving both network efficiency and economic efficiency. However, it is a relatively new topic to study the environmental benefits of DSM. This paper proposes a two-stage scheduling model to comprehensively investigate the environmental benefits of consumers participating in both electricity and carbon emission trading markets through active DSM. A developed zero sum gains-data envelopment analysis (ZSG-DEA) model based multi-criteria allocation scheme for emission allocation is employed. Meanwhile, the carbon emission flow model (CEF) is applied to track the “virtual” carbon flow accompanying power flow. According to case studies on the IEEE 24-bus system and IEEE 118-bus system, the proposed model can effectively achieve carbon emission mitigation and provide consumers extra environmental benefits in some scenarios. This model can be an important guide for governments to establish emission trading schemes.

Hierarchical Deep Learning Machine for Power System Online Transient Stability Prediction

Abstract: This paper develops a hierarchical deep learning machine (HDLM) to efficiently achieve both quantitative and qualitative online transient stability prediction (TSP). For the sake of improving its online efficiency, multiple generators’ fault-on trajectories as well as the two closest data-points in pre-/post-fault stages are acquired by PMUs to form its raw inputs. An anti-noise graphical transient characterization technique is tactfully designed to transform multiplex trajectories into 2-D images, within which system-wide transients are concisely described. Then, following the divide-and-conquer philosophy, the HDLM trains a two-level convolutional neural network (CNN) based regression model. With stability margin regressions hierarchically refined, it manages to perform reliable and adaptive online TSP almost immediately after fault clearance. Test results on the IEEE 39-bus test system and the real-world Guangdong Power Grid in South China demonstrate the HDLM's superior performances on both stability status and stability margin predictions.

Two-Stage Variable Proportion Coefficient Based Frequency Support of Grid-Connected DFIG-WTs

Abstract: This paper proposes a variable proportion coefficient based control scheme for the doubly-fed induction generator based wind turbines (DFIG-WTs) to implement frequency support by regulating rotor speed. To reduce the adverse impact caused by regulating DFIG-WT on the dynamic characteristic of the grid frequency, a two-stage switching control scheme is employed for the proposed method. In the first stage, the variable proportion coefficient is designed for emulating the virtual inertia of the DFIG-WT to provide inertia response. In the second stage, a fuzzy control scheme is employed to design the variable proportion coefficient for both quickly restoring the maximum power point tracking (MPPT) operation of DFIG-WTs and avoiding the secondary frequency drop to system. Case studies are undertaken based on WSCC 9-bus and IEEE 39-bus power systems, respectively. Simulation results show that the proposed control scheme can not only make grid-connected DFIG-WTs provide the friendly frequency support, but also help them to fully use the frequency regulation ability of synchronous generators for quickly restoring the MPPT operation.

Day-Ahead Scheduling of a Local Energy Community: An Alternating Direction Method of Multipliers Approach

Abstract: The paper focuses on the day-ahead operational planning of a grid-connected local energy community (LEC) consisting of an internal low-voltage network and several prosumers including generation units, battery storage systems, and local loads. In order to preserve, as much as possible, the confidentiality of the features of prosumers’ equipment and the production and load forecasts, the problem is addressed by designing a specific distributed procedure based on the alternating direction method of multipliers (ADMM). The distributed procedure calculates the scheduling of the available energy resources to limit the balancing action of the external grid and allocates the internal network losses to the various power transactions. Results obtained for various case studies are compared with those obtained by a centralized optimization approach. The results confirm that, in the considered LEC framework, each of the prosumers achieves a reduction in costs or increases revenues in case it participates to the LEC with respect to the case in which it can only transact with an external energy provider.

Impact of High Renewable Penetration on the Power System Operation Mode: A Data-Driven Approach

Abstract: The high penetration of renewable energy will substantially change the power system operation. Traditionally, the annual operation of a power system can be represented by some typical operation modes and acts as the basis for the power-system-related analysis. The introduction of highly penetrated renewable energy will make the power system operation mode highly diversified and variable. These modes may not follow traditional empirical patterns. In this paper, we propose a data-driven method based on high-dimensional power system operation data (including power flow, unit generation, and load demand) to identify the pattern of the operation modes and analyze the impact of high renewable penetration. Specifically, the proposed data-driven method is composed of simulation, preprocessing, clustering, dimension reduction, and visualization with the aim to provide an intuitive understanding of the operation mode variety under high renewable penetration. In addition, several indices are introduced to quantify the space dispersion, time variation, and seasonal consistency of operation modes. A case study on actual Qinghai provincial power system in China validates the effectiveness of the proposed data-driven method and indicates that the dispersion and time variation of operation mode will significantly increase in the beginning and then saturate with the increase in renewable penetration level. The operation mode is also less correlated with seasons in renewable energy dominated power system.

On An Equivalent Representation of the Dynamics in District Heating Networks for Combined Electricity-Heat Operation

Abstract: Improved coordination between power and district heating systems (DHSs) can play a vital role in accommodating more wind power. However, the dynamics of district heating networks (DHNs) have not yet been fully described for large transfer delays, which remains an obstacle for optimal electricity-heat coordination. This paper solves this problem by proposing a uniform framework in the Laplace domain by modeling heat losses and transfer delays from an electrical-analog perspective. Based on this framework, an equivalent representation of DHNs is proposed, showing the explicit “end-to-end” relationship between heat power generation and demand. The model is applied in combined electricity and heat operation, in which the black-box model is provided by DHS operators for power system operators without revealing the detailed network information. A case study of an integrated electricity-heat system demonstrates the effectiveness and simplicity of the proposed model.

Analytical Reliability Assessment Method for Complex Distribution Networks Considering Post-Fault Network Reconfiguration

Abstract: Analytical methods for evaluating the reliability of simple and radial distribution networks have been well established. Since these analytical methods cannot consider post-fault load transfer between feeders, the reliability indices are significantly underestimated for mesh-constructed distribution networks. To accommodate various application scenarios, Monte-Carlo simulations are widely used for complex distribution networks and heavy computation burden is involved. In this paper, we propose a novel linear programming model which includes precisely assessing reliability and considers post-fault network reconfiguration strategies involving operational constraints. Moreover, this model also can formulate the influences of demand variations, uncertainty of distributed generations and protection failures on the reliability indices. Numerical simulations show that the proposed model yields the same results as the simulation-based algorithm. Specifically, the system average interruption duration indices are reduced when considering post-fault network reconfiguration strategies in all tested systems. Moreover, the proposed model is suitable for inclusion in reliability-constrained operational and planning optimization models for power distribution systems.

A Transactive Energy Framework for Coordinated Energy Management of Networked Microgrids With Distributionally Robust Optimization

Abstract: Networked microgrids (MGs) are considered as an emerging grid design for the future distribution system (DS). The coordination of the networked MGs is critical in order to further enhance the operation efficiency and reliability of the system. In this paper, a transactive energy (TE) framework is proposed for the coordinated energy management of networked MGs in DS. Instead of direct coordination signals and fixed pricing schemes, the distribution network operator (DNO) organizes a transactive market with the MGs to coordinate the energy management in the operation. Further, a distributionally robust optimization (DRO)-based algorithm is developed to provide a robust solution of the detailed scheduling decisions in the proposed TE framework under uncertainty without being too conservative. Case studies with the proposed framework were conducted with the IEEE 33-bus system with three MGs and the IEEE 123-bus system with nine MGs. The results of the case studies show that the proposed TE-based framework can effectively coordinate the energy scheduling of the MGs. The operational cost of the DS is reduced significantly. Meanwhile, the proposed DRO-based algorithm provides a robust but not over-conservative solution for the operation decisions of the DNO and MGs in the proposed framework.

Improved Deep Mixture Density Network for Regional Wind Power Probabilistic Forecasting

Abstract: Unsteady motion of the atmosphere incurs nonlinear and spatiotemporally coupled uncertainties in the wind power prediction (WPP) of multiple wind farms. This brings both opportunities and challenges to wind power probabilistic forecasting (WPPF) of a wind farm cluster or region, particularly when wind power is highly penetrated within the power system. This paper proposes an Improved Deep Mixture Density Network (IDMDN) for short-term WPPF of multiple wind farms and the entire region. In this respect, a deep multi-to-multi (m2m) mapping Neural Network model, which adopts the beta kernel as the mixture component to avoid the density leakage problem, is established to produce probabilistic forecasts in an end-to-end manner. A novel modified activation function and several general training procedures are then introduced to overcome the unstable behavior and NaN (Not a Number) loss issues of the beta kernel function. Verification of IDMDN is based on an open-source dataset collected from seven wind farms, and comparison results show that the proposed model improves the WPPF performance at both wind farm and regional levels. Furthermore, a laconic and accurate probabilistic expression of predicted power at each time step is produced by the proposed model.

Robust Control Scheme for Distributed Battery Energy Storage Systems in Load Frequency Control

Abstract: This paper proposes a robust control scheme to involve the distributed Battery Energy Storage Systems (BESSs) in Load Frequency Control (LFC) through BESS aggregators with sparse communication networks. In order to cope with the uncertainties associated with system operation, a two-layer Model Predictive Control (MPC) is developed so that more efficient control signals are provided to improve the response of BESSs to make larger contribution to the LFC. The outer layer in the proposed structure produces the command signal for the aggregator based on signals which are produced by the inner layer as well as the signal provided from the actual system. These command signals are provided so as to achieve the least value of error in Area Control Error (ACE) with a minimum control effort taking a variety of operational and physical constraints into consideration. Optimization procedures are also carried out to compute the optimal value of weighting coefficients contained in the objective functions. The capability of controller to cope with uncertainties is compared with a conventional single-layer MPC. In addition, the delay caused by propagation channels in delivering control signals to BESSs is modeled, and its impact on the performance of frequency regulation is evaluated. An intelligent fuzzy coordination control is then developed to coordinate the BESS aggregator and conventional power plants to avoid extra power injection/withdrawal by the conventional power plants in case of long delays. Case studies are conducted to illustrate the effectiveness of the proposed structure in controlling distributed BESSs with diverse energy capacities, rated powers, charging/discharging coefficients and time constants; and State of Charges (SoCs).

Real-Time Optimal Power Flow: A Lagrangian Based Deep Reinforcement Learning Approach

Abstract: High-level penetration of intermittent renewable energy sources has introduced significant uncertainties and variabilities into modern power systems. In order to rapidly and economically respond to the changes in power system operating state, this letter proposes a real-time optimal power flow (RT-OPF) approach using Lagrangian -based deep reinforcement learning (DRL) in continuous action domain. A DRL agent to determine RT-OPF decisions is constructed and optimized using the deep deterministic policy gradient. The DRL action-value function is designed to simultaneously model RT-OPF objective and constraints. Instead of using the critic network, the deterministic gradient is derived analytically. The proposed method is tested on the IEEE 118-bus system. Compared with the state-of-the-art methods, the proposed method can achieve a high solution optimality and constraint compliance in real-time.