Search

• A risk-averse deep learning method for real-time emergency load shedding. • Deep learning algorithm to enhance prediction accuracy on load shedding strategy. • A shedding amount inference model to speculate the stability after load shedding. • A risk-averse loss function to minimize the risk of load under-cutting events. • Experiments on multiple power systems to verify the practical values of the method. Emergency load shedding is an effective and frequently used emergency control action for power system transient stability. Solving the full optimization models for load shedding is computational burdensome and thus slow react to the intense system variations from the increasing renewable energy sources and the more active demand-side behavior. Other sensitivity-based methods impair the control accuracy and may not guarantee global optimality. Artificial intelligence methods, as the data-driven approaches, have recently been well-recognized for its real-time decision-making capability to tackle the system variations. The existing artificial intelligence methods for emergency load shedding are based on shallow learning algorithms and can lead to both load under-cutting and over-cutting events. However, when the loads are under-cut, the power system will be exposed to a high risk of post-control instability that can propagate into cascading events, which incurs significantly higher cost than an over-cutting event. Being aware of such unbalanced control costs, this paper proposes a risk-averse deep learning method for real-time emergency load shedding, which trains deep neural network towards the reluctance to load under-cutting events, so as to avoid the huge control cost incurred by control failure. The case studies on two renewable power systems demonstrate that, compared to the state-of-the-art methods, the proposed risk-averse method can significantly improve the control success rate with negligible increase in prediction error, ending up with lower overall control cost. The results verify the enhanced control performance and the practical values of the proposed method. [ABSTRACT FROM AUTHOR]

Deregulation of the retail electricity market has led to the emergence of an increasing number of electricity plans with competitive rates. Electricity customers now have more flexibility in choosing an electricity provider and electricity plan based on individual consumption needs. In this paper, a feature engineering hybrid collaborative filtering-based electricity plan recommender system (FECF-EPRS) is proposed for helping the customer get the right electricity plan. This system is composed of three-segment models for missing feature estimation, feature crosses construction, and electricity plan recommendation. It only takes easy-to-obtain household appliance usage features as inputs and outputs ratings for different plans. Through the test of real electricity market data, the FECF-EPRS shows a greater improvement in terms of recommendation accuracy, which can provide more accurate recommendations to customers and more reasonable pricing references for retailers. • Developing an electricity tariff recommender system for customers with different household energy consumption patterns. • Proposing a missing feature value filling method based the link between different appliances. • Introducing a Feature Crosses method to enhance recommendation accuracy. [ABSTRACT FROM AUTHOR]

Renewable energy resources and transport electrification have become essential components of modern and future distribution planning. Despite the impetus of economic and environmental benefits, the uncertainty brought in poses challenges. In this paper, we propose an integrated expansion planning framework based on a multiobjective mixed-integer nonlinear program. The aim is to minimize the net present value of investments considering feeder routing, substation alterations and construction while maximizing the utilization of proposed charging stations. Distributed generation and load uncertainties, recast in two-stage stochastic programming, are tackled with a scenario generation and reduction technique using a probabilistic approach with Kantorovich metrics. The final number of the scenarios are validated with an alternative clustering method. A flow-based location–allocation theory with user equilibrium traffic assignment model is exploited to site charging stations. The sizing problem is determined using continuous-time Markov chain modeling. The proposed framework is solved with a multiobjective Tchebycheff decomposition-based evolutionary algorithm and tested on a modified 54 bus distribution network and 25 transportation node system. Numerical results demonstrate the capability of the proposed method. Distribution planning authorities can benefit from the presented approach to make intertemporal investment decisions while maintaining the quality of system performance. • Optimal Integrated distribution plan coupled with road electricity is identified. • Uncertainty is handled using scenario generation and reduction approach. • The final branches of scenario tree are verified by k-means clustering method. • Stochastic modeling with Markov Chain solves charging station planning problem. [ABSTRACT FROM AUTHOR]

Networking of microgrids has received increasing attentions in recent years, which requires the uncertainty management associated with variations in the system. In this paper, a two-stage energy management strategy is developed for networked microgrids under the presence of high renewable resources. It decomposes the microgrids energy management into two stages to counteract the intra-day stochastic variations of renewable energy resources, electricity load and electricity prices. In the first stage (hourly time scale), a hierarchical hybrid control method is employed for networked microgrids, aiming to minimize the system operation cost. The mean–variance Markowitz theory is employed to assess the risk of operation cost variability due to the presence of uncertainties. In the second stage (5-min time scale), the components in microgrids are optimally adjusted to minimize the imbalance cost between day-ahead and real-time markets. Simulation study is conducted on an uncoordinated microgrids system as well as on the proposed networked system. According to the simulation results, the proposed method can identify optimal scheduling results, reduce operation costs of risk-aversion, and mitigate the impact of uncertainties. [ABSTRACT FROM AUTHOR]

The thermal unit commitment (UC) problem in power systems can usually be formulated as a mixed-integer quadratic programming (MIQP) problem, which is an NP-hard problem for practical-scale systems and thus is difficult to solve efficiently. In this paper, by projecting the unit generation level onto the interval [ 0 , 1 ] and using reformulation techniques, a novel two-binary-variable (2-bin) MIQP formulation for the UC problem is proposed. The proposed 2-bin formulation is more compact than the state-of-the-art one-binary-variable (1-bin) and three-binary-variable (3-bin) formulations. Moreover, the 2-bin formulation is tighter than the 1-bin and 3-bin formulations in terms of the quadratic cost function, and it is tighter than the 1-bin formulation in terms of linear constraints. The proposed model was tested on 73 instances, including 43 realistic instances and 30 8-unit-based instances, over a scheduling period of 24 h for systems ranging from 10 to 1040 generating units. The simulation results show that our proposed MIQP UC formulation is the tightest and most compact model and can be solved most efficiently. After introducing a sequence of piecewise perspective cuts to approximate the quadratic operational cost function, the three UC MIQP formulations can be approximated by three corresponding mixed-integer linear programming (MILP) formulations. Our experiments show that the proposed 2-bin MILP formulation also performs the best in terms of solution times. [ABSTRACT FROM AUTHOR]

• Static and dynamic image processing architectures are proposed for solar forecast; • Hybrid sky image-based forecasting models are effectively developed and verified; • Noticeable improvements for solar ramp event forecasting have been achieved. With the ever-increased penetration of solar energy in the power grid, solar photovoltaic forecasting has become an indispensable aspect in maintaining power system stability and economic operation. At the operating stage, the forecasting accuracy of renewables has a direct influence on energy scheduling and dispatching. In this paper, we propose a series of novel approaches based on deep whole-sky-image learning architectures for very short-term solar photovoltaic generation forecasting, of which the lookahead windows concern the scales from 4 to 20 min. In particular, multiple deep learning models with the integration of both static sky image units and dynamic sky image stream are explicitly investigated. Extensive numerical studies on various models are carried out, through which the experimental results show that the proposed hybrid static image forecaster provides superior performance as compared to the benchmarking methods (i.e. the ones without sky images), with up to 8.3% improvement in general, and up to 32.8% improvement in the cases of ramp events. In addition, case studies at multiple time scales reveal that sky-image-based models can be more robust to the ramp events in solar photovoltaic generation. [ABSTRACT FROM AUTHOR]