As the power system is facing a transition towards a more intelligent, flexible and interactive system with higher penetration of renewable energy generation, load forecasting, especially short-term load forecasting for individual electric customers plays an increasingly essential role in the future grid planning and operation. Other than aggregated residential load in a large scale, forecasting an electric load of a single energy user is fairly challenging due to the high volatility and uncertainty involved. In this paper, we propose a long short-term memory (LSTM) recurrent neural network (RNN) based framework, which is the latest and one of the most popular techniques of deep learning, to tackle this tricky issue. The proposed framework is tested on a publicly available set of real residential smart meter data, of which the performance is comprehensively compared to various benchmarks including the state-of-the-arts in the field of load forecasting. As a result, the proposed LSTM approach outperforms the other listed rival algorithms in the task of short-term load forecasting for individual residential households.

The integration of renewable energy source (RES) and energy storage systems (ESS) in microgrids has provided potential benefit to end users and system operators. However, intermittent issues of RES and high cost of ESS need to be placed under scrutiny for economic operation of microgrids. This paper presents a two-layer predictive energy management system (EMS) for microgrids with hybrid ESS consisting of batteries and supercapacitors. Incorporating degradation costs of the hybrid ESS with respect to the depth of charge and lifetime, long-term costs of batteries and supercapacitors are modeled and transformed to short-term costs related to real-time operation. In order to maintain high system robustness at minimum operational cost, a hierarchical dispatch model is proposed to determine the scheduling of utilities in microgrids within a finite time horizon, in which the upper layer EMS minimizes the total operational cost and the lower layer EMS eliminates fluctuations induced by forecast errors. Simulation studies demonstrate that different types of energy storages can be utilized at two control layers for multiple decision-making objectives. Scenarios incorporating different pricing schemes, prediction horizon lengths, and forecast accuracies also verify the effectiveness of the proposed EMS structure.

This paper proposes a novel framework for home energy management (HEM) based on reinforcement learning in achieving efficient home-based demand response (DR). The concerned hour-ahead energy consumption scheduling problem is duly formulated as a finite Markov decision process (FMDP) with discrete time steps. To tackle this problem, a data-driven method based on neural network (NN) and Q-learning algorithm is developed, which achieves superior performance on cost-effective schedules for HEM system. Specifically, real data of electricity price and solar photovoltaic (PV) generation are timely processed for uncertainty prediction by extreme learning machine (ELM) in the rolling time windows. The scheduling decisions of the household appliances and electric vehicles (EVs) can be subsequently obtained through the newly developed framework, of which the objective is dual, i.e. to minimize the electricity bill as well as the DR induced dissatisfaction. Simulations are performed on a residential house level with multiple home appliances, an EV and several PV panels. The test results demonstrate the effectiveness of the proposed data-driven based HEM framework.

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.

This letter proposes a data-driven, model-free method for load frequency control (LFC) against renewable energy uncertainties based on deep reinforcement learning (DRL) in continuous action domain. The proposed method can nonlinearly derive control strategies to minimize frequency deviation with faster response speed and stronger adaptability for unmolded system dynamics. It consists of (i) offline optimization of LFC strategies with DRL and continuous action search, and (ii) online control with policy network where features are extracted by stacked denoising auto-encoders (SDAE). Numerical simulations verify the effectiveness and advantages of proposed method over existing approaches.

Lithium-ion (Li-ion) batteries have been widely applied in industrial applications. It is desired to predict the health state of batteries to achieve optimal operation and health management. However, accuracy is the biggest bottleneck for battery health prediction. In this paper, a new hybrid ensemble data-driven method is proposed to accurately predict the state-of-health (SOH) and remaining-useful-life (RUL) of Li-ion batteries. A health indicator is selected as feature inputs to predict the degradation trend of battery, after the Pearson correlation analysis. Two random learning algorithms are integrated to extract the inherent relationship between the extracted health indicator and practical SOH due to their good learning performance. Based on the estimated SOH, the nonlinear autoregressive (NAR) structure is designed to reduce the RUL prediction error of each learning model since the NAR structure makes good use of historical and current information. Finally, in order to quantitatively evaluate the prediction interval of the RUL, a Bootstrap-based uncertainty management method is designed. Test results on two publicly available datasets show that the proposed hybrid data-driven method can accurately predict the SOH and RUL of batteries. The proposed method does not require any other additional hardware or system downtime, which makes it suitable for online practical applications, such as energy storage systems and electric vehicles.