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 novel two-stage game-theoretic residential photovoltaic (PV) panels planning framework for distribution grids with potential PV prosumers. One innovative contribution is that a residential PV panels location-allocation model is integrated with the energy sharing mechanism to increase economic benefits to PV prosumers and meanwhile facilitate the reasonable installation of residential PV panels. The optimization of residential PV panels planning decisions is formulated as a two-stage model. In the first stage, we develop a Stackelberg game based stochastic bi-level energy sharing model to determine the optimal sizing of PV panels with uncertain PV energy output, load demand, and electricity price. Instead of directly solving the proposed bi-level energy sharing problem by using commercial solvers, we develop an efficient descend search algorithm-based solution method which can significantly improve the computation efficiency. In the second stage, we propose a stochastic programming based residential PV panels deployment model for all PV prosumers. This model is formulated as an optimal power flow (OPF) problem to minimize active power loss. Finally, simulations on an IEEE 33-node and 123-node test systems demonstrate the effectiveness of the proposed method.

In this paper, a novel machine learning based data-driven pricing method is proposed for sharing rooftop photovoltaic (PV) generation and energy storage (ES) in an electrically interconnected residential building cluster (RBC). In the studied problem, the energy sharing process is modeled by the leader-followers Stackelberg game where the owner of the rooftop PV system is responsible for pricing self-generated PV energy and operating ES devices. Meanwhile, local electricity consumers in the RBC choose their energy consumption with the given internal electricity prices. To track the stochastic rooftop PV panel outputs, the long short-term memory (LSTM) network based rolling-horizon prediction function is developed to dynamically predict future trends of PV generation. With system information, the predicted information is fed into a Q-learning based decision-making process to find near-optimal pricing strategies. The simulation results verify the effectiveness of the proposed approach in solving energy sharing problems with partial or uncertain information.

The wide implementation of information and communication technologies (ICT) cause power system operations exposed to cyber-attacks. Meanwhile, the tendency of integrated multi energy vectors has worsened this issue with multiple energy coupled. This paper proposes a two-stage risk-averse mitigation strategy for integrated energy systems (IES), incorporating power, natural gas and water systems against false data injection attacks (FDIA) under water-energy nexus. The FDIA on individual sub-systems is modelled through hampering false data integrity to the systems. An innovative two-stage risk-averse distributionally robust optimization (RA-DRO) is proposed to mitigate uneconomic operation and provides a coordinated optimal load shedding scheme for the nexus system security. A coherent risk measure, Conditional Value-at-Risk is incorporated into the RA-DRO to model risk. A Benders decomposition method is used to solve the original NP-hard RA-DRO problem. Case studies are demonstrated on an IES under water-energy nexus and results show that the effectiveness of the method to mitigate risks from potential FDIA and renewable uncertainties. This research provides IES operators an economic system operation tool by optimally coordinating energy infrastructures and implementing reasonable load shedding to enhance cybersecurity.