Abstract
Net electricity consumption (NEC) is the result of the joint action of actual electricity consumption (AEC) and distributed photovoltaic (DPV) generation. The accuracy of NEC prediction affects the retailers' gaming and ultimate interests in electricity market. The ascending DPV installations present new challenges, with the modifications to the NEC curve becoming greater as DPV penetration increases. To track the changes in DPV penetration and improve the prediction accuracy under high penetration of DPV, a monthly NEC prediction model assembled by support vector regression and time series modeling under an online update framework is proposed. First, the DPV features are extracted from a few known solar customers' information to identify whether other customers install DPV or not. Second, an online update framework is proposed and its accuracy is verified by two validations regarding the conversion of non-solar customers to solar customers (namely the change of DPV penetration). Third, a NEC decoupling model based on historical NEC data of months without DPV installation is established. Finally, a monthly NEC prediction model under different DPV penetrations is proposed. Simulation results show that the proposed prediction method with an online update is more accurate than the individual time series model, and the performance of the prediction model is getting better with the increasing DPV penetration.

Abstract
Solar irradiance forecasting is a major priority for the power transmission systems in order to generate and incorporate the performance of massive photovoltaic plants efficiently. As such, prior forecasting techniques that use classical modelling and single deep learning models that undertake feature extraction procedures manually were unable to meet the output demands in specific situations with dynamic variability. Therefore, in this study, we propose an efficient novel hybrid solar irradiance forecasting based on three steps. In step I, we employ a powerful variable input selection strategy named as partial mutual information (PMI) to calculate the linear and non-linear correlations of the original solar irradiance data. In step II, unlike the traditional deep learning models designing their architectures manually, we utilize several deep convolutional neural network (CNN) models optimized by a novel modified whale optimization algorithm in order to compute the forecasting results of the solar irradiance datasets. Finally in step III, we deploy a deep Q-learning reinforcement learning strategy for selecting the best subsets of the combined deep optimized CNN models. Through analysing the forecasting results over two USA solar irradiance stations, it can be inferred that the proposed optimized deep RL-ensemble framework (ODERLEN) outperforms other powerful benchmarked algorithms in different time-step horizons.

Abstract
This paper deals with the bi-level two-stage operation scheduling of hydrogen-based microgrids within a distribution system where the wind and solar generation and load demands are considered as uncertain variables. The distribution system is considered as a leader in the upper level and microgrids as followers in the lower level. Unlike previous approaches, the upper-level is within the day-ahead market and considered a deterministic problem, and the lower-level is considered a stochastic problem and consists of two stages. The first stage determines the purchasing power from the distribution system, while the second stage adjusts the outputs and power dispatch for any realizations of scenarios. This model is transformed from a bi-level to a linear single-level model by applying the Karush- Kuhn-Tucker (KKT) optimally conditions, strong duality, and Fortuny-Amat methods. Several comparisons have been carried out regarding the single clearing price for all microgrids or separate prices for each microgrid. Furthermore, power exchange and dispatch in the distribution system are investigated under the mentioned frameworks.

Abstract
Demand response (DR) is an important technique to explore the demand-side flexibility. The wide deployment of smart meters makes it possible to quantify the baseline load. As an intermediate agent, demand response aggregator needs to obtain the aggregated baseline load (ABL) for the DR event. Previous studies about the household level estimation focus on the estimation method. However, for ABL estimation, customer division is an important issue. A major limitation is the mismatch between the objectives of segmentation and estimation. Therefore, this paper proposes a new closed-loop method for estimating the ABL, which utilizes the contextual bandit with policy gradient to link the segmentation with the estimation. As such, the ABL estimation accuracy can guide the segmentation to divide the customers. The segmentation and estimation optimize collaboratively to improve the ABL estimation accuracy. An ensemble method for combining network's weights during the training process is proposed. Moreover, a pre- and post-event adjustment method is developed to further improve the estimation accuracy. Comprehensive comparisons demonstrate the proposed method can achieve the best estimation performance with regard to the MAPE and RMSE. It improves the estimation accuracy by 7% in terms of MAPE, and 11% in terms of RMSE.

Abstract
This paper proposes an energy hub management model for residential, commercial, and industrial hubs, considering demand response programs (DRPs). The network configuration and AC optimal power flow (ACOPF) constraints have been applied to the model to prevent any unreal power transaction in the system. The cost due to environmental emissions has also been taken into account and the problem is modeled as a dynamic optimization problem, solved using the CPLEX solver in the GAMS software, interfaced with MATLAB/MATPOWER for the power flow analysis. Besides, the problem is studied in two cases as coordinated and uncoordinated operation modes to investigate their impacts on the operating cost, emission, and power losses. The obtained results show that the coordinated operation would lead to reducing the operating cost, power losses, and emission. Moreover, the impacts of the coordinated and uncoordinated operation modes on the load demand-supply under contingent events and disconnection from the upstream grid are assessed. The results derived from the simulation verify the superior performance of the coordinated operation. It is also noted that the DRP leads to mitigating the operating costs.

Abstract
The plug-in hybrid electric vehicles (PHEVs) integration into the electrical network introduces new challenges and opportunities for operators and PHEV owners. On the one hand, PHEVs can decrease environmental pollution. On the other hand, the high penetration of PHEVs in the network without charging management causes harmonics, voltage instability, and increased network problems. In this study, a charging management algorithm is presented to minimize the total cost and flatten the demand curve. The behavior of the PHEV owner in terms of arrival time and leaving time is modeled with a stochastic distribution function. The battery model and hourly power consumption of PHEV are modeled, and the obtained models are applied to determine the battery's state of charge. The proposed method is tested on a sample demand curve with and without a charging management algorithm to verify the efficiency. The results verify the efficiency of the proposed method in decreasing the total cost using the management algorithm for PHEVs, especially when the PHEVs sell the electricity to the network.