In the last decade, a number of severe urban power outages have been caused by extreme natural disasters, e.g., hurricanes, snowstorms and earthquakes, which highlights the need for rethinking current planning principles of urban energy systems and expanding the classical reliability-oriented view. In addition to being reliable to low-impact and high-probability outages, power system should also have high level of resilience to withstand high-impact and low-probability (HILP) events. Compared with power system, multi-energy systems (MESs) have advantages in improving resilience through energy shifting across multiple energy sectors, a variety of generalized energy storage resources and thermal inertia of heat/cooling loads. This paper proposes a resilience-oriented planning method to determine optimal configuration of distribution level MES, e.g., urban energy supply systems, considering comprehensive impacts from supply, network and demand sides in MES. Impacts of energy shifting at supply side, pipe storage at network side and thermal inertia at demand side are described in the same linear modeling framework using energy hub (EH) model. Generalized energy storage resources including centralized and distributed energy storage devices, pipe network storage and building heat capacity are all modeled into centralized energy storage to facilitate an efficient configuration planning of MES.

Solar photovoltaic (PV) power forecasting has become an important issue with regard to the power grid in terms of the effective integration of large-scale PV plants. As the main influence factor of PV power generation, solar irradiance and its accurate forecasting are the prerequisite for solar PV power forecasting. However, previous forecasting approaches using manual feature extraction (MFE), traditional modeling and single deep learning (DL) models could not satisfy the performance requirements in partial scenarios with complex fluctuations. Therefore, an improved DL model based on wavelet decomposition (WD), the Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM) is proposed for day-ahead solar irradiance forecasting. Given the high dependency of solar irradiance on weather status, the proposed model is individually established under four general weather type (i.e., sunny, cloudy, rainy and heavy rainy). For certain weather types, the raw solar irradiance sequence is decomposed into several subsequences via discrete wavelet transformation. Then each subsequence is fed into the CNN based local feature extractor to automatically learn the abstract feature representation from the raw subsequence data. Since the extracted features of each subsequence are also time series data, they are individually transported to LSTM to construct the subsequence forecasting model. In the end, the final solar irradiance forecasting results under certain weather types are obtained via the wavelet reconstruction of these forecasted subsequences. This case study further verifies the enhanced forecasting accuracy of our proposed method via a comparison with traditional and single DL models.

In a typical electricity market, the load aggregator (LA) bids in the wholesale market to purchase electricity and meet the expected demand of its customers in the retail market. However, considering the uncertainty of the wholesale market prices (WMPs), the LA has to undertake all the risks arising from the price volatility in the wholesale market, which may make the LA suffer from financial loss under some scenarios such as price spikes. To this end, firstly, this paper proposes an optimal bidding strategy model for the LA that implements the demand response program (DRP), which enables the LA to reduce the risk of financial loss caused by price volatility. The bidding model is a mixed integer linear programming (MILP) problem, which can be solved efficiently via a commercial solver. Secondly, making a rational and quantitative compensation mechanism is significant for the LA to induce its customers to participate in the DRP while there are few studies investigating it, hence, this paper designs a quantitative compensation mechanism for the LA. Case studies using a dataset from the Thames valley vision (TVV) verify the effectiveness of the proposed bidding model. The results confirm that the implementation of DRP not only brings great profits to LA, but also benefits the other entities in the electricity market.