Optimization for Integrated Electricity and Gas Systems Considering Uncertainties

: (Alternative Format Thesis)

Abstract

In recent years, there is an increasing need for the integration of multi-energy vectors with the traditional power systems due to energy decarbonization and the booming coupling technologies. Integrated electricity and gas system (IEGS) plays a vital part in the modern energy provision through coordinating supply, conversion, storage and consumption. Although the interaction between electricity and gas systems facilitates the economic performance and security, it raises computational and modelling challenges for analysis and accurate modelling of the emerging IEGS. The optimal operation of IEGS is one most significant research topic to ensure the economic and reliable perspectives of IEGS. Nevertheless, the uncertainties introduced from renewable energy resources (RES), integration of smart grid technologies and natural disasters will affect the economic operation and destroy the energy infrastructures. For instance, underestimated uncertain renewable generation could cause network congestion and overestimated renewable generation will lead to a lack of energy supply. Accordingly, non-optimal and even infeasible solutions will be yielded.

This thesis studies the centralized and coordinated operation of IEGS under different types of uncertainties which contributes to optimal operation schemes of IEGS for the economic, reliable, resilient and sustainable perspectives. The proposed studies will greatly contribute to efficient and economic-effective IEGS operation schemes and related industrial applications in the presence of inevitable uncertainties and disasters. The main achievements of this research can be summarized as follows:

(1) This work proposes a two-stage distributionally robust operation model for integrated water-energy systems in a distribution level considering wind uncertainty. The optimization aims to minimize the total operation cost of the overall system. The presence of wind uncertainty inevitably leads to risks in decision making. Accordingly, a coherent risk measure, i.e., conditional value-at-risk, is combined with the optimization objective to determine risk-aversion operation schemes.

(2) To alleviate the impacts of seismic events on both power lines and gas pipelines of IEGS. A two-stage distributionally robust optimization (DRO) model is proposed to enhance the resilience for an IEGS, where the damage on both power lines and gas pipelines are considered. The seismic activities are regarded as uncertain events and the random damage on power lines and pipelines are regarded as uncertainties, which are handled by DRO. A novel model to assess the performance of IEGS against seismic attacks is developed. This damage quantification builds a probabilistic model and estimated by damage scenarios. The proposed novel DRO framework avoids specifying uncertainty distributions but only uses moment information, which is more practical considering that it is normally not possible to gather a sufficiently large amount of distributional information for extreme events.

(3) To address the adverse impact caused by the high integration of intelligent data technologies in IEGS, a two-stage risk mitigation strategy to address the uneconomic operation of IEGS under false data injection attacks (FDIA) considering renewable generation uncertainties. The FDIA mitigation scheme conducts the day-ahead and real time operation, which is more powerful and convenient to be used by system operators to ensure the efficiency and security of the IEGS. FDIA is assumed to attack both electricity and gas meter readings, including i) load measurement of electricity and gas systems and ii) gas density measurement. Uncertainties of renewable resources are considered in the proposed model as they can worsen system operation conditions during FDIA.

(4) The high penetration of renewable generation poses severe challenges to Volt/VAR optimization because of its output uncertainties, leading to voltage deviation and fluctuation. To resolve unacceptable voltage deviation under energy system interdependency, a novel coordinated two-stage multi-objective optimization is proposed for voltage control in the operation of IEGS, considering uncertain renewable generation and multi-vector energy system integration. The optimal voltage is achieved through efficiently coordinating the operation of on-load tap changers (OLTC), photovoltaic systems, and shunt capacitor banks. A conic tractable form with the dual formulation is transformed from the original problem and solved by constraint generation algorithm (CGA).

(5) To investigate the optimal coordinated operation of energy infrastructures in IEGS meanwhile ensure the gas quality, a co-optimization for both gas quality and system operation in an IEGS is proposed. The renewable uncertainty is captured by DRO approach with Kullback-Leibler (KL) divergence-based ambiguity set to ensure both the system robustness and tractability. The key indices to quantify the gas quality include gross calorific value (GCV), specific gravity (SG), Wobbe Index (WI), and Combustion Potential (CP). Apart from ensuring the indices to meet the related standards, the injected gas from power-to-gas (P2G) facility to gas system is mixed with nitrogen and Liquid Petroleum Gas (LPG) for maintaining the overall gas quality.

Date of Award	16 Jun 2021
Original language	English
Awarding Institution	University of Bath
Supervisor	Chenghong Gu (Supervisor) & Kang Ma (Supervisor)