Description:
This paper presents a new robust self-scheduling strategy for virtual power plants (VPPs) considering the uncertainty sources of electricity prices, wind generations, and loads. Multi-horizon information-gap decision theory (MH-IGDT) as a non-deterministic and non-probabilistic uncertainty modeling framework is proposed here to specifically model the uncertainty sources considering their various uncertainty horizons. Since each uncertain parameter tends to optimize its uncertainty horizon competitively for a particular value of the uncertainty budget, the proposed MH-IGDT model is formulated as a multi-objective optimization problem. To solve this multi-objective problem, enhanced normalized normal constraint (ENNC) method is presented, which can obtain efficient uniformly-distributed Pareto optimal solutions. The proposed ENNC includes augmented normalized normal constraint method and lexicographic optimization technique to enhance the search performance in the objective space. To address the unsolved issue of being risk-averse or risk-seeker for a VPP in the market, a bi-directional decision-making approach is presented. This decision maker comprises an ex-ante performance evaluation method and a forward-backward dynamic programming approach to hourly find the best Pareto solution within the generated risk-averse and risk-seeker Pareto frontiers. Simulation results of the proposed self-scheduling strategy are presented for a VPP including dispatchable/non-dispatchable units, storages, and loads.
Description:
This paper addresses the planning problem of residential micro combined heat and power (micro-CHP) systems (including micro-generation units, auxiliary boilers, and thermal storage tanks) considering the associated technical and economic factors. Since the accurate values of the thermal and electrical loads of this system cannot be exactly predicted for the planning horizon, the thermal and electrical load uncertainties are modeled using a two-stage adaptive robust optimization method based on a polyhedral uncertainty set. A solution method, which is composed of column-and-constraint generation (C&CG) algorithm and block coordinate descent (BCD) method, is proposed to efficiently solve this adaptive robust optimization model. Numerical results from a practical case study show the effective performance of the proposed adaptive robust model for residential micro-CHP planning and its solution method.
Description:
Summary This paper presents a new adaptive robust operation optimization approach for energy hub (EH) to identify the optimal decisions on purchased energy carriers, upstream network interactions, and storing/conversion of the energy resources considering uncertainties. In this regard, a linearized framework for EH operation is first introduced. The proposed model is used to develop an EH including electrical energy, natural gas, and direct heat as inputs and electricity and heat demands as outputs. The electrical input energy is provided considering both purchased energy from upstream market and a photovoltaic (PV) generation, operated by the EH operator (EHO). The proposed approach characterizes the uncertain nature of loads, energy prices, and PV generations through polyhedral uncertainty sets, while the robustness of the proposed model can be controlled using the budget of uncertainty. The proposed adaptive robust model is formulated as a min‐max‐min optimization problem, which cannot be solved directly through an off‐the‐shelf optimization package. Thus, a new method, consisting decomposition + primal cutting plane + duality theory + exact linearization + post‐optimization analysis, is introduced to determine the EH optimal solution. The performance of the proposed approach is evaluated through a comprehensive case study.
A quantitative risk assessment model involving frequency and threat degree under line-of-business services for infrastructure of emerging sensor networks
Description:
This study presents a reliability-constrained adaptive-robust multi-resolution model for generation maintenance scheduling (GMS) problem considering the uncertainty sources of electricity demand, wind power generation, and equipment unavailabilities. In the proposed tri-level adaptive-robust model, a polyhedral uncertainty set is used to model the electricity demand and wind power generation fluctuations. In addition, equipment unavailabilities as discrete uncertainty sources are modelled in the reliability sub-problem where the expected energy not supplied is determined as a reliability criterion. Accordingly, the proposed model obtains a robust maintenance schedule for generating units immunised against the worst realisation of electricity demand and wind power generation while satisfying the reliability constraint considering equipment unavailabilities. Moreover, maintenance and operation periods are specifically modelled using different resolutions in the proposed multi-resolution GMS approach. To solve the proposed reliability-constrained adaptive-robust multi-resolution model, a new solution approach including Benders cut, reliability cut, and block coordinate descent method is presented. Numerical results on two test systems show the effectiveness of both the proposed GMS model and the proposed solution approach.
Description:
Solar generation systems are globally extending in terms of scale and number, which highlights the increasing importance of solar power forecast. In this paper, a day-ahead solar power prediction method is proposed including 1) a novel feature selecting/clustering approach based on relevancy and redundancy criteria and 2) an innovative hybrid-classification-regression forecasting engine. The proposed feature selecting/clustering approach filters out irrelevant features and partitions relevant features to two separate subsets to decrease the redundancy of features. Each of these two subsets is separately trained by one forecasting engine and the final solar power prediction of the proposed method is obtained by a relevancy-based combination of these two forecasts. The proposed forecasting engine classifies the historical data based on the learnability of its constituent regression models and assigns each class of training samples to one regression model. Each regression model predicts the outputs of the test samples that belong to its class. The effectiveness of the proposed solar power prediction method is illustrated by testing on two real-world solar farms.