Research On Models And Methods Of Generation Scheduling Optimization Incorporating Low-Carbon Factor

Based on low-carbon environment, generation scheduling are being promoted as a way of considering carbon cap-and-trade and carbon tax, balancing robust profit and price risk and incorporating wind farm. The aim is to reduce carbon emission, stabilize wind fluctuation and reduce generation cost. However, the traditional optimal dispatch schemes are unable to satisfy the requirement of power systems stability and economy, due to utmost complex elements such as low-carbon policy, electricity price risk and wind power fluctuation. Thus, a flexible, reliable and economic optimal dispatch approach for low-carbon environment is proposed in this thesis. Firstly, the strategy of solving the cost-based unit commitment (CBUC) and profit-based unit commitment (PBUC) models was studied. Secondly, the generation scheduling models considering carbon cap-and-trade and carbon tax were proposed, and the solving algorithms are developed. Thirdly, based on robust profit and value at risk, the generation scheduling models considering price fluctuation were presented. Finally, wind farm diversification was applied to stabilize wind power fluctuation, and the unit commitment based on wind power fluctuation and carbon cap-and-trade was established.This thesis contains six chapters. Chapter one is introduction. Chapter two to five are devoted to the main works. Chapter six summarizes the thesis. The main achievements are listed as follows:Chapter two discussed cost-based unit commitment and profit-based unit commitment models, and studied the strategy of solving the two models. Firstly, a more compact second-order cone relaxation method was used to establish mixed integer second-order cone programming (MISOCP) form of two models. Further, a mixed integer linear programming (MILP) form was casted by using second-order cone linearization method. The simulation results showed that the MILP model in this paper is a better formulation to solve the UC problem, and can obtain high quality solutions.Chapter three proposed carbon cap-and-trade and carbon tax generation scheduling models and methods, and compared the carbon cap-and-trade and carbon tax mechanism. Firstly, the carbon cap-and-trade cost-based unit commitment (CC-CBUC) model and carbon cap-and-trade profit-based unit commitment (CC-PBUC) model were proposed. The proposed models considered not only carbon emission allowance constraint but also carbon emission trading. A new method was proposed for solving the CC-CBUC and CC-PBUC problems based on piecewise linearization and second-order cone linearization. Secondly, the carbon tax profit-based unit commitment (CT-PBUC) model was proposed. The proposed model can analyze the effects of GENCO’s profit and emission in different carbon tax scenarios. By using the second-order conic linearization approximation method, the CT-PBUC model was transformed into a MILP formulation. Finally, the carbon cap-and-trade and carbon tax mechanism were compared from reducing emission and restraining demand. The effectiveness and validity of proposed model and method were illustrated by 10 unit systems over 24 hours.Chapter four proposed conditional robust profit and carbon cap-and-trade generation scheduling model and method. Firstly, considering the influence of GENCO’s profit because of electricity price fluctuation, the profit was defined by conditional robust profit (CRP) and worst-case conditional robust profit (WCRP). The PBUC model was extended to conditional robust profit-based unit commitment (CR-PBUC) and worst-case conditional robust profit-based unit commitment (WCR-PBUC). Furthermore, the conditional robust profit and carbon cap-and-trade PBUC (CRPCC-PBUC) model was proposed. The different price scenarios were simulated by Monte Carlo method. The proposed models were transformed as MISOCP formulation to solve. The effectiveness and validity of proposed model and method were illustrated by 10-unit 24-period system. GENCO can use the model to optimize power generation, and balance between profit and risk, and sell or buy quota strategy according to the carbon price.Chapter five proposed wind farm diversification method to stabilize the wind fluctuation, and established wind farm carbon cap-and-trade cost-based unit commitment (WTCC-CBUC) model. Firstly, the thesis applied Copula function and mean-variance model to study the wind speed spatial correlation and optimal wind power allocation. But traditional mean-variance wind farm diversification method is sensitive to the input data. To overcome the problem of lack of robustness, a novel wind farm diversification method was proposed based on robust optimization model. Under box and ellipsoidal uncertainty structures, the proposed robust optimization model can be formulated as a coupled problem composed of a linear programming problem and a conic quadratic programming problem. This model can be efficiently solved by interior point algorithm. Finally, the wind farm carbon cap-and-trade cost-based unit commitment model was proposed. Using second-order cone and piecewise linearization method, the proposed model was formulated as a mixed integer linear programming. The 10 thermal power units and 1 wind farm simulation verified the effect of systems operation cost and carbon emission because of wind power fluctuation and carbon price variation.