Studies On Early Warning And Decision Making For Large-Scale Wind Power Ramping

With the energy issues and environmental issues becoming constantly prominent within the scope of the global,it becomes increasingly urgent to seek sustainable energy as new kinds of power supply.As a kind of sustainable energy generations with mature technology and large development potential,wind power generation has been developed rapidly in recent years,its penetration in power system continues to improve.Wind power has the characteristics of economy and cleanliness,but at the same time,the inherent volatility and uncertainty of wind power bring new challenges to the operational safety and stability of power system.Especially in China,the development of wind power has the characteristics of large-scale and high-concentration due to the imbalance of geographical distribution of wind energy,which increases the potential possibility of occurrence of wind power ramp events(WPREs).With the characteristics of strong volatility and strong uncertainty,WPREs cause the power system losing plenty of power supply in a short time and seriously affect the active power balance,which not only places higher demands on the adequacy and flexibility of power system,but also seriously affects the effectiveness of existing wind power control measures.Therefore,in order to accurately perceive and effectively prevent and control the adverse effects of WPREs on power system and to improve the reliability and economy of power system operation in the process of WPREs,it has important theoretical value and urgent practical significance to research the early warning and decision making on WPREs.The purpose of early warning and decision making on WPREs is to analysis and judge the harm brought by WPREs according to the forecast information and further issue a corresponding warning signal for power system operators before the occurrence of WPREs,and to guide the operators to develop the optimal control decision strategy,which can reduce the adverse effects caused by WPREs and improve the reliability and economy of power system operation in the process of WPREs.In this thesis,based on a comprehensive summary of the existing related researches,an in-depth research on risk assessment,early warning and coordinated scheduling strategy of WPREs has been conducted.The whole research process follows the ideas that from simple to complex and give consideration to both theory and practice,many theories including probability theory,scenario reduction,interval analysis and information gap decision theory(IGDT)are introduced,multifaceted contents including model establishment,formula derivation and optimization problems solving are involved,furthermore,the effectiveness of proposed methods are verified by simulated and practical cases.Main contributions and innovations are described as follows:1)An assessment method on the adverse impact caused by WPREs on operation adequacy of power system is proposed.Based on the analysis on characteristics and status of prediction for WPRE,a WPRE model is proposed to establish the quantitative relationship between charactering quantities and wind power curve.The uncertainties of WPRE and load forecasting are considered simultaneously,and the scenario reduction method is used to generate the typical scenarios set of net load during WPRE,which can reduce the computation load without losing precision.An assessment indicators system is established to reflect the operation adequacy of power system,including overall adequacy,loss of load probability,loss of load expectation and buffered loss of load probability,and mathematical calculating method of buffered load loss probability is derivated.Futhermore,the system status can be categorized as being adequacy,load loss and marginal according to the mathematical and physical meaning of these indicators,the corresponding probability of each status and severity of load loss can be calculated respectively.The assessment results of typical events based on the Gansu grid verify the effectiveness and rationality of proposed assessment method and indicator system,and the roles of ramp rate of conventional units and proportion of reserve capacity in preventing the adverse effects of WPRE are analyzed in further.2)A multi-level early warning method for WPREs is proposed.According to the characteristics of WPRE,interval numbers are introduced to describe the uncertainty of wind power.considering the impact of large volatility of WPRE on power control,the permissible intervals of ramping amplitude are analyzed to determine the boundaries between different warning levels,which are corresponding to different power control measures(automatic control including primary and secondary frequency regulation,scheduling control with spinning and non-spinning reserve,forced control with load shedding)to keep the power balance with the static frequency deviation maintained within the allowable range.By comparing these boundaries with the prediction results of wind power,the multi-level early warning for WPRE can be realized.Moreover,by using interval ranking method,the probability distributions of different warning levels can be provided.Simulation results and comparative analysis demonstrate that,both the boundaries and probability distributions of different warning levels,with clear physical meaning,can be provided by the proposed method,the simple and rapid characteristics are conducive to real-time scrolling operation,which all verify the effectiveness of the proposed method.3)An information gap decision theory(IGDT)robust model-based coordinated scheduling strategy method for WPREs is proposed.Due to the strong volatility,scheduling strategy with only conventional units can not meet the requirement of power balance.Based on analysis on costs and constraints of all participants(conventional units,wind farms and demand side),a coordinated scheduling of these participants is proposed,the adverse effects of wind power ramp events can be reduced and the economy of strategy can be ensured.Considering the strong uncertainty of WPREs,IGDT is introduced to analyze the quantitative relationship between the expected cost of decision makers and the maximum allowable uncertainty degree of wind power,the worst ramping scenario is established furthermore to propose a scheduling model solved by IGDT method.Simulation results on two cases under different scenarios demonstrate that the proposed method can provide strategies considering both robustness and economy.Corresponding strategy can be made according to the expected cost of decision makers and forecast information of WPREs,which can ensure that:the cost of scheduling strategy is not larger than the expected cost when the wind power fluctuating in the scope of corresponding robust interval,and when the actual ramping scenario is worse,it can ensure that the scheduling strategy has a better economy.