Model And Algorithm For Risk-Based Dispatch In Power Systems Considering Credible Contingencies

The traditional Security-Constrained Dispatch (SCD) is a deterministic model, which could neither forecast nor control the operating risk of a power system. On the contrary, the solutions made by Risk-based Dispatch (RD) that considering credible contingencies are required not only meeting the security constraints, but also controlling the operating risk within the maximum tolerable limit. RD is therefore very essential to improve operators’cognitive ability of power systems’operating state, and could increase the capability of power systems to prevent and withstand the operating risk. This dissertation focuses on the model and algorithm of risk-based dispatch that incorporating the base case and post-contingency short/long-term emergency state, the main contributions includes five parts:(1) An off-line multi-objective based method is presented to assess economy, security and risk of the dispatch strategies. It is observed that results obtained by the traditional security-constrained dispatch methods are only special cases among the Pareto front. The proposed method could serve as a novel decision-making tool for operators. Based on the Pareto front, operators could clearly observe and easily analyze the generation cost, security degree and risk level of the various dispatch solutions, thus assisting to make a more security, cheaper and lower risk generation schedule.(2) We propose a Preventive and Corrective Coordinated Risk-based Dispatch (PCCRD) that minimizing the expected risk of the total credible contingencies. The preventive control of generating units is extended from the normal state to the post-contingency short-term emergency state, so as to mitigate operating risk during the post-contingency emergency state, as well as easing the burden of operators that they had to take corrective actions in order to remove the security issues. The effect of weather conditions on the outage probability of transmission lines is included in the model, therefore the RD model is able to trace the system’s operating risk under different weather state. Under a high risk adverse or extreme weather state, PCCRD could dynamically adjust the units’output in order to keep the operating risk within the limits. The solution obtained by the PCCRD consists of the base case preventive generating schedule, as well as the post-contingency generation resdispatch and load curtailment.(3) Aiming at reflecting operators’risk-averse preference to the worst-case contingency and keeping the risk value of the worst-case contingency acceptable, an adjustable Risk-based Dispatch via Worst-Case optimization (WCRD) is proposed. The model is formulated as a min-max-min programming with multi-state control actions. Risk factors are both introduced to the objective function and the transmission constraints, a tradeoff between system economy and operating risk can thus be achieved by appropriately setting their values. The proposed approach mitigates risk of the worst-case contingency and allows the flexibility of adjustable preventive/corrective solutions.(4) An Enhanced Risk-based Dispatch (ERD) is proposed that incorporates the fast response capability that distributed battery energy storage could be used to implement post-contingency corrective control actions. Immediately after a contingency, the injections of distributed batteries could be adjusted to alleviate overloads and reduce flows below their short-term emergency rating. This ensures that the post-contingency system remains stable until the operator has redispatched the generation. ERD is solved using a Benders decomposition based iterative algorithm. An improved convergence rule is presented which could check the optimality of operating risk but without loss of economy.(5) Oriented to the hierarchical multi-area power systems with multi-level dispatch centers, a Coordinated Decentralized Risk-based Dispatch (CDRD) approach based on the analytical target cascading technique is proposed. We analyzed the paradigm of the CDRD and build an area-decomposition scheme. The upper- and lower-level dispatch problems are respectively formulated as a master problem and a set of sub-problems, where the master problem handles the coordination of exchange power flows across the tie-lines, while each sub-problem determines local dispatch solutions. The master problem is solved iteratively with the sub-problems until the convergence requirement is met. CDRD could thus manage the operating risk of each area power system dispersedly and autonomously, meanwhile the coordination mechanism guarantees the economy of the whole system.