【作者】 刘思革；

【导师】 周孝信；

【作者基本信息】 上海交通大学 ， 电力系统及其自动化， 2009， 博士

【摘要】 基于风险评估理论的输电网络扩展规划研究,是在传统的确定性规划方法中引入风险评估理论和技术,其目的是通过在规划过程中增加新的考虑因素,对确定性规划方案进一步进行风险和经济的综合分析,选择得出最佳方案。本文探讨了风险评估理论在输电网规划中的应用,建立了电网规划模型,给出了电网规划方案的风险评估方法。对电压安全风险、节点停电风险和投资风险进行了量化,并对相关不确定性因素进行了模糊化处理,解决了传统确定性规划方法不能够准确反映电网系统中所具有随机性、概率性和不确定性的问题。本文的主要研究成果如下:(1)提出引入风险评估技术的单阶段和多阶段输电网规划模型。目标函数为规划方案的总投资成本和运行成本之和最小,满足常规的输电网规划约束条件。具体计算方法为,采用改进遗传算法求解,并采用风险评估技术对已获得的规划方案集逐一进行风险评估,根据确定性成本和风险费用选择最优方案。(2)提出考虑电压安全风险的输电网规划方法。由于电压安全准则是规划的基本准则,在前述引入风险评估的输电网规划方法基础上,有必要引入电压安全裕度指标。以电压安全裕度和负荷削减费用为目标函数,建立了多目标最优负荷削减模型,并采用连续模糊线性规划法求解。根据确定性成本和考虑电压安全裕度指标后的风险费用选择最优方案。(3)提出考虑模糊不确定性因素的概率规划方法。在引入风险评估的输电网规划方法基础上,考虑相关不确定性因素模糊化的问题。规划模型以模糊投资成本和模糊运行成本之和最小为目标函数,满足电网安全运行约束。具体计算方法,采用改进遗传算法求解,并采用模糊风险评估技术对已获得的规划方案集逐一进行风险评估,根据模糊投资成本、模糊运行成本和模糊风险费用选择最优方案。(4)提出考虑负荷节点停电风险的输电网规划方法。采用基于调查法的节点停电损失模型来衡量节点的停电损失水平。规划模型的目标函数为输电投资成本和运行成本之和最小,满足电网安全运行约束和节点停电风险的随机约束。具体计算方法,采用蒙特卡罗模拟和改进遗传算法相结合的方法求解该模型,根据给定置信水平下的目标函数值最小确定最优方案。(5)提出考虑条件风险价值评估(CVaR)的输电网规划方法。将节点功率的不确定性作为随机扰动,通过概率最优潮流求取该扰动下输电投资收益的随机分布,进而计算CVaR风险指标。规划模型的目标函数为CVaR风险指标、输电投资成本和越限惩罚之和最小,满足电网安全运行约束和输电投资收益率期望值约束。具体计算方法,采用改进遗传算法求解规划模型,并采用随机扰动法评估电网风险水平,根据给定置信水平下的目标函数值最小确定最优方案。本文最后对算例进行统一的分析和比较,比较各种方法获得的规划结果间的差异,分析总结各规划方法的优缺点和侧重点,指出基于风险评估理论的规划方法和确定性规划方法的区别,并验证所提出的规划模型和求解算法的可行性、有效性和实用性。更多还原

【Abstract】 Transmission expansion planning problems based on risk assessment theories and techniques use comprehensive analysis of economic and risk to search optimum solution of transmission expansion planning with the consideration of new constrains.This dissertation has discussed the risk assessment theory and its application on transmission expansion planning. According to the risk calculation of voltage security, outage and investment and fuzzification of uncertainty factors, the new planning methods accurately reflected the characteristics of random and probabilistic in transmission systems that can not be found by deterministic methods. The main contributions of this dissertation are as follows:(1) Improved single-stage and multiple-stage transmission network planning models integrating with risk assessment techniques are proposed. Transmission investment cost and operation cost are involved in the proposed deterministic transmission planning models and an improved genetic algorithm is applied for solving. The risk assessment techniques are used to evaluate each solution that is obtained by the deterministic approach. According to the sum of deterministic cost and risk cost of each plan, the optimal plan is obtained.(2) A novel transmission planning approach that considers voltage security is proposed. Voltage security is the basic criterion for transmission system, so the voltage stable index (VSI) is proposed and applied on the risk assessment of transmission planning. An optimal load curtailment model that includes the security margin and cost is proposed that can be solved by a successive fuzzy linear programming method. The optimal transmission plan is decided by the sum of deterministic cost and risk cost.(3) A novel transmission planning model that considers uncertainties and integrates risk assessment techniques is proposed. Fuzzy transmission investment cost and fuzzy operation cost are involved in the transmission planning model and an improved genetic algorithm is applied for solving. The risk assessment techniques considering uncertainties are used to evaluate each solution that is obtained by the provirus method. According to the sum of fuzzy transmission investment cost, fuzzy operation cost and fuzzy risk cost, the proposed approach decides the optimal plan.(4) A novel transmission planning approach that considers outage risk of load bus is proposed. The model of outage cost is proposed to evaluate the outage risk level of load bus. The objective of transmission planning is the sum of investment cost and operation cost, integrating with security operation constrains and chance constraint of outage risk of load bus. The Monte Carlo simulation method and improved genetic algorithm are combined to solve this model and decided the optimal planning scheme according to a certain trust level.(5) A novel transmission planning approach that considers conditional values at risk (CVaR) is proposed. The CVaR index is adopted to evaluate the transmission investment risk. In the analysis of CVaR, the uncertainty of nodal power is taken as a random disturb and caused by which, the random distribution of transmission investment profit is obtained by solving the probabilistic power flow. After the CVaR index is adopted in objective functions and limitation of rate of return on transmission investment is taken as one of constraints, the transmission planning model is built up and solved by genetic method to achieve the optimal transmission planning.Analysis and comparison on the test cases illustrate the differences between the proposed methods and the differences between risk methods and deterministic methods and prove the feasibility, validity and practicability of the proposed planning models and solving algorithms in this dissertation.更多还原