【作者】 崔勇；

【导师】 黄元生；

【作者基本信息】 华北电力大学 ， 技术经济及管理， 2014， 博士

【摘要】 电力市场环境中,无功服务已经被认定为最重要的服务之一。首先,电力系统的稳定性需要通过适量且分布合理的无功服务予以支持。其次,无功服务的优化配置利用可以进一步改善有功的合理分配和交易,以达到实现提升电力系统的效率的目的。再者,电力市场环境下用于降损的最直接手段就是调节无功分布,通过无功功率的合理分配,可以实现降低系统的电能损耗。从电力服务角度上看,无功支持可以改善电压波形,减少系统中出现的过电压对电网安全和设备绝缘寿命的影响。随着分布式新能源发电不断并入电网中运行,由于分布式能源分散性、随机性和间歇性的特点,出现了谐波污染、电压失稳及电压闪变等问题,从而导致电网的安全稳定面临新的考验,如何充分调配并发挥无功资源设备对电力系统稳定调节的作用显得十分重要,同时为实现电力系统中各供应商的经济利益和国家能源节约的整体效益,采用无功定价及无功优化措施实现激励供应商积极改善无功服务支持的目标,最终通过资源的整合调配和高效利用达到经济效益最优化。本文针对分布式能源不断并网运行的背景下,基于电力市场的特点对无功负荷预测、无功定价及无功优化模型问题展开了研究,并对无功优化后的效果进行了评价,具体研究内容包括以下几个方面：(1)在对负荷预测研究的基础上,综合分析各种预测方法,结合无功负荷的特点提出了一种时间序列与支持向量机组合无功预测模型,利用该模型对无功负荷进行预测,采用均方根相对误差RMSRE对预测效果进行评估,算例证明该组合模型精度高,对无功负荷预测有较强的适用性。(2)充分考虑分布式能源接入电网后系统安全稳定问题,在无功定价经典模型的基础上,建立了基于电压稳定的无功定价新模型,通过计算仿真验证了新型模型具有合理引导消费和提升电力系统安全稳定的双重效果。新型模型具有非线性特点,建立新型改进算法求解(结合免疫算法的改进多中心-校正内点法),通过多种算法对比,该计算准确度高且速度快。(3)本文建立了考虑分布式新能源发电机组组合的无功经济调度模型,利用该模型对电力系统无功经济调度优化计算,以IEEE-14节点为例,分析计算了在考虑或不考虑网络潮流约束及N-1故障集等多种情况下的发电机出力与无功优化结果,得到了满足电力系统安全约束且经济性明显的发电计划。(4)文中最后对电网无功优化的效果进行了评价,综合分析了多种评价方法,结合无功服务的特点以及数据特点,构建了无功效果评价的指标体系,采用了基于自主决策视角和数据包络分析结合的方法对无功优化的效果进行了综合评价,计算结论表明无功优化后的效果是明显的。更多还原

【Abstract】 A reactive service is generally regarded as one of the most important services in the electricity market services. Firstly, the stability of the power system needs to be supported by an appropriate amount and rational allocation of reactive services. Secondly, optimizing reactive service configuration can further improve the rational allocation and active transactions for the purpose of improving the efficiency of the power system. Furthermore, the most direct means for reducing losses in the electricity market environment is to regulate the distribution of reactive power, which can be achieved to reduce the power losses of the system. From the point of view of electricity services, reactive support can improve the voltage waveform, diminishing the impact on the grid overvoltage for safety and equipment insulation life. With the new distributed energy generation and constantly running into the grid, due to the distributed energy dispersed, random and intermittent characteristics, appeared harmonic pollution, voltage instability and voltage flicker and other issues, leading to grid security stability are facing new challenges, how to make the deployment of resources and play equipment for reactive power system stability is very important regulatory role, as well as economic benefits and overall effectiveness of the power system of the country each supplier of energy saving, the use of reactive power pricing and Reactive Power Optimization measures to achieve an incentive for vendors to actively improve reactive support service goals, and ultimately economic optimization by integrating the deployment and efficient use of resources to achieve. To sum up the electricity market environment reactive and reactive pricing optimization model is a very important issue.In this study, distributed energy grid run continuously in the background, based on the characteristics of the electricity market for reactive power load forecasting, pricing and reactive power optimization model launched a research question, which lead a evaluation about the effect on reactive optimizing.The study includes the following aspects:(1) On the basis of load forecasting study, a comprehensive analysis of the various forecasting methods, presents a forecasting model that combine a time series with support vector machines, using the model to predict the reactive load, and using the relative root mean square error of prediction RMSRE to assess the effect, examples prove that the combination of high accuracy of the model of reactive power load forecasting has strong applicability.(2) After full consideration of distributed energy systems connected to the grid security and stability issues, based on the classic reactive pricing model, established a new reactive pricing model based on the stable voltage, by calculating a new simulation model to guide rational consumption and enhance the power system security and stability of double effect. The new model has a non-linear characteristics, the establishment of a new algorithm into (a combination of immune algorithm to improve multi-center-Correction interior-point method).This method has high accuracy and efficiency by comparing a variety of algorithms.(3) This paper established a reactive economic dispatch model that considers new energy generators distributed portfolio, using the model of economic dispatch reactive power optimization calculation to IEEE-14node example, the analysis calculated the trend with or without considering the network constraints and generator reactive power output and a variety of situations, such as N-1fault set optimization results, got to meet the power system security constraints and generating significant economic plan.(4) Finally, the paper reactive power optimization results were evaluated, a comprehensive analysis of a variety of evaluation methods, combined with the characteristics of reactive services and data features, build reactive effect evaluation index system, using a perspective based on their own decisions and packet DEA method of combining the effect of reactive power optimization conducted a comprehensive evaluation, the conclusions show that the effect of reactive power is obvious after optimizing reactive.更多还原