风电并网电力系统随机潮流计算方法及安全评估应用

发布时间：2018-06-01 12:46

本文选题：风电 + 随机潮流　；参考：《华北电力大学(北京)》2014年博士论文

【摘要】：随着电网运行环境的改变,系统随机性不断增强,尤其是风电这一非常规随机因素的大规模并网,系统状态将更加复杂多变,这对电网分析及安全评估工作提出了更高要求。随机潮流(Probabilistic Load Flow, PLF)可综合考虑多种随机因素分析出系统状态量的概率分布信息,从而更深刻地揭示系统运行特性。因此,随机潮流可作为新环境条件下电网分析的一种有力工具。鉴于此,本文以风电并网为背景,对电力系统随机潮流计算方法及其安全评估应用问题展开研究。本文将风电并网电力系统的随机潮流问题分为中长期和短期两类。其中,中长期随机潮流(PLF for Long and Medium Term, LMT-PLF)计算时考虑风电出力从零到满发随机波动,短期随机潮流(PLF for Short Term, ST-PLF)计算中风电出力随机性则处理为风电功率预测误差随机性。围绕上述两类随机潮流的计算方法及安全评估应用,开展的主要工作归纳如下。针对风场出力在中长期内波动可能较大,现有解析法由于采用线性化PLF模型而难以确保LMT-PLF结果精度,提出了一种基于多点线性化模型的LMT-PLF解析计算方法。建立了适用于风电并网系统的多点线性化LMT-PLF模型；采用半不变量法求解所建模型,推导得出了多点线性化条件下系统状态量半不变量求解方法,并通过引入C型Gram-Charlier级数,避免了在风电的系统中采用A型Gram-Charlier级数逼近状态量概率密度函数会出现负值的问题。算例分析表明,所提方法与现有解析法相比,结果精度得到了较大提高,且对风电并网规模增加具有较好的适应性。鉴于PLF现有方法均采用单一平衡节点而难以适应风电并网系统中长期内可能存在的较大范围出力波动,将动态潮流思想引入PLF,研究了一种计及频率调节的LMT-PLF计算方法。方法首先结合系统频率调整方式,建立系统注入随机波动时动态不平衡功率的分配模型。然后,与现有解析法PLF模型联合推导,得出了计及调频作用的PLF计算模型,且计及了机组调频能力不足的影响。针对模型特点,提出了混合求解算法。算例结果表明,所提方法可保证网内各电厂出力按系统实际在合理范围内波动,有效避免了现有方法用于LMT-PLF时平衡机组出现功率越限甚至负功率的情形,从而亦保证了LMT-PLF结果的正确性。为解决PLF当前方法均假定风电出力等随机变量的概率分布模型已知而实际系统却难以提供这一矛盾,基于序列运算理论提出了ST-PLF的一种实用计算方法。方法首先结合PLF问题的特点,分析了现有序列运算理论直接应用于PLF的可行性,并据此对现有序列运算进行了扩展；然后基于所提扩展运算,以扩展概率性序列描述系统随机性并基于历史数据实现随机变量的序列化,通过系统各输入变量对应序列的直接运算快速分析潮流的概率分布,克服了现有方法对基础数据要求过高的不足。测试表明,所提方法可有效实现系统注入随机量概率分布未知情形下的ST-PLF计算,提高了现有方法的可实施性。为有效评价风电并网系统的电压随机波动情况,从描述电压随机分布的特征出发,基于PLF结果分别定义了电压分布指数、偏度指数、保持指数三个电压波动评价指标。最后,以LMT-PLF为例验证所提方法,结果表明,所提指标可较好地从整体和局部两个方面反映风电并网系统的电压随机分布情况。以测试系统为例,详细分析了风电场尺度参数、形状参数、风电接入位置等因素对电网电压波动的影响,为风电的合理并网提供参考。为快速有效评估风电并网系统的静态安全水平,将PLF引入安全评估。方法以解析法PLF为电网状态分析手段,保证评估高效性。从概率和风险两个角度定义了基于PLF结果的安全指标,实现系统安全的量化评价。基于元件级安全指标进一步给出了系统薄弱环节辨识方法。最后,以ST-PLF为例对方法进行验证,结果表明,与传统不确定性评估方法相比,所提方法在保证评估结果准确性的同时能大幅提高评估效率,可为间歇性能源并网系统的安全评估提供借鉴。
[Abstract]:With the change of the operating environment of the power grid, the randomness of the system is constantly enhanced, especially the large-scale grid connection of the wind power, an unconventional random factor, and the system state will be more complex and changeable, which puts forward higher requirements for the analysis of power grid and safety assessment. The random flow (Probabilistic Load Flow, PLF) can consider a variety of random factor analysis The probability distribution information of the system state quantity is given to reveal the characteristics of the system more deeply. Therefore, the random flow can be used as a powerful tool for the analysis of the power grid under the new environment conditions. The random power flow problem of electric grid connected power system is divided into two classes of medium and long term. In the middle and long term random flow (PLF for Long and Medium Term, LMT-PLF) calculation, the wind power output from zero to full random fluctuation is considered, and the short-term random flow (PLF for Short Term, ST-PLF) is used to calculate the randomness of the stroke power output. The main work carried out around the two types of stochastic power flow and the application of the safety assessment are summarized as follows. In view of the possibility that the wind field output may fluctuate in the middle and long term, the existing analytical method is difficult to ensure the accuracy of the LMT-PLF results because of the linear PLF model, and a LM based on the multi point linearization model is proposed. T-PLF analytical calculation method. A multi point linearized LMT-PLF model suitable for wind power grid connected system is established. The semi invariants method is used to solve the built model. The method of solving the semi invariants of the system state quantity under multi point linearization is derived, and the C Gram-Charlier series is introduced to avoid the A Gram-Ch in the wind power system. The arlier series approximated the problem of the negative value of the probability density function of the state quantity. The numerical example shows that the proposed method is more accurate than the existing analytical method, and has a better adaptability to the increase of the scale of the wind power grid. In view of the fact that the existing methods of PLF use a single balance node, it is difficult to adapt to the wind power grid connected system. In the long run, a large range of force fluctuation may be present, the idea of dynamic power flow is introduced to PLF, and a LMT-PLF calculation method with frequency regulation is studied. First, the distribution model of dynamic unbalanced power is established by combining the system frequency adjustment mode with the system frequency adjustment. Then, the method is derived from the existing analytical PLF model. The PLF calculation model considering the function of FM is given and the influence of the lack of frequency modulation capacity of the unit is taken into account. A mixed solution algorithm is proposed for the characteristics of the model. The calculation example shows that the proposed method can guarantee the power plant output in the network to fluctuate in a reasonable range according to the actual system, and effectively avoids the presence of the existing method for the emergence of the balance unit. The correctness of the LMT-PLF results is guaranteed by the case of the power limit or even the negative power. In order to solve the probability distribution model of random variables such as the assumption of wind power output in PLF current methods, the actual system is difficult to provide this contradiction. Based on the sequence operation theory, a practical calculation method of ST-PLF is proposed. First, the method is combined with PLF. The characteristic of the problem is to analyze the feasibility of the existing sequence operation theory directly applied to PLF, and then extend the existing sequence operation. Then based on the extended operation, the randomness of the system is described by extending the probability sequence and the sequence of random variables is realized based on the historical data, and the sequence of the system is corresponding to the input variables. The direct operation can quickly analyze the probability distribution of the power flow, and overcome the shortage of the existing methods on the basic data. The test shows that the proposed method can effectively realize the ST-PLF calculation under the unknown probability distribution of the system injected random quantity, and improve the feasibility of the existing methods. In the case of describing the characteristics of the random distribution of the voltage, the voltage distribution index, the deflection index and the index of maintaining the three voltage fluctuation are respectively defined based on the PLF results. Finally, the proposed method is verified with LMT-PLF as an example. The results show that the proposed index can better reflect the voltage of the wind power grid connected system from the whole and the local two aspects. Taking the test system as an example, the influence of wind electric field scale parameters, shape parameters, wind power access position and other factors on the voltage fluctuation of the power grid is analyzed in detail, which provides a reference for the rational grid connection of wind power. The static security level of the wind power grid system is quickly and effectively evaluated, and PLF is introduced to the safety assessment method, and the analytical method PLF is used as the electricity. Network state analysis means to ensure the efficiency of evaluation. From two angles of probability and risk, we define the security index based on the results of PLF and realize the quantitative evaluation of system security. Based on the component level security index, the identification method of the system weak link is given. Finally, the method is verified with ST-PLF as an example. The results show that the method is uncertain with the traditional method. Compared with the method of sex assessment, the proposed method can greatly improve the evaluation efficiency while ensuring the accuracy of the evaluation results, and can be used for reference for the safety assessment of the intermittent energy grid system.
【学位授予单位】：华北电力大学(北京)
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TM614;TM744

【相似文献】