电铁牵引负荷扰动电流概率分布模型研究

发布时间：2018-11-16 19:19

【摘要】：电气化铁路牵引负荷是电力系统中重要的单相负荷,同时,也是电力系统主要的谐波源之一,随着电能质量监测系统的不断完善,研究基于实测数据的电铁牵引负荷扰动电流概率分布模型具有重要意义。本文基于实测数据的分布直方图通过曲线拟合的方法求解牵引负荷的概率分布,本文所提方法可用于求解电压、电流等不同测量数据的概率分布。针对最大熵法可能因陷入局部最优而导致拟合效果差的问题,提出了一种基于实测数据分布特征的改进最大熵法,使得所求概率分布更加符合实际的分布特性;针对云变换法可能出现不能收敛的问题,提出了考虑负数区间拟合的云变换法,将云变换法用于负数部分的拟合,降低了云变换法对参数的要求,提高了算法的收敛性;最后根据云变换法算法精度高,而最大熵法概率密度方程简单的特点,提出了一种最大熵法和云变换法联合求解概率分布的云熵法,该方法既有云变换法的高精度,又有概率密度函数较为简单的特点。文中采用拟合误差参数对以上方法的拟合效果进行了评价,并采用不同数据验证了所提出方法的正确性。本文提出利用云熵法求得的概率密度函数组成实测数据的测试数据库,克服了大量实测数据占用空间大、使用不便等问题;利用牵引站的列车运行时刻表等信息,提出了一种预测新建站或未监测牵引站谐波的方法,基于云熵法求得预测数据库作为测试数据库的补充。基于某区域电网搭建ETAP仿真模型,利用云熵法建立的概率分布模型求得牵引负荷某概率下的负荷水平作为仿真参数,分析了牵引负荷对该区域电网电能质量的影响,指出了可能出现的电能质量问题,给出了相应的治理建议。
[Abstract]:The traction load of electrified railway is an important single-phase load in the power system, at the same time, it is also one of the main harmonic sources in the power system. With the continuous improvement of the power quality monitoring system, It is of great significance to study the probability distribution model of electric traction load disturbance current based on the measured data. Based on the distribution histogram of measured data, the probability distribution of traction load is solved by curve fitting method. The method proposed in this paper can be used to solve the probability distribution of different measurement data such as voltage and current. Aiming at the problem that the maximum entropy method may lead to poor fitting effect due to falling into local optimum, an improved maximum entropy method based on the distribution characteristics of measured data is proposed, which makes the probability distribution more in line with the actual distribution characteristics. In view of the problem that cloud transformation method may not converge, a cloud transformation method considering negative number interval fitting is put forward. The cloud transformation method is used to fit the negative part, which reduces the requirement of cloud transform method to parameters and improves the convergence of the algorithm. Finally, according to the high accuracy of cloud transformation method and the simple probability density equation of maximum entropy method, a cloud entropy method is proposed to solve the probability distribution by combining the maximum entropy method and cloud transformation method. This method has the high accuracy of cloud transformation method. The probability density function is simple. In this paper, the fitting effect of the above methods is evaluated by fitting error parameters, and the correctness of the proposed method is verified by different data. In this paper, the probability density function obtained by cloud entropy method is used to form the test database of measured data, which overcomes the problems of large amount of measured data occupying large space and inconvenient use. Based on the information of train running schedule of traction station, a method of predicting harmonics in new or unmonitored traction station is proposed. The prediction database is obtained by cloud entropy method as a supplement to the test database. Based on the ETAP simulation model of a regional power network, the probability distribution model established by cloud entropy method is used to obtain the load level under a certain probability of traction load as simulation parameter, and the influence of traction load on the power quality of regional power network is analyzed. The possible power quality problems are pointed out and the corresponding treatment suggestions are given.
【学位授予单位】：华北电力大学(北京)
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM711;TM74

【参考文献】