技术领域technical field
本发明涉及光伏发电技术领域，具体的说是一种基于光伏发电站精细化出力的并网退网选择方法。The invention relates to the technical field of photovoltaic power generation, in particular to a selection method for grid connection and grid withdrawal based on the refined output of photovoltaic power stations.
背景技术Background technique
随着全球化石能源的紧缺，清洁、可再生、蕴藏量大的太阳能得到了快速的发展。但由于太阳能具有间歇性、随机性与波动性的特点，光伏发电输出功率的波动势必给大规模电站集中并网带来困难。通过对光伏发电出力曲线进行科学描述，光伏发电出力曲线为：根据不同时刻采样点，采集的该时刻输出功率值，绘出的输出功率曲线。电网调度人员可以较为准确地掌握光伏出力特性规律，对平抑光伏出力波动，减小并网难度具有一定意义。因此，分析与掌握光伏发电的输出特性是解决光伏发电站集中并网问题的基础与前提。With the shortage of global fossil energy, clean, renewable, and abundant solar energy has developed rapidly. However, due to the intermittent, random and volatile characteristics of solar energy, the fluctuation of the output power of photovoltaic power generation will inevitably bring difficulties to the centralized grid connection of large-scale power stations. Through the scientific description of the photovoltaic power generation output curve, the photovoltaic power generation output curve is: the output power curve drawn according to the output power value collected at the time of sampling points at different times. Grid dispatchers can more accurately grasp the characteristics and laws of photovoltaic output, which is of certain significance for stabilizing fluctuations in photovoltaic output and reducing the difficulty of grid connection. Therefore, analyzing and mastering the output characteristics of photovoltaic power generation is the basis and premise for solving the problem of centralized grid connection of photovoltaic power stations.
随着大量光伏发电站投入运行，对光伏发电站出力的数据分析工作也逐渐开展，但研究尚未形成一套科学的出力特性描述体系。现有文献主要从单个电站出力特性量化描述与集群光伏发电站特性分析两个角度对光伏出力进行研究。在单个电站出力特性量化描述方面，有的文献提出平均波动强度、反向波动次数以及移动波动密度三个指标来对单个光伏发电站的出力波动性进行描述，虽然提出了出力波动的量化方法，但对波动性的计算仍不够精确。现有技术还指出光伏发电站可以通过光伏发电站中太阳能电池板的数目与分散因数来量化电站的波动特性，但研究结果只适用光伏发电站各太阳能电池板的大小、分布方向与空间间隔均相同的条件下，难以推广。在光伏发电站集群聚合特性研究方面，现有文献研究指出了随着光伏发电站集群规模的扩大，集群整体的出力将逐渐趋于平滑，但是缺少对集群电站出力平滑效应的定量描述。现有研究提出的光伏发电站出力描述指标较为笼统，大部分研究沿用了风电出力特性描述的思路，未考虑光伏独有的出力特性，因此有待对其进行进一步的研究。此外，光伏发电站集群聚合出力特性的研究尚停留在定性分析层面，缺乏定量研究。With a large number of photovoltaic power stations put into operation, the data analysis of photovoltaic power station output is gradually carried out, but the research has not yet formed a set of scientific output characteristic description system. The existing literature mainly studies the photovoltaic output from two perspectives: the quantitative description of the output characteristics of a single power station and the characteristic analysis of the cluster photovoltaic power station. In terms of quantitative description of the output characteristics of a single photovoltaic power station, some literatures propose three indicators, the average fluctuation intensity, the number of reverse fluctuations, and the moving fluctuation density, to describe the output fluctuation of a single photovoltaic power station. But the calculation of volatility is still not precise enough. The prior art also pointed out that the photovoltaic power station can quantify the fluctuation characteristics of the power station through the number and dispersion factor of the solar panels in the photovoltaic power station, but the research results only apply to the size, distribution direction and spatial interval of each solar panel in the photovoltaic power station. Under the same conditions, it is difficult to promote. In terms of research on the aggregation characteristics of photovoltaic power station clusters, existing literature studies have pointed out that with the expansion of the scale of photovoltaic power station clusters, the overall output of the cluster will gradually become smoother, but there is a lack of quantitative description of the output smoothing effect of cluster power stations. The output description indicators of photovoltaic power plants proposed by existing studies are relatively general. Most studies follow the idea of wind power output characteristics description, and do not consider the unique output characteristics of photovoltaics, so further research is needed. In addition, the research on the aggregated output characteristics of photovoltaic power station clusters is still at the qualitative analysis level, and there is a lack of quantitative research.
在现有技术中，在选择并网和退网的发电站时，都是随机选择。在并网时，若并网的发电站发电出力波动大，则并入将会造成对电网冲击，导致电网电压出现波动。退网时，若退出的发电网络为较为稳定的处理发电站，则会降低原电网的稳定性。并网退网随时可以发生，则现有技术中，发电站处理还停留在以天数为单位的阶段，不能满足并网退网需求。In the prior art, when selecting power stations for grid connection and grid disconnection, they are all randomly selected. When connecting to the grid, if the power generation output of the grid-connected power station fluctuates greatly, the integration will cause an impact on the grid, resulting in fluctuations in the grid voltage. When withdrawing from the grid, if the withdrawn power generation network is a relatively stable processing power station, the stability of the original grid will be reduced. Grid connection and withdrawal can happen at any time, but in the existing technology, the processing of power stations is still in the stage of days, which cannot meet the needs of grid connection and withdrawal.
发明内容Contents of the invention
针对上述问题，本发明提供了一种基于光伏发电站精细化出力的并网退网选择方法，从单个光伏发电站出力特征提取与集群聚合出力特性分析两个层面对光伏发电的输出特性进行研究。在单个电站层面，对其出力曲线进行特征提取，进一步对局部进行精细化描述，最终得到并网退网光伏发电站的选择，减小并网退网对电网的冲击，使电网供电更加稳定。In view of the above problems, the present invention provides a grid-connected and off-grid selection method based on the refined output of photovoltaic power stations, and studies the output characteristics of photovoltaic power generation from two levels: extraction of output characteristics of a single photovoltaic power station and analysis of cluster aggregate output characteristics . At the level of a single power station, the feature extraction of its output curve is carried out, and the local fine description is further carried out, and finally the selection of grid-connected and de-grid photovoltaic power stations is obtained, which reduces the impact of grid-connected and de-grid on the grid, and makes the power supply of the grid more stable.
为达到上述目的，本发明采用的具体技术方案如下：In order to achieve the above object, the concrete technical scheme that the present invention adopts is as follows:
一种基于光伏发电站精细化出力的并网退网选择方法，其关键技术在于包括以下步骤：A grid-connected and off-grid selection method based on the refined output of photovoltaic power stations, the key technology of which is to include the following steps:
S1：选定预选并网光伏发电站，得到光伏发电站集群，并获取光伏发电站集群数据和光伏发电站集群中所有光伏发电站的历史出力数据；S1: Select the pre-selected grid-connected photovoltaic power station, obtain the photovoltaic power station cluster, and obtain the photovoltaic power station cluster data and the historical output data of all photovoltaic power stations in the photovoltaic power station cluster;
S2：选取光伏发电站的日出力特征指标，并分别计算单个光伏发电站的所有日出力特征指标值；S2: Select the daily output characteristic index of the photovoltaic power station, and calculate all the daily output characteristic index values of a single photovoltaic power station;
S3：对光伏发电站日出力特征指标进行划分，得到子时段出力特征指标值；S3: Divide the daily output characteristic index of the photovoltaic power station to obtain the sub-period output characteristic index value;
S4：设定所有特征指标值的权重，得到每个光伏发电站并退网冲击系数，根据并退网冲击系数，对并网和退网的光伏发电站进行预选择，得到最终光伏发电站集群；S4: Set the weights of all characteristic index values to obtain the impact coefficient of grid connection and withdrawal of each photovoltaic power station. According to the impact coefficient of grid connection and withdrawal, pre-select the photovoltaic power stations that are connected to the grid and withdraw from the grid to obtain the final cluster of photovoltaic power stations ;
S5：根据步骤S4得到的最终光伏发电站集群，计算当前光伏发电站集群出力特征指标值，对当前电网进行评估，得到最终并网和退网的选择结果。S5: According to the final photovoltaic power station cluster obtained in step S4, calculate the output characteristic index value of the current photovoltaic power station cluster, evaluate the current power grid, and obtain the final selection results of grid connection and grid withdrawal.
通过上述方法，单个光伏电站与区域光伏发电集群为对象，分别针对单个站点出力曲线进行特征提取，研究其特性评价指标。实现精细化分析，提高分析精度。选择并网光伏发电站时，选择并退网冲击系数较小的光伏发电站。在选择退网光伏发电站时，选择并退网冲击系数较大的光伏发电站。提高电网稳定性。在此基础上针对区域光伏集群聚合特性进行分析，研究集群聚合平滑效应的量化指标，并揭示平滑效应的产生机理，通过实际数据研究区域电站数量和区域直径与平滑效应之间关系，对电网总体出力进行评估，得到所有光伏发电站之间的相关性。Through the above method, a single photovoltaic power station and a regional photovoltaic power generation cluster are used as objects, and feature extraction is performed on the output curve of a single station, and its characteristic evaluation indicators are studied. Realize refined analysis and improve analysis accuracy. When choosing a grid-connected photovoltaic power station, choose a photovoltaic power station with a small impact factor for grid-connected and grid-connected photovoltaic power stations. When choosing a photovoltaic power station to withdraw from the network, choose a photovoltaic power station with a large impact factor for the withdrawal from the network. Improve grid stability. On this basis, analyze the characteristics of regional photovoltaic cluster aggregation, study the quantitative indicators of cluster aggregation smoothing effect, and reveal the mechanism of smoothing effect, and use actual data to study the relationship between the number of regional power stations and the diameter of the region and the smoothing effect. Efforts are made to evaluate and obtain correlations between all photovoltaic power plants.
进一步的，在步骤S1中的光伏发电站集群数据至少包括光伏发电站数目、区域地理范围、集群区域直径和所有光伏发电站的相对距离；Further, the photovoltaic power station cluster data in step S1 at least includes the number of photovoltaic power stations, the geographical range of the region, the diameter of the cluster area and the relative distance of all photovoltaic power stations;
所述光伏发电站的历史出力数据为选择并网或者退网时刻前x天的光伏出力数据；所述出力数据至少包括光伏发电输出功率采样数据和光伏发电出力曲线。The historical output data of the photovoltaic power station is the photovoltaic output data of x days before the grid connection or grid withdrawal time; the output data at least includes photovoltaic power generation output power sampling data and photovoltaic power generation output curve.
再进一步描述，步骤S2中光伏发电站的日出力特征至少包括：日平均出力指标、日出力分布偏度指标和日平均波动率指标；To further describe, the daily output characteristics of the photovoltaic power station in step S2 at least include: daily average output index, daily output distribution skewness index and daily average volatility index;
所述日平均出力指标的计算公式为：The daily average output index The calculation formula is:
其中，n为昼间采样点个数；Pt为光伏发电站的输出功率序列，t＝1,2,3…M；Pbase为光伏发电站的装机容量；Among them, n is the number of sampling points during the day; Pt is the output power sequence of the photovoltaic power station, t=1,2,3...M; Pbase is the installed capacity of the photovoltaic power station;
日平均出力指标描述的是光伏电站的日平均出力水平，该指标与天气类型相关，不同的天气类型下光伏出力的大小存在较大不同。The daily average output index describes the daily average output level of the photovoltaic power station. This index is related to the weather type, and the photovoltaic output is quite different under different weather types.
所述日出力分布偏度指标的计算公式为：The calculation formula of the daily output force distribution skewness index is:
其中，表示日出力分布偏度指标；日发出的视在功率；in, Indicates the daily output force distribution skewness index; Apparent power emitted by day;
日出力分布偏度指标表征光伏电站当日出力分布的偏斜程度，无云层遮挡等影响的光伏发电出力曲线，相对于正态分布，其出力分布峰度偏向较大数值方向，偏度为负值，随着遮挡作用的逐渐增强，曲线整体的波动逐渐增大，光伏出力分布向低输出功率方向偏移，偏度值逐渐增加。The daily output force distribution skewness index represents the degree of skewness of the daily output power distribution of the photovoltaic power plant. The photovoltaic power generation output curve without cloud cover and other influences, compared with the normal distribution, the kurtosis of the output distribution is biased towards a larger value direction, and the skewness is negative. , with the gradual enhancement of the shading effect, the overall fluctuation of the curve gradually increases, the photovoltaic output distribution shifts to the direction of low output power, and the skewness value gradually increases.
所述日平均波动率指标σ的计算公式为：The formula for calculating the daily average volatility index σ is:
设定与固有波动方向相反的光伏发电出力曲线波动为有效波动；其中，光伏出力的固有波动整体呈抛物线形状；N为有效波动的次数；ΔPi为第i次有效波动的有效波动幅值，该有效波动量等于每次有效波动的极小值点与其相邻最近的极大值点功率差值的绝对值。Set the photovoltaic power generation output curve fluctuation opposite to the inherent fluctuation direction as effective fluctuation; among them, the inherent fluctuation of photovoltaic output is in the shape of a parabola as a whole; N is the number of effective fluctuations; ΔPi is the effective fluctuation amplitude of the i-th effective fluctuation, the The effective fluctuation amount is equal to the absolute value of the power difference between the minimum value point of each effective fluctuation and the nearest adjacent maximum value point.
通过日平均波动率指标，进一步克服了光伏出力本身呈现抛物线形的波动特性和受到云层遮挡等方面的影响，提高了固有波动特性与随机波动特性分辨率。对光伏发电站波动量的具体描述，提出有效波动率的概念来对光伏出力的波动水平进行描述。Through the daily average volatility index, it further overcomes the parabolic fluctuation characteristics of photovoltaic output itself and the influence of cloud cover, etc., and improves the resolution of inherent fluctuation characteristics and random fluctuation characteristics. For the specific description of the fluctuation of photovoltaic power station, the concept of effective fluctuation rate is proposed to describe the fluctuation level of photovoltaic output.
日平均有效波动率是表征光伏电站当日出力波动情况的重要特征，间接反映天气状态的变化情况。该特征的值越小，表示光伏电站出力的当日平均波动性越小，天气状态越稳定。The daily average effective fluctuation rate is an important feature that characterizes the daily output fluctuation of photovoltaic power plants, and indirectly reflects the changes in weather conditions. The smaller the value of this feature, the smaller the daily average fluctuation of the output of the photovoltaic power plant and the more stable the weather state.
再进一步描述，在步骤S3中光伏发电站日出力特征指标包括子时段平均出力指标和子时段波动率指标；以日为整体划分成Y个时间段，Y个时间段中的任意一个时间段为子时段；所述子时段波动率指标包括子时段波动率的均值指标和子时段波动率极大值指标。To further describe, in step S3, the daily output characteristic index of the photovoltaic power station includes the sub-period average output index and the sub-period volatility index; the day is divided into Y time periods as a whole, and any one of the Y time periods is a time period; the sub-period volatility index includes a sub-period volatility average index and a sub-period volatility maximum value index.
再进一步描述，所述子时段平均出力指标的计算公式为：To further describe, the calculation formula of the sub-period average output index is:
所述时段平均出力指标计算公式为：The average output index of the period The calculation formula is:
其中，my为第y个子时段的采样点个数；y＝1,2,3…Y，m1+m2+m3+…+mY＝n，Pbase为光伏发电站的装机容量；n为昼间采样点个数，Pt为光伏发电站的输出功率序列；时段平均出力指标可以分时段确定出力曲线大致的幅值大小，有助于实现对多天气状态下光伏出力更加精确化的科学描述。Among them, my is the number of sampling points in the yth sub-period; y=1,2,3...Y, m1 +m2 +m3 +...+mY ＝n, Pbase is the installed capacity of the photovoltaic power station ; n is the number of sampling points during the day, and Pt is the output power sequence of the photovoltaic power station; the average output index of the period can determine the approximate amplitude of the output curve in different periods, which is helpful to realize more accurate photovoltaic output under multi-weather conditions A scientific description of it.
设子时段有效波动率序列Sy为：Let the sub-period effective volatility sequence Sy be:
其中，yl为第y个子时段的有效波动次数；ΔP为有效波动的有效波动量，为第y个子时段yl次波动时的有效波动量；Among them, yl is the number of effective fluctuations in the yth sub-period; ΔP is the effective fluctuation amount of effective fluctuations, is the effective fluctuation amount when the yth sub-period fluctuates for y times;
所述子时段波动率的均值指标计算公式为：Mean indicator of volatility for the sub-period The calculation formula is:
所述子时段波动率极大值指标计算公式为：The sub-period volatility maximum value index The calculation formula is:
若某时段平均波动率较高，说明在此时段光伏电站受云层移动等天气变化的影响整体较大。可以描述子时段波动的最大程度，该特征值越大，说明该时段光伏出力的波动对电网的冲击越大，会对电网产生较大的影响。If the average fluctuation rate in a certain period is high, it means that the photovoltaic power station is affected by weather changes such as cloud movement and other weather changes during this period. It can describe the maximum degree of sub-period fluctuations. The larger the eigenvalue, the greater the impact of photovoltaic output fluctuations on the grid during this period, and it will have a greater impact on the grid.
在步骤S4中并退网冲击系数α的计算公式为：In step S4, the formula for calculating the impact coefficient α of parallel network disconnection is:
其中，λ1+λ2+λ3+λ4+λ5+λ6，λ1，λ2，λ3，λ4，λ5，λ6的取值范围为0～1。λ1，λ2，λ3，λ4，λ5，λ6为权重系数；Wherein, λ1 +λ2 +λ3 +λ4 +λ5 +λ6 , λ1 , λ2 , λ3 , λ4 , λ5 , and λ6 range from 0 to 1. λ1 , λ2 , λ3 , λ4 , λ5 , λ6 are weight coefficients;
为光伏发电站的日出力特征的日平均出力指标；表示光伏发电站的日出力特征的日出力分布偏度指标；σ为光伏发电站的日出力特征日平均波动率指标；为光伏发电站日出力特征指标的子时段平均出力指标；为光伏发电站日出力特征指标的子时段波动率的均值指标；为光伏发电站日出力特征指标的子时段波动率极大值指标。 is the daily average output index of the daily output characteristics of the photovoltaic power station; The daily output power distribution skewness index indicating the daily output power characteristics of the photovoltaic power station; σ is the daily average volatility index of the daily output power characteristics of the photovoltaic power station; is the sub-period average output index of the daily output characteristic index of the photovoltaic power station; is the average index of the sub-period volatility of the daily output characteristic index of the photovoltaic power station; It is the sub-period volatility maximum index of the daily output characteristic index of the photovoltaic power station.
通过设定权重系数，得到每个变量对电网冲击的比例。最终得到的并退网冲击系数α。需要并网或退网时，通过并退网冲击系数α，来进行选择，并网时，降低对原电网的冲击。退网选择后，使原电网更加稳定。By setting the weight coefficient, the proportion of each variable to the grid impact is obtained. Finally, the impact coefficient α of network disconnection is obtained. When it is necessary to connect to the grid or withdraw from the grid, the selection is made through the impact coefficient α of the grid connection and withdrawal. When connecting to the grid, the impact on the original grid is reduced. After the option of withdrawing from the grid, the original grid will be more stable.
再进一步描述，步骤S5中光伏发电站集群出力特征指标包括集群平滑效应系数ξcluster和光伏发电站出力不一致性系数K；To further describe, in step S5, the output characteristic index of the photovoltaic power station cluster includes the cluster smoothing effect coefficient ξcluster and the output inconsistency coefficient K of the photovoltaic power station;
所述集群平滑效应系数ξcluster计算公式为：The calculation formula of the cluster smoothing effect coefficient ξcluster is:
其中，表示集群中第l个光伏发电站的有效波动率；in, Indicates the effective volatility of the lth photovoltaic power station in the cluster;
σcluster表示集群总出力的有效波动率；σcluster represents the effective volatility of the total output of the cluster;
M表示光伏发电站集群中光伏发电站的个数；M represents the number of photovoltaic power stations in the photovoltaic power station cluster;
若ξcluster大于1，说明光伏电站集群出力相对于各单个光伏发电站具有平滑效果，且ξcluster越大，说明功率波动的平滑效果愈显著。If ξcluster is greater than 1, it means that the output of the photovoltaic power station cluster has a smoothing effect relative to each single photovoltaic power station, and the larger the ξcluster , the more significant the smoothing effect of power fluctuations.
光伏发电站出力不一致性系数K的计算公式为：The formula for calculating the output inconsistency coefficient K of the photovoltaic power station is:
其中，P1,P2表示两个不同的光伏发电站的出力序列；Among them, P1 and P2 represent the output sequences of two different photovoltaic power stations;
fi(P1,P2)的表达式为：The expression of fi (P1 ,P2 ) is:
fi(P1,P2)表示在第i-1个采样点与第i个采样点之间两光伏发电出力曲线趋势的不一致性。fi (P1 , P2 ) represents the inconsistency of the trend of the two photovoltaic power generation output curves between the i-1th sampling point and the i-th sampling point.
不一致性系数K表示两个出力曲线上趋势不一致的采样间隔个数占总间隔个数的比值，是一个大于0小于1的数，由于两出力曲线的固有波动具有相同的趋势，针对该波动不一致性系数为0，该指标仅反映随机波动的影响，因此可用来解释电站集群聚合的平滑效应，K值越大说明这两个光伏电站集群聚合的平滑效应越好。The inconsistency coefficient K represents the ratio of the number of sampling intervals with inconsistent trends to the total number of intervals on the two output curves, which is a number greater than 0 and less than 1. Since the inherent fluctuations of the two output curves have the same trend, the fluctuations are inconsistent The coefficient of K is 0, and this index only reflects the influence of random fluctuations, so it can be used to explain the smoothing effect of cluster aggregation of power plants. The larger the value of K, the better the smoothing effect of cluster aggregation of two photovoltaic power plants.
由于光伏发电出力曲线的总体趋势是先上升后下降的，因此不存在两条完全互补的出力曲线，即任意两条出力曲线的不一致性系数不可能达到1。经过大量数据计算，当不一致性系数接近0.5时就可认为它们之间的相关性较差，互补性较好。通过上述设计，可以得到最终电网中所有光伏发电站的相关性和不一致系数。Since the overall trend of the photovoltaic power generation output curve is to rise first and then decline, there are no two completely complementary output curves, that is, the inconsistency coefficient of any two output curves cannot reach 1. After a large amount of data calculation, when the inconsistency coefficient is close to 0.5, it can be considered that the correlation between them is poor and the complementarity is good. Through the above design, the correlation and inconsistency coefficients of all photovoltaic power stations in the final grid can be obtained.
再进一步描述，在步骤S5中对当前电网进行评估包括：To further describe, evaluating the current power grid in step S5 includes:
计算两两光伏发电站的光伏发电站出力不一致性系数K，并得到该光伏发电站出力不一致性系数K与该两个光伏发电站的相对距离的矩阵关系；Calculate the photovoltaic power station output inconsistency coefficient K of two photovoltaic power stations, and obtain the matrix relationship between the photovoltaic power station output inconsistency coefficient K and the relative distance between the two photovoltaic power stations;
采用多项式对集群区域直径与集群平滑效应系数的关系进行拟合，该第一拟合关系式为：ξcluster＝-3.717d2+4.685d+0.4053；其中d表示集群区域直径；A polynomial is used to fit the relationship between the cluster area diameter and the cluster smoothing effect coefficient, the first fitting relational expression is: ξcluster =-3.717d2 +4.685d+0.4053; where d represents the cluster area diameter;
采用多项式对集群平滑效应系数、集群区域直径和电站数量进行拟合；该第二拟合关系式为：其中d表示集群区域直径；num表示该区域电站数量。本发明的有益效果：通过历史数据，计算每个光伏发电站的特征值，使并退网选择可靠。并网选择时，对原电网冲击小。退网选择时，使原电网更加稳定。选择可靠性好。并通过计算当前光伏发电站集群出力特征指标值，对当前电网进行评估。A polynomial is used to fit the cluster smoothing effect coefficient, the cluster area diameter and the number of power stations; the second fitting relationship is: Where d represents the diameter of the cluster area; num represents the number of power stations in the area. Beneficial effects of the present invention: through historical data, the characteristic value of each photovoltaic power station is calculated, so that the selection of grid connection and withdrawal is reliable. When grid connection is selected, the impact on the original grid is small. When withdrawing from the grid, the original grid is more stable. Choose reliability. And evaluate the current power grid by calculating the output characteristic index value of the current photovoltaic power station cluster.
附图说明Description of drawings
图1是本发明的选择方法流程图；Fig. 1 is the selection method flowchart of the present invention;
图2是不同天气条件下的日平均出力曲线图；Fig. 2 is the daily average output curve under different weather conditions;
图3是不同波动状况下日出力分布偏度对比图；Figure 3 is a comparison chart of daily output force distribution skewness under different fluctuation conditions;
图4是出力曲线波动情况与出力分布偏度对应关系示意图；Figure 4 is a schematic diagram of the corresponding relationship between the fluctuation of the output curve and the skewness of the output distribution;
图5是光伏发电站出力有效波动示意图；Figure 5 is a schematic diagram of the effective fluctuation of the output of a photovoltaic power station;
图6是光伏发电集群聚合特性示意图；Fig. 6 is a schematic diagram of aggregation characteristics of photovoltaic power generation clusters;
图7是E、C两电站2015年3月1日出力曲线；Figure 7 is the output curves of E and C power stations on March 1, 2015;
图8是五个光伏电站出力曲线图；Figure 8 is a graph of the output of five photovoltaic power plants;
图9是集群区域直径与平滑效应系数拟合曲线图；Fig. 9 is a fitting curve diagram of cluster area diameter and smoothing effect coefficient;
图10是平滑效应系数与集群区域直径、区域电站个数拟合曲线图；Fig. 10 is a fitting curve diagram of the smoothing effect coefficient, the diameter of the cluster area, and the number of regional power stations;
具体实施方式Detailed ways
下面结合附图对本发明的具体实施方式以及工作原理作进一步详细说明。The specific implementation manner and working principle of the present invention will be further described in detail below in conjunction with the accompanying drawings.
一种基于光伏发电站精细化出力的并网退网选择方法，结合图1可看出，包括以下步骤：A grid-connected and de-grid selection method based on the refined output of photovoltaic power stations can be seen from Figure 1, including the following steps:
S1：选定预选并网光伏发电站，得到光伏发电站集群，并获取光伏发电站集群数据和光伏发电站集群中所有光伏发电站的历史出力数据；S1: Select the pre-selected grid-connected photovoltaic power station, obtain the photovoltaic power station cluster, and obtain the photovoltaic power station cluster data and the historical output data of all photovoltaic power stations in the photovoltaic power station cluster;
S2：选取光伏发电站的日出力特征指标，并分别计算单个光伏发电站的所有日出力特征指标值；S2: Select the daily output characteristic index of the photovoltaic power station, and calculate all the daily output characteristic index values of a single photovoltaic power station;
S3：对光伏发电站日出力特征指标进行划分，得到子时段出力特征指标值；S3: Divide the daily output characteristic index of the photovoltaic power station to obtain the sub-period output characteristic index value;
S4：设定所有特征指标值的权重，得到每个光伏发电站并退网冲击系数，根据并退网冲击系数，对并网和退网的光伏发电站进行预选择；S4: Set the weights of all characteristic index values to obtain the impact coefficient of grid connection and withdrawal of each photovoltaic power station, and pre-select grid-connected and grid-disconnected photovoltaic power stations according to the impact coefficient of grid connection and withdrawal;
S5：根据步骤S4得到的最终光伏发电站集群，计算当前光伏发电站集群出力特征指标值，对当前电网进行评估，得到最终并网和退网的选择结果。S5: According to the final photovoltaic power station cluster obtained in step S4, calculate the output characteristic index value of the current photovoltaic power station cluster, evaluate the current power grid, and obtain the final selection results of grid connection and grid withdrawal.
在本实施例中，在步骤S1中的光伏发电站集群数据包括光伏发电站数目、区域地理范围和所有光伏发电站的相对距离；In this embodiment, the photovoltaic power station cluster data in step S1 includes the number of photovoltaic power stations, the geographical range of the region and the relative distance of all photovoltaic power stations;
所述光伏发电站的历史出力数据为选择并网或者退网时刻前x天的光伏出力数据；所述出力数据至少包括光伏发电输出功率采样数据和光伏发电出力曲线。The historical output data of the photovoltaic power station is the photovoltaic output data of x days before the grid connection or grid withdrawal time; the output data at least includes photovoltaic power generation output power sampling data and photovoltaic power generation output curve.
在本实施例中，历史数据为2015年3月至2016年3月采集的数据，采样时间间隔为15min。In this embodiment, the historical data is data collected from March 2015 to March 2016, and the sampling time interval is 15 minutes.
在步骤S2中光伏发电站的日出力特征包括：日平均出力指标、日出力分布偏度指标和日平均波动率指标；In step S2, the daily output characteristics of the photovoltaic power station include: daily average output index, daily output distribution skewness index and daily average volatility index;
在本实施例中，日平均出力指标的计算公式为：In this embodiment, the daily average output index The calculation formula is:
其中，n为昼间采样点个数；Pt为光伏发电站的输出功率序列，t＝1,2,3…M；Pbase为光伏发电站的装机容量；Among them, n is the number of sampling points during the day; Pt is the output power sequence of the photovoltaic power station, t=1,2,3...M; Pbase is the installed capacity of the photovoltaic power station;
日平均出力指标描述的是光伏电站的日平均出力水平，该指标与天气类型相关，不同的天气类型下光伏出力的大小存在较大不同。如图2所示，阴天、雨雪天气云层遮挡作用较强，光伏电站出力水平不高，日平均出力仅为0.236与0.049，晴天地表太阳辐照度较高，出力水平较高，日平均出力达到0.436。The daily average output index describes the daily average output level of the photovoltaic power station. This index is related to the weather type, and the photovoltaic output is quite different under different weather types. As shown in Figure 2, the cloud cover effect is strong in cloudy, rainy and snowy days, the output level of photovoltaic power plants is not high, and the daily average output is only 0.236 and 0.049. The output reached 0.436.
日出力分布偏度指标的计算公式为：The formula for calculating the skewness index of the daily output force distribution is:
其中，表示日出力分布偏度指标；日发出的视在功率；in, Indicates the daily output force distribution skewness index; Apparent power emitted by day;
日出力分布偏度指标表征光伏电站当日出力分布的偏斜程度，无云层遮挡等影响的光伏发电出力曲线，相对于正态分布，其出力分布峰度偏向较大数值方向，偏度为负值，随着遮挡作用的逐渐增强，曲线整体的波动逐渐增大，光伏出力分布向低输出功率方向偏移，偏度值逐渐增加。The daily output force distribution skewness index represents the degree of skewness of the daily output power distribution of the photovoltaic power plant. The photovoltaic power generation output curve without cloud cover and other influences, compared with the normal distribution, the kurtosis of the output distribution is biased towards a larger value direction, and the skewness is negative. , with the gradual enhancement of the shading effect, the overall fluctuation of the curve gradually increases, the photovoltaic output distribution shifts to the direction of low output power, and the skewness value gradually increases.
从图3可以看出，第一张子图中曲线a、b出力平滑，偏度分别为-0.42和-0.39，而第二张子图中出力波动剧烈，曲线c、d偏度值较上述数值均有较大增加，达到0.46和0.64。It can be seen from Fig. 3 that the output of curves a and b in the first sub-graph is smooth, and the skewness is -0.42 and -0.39 respectively, while the output of the second sub-graph fluctuates violently, and the skewness values of curves c and d are higher than the above values. Large increases, reaching 0.46 and 0.64.
从图4可以看出，光伏出力分布曲线的偏度是表征出力曲线整体波动剧烈程度的统计特征，曲线出力波动越剧烈，其出力分布偏度值越大，出力曲线波动与出力分布偏度值的对应关系。It can be seen from Figure 4 that the skewness of the photovoltaic output distribution curve is a statistical feature that characterizes the intensity of the overall fluctuation of the output curve. corresponding relationship.
从图5可以看出，光伏发电出力曲线中与固有波动方向相反的出力曲线波动为有效波动，每次有效波动的极小值点与其相邻最近的极大值点功率差值的绝对值为该次波动的有效波动量。It can be seen from Figure 5 that the fluctuation of the output curve in the output curve of photovoltaic power generation opposite to the inherent fluctuation direction is an effective fluctuation, and the absolute value of the power difference between the minimum value point of each effective fluctuation and the nearest adjacent maximum value point is The effective fluctuation amount of this fluctuation.
所述日平均波动率指标σ的计算公式为：The formula for calculating the daily average volatility index σ is:
设定与固有波动方向相反的光伏发电出力曲线波动为有效波动；N为有效波动的次数；ΔPi为第i次波动的有效波动的有效波动量，该有效波动量等于每次有效波动的极小值点与其相邻最近的极大值点功率差值的绝对值。Set the photovoltaic power generation output curve fluctuation opposite to the inherent fluctuation direction as effective fluctuation; N is the number of effective fluctuations; ΔPi is the effective fluctuation amount of the i-th fluctuation, which is equal to the extreme The absolute value of the power difference between a small point and its nearest adjacent maximum point.
通过日平均波动率指标，进一步克服了光伏出力本身呈现抛物线形的波动特性和受到云层遮挡等方面的影响，提高了固有波动特性与随机波动特性分辨率。对光伏发电站波动量的具体描述，提出有效波动率的概念来对光伏出力的波动水平进行描述。日平均有效波动率是表征光伏电站当日出力波动情况的重要特征，间接反映天气状态的变化情况。该特征的值越小，表示光伏电站出力的当日平均波动性越小，天气状态越稳定。Through the daily average volatility index, it further overcomes the parabolic fluctuation characteristics of photovoltaic output itself and the influence of cloud cover, etc., and improves the resolution of inherent fluctuation characteristics and random fluctuation characteristics. For the specific description of the fluctuation of photovoltaic power station, the concept of effective fluctuation rate is proposed to describe the fluctuation level of photovoltaic output. The daily average effective fluctuation rate is an important feature that characterizes the daily output fluctuation of photovoltaic power plants, and indirectly reflects the changes in weather conditions. The smaller the value of this feature, the smaller the daily average fluctuation of the output of the photovoltaic power plant and the more stable the weather state.
在步骤S3中光伏发电站日出力特征指标包括子时段平均出力指标和子时段波动率指标；其中，以日为整体划分成Y个时间段，Y个时间段中的任一一个时间段为子时段；所述子时段波动率指标包括子时段波动率的均值指标和子时段波动率极大值指标。In step S3, the daily output characteristic index of the photovoltaic power station includes the average output index of the sub-period and the volatility index of the sub-period; wherein, the day is divided into Y time periods as a whole, and any one of the Y time periods is the sub-period time period; the sub-period volatility index includes a sub-period volatility average index and a sub-period volatility maximum value index.
在本实施例中，设置有2个子时段，分别为上述子时段和下午子时段。In this embodiment, two sub-periods are set, which are the above-mentioned sub-period and the afternoon sub-period.
所述子时段平均出力指标的计算公式为：The formula for calculating the average output index of the sub-period is:
上午子时段平均出力指标计算公式为：Average output index in the morning sub-period The calculation formula is:
上午子时段平均出力指标计算公式为：Average output index in the morning sub-period The calculation formula is:
其中，m1为上午子时段的采样点个数；m2为下午子时段的采样点个数Among them, m1 is the number of sampling points in the morning sub-period; m2 is the number of sampling points in the afternoon sub-period
在本实施例中，y＝1时，表示上午子时段；y＝2时，表示下午子时段；In this embodiment, when y=1, it represents the morning sub-period; when y=2, it represents the afternoon sub-period;
m1+m2＝n，Pbase为光伏发电站的装机容量；n为昼间采样点个数，Pt为光伏发电站的输出功率序列；m1 +m2 =n, Pbase is the installed capacity of the photovoltaic power station; n is the number of sampling points during the day, Pt is the output power sequence of the photovoltaic power station;
设上午子时段有效波动率序列S1为：Suppose the effective volatility sequence S1 in the morning sub-period is:
设上午子时段有效波动率序列S2为：Suppose the effective volatility sequence S2 in the morning sub-period is:
其中，1l为上午子时段的有效波动次数；2l为上午子时段的有效波动次数；ΔP为有效波动的有效波动量，为上午子时段l次波动时的有效波动量；为下午子时段l次波动时的有效波动量；Among them, 1l is the number of effective fluctuations in the morning sub-period; 2l is the number of effective fluctuations in the morning sub-period; ΔP is the effective fluctuation amount of effective fluctuations, is the effective fluctuation amount when there is one fluctuation in the morning sub-period; is the effective fluctuation amount when there is one fluctuation in the afternoon sub-period;
则上午子时段波动率的均值指标计算公式为：Then the average index of volatility in the morning sub-period The calculation formula is:
则下午子时段波动率的均值指标计算公式为：Then the average index of volatility in the afternoon sub-period The calculation formula is:
上午子时段波动率极大值指标计算公式为：Morning sub-session volatility maximum value indicator The calculation formula is:
上午子时段波动率极大值指标计算公式为：Morning sub-session volatility maximum value indicator The calculation formula is:
在步骤S4中并退网冲击系数α的计算公式为：In step S4, the formula for calculating the impact coefficient α of parallel network disconnection is:
在步骤S4中并退网冲击系数α的计算公式为：In step S4, the formula for calculating the impact coefficient α of parallel network disconnection is:
其中，λ1+λ2+λ3+λ4+λ5+λ6＝，1λ1，λ2，λ3，λ4，λ5，λ6的取值范围为0～1。λ1，λ2，λ3，λ4，λ5，λ6为权重系数。Wherein, λ1 +λ2 +λ3 +λ4 +λ5 +λ6 =, 1λ1 , λ2 , λ3 , λ4 , λ5 , and λ6 range from 0 to 1. λ1 , λ2 , λ3 , λ4 , λ5 , λ6 are weight coefficients.
在本实施例中，λ1＝0，λ2＝0，λ3＝0，λ4＝0，λ5＝0.85，λ6＝0.15。In this embodiment, λ1 =0, λ2 =0, λ3 =0, λ4 =0, λ5 =0.85, and λ6 =0.15.
从图6可以看出，光伏发电出力受其固有波动的影响较大，出力曲线大致呈现“单峰”分布，其平滑效应本质上是区域光伏电站集群聚合后，对集群中各电站随机波动的平滑效果。因此，随着平滑效应的增强，集群出力特性会形成较为平滑的出力曲线，有利于减少单个电站出力波动对电网造成的影响，但曲线仍保持固有的“单峰”波形基本不变。It can be seen from Figure 6 that the output of photovoltaic power generation is greatly affected by its inherent fluctuations, and the output curve roughly presents a "single peak" distribution. smoothing effect. Therefore, with the enhancement of the smoothing effect, the cluster output characteristics will form a relatively smooth output curve, which is beneficial to reduce the impact of single power station output fluctuations on the power grid, but the curve still maintains the inherent "single peak" waveform basically unchanged.
步骤S4中光伏发电站集群出力特征指标包括集群平滑效应系数ξcluster和光伏发电站出力不一致性系数K；In step S4, the output characteristic index of the photovoltaic power station cluster includes the cluster smoothing effect coefficient ξcluster and the output inconsistency coefficient K of the photovoltaic power station;
所述集群平滑效应系数ξcluster计算公式为：The calculation formula of the cluster smoothing effect coefficient ξcluster is:
其中，表示集群中第l个光伏发电站的有效波动率；in, Indicates the effective volatility of the lth photovoltaic power station in the cluster;
σcluster表示集群总出力的有效波动率；σcluster represents the effective volatility of the total output of the cluster;
M表示光伏发电站集群中光伏发电站的个数；M represents the number of photovoltaic power stations in the photovoltaic power station cluster;
光伏发电站出力不一致性系数K的计算公式为：The formula for calculating the output inconsistency coefficient K of the photovoltaic power station is:
其中，P1,P2表示两个不同的光伏发电站的出力序列；Among them, P1 and P2 represent the output sequences of two different photovoltaic power stations;
fi(P1,P2)的表达式为：The expression of fi (P1 ,P2 ) is:
fi(P1,P2)表示在第i-1个采样点与第i个采样点之间两光伏发电出力曲线趋势的不一致性。如图8所示，两采样点间二者出力曲线趋势不一致时fi(P1,P2)取1，反之取0。fi (P1 , P2 ) represents the inconsistency of the trend of the two photovoltaic power generation output curves between the i-1th sampling point and the i-th sampling point. As shown in Fig. 8, fi (P1 , P2 ) takes 1 when the output curve trends of the two sampling points are inconsistent, and takes 0 otherwise.
不一致性系数K表示两出力曲线上趋势不一致的采样间隔个数占总间隔个数的比值，是一个大于0小于1的数，由于两出力曲线的固有波动具有相同的趋势，针对该波动不一致性系数为0，该指标仅反映随机波动的影响，因此可用来解释电站集群聚合的平滑效应，K值越大说明这两个光伏电站集群聚合的平滑效应越好。The inconsistency coefficient K represents the ratio of the number of sampling intervals with inconsistent trends on the two output curves to the total number of intervals, which is a number greater than 0 and less than 1. Since the inherent fluctuations of the two output curves have the same trend, for this fluctuation inconsistency The coefficient is 0, and this index only reflects the influence of random fluctuations, so it can be used to explain the smoothing effect of the cluster aggregation of power plants. The larger the K value, the better the smoothing effect of the cluster aggregation of the two photovoltaic power plants.
由于光伏发电出力曲线的总体趋势是先上升后下降的，因此不存在两条完全互补的出力曲线，换言之任意两条出力曲线的不一致性系数不可能达到1。经过大量数据计算，当不一致性系数接近0.5时就可认为它们之间的相关性较差，互补性较好。Since the overall trend of the photovoltaic power generation output curve is to rise first and then decline, there are no two completely complementary output curves. In other words, the inconsistency coefficient of any two output curves cannot reach 1. After a large amount of data calculation, when the inconsistency coefficient is close to 0.5, it can be considered that the correlation between them is poor and the complementarity is good.
在步骤S5中对当前电网进行评估包括：Evaluating the current grid in step S5 includes:
计算两两光伏发电站的光伏发电站出力不一致性系数K，并得到该光伏发电站出力不一致性系数K与该两个光伏发电站的相对距离的矩阵关系；Calculate the photovoltaic power station output inconsistency coefficient K of two photovoltaic power stations, and obtain the matrix relationship between the photovoltaic power station output inconsistency coefficient K and the relative distance between the two photovoltaic power stations;
采用多项式对集群区域直径与集群平滑效应系数的关系进行拟合，该第一拟合关系式为：ξcluster＝-3.717d2+4.685d+0.4053；其中d表示集群区域直径；A polynomial is used to fit the relationship between the cluster area diameter and the cluster smoothing effect coefficient, the first fitting relational expression is: ξcluster =-3.717d2 +4.685d+0.4053; where d represents the cluster area diameter;
采用多项式对集群平滑效应系数、集群区域直径和电站数量进行拟合；该第二拟合关系式为：其中d表示集群区域直径；num表示该区域电站数量。为验证本发明提取的整体、局部特征对实际光伏发电出力曲线的描述效果，以及光伏发电集群聚合特性分析理论在实测数据上的应用效果。对前述理论采用Matlab软件进行仿真计算。在本实施例中，计算所采用数据为云南某地区A～E五个光伏电站的实测数据。仿真数据的采集时间段为2015年3月至2016年3月，采样时间间隔为15min。A polynomial is used to fit the cluster smoothing effect coefficient, the cluster area diameter and the number of power stations; the second fitting relationship is: Where d represents the diameter of the cluster area; num represents the number of power stations in the area. In order to verify the description effect of the overall and local features extracted by the present invention on the actual photovoltaic power generation output curve, and the application effect of the photovoltaic power generation cluster aggregation characteristic analysis theory on the measured data. The aforementioned theory is simulated and calculated using Matlab software. In this embodiment, the data used for calculation are the actual measurement data of five photovoltaic power plants A-E in a certain area of Yunnan. The simulation data collection period is from March 2015 to March 2016, and the sampling interval is 15 minutes.
以云南某地区E、C两电站的输出特性曲线进行实例计算分析。从图7可以看出，分别为E、C两电站2015年3月1日的输出功率曲线。Taking the output characteristic curves of E and C power stations in a certain area of Yunnan Province as an example, calculation and analysis are carried out. It can be seen from Fig. 7 that the output power curves of E and C power stations on March 1, 2015 are respectively.
从图7可以看出，E电站当日全天出力均有波动，波动分布较为平均。C电站在上午时段出力平稳，但下午由于天气状态的改变光伏发电出力曲线具有比较剧烈的波动，例如晴转多云。以本文提取的特征对两条曲线进行描述，指标计算结果如表1。It can be seen from Figure 7 that the output of power station E fluctuates throughout the day, and the fluctuation distribution is relatively even. The output of power plant C is stable in the morning, but in the afternoon, due to changes in weather conditions, the output curve of photovoltaic power generation has relatively severe fluctuations, such as sunny to cloudy. The two curves are described with the features extracted in this paper, and the index calculation results are shown in Table 1.
表1两出力曲线指标计算结果Table 1 Calculation results of two output curve indicators
从表1中可以看出两电站平均出力基本相同，日平均波动率数值反应两曲线出力波动总量相近，但从出力分布的偏度可以看出C电站曲线的波动程度略为剧烈。由于并无出现雨雪转晴等天气，E电站出力曲线上下午时段较为对称，各时段平均出力十分接近，C电站上午时段平均出力略高于下午。从子时段波动率均值以及最大值可看出，E电站全天均有少量云层遮挡，上下午均有波动，波动分布较为均匀。It can be seen from Table 1 that the average output of the two power stations is basically the same, and the daily average fluctuation rate value reflects that the total output fluctuations of the two curves are similar. However, it can be seen from the skewness of the output distribution that the degree of fluctuation of the C power station curve is slightly severe. Since there is no weather such as rain, snow and clear weather, the output curve of E power station is relatively symmetrical in the morning and afternoon, and the average output of each time period is very close. The average output of C power station in the morning is slightly higher than that in the afternoon. It can be seen from the average value and maximum value of the fluctuation rate in the sub-period that the E power station is covered by a small amount of clouds throughout the day, and there are fluctuations in the morning and afternoon, and the fluctuation distribution is relatively uniform.
C电站上午时段曲线光滑平稳，并未出现明显波动，平均波动率较低仅为0.008，波动率极大值也仅为0.01，而下午时段由于天气状态的变化使得C电站出现较为剧烈的出力波动，波动率均值为0.133，最大波动率达到了0.174，此时该电站出力的剧烈波动会对电网造成较大影响。两曲线整体形状相似，整体特征较为相近，但从局部特征的计算结果来看，二者对电网的影响相差很大，应按照不同出力情况理。The curve of Power Station C in the morning is smooth and stable, without obvious fluctuations. The average fluctuation rate is as low as 0.008, and the maximum fluctuation rate is only 0.01. In the afternoon, due to changes in weather conditions, Power Station C has relatively severe output fluctuations. , the average volatility rate is 0.133, and the maximum volatility rate reaches 0.174. At this time, the violent fluctuation of the power station output will have a great impact on the power grid. The overall shape of the two curves is similar, and the overall characteristics are relatively similar. However, from the calculation results of local characteristics, the influence of the two curves on the power grid is very different, and should be treated according to different output conditions.
故在对整体特征较为相近的E电站和C电站进行预选择时，根据并退网冲击系数α可以计算得到：Therefore, when pre-selecting the E power station and the C power station with relatively similar overall characteristics, according to the impact coefficient α of grid disconnection, it can be calculated as follows:
E电站上午子时段并退网冲击系数：α＝0.85×0.048+0.15×0.08＝0.0528The impact coefficient of E power station paralleling and withdrawing from the grid during the morning sub-period: α＝0.85×0.048+0.15×0.08＝0.0528
E电站下午子时段并退网冲击系数：α＝0.85×0.083+0.15×0.105＝0.0863The impact coefficient of E power station’s grid disconnection in the afternoon sub-period: α＝0.85×0.083+0.15×0.105＝0.0863
C电站上午子时段并退网冲击系数：α＝0.85×0.008+0.15×0.010＝0.0083Impact coefficient of power station C disconnecting from the network during the morning sub-period: α＝0.85×0.008+0.15×0.010＝0.0083
C电站下午子时段并退网冲击系数：α＝0.85×0.133+0.15×0.174＝0.13915Impact coefficient of power station C disconnecting from the grid during the afternoon sub-period: α＝0.85×0.133+0.15×0.174＝0.13915
通过上述计算，在上午子时段时，若要选择并网的光伏发电站时，选择对电网冲击小的C电站，降低对电网的冲击。若要选择退网的光伏发电站时，选择对电网冲击大的E电站；E电站退网后，原电网更加稳定。Through the above calculations, in the morning sub-period, if you want to choose a grid-connected photovoltaic power station, choose C power station with less impact on the grid to reduce the impact on the grid. If you want to choose a photovoltaic power station that will be withdrawn from the grid, choose the E power station that will have a large impact on the grid; after the E station is withdrawn from the grid, the original grid will be more stable.
在下午子时段时，若要选择并网的光伏发电站时，选择对电网冲击小的E电站，降低对电网的冲击。若要选择退网的光伏发电站时，选择对电网冲击大的C电站，C电站退网后，原电网更加稳定。In the afternoon sub-period, if you want to choose a grid-connected photovoltaic power station, choose the E power station with little impact on the grid to reduce the impact on the grid. If you want to choose a photovoltaic power station that will be withdrawn from the grid, choose the C power station that will have a greater impact on the grid. After the C power station is withdrawn from the grid, the original grid will be more stable.
光伏出力的平滑效应主要是由空间分布上的气象资源差异引起的，而影响空间分布效应有两个主导因素：光伏电站的数目和区域地理范围，在本实施例中用区域直径表示。通过计算电站集群的平滑效应系数以及各电站之间的不一致性系数完成对平滑效应的量化分析。The smoothing effect of photovoltaic output is mainly caused by the difference in meteorological resources in spatial distribution, and there are two dominant factors affecting the spatial distribution effect: the number of photovoltaic power plants and the geographical scope of the region, which is represented by the regional diameter in this embodiment. The quantitative analysis of the smoothing effect is completed by calculating the smoothing effect coefficient of the power station cluster and the inconsistency coefficient among the power stations.
在本实施例中，以E电站、C电站以及云南的A电站、C电站、D电站进行整体计算分析。In this embodiment, the overall calculation and analysis are carried out with E power station, C power station and Yunnan A, C and D power stations.
为叙述方便ABCDE分别叙述为PVF1～PVF5。各个光伏电站间的相对距离如表2所示For the convenience of description, ABCDE are respectively described as PVF1～PVF5. The relative distance between each photovoltaic power station is shown in Table 2
表2各个光伏电站间的相对距离(km)Table 2 The relative distance between each photovoltaic power station (km)
选取上述5个光伏电站2015年3月19日有功出力时间序列为研究对象，5个光伏电站出力变化如图8所示。The time series of active power output of the above five photovoltaic power plants on March 19, 2015 was selected as the research object, and the output changes of the five photovoltaic power plants are shown in Figure 8.
为定量衡量不同光伏电站出力的相关性，分别计算这5个光伏电站两两间的趋势不一致性系数，计算结果如表3所示。In order to quantitatively measure the correlation of the output of different photovoltaic power plants, the trend inconsistency coefficients between the five photovoltaic power plants are calculated respectively, and the calculation results are shown in Table 3.
表3光伏电站间不一致性系数矩阵Table 3 Inconsistency coefficient matrix among photovoltaic power plants
PVF1和PVF2地理位置相距较近，仅相隔5.72km，两者的趋势不一致性系数为0.16，呈现出一定的互补性，但是互补性不高；PVF1和PVF3之间的距离增大到28.67km,它们两者的趋势不一致性系数为0.30具有较高的互补性；观察不一致性系数矩阵可知，其最大值出现在PVF3和PVF5之间，它们两者相距82.41km，因此由上述分析可知区域直径是影响光伏平滑效应的重要因素。The geographical location of PVF1 and PVF2 is relatively close, only 5.72km apart, and the trend inconsistency coefficient between the two is 0.16, showing a certain degree of complementarity, but the complementarity is not high; the distance between PVF1 and PVF3 increases to 28.67km, The trend inconsistency coefficient of both of them is 0.30, which has high complementarity; observing the inconsistency coefficient matrix shows that its maximum value appears between PVF3 and PVF5, and the distance between them is 82.41km. Therefore, the above analysis shows that the area diameter is Important factors affecting photovoltaic smoothing effect.
以PVF1为基础，逐渐加入PVF2～5，分别计算不同光伏集群下的平滑效应系数ξcluster，计算结果如表4所示。表中：PVF1～2表示PVF1和PVF2；PVF1～3表示PVF1、PVF2和PVF3；PVF1～4表示PVF1、PVF2、PVF3和PVF4，PVF1～5表示5个光伏电站组成的光伏电站集群。可以看出，随着光伏电站数目的增多，平滑效应系数呈现出逐渐增大的趋势，说明随着光伏范围的扩大，不同区域光伏电站的加入，总体出力平滑效应得到了体现。Based on PVF1, gradually add PVF2 to 5, and calculate the smoothing effect coefficient ξcluster under different photovoltaic clusters respectively. The calculation results are shown in Table 4. In the table: PVF1~2 represent PVF1 and PVF2; PVF1~3 represent PVF1, PVF2 and PVF3; PVF1~4 represent PVF1, PVF2, PVF3 and PVF4; PVF1~5 represent a photovoltaic power station cluster composed of 5 photovoltaic power stations. It can be seen that with the increase of the number of photovoltaic power plants, the smoothing effect coefficient presents a gradually increasing trend, indicating that with the expansion of the photovoltaic range and the addition of photovoltaic power plants in different regions, the overall output smoothing effect has been reflected.
表4平滑效应影响因素分析Table 4 Analysis of influencing factors of smoothing effect
以五个光伏电站2015年3月2日数据为例，量化分析区域直径、区域电站数目对光伏电站集群聚合出力特性的影响。对不同区域直径的光伏电站集群与其相对应的平滑效应系数进行数据分析，其中平滑效应系数为各区域直径数值下的均值。采用多项式拟合方法对二者关系进行拟合。由于光伏出力的平滑效应主要是由地理分布上的气象资源差异引起的，而集群区域的覆盖面积可直接反应集群内地理分布的差异的大小，因此，本实施例选取可反应集群覆盖区域的区域直径二次式来与平滑效应系数进行拟合。Taking the data of five photovoltaic power stations on March 2, 2015 as an example, the influence of the regional diameter and the number of regional power stations on the aggregation output characteristics of photovoltaic power station clusters was quantitatively analyzed. Perform data analysis on photovoltaic power plant clusters with different regional diameters and their corresponding smoothing effect coefficients, where the smoothing effect coefficient is the average value under the diameter values of each region. The polynomial fitting method was used to fit the relationship between the two. Since the smoothing effect of photovoltaic output is mainly caused by the difference in geographical distribution of meteorological resources, and the coverage area of the cluster area can directly reflect the size of the difference in geographical distribution within the cluster, this embodiment selects an area that can reflect the cluster coverage area The diameter quadratic is used to fit the smoothing effect coefficients.
拟合表达式为：ξcluster＝-3.717d2+4.685d+0.4053；The fitting expression is: ξcluster =-3.717d2 +4.685d+0.4053;
式中d表示集群区域直径。拟合表达式对数据拟合的误差平方和(SSE)为0.1195，R2系数为0.9034，说明表达式具有较好的拟合效果。where d is the diameter of the cluster area. The sum of squared error (SSE) of the fitting expression to the data fitting was 0.1195, and theR2 coefficient was 0.9034, indicating that the expression had a good fitting effect.
从图9曲线可以看出，在集群区域直径较小时，电站间不一致性系数的系数较小，平滑效应较弱。随着集群区域直径的扩大，平滑效应系数也逐渐增大，集群呈现较好的空间平滑效果，但当距离达到60km后，平滑效应会出现饱和现象，甚至有下降的趋势。结合表2可以看出，E电站距离A、D和E电站均在60km以内，则在下午子时段并网时，则可选择E电站。It can be seen from the curve in Figure 9 that when the diameter of the cluster area is small, the coefficient of the inconsistency coefficient between power stations is small, and the smoothing effect is weak. As the diameter of the cluster area expands, the coefficient of the smoothing effect gradually increases, and the cluster presents a better spatial smoothing effect. However, when the distance reaches 60 km, the smoothing effect will appear saturated and even have a downward trend. It can be seen from Table 2 that E power station is within 60km from A, D and E power stations, so E power station can be selected when connecting to the grid in the afternoon sub-period.
在拟合区域直径与平滑效应系数关系的基础上，分析区域电站数量和区域直径对平滑效应的共同影响。仍采取多项式拟合，对平滑效应系数与区域直径以及电站数量的关系进行拟合。On the basis of fitting the relationship between the area diameter and the smoothing effect coefficient, the joint influence of the number of regional power stations and the area diameter on the smoothing effect is analyzed. Polynomial fitting is still adopted to fit the relationship between the smoothing effect coefficient and the area diameter and the number of power stations.
拟合表达式为：The fitting expression is:
式中d表示集群区域直径；num表示区域电站数量。拟合表达式对数据拟合的误差平方和(SSE)为0.037，R2系数为0.97，说明表达式可以较好地反应平滑效应系数与两影响因素之间的关系。In the formula, d represents the diameter of the cluster area; num represents the number of regional power stations. The sum of squared error (SSE) of the fitting expression to the data fitting was 0.037, and the R2 coefficient was 0.97, indicating that the expression could better reflect the relationship between the smoothing effect coefficient and the two influencing factors.
从图10拟合结果图可以看出随着区域电站个数与集群区域直径的增加，平滑效应逐渐增强，且平滑效应受集群区域直径的影响更为强烈。综上所述，集群区域直径与区域电站数量这两个影响因子对平滑效应有较强的解释作用，三者关系的拟合结果可用于光伏电站并网和退网的选择，对平抑区域光伏电站整体出力波动具有一定的参考意义。From the fitting results in Figure 10, it can be seen that as the number of regional power stations and the diameter of the cluster area increase, the smoothing effect gradually increases, and the smoothing effect is more strongly affected by the diameter of the cluster area. To sum up, the two influencing factors, the diameter of the cluster area and the number of regional power stations, have a strong explanatory effect on the smoothing effect. The fitting results of the relationship between the three can be used for the selection of grid-connected and withdrawn photovoltaic power stations, and are important for stabilizing regional photovoltaic power generation. The overall output fluctuation of the power station has certain reference significance.
结合图9和图10，可以得到，在本实施例中，在选择并网发电站时，当发电站数量一定时，尽量选择距离接近60km的光伏发电站。则在下午子时段并网时，则可选择E电站。下午退网时，选择与E电站超出60km的C电站，与步骤S4的预选则相同。Combining FIG. 9 and FIG. 10 , it can be obtained that in this embodiment, when selecting a grid-connected power station, when the number of power stations is constant, try to choose a photovoltaic power station with a distance close to 60km. Then, when connecting to the grid in the afternoon sub-period, E power station can be selected. When disconnecting from the network in the afternoon, select the C power station which is 60km away from the E power station, which is the same as the preselection in step S4.
应当指出的是，上述说明并非是对本发明的限制，本发明也并不仅限于上述举例，本技术领域的普通技术人员在本发明的实质范围内所做出的变化、改性、添加或替换，也应属于本发明的保护范围。It should be noted that the above description is not intended to limit the present invention, and the present invention is not limited to the above-mentioned examples. Those skilled in the art may make changes, modifications, additions or replacements within the scope of the present invention. It should also belong to the protection scope of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810146594.5ACN108321840B (en) | 2018-02-12 | 2018-02-12 | Selection method of grid-connected and off-grid based on refined output of photovoltaic power stations |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810146594.5ACN108321840B (en) | 2018-02-12 | 2018-02-12 | Selection method of grid-connected and off-grid based on refined output of photovoltaic power stations |
| Publication Number | Publication Date |
|---|---|
| CN108321840Atrue CN108321840A (en) | 2018-07-24 |
| CN108321840B CN108321840B (en) | 2020-11-27 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810146594.5AExpired - Fee RelatedCN108321840B (en) | 2018-02-12 | 2018-02-12 | Selection method of grid-connected and off-grid based on refined output of photovoltaic power stations |
| Country | Link |
|---|---|
| CN (1) | CN108321840B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111293681A (en)* | 2020-01-22 | 2020-06-16 | 重庆大学 | A Quantitative Evaluation Method for Output Volatility of PV Stations Based on RankBoost |
| CN112003274A (en)* | 2020-08-13 | 2020-11-27 | 国网江苏省电力有限公司无锡供电分公司 | Photovoltaic output prediction method and system |
| CN113159523A (en)* | 2021-03-30 | 2021-07-23 | 国家电网有限公司 | Method for quantitatively analyzing cluster effect of photovoltaic power station based on time domain correlation |
| CN114880378A (en)* | 2022-05-18 | 2022-08-09 | 国网福建省电力有限公司 | Wide-area distributed photovoltaic global output estimation method |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184475A (en)* | 2011-05-11 | 2011-09-14 | 浙江大学 | Optimizing and dispatching method for microgrid economical operation on basis of multiple time scale coordination |
| CN104242337A (en)* | 2014-08-14 | 2014-12-24 | 广东易事特电源股份有限公司 | Real-time coordination and control method of photovoltaic micro-grid system |
| CN105262131A (en)* | 2015-10-22 | 2016-01-20 | 华南理工大学 | Black-start system and black-start method applicable to light storage micro grid |
| WO2016179928A1 (en)* | 2015-05-14 | 2016-11-17 | 中国电力科学研究院 | Method for detecting islanding protection performance of inverter |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184475A (en)* | 2011-05-11 | 2011-09-14 | 浙江大学 | Optimizing and dispatching method for microgrid economical operation on basis of multiple time scale coordination |
| CN104242337A (en)* | 2014-08-14 | 2014-12-24 | 广东易事特电源股份有限公司 | Real-time coordination and control method of photovoltaic micro-grid system |
| WO2016179928A1 (en)* | 2015-05-14 | 2016-11-17 | 中国电力科学研究院 | Method for detecting islanding protection performance of inverter |
| CN105262131A (en)* | 2015-10-22 | 2016-01-20 | 华南理工大学 | Black-start system and black-start method applicable to light storage micro grid |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111293681A (en)* | 2020-01-22 | 2020-06-16 | 重庆大学 | A Quantitative Evaluation Method for Output Volatility of PV Stations Based on RankBoost |
| CN111293681B (en)* | 2020-01-22 | 2023-04-25 | 重庆大学 | Photovoltaic field station output fluctuation quantitative evaluation method based on RankBoost |
| CN112003274A (en)* | 2020-08-13 | 2020-11-27 | 国网江苏省电力有限公司无锡供电分公司 | Photovoltaic output prediction method and system |
| CN112003274B (en)* | 2020-08-13 | 2022-07-22 | 国网江苏省电力有限公司无锡供电分公司 | Photovoltaic output prediction method and system |
| CN113159523A (en)* | 2021-03-30 | 2021-07-23 | 国家电网有限公司 | Method for quantitatively analyzing cluster effect of photovoltaic power station based on time domain correlation |
| CN114880378A (en)* | 2022-05-18 | 2022-08-09 | 国网福建省电力有限公司 | Wide-area distributed photovoltaic global output estimation method |
| CN114880378B (en)* | 2022-05-18 | 2025-02-07 | 国网福建省电力有限公司 | A method for estimating global output of wide-area distributed photovoltaics |
| Publication number | Publication date |
|---|---|
| CN108321840B (en) | 2020-11-27 |
| Publication | Publication Date | Title |
|---|---|---|
| WO2023201552A1 (en) | County-wide photovoltaic prediction method based on cluster division and data enhancement | |
| CN110008982B (en) | Meteorological monitoring point selection method based on photovoltaic power generation output clustering | |
| CN107093007B (en) | A reliability assessment method for distribution network considering the continuous load capacity of photovoltaic storage | |
| CN112884601B (en) | Power system operation risk assessment method based on weather division strategy | |
| Kan et al. | The linkage between renewable energy potential and sustainable development: Understanding solar energy variability and photovoltaic power potential in Tibet, China | |
| CN110909911B (en) | Aggregation method of multidimensional time series data considering space-time correlation | |
| CN103683274B (en) | Regional long-term wind power generation capacity probability prediction method | |
| CN105184423B (en) | A kind of wind power plant cluster wind speed forecasting method | |
| CN108321840B (en) | Selection method of grid-connected and off-grid based on refined output of photovoltaic power stations | |
| CN114662922B (en) | Resident demand response potential evaluation method and system considering photovoltaic uncertainty | |
| CN104578157A (en) | Load flow calculation method of distributed power supply connection power grid | |
| CN111275238A (en) | Generation method of photovoltaic output sequence of large-scale power station based on hourly clear sky index | |
| CN108960526A (en) | A kind of distributed photovoltaic based on region equivalent goes out force prediction method and system | |
| CN105335560A (en) | Photovoltaic generation power volatility and automatic generation control reserve demand computing method thereof | |
| CN112165084A (en) | Multi-time-scale optimization method considering PV-load bilateral forecast uncertainty | |
| CN118536755A (en) | New energy base DC (direct current) external power supply system typical planning scene generation method | |
| CN115204444A (en) | Photovoltaic power prediction method based on improved cluster analysis and fusion integration algorithm | |
| CN118611033A (en) | A distributed photovoltaic spatiotemporal distribution prediction method and system | |
| Sun et al. | Measuring dynamics of solar energy resource quality: Methodology and policy implications for reducing regional energy inequality | |
| Wang et al. | A two-step load disaggregation algorithm for quasi-static time-series analysis on actual distribution feeders | |
| CN117592255A (en) | Wind-light output sequence modeling method considering space-time correlation | |
| Xiao et al. | A statistical photovoltaic power forecast model (spf) based on historical power and weather data | |
| Ray et al. | Performance assessment of prospective pv systems in queensland and new south wales of australia | |
| CN110212591A (en) | A kind of distributed photovoltaic irradiation level measurement points distributing method based on compressed sensing technology | |
| Guwaeder et al. | A study of the monthly insolation in libya |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee | Granted publication date:20201127 |