WO2017054537A1 - 一种长时间尺度光伏出力时间序列建模方法及装置 - Google Patents

一种长时间尺度光伏出力时间序列建模方法及装置 Download PDF

Info

Publication number
WO2017054537A1
WO2017054537A1 PCT/CN2016/087809 CN2016087809W WO2017054537A1 WO 2017054537 A1 WO2017054537 A1 WO 2017054537A1 CN 2016087809 W CN2016087809 W CN 2016087809W WO 2017054537 A1 WO2017054537 A1 WO 2017054537A1
Authority
WO
WIPO (PCT)
Prior art keywords
weather
time series
output
photovoltaic
probability
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2016/087809
Other languages
English (en)
French (fr)
Inventor
王伟胜
刘纯
李驰
黄越辉
王跃峰
董存
刘德伟
张楠
礼晓飞
高云峰
马烁
许晓艳
许彦平
潘霄峰
李丽
王江元
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Electric Power Research Institute Co Ltd CEPRI
CLP Puri Zhangbei Wind Power Research and Test Ltd
State Grid Corp of China SGCC
Original Assignee
China Electric Power Research Institute Co Ltd CEPRI
CLP Puri Zhangbei Wind Power Research and Test Ltd
State Grid Corp of China SGCC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Electric Power Research Institute Co Ltd CEPRI, CLP Puri Zhangbei Wind Power Research and Test Ltd, State Grid Corp of China SGCC filed Critical China Electric Power Research Institute Co Ltd CEPRI
Priority to US15/751,471 priority Critical patent/US20180240200A1/en
Publication of WO2017054537A1 publication Critical patent/WO2017054537A1/zh
Anticipated expiration legal-status Critical
Priority to US15/963,073 priority patent/US10290066B2/en
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24S—SOLAR HEAT COLLECTORS; SOLAR HEAT SYSTEMS
    • F24S2201/00—Prediction; Simulation

Definitions

  • the invention relates to a modeling technology, in particular to a long-time scale photovoltaic output time series modeling method and device.
  • Photovoltaic power generation is a renewable energy technology with the greatest potential and application value after wind power. With the support of a series of supporting policies in China, photovoltaic power generation has developed rapidly. As photovoltaic power generation accounts for the increase in the proportion of the entire power system power supply, a deep understanding of the characteristics of the photovoltaic output itself can accurately grasp the impact of photovoltaic grid-connected power system, so that the power system can more effectively solve the photovoltaic access problem.
  • the existing weather simulation technology can only realize the PV's annual/month electricity forecast, and can't realize the long-term power forecast. It can't directly obtain the time series that can be used for the power system time series production simulation. Therefore, it is necessary to build the PV output time series.
  • the model in order to accurately grasp the law of the change of photovoltaic power generation output, provides essential basic data for time series production simulation with large-scale new energy, annual new energy consumption capacity analysis, and annual plan formulation.
  • an embodiment of the present invention provides a long-time scale photovoltaic output time series modeling method, which analyzes the characteristics of a photovoltaic output time series, and uses a Markov chain to simulate various weather transition processes to obtain a transition probability.
  • a new method for constructing future PV output scenarios is proposed.
  • Embodiments of the present invention provide a modeling method for a long-term scale photovoltaic output time series.
  • the method includes:
  • the calculating the transition probability between various types of weather separately includes: using a Markov chain to simulate various weather transition processes, and obtaining a transition probability between each weather type, and expressing the same
  • the formula is:
  • P k is the probability of a sunny day shifting to other weather types; k is the weather type, N k is the number of transitions; N 1 is the number of sunny weather occurrences.
  • the method further includes: using a probability calculation method for transferring other weather types by using the sunny day, and sequentially obtaining transition probabilities between other weather types.
  • the generating a simulated time series of a preset time-scale photovoltaic output includes: sequentially randomly selecting weather within a preset time scale according to a transition probability between the weather types. The type and the corresponding relative output; and calculate the product of the relative output force and the predetermined threshold to generate a simulated time series of the photovoltaic output;
  • the simulated time series is a graph for reflecting the probability distribution characteristics and autocorrelation characteristics of the multi-time scale photovoltaic output And changes in short-term fluctuation characteristics;
  • the short-term fluctuation characteristic is: a maximum probability distribution characteristic of the photovoltaic output when the time scale is t, 15 min ⁇ t ⁇ 60 min;
  • the maximum probability distribution characteristic is a difference between the maximum output and the minimum output in the time scale t; The maximum output occurs after the minimum output, then the difference is positive, and if it occurs before the minimum output, the difference is negative.
  • the verifying the validity of the simulated time series includes:
  • the probability distribution feature C f , the short-term fluctuation feature C d and the autocorrelation feature C r of the simulated time series are respectively defined;
  • the validity of the time series is quantitatively evaluated by using the root mean square error RMSE of each feature.
  • the expression is:
  • a unit vector representing a function value of each feature of the simulated time series
  • y i represents a characteristic function value of a historical time series corresponding to each feature of the simulated time series
  • n is a length of a set of characteristic function values of the time series
  • the embodiment of the invention provides a modeling device for a long-term scale photovoltaic output time series, wherein the device comprises:
  • the data acquisition unit is configured to collect historical data of the photovoltaic power station, and select a photovoltaic output with a length of one year and a time resolution of 15 minutes;
  • the acquiring unit is configured to obtain, from the weather station, a weather type corresponding to each day of the photovoltaic output, the weather type including at least one of sunny, cloudy, cloudy, and changing weather;
  • a processing unit configured to calculate a transition probability between various types of weather
  • Generating unit configured to generate a simulated time series of preset time scale photovoltaic outputs
  • An evaluation unit configured to verify the validity of the simulated time series.
  • the processing unit is further configured to: simulate a transition process of various weathers by using a Markov chain, and obtain a transition probability between each weather type, and the expression is:
  • P k is the probability of a sunny day shifting to other weather types; k is the weather type, N k is the number of transitions; N 1 is the number of sunny weather occurrences.
  • the apparatus further includes: a probability acquiring unit configured to: use the probability calculation method of transferring other weather types by the sunny day, and sequentially obtain transition probabilities between other weather types.
  • the generating unit is further configured to sequentially randomly select a weather type and a corresponding relative output in a preset time scale according to a transition probability between the weather types; and Calculating a product of a relative output force and a predetermined threshold to generate a simulated time series of photovoltaic output;
  • the simulated time series is a graph for reflecting a change in probability distribution characteristics, autocorrelation characteristics, and short-term fluctuation characteristics of the multi-time scale photovoltaic output;
  • the short-term fluctuation characteristic is: a maximum probability distribution characteristic of the photovoltaic output when the time scale is t, 15 min ⁇ t ⁇ 60 min;
  • the maximum probability distribution characteristic is a difference between the maximum output and the minimum output force in the time scale t; the maximum output force occurs after the minimum output force, and the difference is positive, and if it occurs before the minimum output force, the difference is negative.
  • the evaluation unit is further configured to:
  • the probability distribution feature C f , the short-term fluctuation feature C d and the autocorrelation feature C r of the simulated time series are respectively defined;
  • the validity of the time series is quantitatively evaluated by using the root mean square error RMSE of each feature.
  • the expression is:
  • I a unit vector, an analog time series of feature function values
  • y i represents the time series simulation of the characteristics corresponding to the historical time series of characteristic function value
  • n is the length of the time series of feature set of function values
  • the RMSE is less than ⁇ and ranges from 0.1 to 0.2.
  • the beneficial effects that can be achieved by using the embodiments of the present invention are: using Markov chain to simulate various weather transfer processes, calculating the transition probability between weather types; simulating the randomness and volatility of photovoltaics, etc.
  • the deterministic characteristics, the construction results are more in line with the photovoltaic output characteristics than other methods, and truly and accurately characterize the future output of photovoltaics. It can generate annual and monthly PV output simulation time series according to the stochastic fluctuation law of photovoltaic timing according to demand, and provide essential basic data for time series production simulation with large-scale new energy, annual new energy consumption capacity analysis, and annual plan formulation. .
  • FIG. 1 is a flowchart of a long-time photovoltaic output time series modeling method according to an embodiment of the present invention
  • FIG. 5 are schematic diagrams showing comparison of parameters of a historical time series and an analog time series according to an embodiment of the present invention
  • Figure 3 is a schematic diagram of a 15min probability distribution
  • Figure 4 is a schematic diagram of a 60 min probability distribution
  • Figure 5 is a schematic diagram of the comparison of autocorrelation coefficients.
  • a method for modeling a long-term scale photovoltaic output time series includes:
  • Step 101 Collect historical data of the photovoltaic power station, and select a photovoltaic output with a length of one year and a time resolution of 15 minutes;
  • Step 102 Obtain weather types of the day corresponding to the photovoltaic output from the weather station, including sunny, cloudy, cloudy, and changing weather;
  • Step 103 Calculate the transition probability between various types of weather separately; simulate a transition process of various weathers by using a Markov chain, and obtain a transition probability between each weather type, and the expression is:
  • P k is the probability of a sunny day shifting to other weather types; k is the weather type, N k is the number of transitions; N 1 is the number of sunny weather occurrences.
  • the transition probability between other weather types is sequentially obtained.
  • the subscript 1 is the cloudy weather type
  • the subscript 2 is the sunny weather type
  • the subscript 3 is the cloudy weather type
  • the subscript 4 is the changing weather type
  • P (1-1) , P (1-2) , P (1-3) , and P (1-4) indicate the probability of cloudy weather shifting to other weather types
  • N (1-4) indicate the number of times the cloudy shifts to other weather types
  • N (1) indicates the number of cloudy weather types.
  • the probability of transition between cloudy and changing weather can be calculated.
  • Step 104 Generate a simulated time series of the scaled photovoltaic output in a preset time period
  • the predetermined threshold is based on historical PV data and historical time
  • the inter-sequence is a reference basis, and the set standard value is customized
  • the simulated time series is a graph for reflecting a probability density function (PDF) and an autocorrelation function (autocorrelation function) of the multi-time scale photovoltaic output. ACF) and changes in short-term fluctuation characteristics;
  • the short-term fluctuation characteristic is: a maximum probability distribution characteristic of the photovoltaic output when the time scale is t, 15 min ⁇ t ⁇ 60 min;
  • the maximum probability distribution characteristic is a difference between the maximum output and the minimum output force in the time scale t; the maximum output force occurs after the minimum output force, and the difference is positive, and if it occurs before the minimum output force, the difference is negative.
  • Step 105 Verify the validity of the simulated time series. As shown in the various schematic diagrams of Figures 2-5.
  • Step 1051 respectively defining a probability distribution feature C f of the simulated time series, a short-term fluctuation feature C d and an autocorrelation feature C r ;
  • Step 1052 quantitatively evaluate the validity of the time series by using the root mean square error RMSE of each feature, and the expression is:
  • a unit vector representing a function value of each feature of the simulated time series
  • y i represents a characteristic function value of a historical time series corresponding to each feature of the simulated time series
  • n is a length of a set of characteristic function values of the time series
  • Figure 2 is a schematic diagram of probability distribution, as shown in Figure 2, when Time, the probability distribution characteristic function value of the simulated time series is represented.
  • y i represents the historical time series probability distribution characteristic function value corresponding to the simulated time series probability distribution feature
  • FIG. 3 is a 15 min probability distribution diagram
  • FIG. 4 is 60 min.
  • y i represents the historical time series short-term fluctuation characteristic function value corresponding to the short-time fluctuation characteristic of the simulated time series;
  • FIG. 1 represents the historical time series short-term fluctuation characteristic function value corresponding to the short-time fluctuation characteristic of the simulated time series
  • FIG. 5 is a schematic diagram of the comparison of the auto-correlation coefficients, such as Figure 5, when When, it represents the analog value of the autocorrelation function wherein the time series, and the analog case y i denotes a time series from the time history of the sequence corresponding to relevant characteristics Autocorrelation function value;
  • a modeling device for a long-term scale photovoltaic output time series comprising:
  • the data acquisition unit is configured to collect historical data of the photovoltaic power station, and select a photovoltaic output with a length of one year and a time resolution of 15 minutes;
  • An acquisition unit configured to obtain, from the weather station, a weather type corresponding to each day of the photovoltaic output, including sunny, cloudy, cloudy, and changing weather;
  • a processing unit configured to calculate a transition probability between various types of weather
  • Generating unit configured to generate a simulated time series of preset time scale photovoltaic outputs
  • An evaluation unit configured to verify the validity of the simulated time series.
  • the processing unit is further configured to: simulate a transition process of various weathers by using a Markov chain, and obtain a transition probability between each weather type, and the expression is:
  • P k is the probability of a sunny day shifting to other weather types; k is the weather type, N k is the number of transitions; N 1 is the number of sunny weather occurrences.
  • the apparatus further includes: a probability acquiring unit configured to: use the probability calculation method of transferring other weather types by the sunny day, and sequentially obtain transition probabilities between other weather types.
  • the generating unit is further configured to sequentially randomly select a weather type and a corresponding relative output in a preset time scale according to a transition probability between the weather types; and Calculating a product of a relative output force and a predetermined threshold to generate a simulated time series of photovoltaic output;
  • the simulated time series is a graph for reflecting a change in probability distribution characteristics, autocorrelation characteristics, and short-term fluctuation characteristics of the multi-time scale photovoltaic output;
  • the short-term fluctuation characteristic is: a maximum probability distribution characteristic of the photovoltaic output when the time scale is t, 15 min ⁇ t ⁇ 60 min;
  • the maximum probability distribution characteristic is a difference between the maximum output and the minimum output force in the time scale t; the maximum output force occurs after the minimum output force, and the difference is positive, and if it occurs before the minimum output force, the difference is negative.
  • the evaluation unit is further configured to:
  • the probability distribution feature C f , the short-term fluctuation feature C d and the autocorrelation feature C r of the simulated time series are respectively defined;
  • the validity of the time series is quantitatively evaluated by using the root mean square error RMSE of each feature.
  • the expression is:
  • a unit vector representing a function value of each feature of the simulated time series
  • y i represents a characteristic function value of a historical time series corresponding to each feature of the simulated time series
  • n is a length of a set of characteristic function values of the time series
  • the Markov chain is used to simulate various weather transfer processes to calculate the transition probability between weather types; the uncertainty characteristics of photovoltaic randomness and volatility are simulated, and the construction results are compared with other methods. More in line with the photovoltaic output characteristics, a true and accurate representation of the future output of photovoltaics. It can generate annual and monthly PV output simulation time series according to the stochastic fluctuation law of photovoltaic timing according to demand, and provide essential basic data for time series production simulation with large-scale new energy, annual new energy consumption capacity analysis, and annual plan formulation. .

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Marketing (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Development Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Supply And Distribution Of Alternating Current (AREA)
  • Photovoltaic Devices (AREA)

Abstract

一种长时间尺度光伏出力时间序列建模方法和装置，该方法包括：采集光伏电站的历史数据，选取时间长度为1年且时间分辨率为15min的光伏出力（101）；从气象站获取光伏出力对应的各日天气类型（102），分别计算各天气之间的转移概率（103）；生成预设时间尺度内光伏出力的模拟时间序列（104）并验证其有效性（105）。该方法可以根据不同需求获取年度、月度符合光伏时序随机波动规律的模拟时间序列，为含大规模新能源的时序生产模拟仿真提供有利条件和数据支持。

Description

一种长时间尺度光伏出力时间序列建模方法及装置 技术领域
本发明涉及建模技术，尤其涉及一种长时间尺度光伏出力时间序列建模方法及装置。
背景技术
光伏发电是继风电后具有最大潜力和应用价值的可再生能源技术，在我国一系列配套政策支持下，光伏发电发展迅速。随着光伏发电占整个电力系统电源比重的增加，深刻认识光伏出力本身所具有的特性规律可以准确把握光伏并网对电力系统的影响，使电力系统可以更有效地解决光伏接入难题。
现有的天气模拟技术仅能实现光伏的年/月电量预测，无法实现长时间尺度的功率预测，不能直接得到可用于电力系统时序生产模拟仿真的时间序列，因此需要对光伏出力时间序列进行建模，以准确把握光伏发电出力变化规律，为含大规模新能源的时序生产模拟仿真、年度新能源消纳能力分析、年度计划制定提供必不可少的基础数据。
发明内容
为了实现上述目的，本发明实施例提供一种长时间尺度光伏出力时间序列建模方法，通过分析光伏出力时间序列的特性，并采用马尔科夫链模拟各类天气的转移过程，获取转移概率，来生成模拟光伏序列，为构造未来光伏出力场景提出了新的方法。
本发明实施例是采用下述技术方案实现的：
本发明实施例提供了一种长时间尺度光伏出力时间序列的建模方法， 所述方法包括：
采集光伏电站的历史数据，选取时间长度为1年且时间分辨率为15min的光伏出力；
获取光伏出力对应的各日天气类型，所述天气类型包括晴天、多云、阴天、变化天气中的至少一种；
分别计算各类天气之间的转移概率；
生成预设时间内尺度光伏出力的模拟时间序列；
验证所述模拟时间序列的有效性。
在本发明实施例一实施方式中，所述分别计算各类天气之间的转移概率，包括：采用马尔科夫链模拟各类天气的转移过程，获取各天气类型之间的转移概率，其表达式为：
Figure PCTCN2016087809-appb-000001
式(1)中，Pk为晴天转移为其他天气类型的概率；k表示天气类型，Nk为转移次数；N1为出现晴天天气的次数。
在本发明实施例一实施方式中，还包括：利用所述晴天转移其他天气类型的概率计算方法，依次获得其他天气类型之间的转移概率。
在本发明实施例一实施方式中，所述生成预设时间内尺度光伏出力的模拟时间序列，包括：根据所述各天气类型之间的转移概率，序贯随机抽取预设时间尺度内的天气类型以及对应的相对出力；并计算相对出力与预定阈值的乘积，生成光伏出力的模拟时间序列；所述模拟时间序列为曲线图，用于反映多时间尺度光伏出力的概率分布特征、自相关特征和短时波动特征的变化；
所述短时波动特征为：时间尺度为t时，光伏出力的最大概率分布特征，15min≤t≤60min；
所述最大概率分布特征为该时间尺度t内最大出力与最小出力的差值； 所述最大出力出现在最小出力之后，则差值为正，若出现在所述最小出力之前，则差值为负。
在本发明实施例一实施方式中，所述验证所述模拟时间序列的有效性，包括：
分别定义模拟时间序列的概率分布特征Cf、短时波动特征Cd和自相关特征Cr；
采用各项特征的均方根误差RMSE，定量评价时间序列的有效性，其表达式为：
Figure PCTCN2016087809-appb-000002
其中，
Figure PCTCN2016087809-appb-000003
为单位向量，表示模拟时间序列各项特征的函数值；yi表示与模拟时间序列各项特征相对应的历史时间序列各项特征函数值；n为时间序列各项特征函数值集合的长度,所述RMSE小于ε，其取值范围为0.1～0.2。
本发明实施例提供了一种长时间尺度光伏出力时间序列的建模装置，其中，所述装置包括：
数据采集单元，配置为采集光伏电站的历史数据，选取时间长度为1年且时间分辨率为15min的光伏出力；
获取单元，配置为从气象站获取光伏出力对应的各日天气类型，所述天气类型包括晴天、多云、阴天、变化天气中的至少一种；
处理单元，配置为分别计算各类天气之间的转移概率；
生成单元，配置为生成预设时间尺度光伏出力的模拟时间序列；
评估单元，配置为验证模拟时间序列的有效性。
在本发明实施例一实施方式中，所述处理单元，进一步配置为：采用马尔科夫链模拟各类天气的转移过程，获取各天气类型之间的转移概率，其表达式为：
Figure PCTCN2016087809-appb-000004
式(1)中，Pk为晴天转移为其他天气类型的概率；k表示天气类型，Nk为转移次数；N1为出现晴天天气的次数。
在本发明实施例一实施方式中，所述装置还包括：概率获取单元，配置为：利用所述晴天转移其他天气类型的概率计算方法，依次获得其他天气类型之间的转移概率。
在本发明实施例一实施方式中，所述生成单元，进一步配置为：根据所述各天气类型之间的转移概率，序贯随机抽取预设时间尺度内的天气类型以及对应的相对出力；并计算相对出力与预定阈值的乘积，生成光伏出力的模拟时间序列；所述模拟时间序列为曲线图，用于反映多时间尺度光伏出力的概率分布特征、自相关特征和短时波动特征的变化；
所述短时波动特征为：时间尺度为t时，光伏出力的最大概率分布特征，15min≤t≤60min；
所述最大概率分布特征为该时间尺度t内最大出力与最小出力的差值；所述最大出力出现在最小出力之后，则差值为正，若出现在所述最小出力之前，则差值为负。
在本发明实施例一实施方式中，所述评估单元，进一步配置为：
分别定义模拟时间序列的概率分布特征Cf、短时波动特征Cd和自相关特征Cr；
采用各项特征的均方根误差RMSE，定量评价时间序列的有效性，其表达式为：
Figure PCTCN2016087809-appb-000005
其中，
Figure PCTCN2016087809-appb-000006
为单位向量，表示模拟时间序列各项特征的函数值；yi表示与模拟时间序列各项特征相对应的历史时间序列各项特征函 数值；n为时间序列各项特征函数值集合的长度,所述RMSE小于ε，其取值范围为0.1～0.2。
与现有技术比，采用本发明实施例能达到的有益效果是：采用马尔科夫链模拟各类天气的转移过程，计算天气类型间的转移概率；模拟了光伏的随机性与波动性等不确定性特点，构造结果相比其它方法更符合光伏出力特性，真实准确地表征了光伏的未来出力情况。可以根据需求产生年度、月度符合光伏时序随机波动规律的光伏出力模拟时间序列，为含大规模新能源的时序生产模拟仿真、年度新能源消纳能力分析、年度计划制定提供必不可少的基础数据。
附图说明
图1为本发明实施例提供的一种长时间尺度光伏出力时间序列建模方法流程图；
图2-图5为本发明实施例提供的历史时间序列与模拟时间序列的参数对比的各个示意图；其中，
图2为概率分布示意图；
图3为15min概率分布示意图；
图4为为60min概率分布示意图；
图5为自相关系数对比示意图。
具体实施方式
下面结合附图对本发明的具体实施方式作进一步的详细说明。
如图1所示，本发明实施例的一种长时间尺度光伏出力时间序列的建模方法，所述方法包括：
步骤101、采集光伏电站的历史数据，选取时间长度为1年且时间分辨率为15min的光伏出力；
步骤102、从气象站获取光伏出力对应的各日天气类型，包括晴天、多云、阴天和变化天气；
步骤103、分别计算各类天气之间的转移概率；采用马尔科夫链模拟各类天气的转移过程，获取各天气类型之间的转移概率，其表达式为：
Figure PCTCN2016087809-appb-000007
式(1)中，Pk为晴天转移为其他天气类型的概率；k表示天气类型，Nk为转移次数；N1为出现晴天天气的次数。
利用所述晴天转移其他天气类型的概率计算方法，依次获得其他天气类型之间的转移概率。
例如计算多云天气转移其他天气类型的概率，其表达式为：
Figure PCTCN2016087809-appb-000008
Figure PCTCN2016087809-appb-000009
Figure PCTCN2016087809-appb-000010
Figure PCTCN2016087809-appb-000011
式中，下标为1的为多云天气类型，下标为2的为晴天天气类型，下标为3的为阴天天气类型，下标为4的为变化天气类型；P(1-1)、P(1-2)、P(1-3)、P(1-4)分别表示多云天气转移到其他天气类型的概率，N(1-1)、N(1-2)、N(1-3)、N(1-4)分别表示多云转移到其他天气类型的次数，N(1)表示出现多云天气类型的次数，同理可计算阴天和变化天气的转移概率。
步骤104、生成预设时间内尺度光伏出力的模拟时间序列；
根据所述各天气类型之间的转移概率，序贯随机抽取预设时间尺度内的天气类型以及对应的相对出力；并计算相对出力与预定阈值的乘积，生成光伏出力的模拟时间序列；所述预定阈值是根据历史光伏数据和历史时 间序列为参考依据，自定义设定的标准值；所述模拟时间序列为曲线图，用于反映多时间尺度光伏出力的概率分布特征(probability density function，PDF)、自相关特征(autocorrelation function，ACF)和短时波动特征的变化；
所述短时波动特征为：时间尺度为t时，光伏出力的最大概率分布特征，15min≤t≤60min；
所述最大概率分布特征为该时间尺度t内最大出力与最小出力的差值；所述最大出力出现在最小出力之后，则差值为正，若出现在所述最小出力之前，则差值为负。
步骤105、验证所述模拟时间序列的有效性。如图2-图5的各个示意图所示。
这里，本步骤的具体处理过程包括：
步骤1051：分别定义模拟时间序列的概率分布特征Cf、短时波动特征Cd和自相关特征Cr；
步骤1052：采用各项特征的均方根误差RMSE，定量评价时间序列的有效性，其表达式为：
Figure PCTCN2016087809-appb-000012
其中，
Figure PCTCN2016087809-appb-000013
为单位向量，表示模拟时间序列各项特征的函数值；yi表示与模拟时间序列各项特征相对应的历史时间序列各项特征函数值；n为时间序列各项特征函数值集合的长度,所述RMSE小于ε，其取值范围为0.1～0.2。
图2为概率分布示意图，如图2所示，当
Figure PCTCN2016087809-appb-000014
时，表示模拟时间序列的概率分布特征函数值，此时yi表示与模拟时间序列概率分布特征相对应的历史时间序列概率分布特征函数值；图3为15min概率分布示意图，图4 为为60min概率分布示意图，如图3和图4所示,当
Figure PCTCN2016087809-appb-000015
时，表示模拟时间序列的短时波动特征函数值，此时yi表示与模拟时间序列短时波动特征相对应的历史时间序列短时波动特征函数值；图5为自相关系数对比示意图，如图5所示，当
Figure PCTCN2016087809-appb-000016
时，表示模拟时间序列的自相关特征函数值，此时yi表示与模拟时间序列自相关特征相对应的历史时间序列自相关特征函数值；
本发明实施例的一种长时间尺度光伏出力时间序列的建模装置，所述装置包括：
数据采集单元，配置为采集光伏电站的历史数据，选取时间长度为1年且时间分辨率为15min的光伏出力；
获取单元，配置为从气象站获取光伏出力对应的各日天气类型，包括晴天、多云、阴天和变化天气；
处理单元，配置为分别计算各类天气之间的转移概率；
生成单元，配置为生成预设时间尺度光伏出力的模拟时间序列；
评估单元，配置为验证模拟时间序列的有效性。
在本发明实施例一实施方式中，所述处理单元，进一步配置为：采用马尔科夫链模拟各类天气的转移过程，获取各天气类型之间的转移概率，其表达式为：
Figure PCTCN2016087809-appb-000017
式(1)中，Pk为晴天转移为其他天气类型的概率；k表示天气类型，Nk为转移次数；N1为出现晴天天气的次数。
在本发明实施例一实施方式中，所述装置还包括：概率获取单元，配置为：利用所述晴天转移其他天气类型的概率计算方法，依次获得其他天气类型之间的转移概率。
在本发明实施例一实施方式中，所述生成单元，进一步配置为：根据所述各天气类型之间的转移概率，序贯随机抽取预设时间尺度内的天气类型以及对应的相对出力；并计算相对出力与预定阈值的乘积，生成光伏出力的模拟时间序列；所述模拟时间序列为曲线图，用于反映多时间尺度光伏出力的概率分布特征、自相关特征和短时波动特征的变化；
所述短时波动特征为：时间尺度为t时，光伏出力的最大概率分布特征，15min≤t≤60min；
所述最大概率分布特征为该时间尺度t内最大出力与最小出力的差值；所述最大出力出现在最小出力之后，则差值为正，若出现在所述最小出力之前，则差值为负。
在本发明实施例一实施方式中，所述评估单元，进一步配置为：
分别定义模拟时间序列的概率分布特征Cf、短时波动特征Cd和自相关特征Cr；
采用各项特征的均方根误差RMSE，定量评价时间序列的有效性，其表达式为：
Figure PCTCN2016087809-appb-000018
其中，
Figure PCTCN2016087809-appb-000019
为单位向量，表示模拟时间序列各项特征的函数值；yi表示与模拟时间序列各项特征相对应的历史时间序列各项特征函数值；n为时间序列各项特征函数值集合的长度,所述RMSE小于ε，其取值范围为0.1～0.2。
最后应当说明的是：以上实施例仅用以说明本发明的技术方案而非对其限制，尽管参照上述实施例对本发明进行了详细的说明，所属领域的普通技术人员应当理解：依然可以对本发明的具体实施方式进行修改或者等同替换，而未脱离本发明精神和范围的任何修改或者等同替换，其均应涵盖在本发明的权利要求范围当中。
工业实用性
采用本发明实施例，通过采用马尔科夫链模拟各类天气的转移过程，以计算天气类型间的转移概率；模拟了光伏的随机性与波动性等不确定性特点，构造结果相比其它方法更符合光伏出力特性，真实准确地表征了光伏的未来出力情况。可以根据需求产生年度、月度符合光伏时序随机波动规律的光伏出力模拟时间序列，为含大规模新能源的时序生产模拟仿真、年度新能源消纳能力分析、年度计划制定提供必不可少的基础数据。

Claims (10)

  1. 一种长时间尺度光伏出力时间序列的建模方法，所述方法包括：
    采集光伏电站的历史数据，选取时间长度为1年且时间分辨率为15min的光伏出力；
    获取光伏出力对应的各日天气类型，所述天气类型包括晴天、多云、阴天、变化天气中的至少一种；
    分别计算各类天气之间的转移概率；
    生成预设时间尺度内光伏出力的模拟时间序列；
    验证所生成模拟时间序列的有效性。
  2. 如权利要求1所述的方法，其中，所述分别计算各类天气之间的转移概率，包括：采用马尔科夫链模拟各类天气的转移过程，获取各天气类型之间的转移概率，其表达式为：
    Figure PCTCN2016087809-appb-100001
    式(1)中，Pk为晴天转移为其他天气类型的概率；k表示天气类型，Nk为转移次数；N1为出现晴天天气的次数。
  3. 如权利要求2所述的方法，其中，所述方法还包括：利用所述晴天转移其他天气类型的概率计算方法，依次获得其他天气类型之间的转移概率。
  4. 如权利要求1或3所述的方法，其中，所述生成预设时间内尺度光伏出力的模拟时间序列，包括：根据所述各天气类型之间的转移概率，序贯随机抽取预设时间尺度内的天气类型以及对应的相对出力；并计算相对出力与预定阈值的乘积，生成光伏出力的模拟时间序列；所述模拟时间序列为曲线图，用于反映多时间尺度光伏出力的概率分布特征、自相关特征和短时波动特征的变化；
    所述短时波动特征为：时间尺度为t时，光伏出力的最大概率分布特征，15min≤t≤60min；
    所述最大概率分布特征为该时间尺度t内最大出力与最小出力的差值；所述最大出力出现在最小出力之后，则差值为正，若出现在所述最小出力之前，则差值为负。
  5. 如权利要求1所述的方法，其中，所述验证所述模拟时间序列的有效性，包括：
    分别定义模拟时间序列的概率分布特征Cf、短时波动特征Cd和自相关特征Cr；
    采用各项特征的均方根误差RMSE，定量评价时间序列的有效性，其表达式为：
    Figure PCTCN2016087809-appb-100002
    其中，
    Figure PCTCN2016087809-appb-100003
    Figure PCTCN2016087809-appb-100004
    为单位向量，表示模拟时间序列各项特征的函数值；yi表示与模拟时间序列各项特征相对应的历史时间序列各项特征函数值，n为时间序列各项特征函数值集合的长度,所述RMSE小于ε，其取值范围为0.1～0.2。
  6. 一种长时间尺度光伏出力时间序列的建模装置，所述装置包括：数据采集单元，配置为采集光伏电站的历史数据，选取时间长度为1年且时间分辨率为15min的光伏出力；
    获取单元，配置为从气象站获取光伏出力对应的各日天气类型，所述天气类型包括晴天、多云、阴天和变化天气中的至少一种；
    处理单元，配置为分别计算各类天气之间的转移概率；
    生成单元，配置为生成预设时间尺度光伏出力的模拟时间序列；
    评估单元，配置为验证模拟时间序列的有效性。
  7. 如权利要求6所述的装置，其中，所述处理单元，进一步配置为： 采用马尔科夫链模拟各类天气的转移过程，获取各天气类型之间的转移概率，其表达式为：
    Figure PCTCN2016087809-appb-100005
    式(1)中，Pk为晴天转移为其他天气类型的概率；k表示天气类型，Nk为转移次数；N1为出现晴天天气的次数。
  8. 如权利要求7所述的装置，其中，所述装置还包括：概率获取单元，配置为：利用所述晴天转移其他天气类型的概率计算方法，依次获得其他天气类型之间的转移概率。
  9. 如权利要求6或8所述的装置，其中，所述生成单元，进一步配置为：根据所述各天气类型之间的转移概率，序贯随机抽取预设时间尺度内的天气类型以及对应的相对出力；并计算相对出力与预定阈值的乘积，生成光伏出力的模拟时间序列；所述模拟时间序列为曲线图，用于反映多时间尺度光伏出力的概率分布特征、自相关特征和短时波动特征的变化；
    所述短时波动特征为：时间尺度为t时，光伏出力的最大概率分布特征，15min≤t≤60min；
    所述最大概率分布特征为该时间尺度t内最大出力与最小出力的差值；所述最大出力出现在最小出力之后，则差值为正，若出现在所述最小出力之前，则差值为负。
  10. 如权利要求6所述的装置，其中，所述评估单元，进一步配置为：
    分别定义模拟时间序列的概率分布特征Cf、短时波动特征Cd和自相关特征Cr；
    采用各项特征的均方根误差RMSE，定量评价时间序列的有效性，其表达式为：
    Figure PCTCN2016087809-appb-100006
    其中，
    Figure PCTCN2016087809-appb-100007
    Figure PCTCN2016087809-appb-100008
    为单位向量，表示模拟时间序列各项特征的函数值；yi表示与模拟时间序列各项特征相对应的历史时间序列各项特征函数值；n为时间序列各项特征函数值集合的长度,所述RMSE小于ε，其取值范围为0.1～0.2。
PCT/CN2016/087809 2015-09-30 2016-06-30 一种长时间尺度光伏出力时间序列建模方法及装置 Ceased WO2017054537A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US15/751,471 US20180240200A1 (en) 2015-09-30 2016-06-30 Method and device for modeling a long-time-scale photovoltaic output time sequence
US15/963,073 US10290066B2 (en) 2015-09-30 2018-04-25 Method and device for modeling a long-time-scale photovoltaic output time sequence

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510639474.5A CN106557828A (zh) 2015-09-30 2015-09-30 一种长时间尺度光伏出力时间序列建模方法和装置
CN201510639474.5 2015-09-30

Related Child Applications (2)

Application Number Title Priority Date Filing Date
US15/751,471 A-371-Of-International US20180240200A1 (en) 2015-09-30 2016-06-30 Method and device for modeling a long-time-scale photovoltaic output time sequence
US15/963,073 Continuation US10290066B2 (en) 2015-09-30 2018-04-25 Method and device for modeling a long-time-scale photovoltaic output time sequence

Publications (1)

Publication Number Publication Date
WO2017054537A1 true WO2017054537A1 (zh) 2017-04-06

Family

ID=58417673

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/087809 Ceased WO2017054537A1 (zh) 2015-09-30 2016-06-30 一种长时间尺度光伏出力时间序列建模方法及装置

Country Status (3)

Country Link
US (2) US20180240200A1 (zh)
CN (1) CN106557828A (zh)
WO (1) WO2017054537A1 (zh)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109449993A (zh) * 2018-12-26 2019-03-08 西安交通大学 一种面向多能互补电力系统的多时间尺度生产模拟方法
CN109492315A (zh) * 2018-11-19 2019-03-19 西安交通大学 一种基于Copula函数的时空相关风光序列模拟方法
CN110544016A (zh) * 2019-08-09 2019-12-06 国网江苏省电力有限公司电力科学研究院 评估气象因素对电力设备故障概率影响程度的方法和设备
CN110717277A (zh) * 2019-10-14 2020-01-21 河北工业大学 一种时序风速模拟方法
CN112116127A (zh) * 2020-08-20 2020-12-22 中国农业大学 一种基于气象过程与功率波动关联的光伏功率预测方法
CN115189421A (zh) * 2022-08-01 2022-10-14 长沙理工大学 基于光伏电站出力特性分析的晴空日判定方法
CN115864387A (zh) * 2022-12-09 2023-03-28 国网宁夏电力有限公司经济技术研究院 与电网负荷时序匹配的风力发电功率品质划分方法及系统
CN119515018A (zh) * 2025-01-21 2025-02-25 广东华成电力能源股份有限公司 一种动态柔性光伏箱变的运行成本评估系统及方法

Families Citing this family (27)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11068563B2 (en) * 2011-07-25 2021-07-20 Clean Power Research, L.L.C. System and method for normalized ratio-based forecasting of photovoltaic power generation system degradation with the aid of a digital computer
FR3019274B1 (fr) * 2014-03-26 2020-05-22 Sun'r Procede de production d'energie electrique adapte aux cultures
CN107093007B (zh) * 2017-03-31 2020-12-22 华南理工大学 一种考虑光储持续带载能力的配电网可靠性评估方法
CN108601035B (zh) * 2018-04-26 2022-04-22 南京邮电大学 基于太阳能收集模型的WMSNs节点调度方法
CN109460893B (zh) * 2018-09-26 2024-08-09 中国电力科学研究院有限公司 一种光伏电站天气类型相关性指标计算方法和系统
CN109524983B (zh) * 2018-10-25 2022-04-01 国家电网有限公司 一种基于典型状态的光伏出力建模方法
CN109711870A (zh) * 2018-12-13 2019-05-03 江苏中科瀚星数据科技有限公司 一种居民区电动汽车充电负载预测与配置方法
CN109524993B (zh) * 2018-12-19 2020-06-02 中国农业大学 用于中长期优化调度的风电光伏典型周出力场景生成方法
US10977341B2 (en) * 2019-01-18 2021-04-13 Hide Housing Corporation Insolation probability distribution analysis method, insolation probability distribution analysis system, insolation probability distribution analysis program product, insolation normalization statistical analysis method, insolation normalization statistical analysis system, and insolation normalization statistical analysis program product
JP6562491B1 (ja) 2019-01-18 2019-08-21 株式会社ヒデ・ハウジング 期待値計算による発電量精密予測法、期待値計算による発電量精密予測システムおよび期待値計算による発電量精密予測プログラム
CN110378504B (zh) * 2019-04-12 2023-04-07 东南大学 一种基于高维Copula技术的光伏发电爬坡事件概率预测方法
CN110334870B (zh) * 2019-07-09 2022-05-31 福州大学 基于门控循环单元网络的光伏电站短期功率预测方法
CN110610262B (zh) * 2019-08-27 2023-06-27 天津大学 一种计及天气要素的长时间尺度光伏时序出力生成方法
CN113052351A (zh) * 2019-12-26 2021-06-29 中国电力科学研究院有限公司 一种光伏电站出力序列预测方法和系统
CN111861259B (zh) * 2020-07-30 2022-07-15 中国大唐集团科学技术研究院有限公司华东电力试验研究院 一种考虑时序性的负荷建模方法、系统、存储介质
CN112149905A (zh) * 2020-09-25 2020-12-29 福州大学 一种基于小波变换和小波神经网络的光伏电站短期功率预测方法
CN112671035A (zh) * 2020-12-21 2021-04-16 北京华能新锐控制技术有限公司 一种基于风电预测的虚拟电厂储能容量配置方法
CN112861363B (zh) * 2021-02-23 2022-09-27 国网冀北电力有限公司智能配电网中心 光伏发电状态估计方法、装置及终端设备
CN112907108B (zh) * 2021-03-15 2022-07-01 福州大学 一种海上风电场多阶段装机容量规划方法
CN113779861B (zh) * 2021-07-23 2023-08-22 国网河北省电力有限公司电力科学研究院 光伏功率的预测方法及终端设备
TWI783605B (zh) * 2021-08-02 2022-11-11 崑山科技大學 太陽能系統發電量預測方法
CN113570267B (zh) * 2021-08-02 2024-08-02 福州万山电力咨询有限公司 一种确定分布式光伏发电自发自用比例的方法及终端
CN117091256B (zh) * 2023-09-22 2026-03-10 长沙理工大学 一种光伏空调用能控制方法、装置、设备及存储介质
CN117411436B (zh) * 2023-12-15 2024-03-15 国网浙江省电力有限公司金华供电公司 光伏组件状态检测方法、系统及存储介质
CN118134025B (zh) * 2024-02-06 2024-09-03 水电水利规划设计总院有限公司 多时间尺度贝叶斯风电光伏出力预测方法
CN119203620B (zh) * 2024-11-28 2025-05-09 国网江西省电力有限公司电力科学研究院 一种光伏出力时间序列数据集生成方法及系统
CN120387275B (zh) * 2025-03-19 2025-11-25 山东大学 考虑多要素影响下的细粒度光伏出力仿真方法及系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110066401A1 (en) * 2009-09-11 2011-03-17 Wattminder, Inc. System for and method of monitoring and diagnosing the performance of photovoltaic or other renewable power plants
CN103810534A (zh) * 2013-12-11 2014-05-21 广西电网公司电力科学研究所 一种光伏出力预测方法
CN104182914A (zh) * 2014-09-05 2014-12-03 国家电网公司 一种基于波动特性的风电出力时间序列建模方法
CN104182889A (zh) * 2014-08-18 2014-12-03 国家电网公司 一种历史风电出力数据处理与波动辨识方法
CN104218574A (zh) * 2014-08-29 2014-12-17 国家电网公司 反映太阳辐射强度变化特性的光伏出力随机模型建模方法
CN104616078A (zh) * 2015-02-03 2015-05-13 河海大学 基于Spiking神经网络的光伏系统发电功率预测方法

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8200370B2 (en) * 2008-12-04 2012-06-12 American Power Conversion Corporation Energy reduction
DE112011105049T5 (de) * 2011-03-18 2013-12-19 Fujitsu Limited Betriebsplanvorbereitungsverfahren, Betriebsplanvorbereitungsvorrichtung und Betriebsplanvorbereitungsprogramm
JP5842994B2 (ja) * 2012-03-12 2016-01-13 富士通株式会社 運転計画作成方法、運転計画作成プログラムおよび運転計画作成装置
CN102880989B (zh) * 2012-09-05 2016-04-20 中国电力科学研究院 一种风电出力时间序列建模方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110066401A1 (en) * 2009-09-11 2011-03-17 Wattminder, Inc. System for and method of monitoring and diagnosing the performance of photovoltaic or other renewable power plants
CN103810534A (zh) * 2013-12-11 2014-05-21 广西电网公司电力科学研究所 一种光伏出力预测方法
CN104182889A (zh) * 2014-08-18 2014-12-03 国家电网公司 一种历史风电出力数据处理与波动辨识方法
CN104218574A (zh) * 2014-08-29 2014-12-17 国家电网公司 反映太阳辐射强度变化特性的光伏出力随机模型建模方法
CN104182914A (zh) * 2014-09-05 2014-12-03 国家电网公司 一种基于波动特性的风电出力时间序列建模方法
CN104616078A (zh) * 2015-02-03 2015-05-13 河海大学 基于Spiking神经网络的光伏系统发电功率预测方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LI, CHI;: "Study on The Modeling Method of New Energy Power Time Series Based on Fluctuation Characteristics", SCIENCE -ENGINEERING (B), CHINA MASTER'S THESES FULL-TEXT DATABASE, 15 December 2015 (2015-12-15), ISSN: 1674-0246 *

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109492315A (zh) * 2018-11-19 2019-03-19 西安交通大学 一种基于Copula函数的时空相关风光序列模拟方法
CN109449993A (zh) * 2018-12-26 2019-03-08 西安交通大学 一种面向多能互补电力系统的多时间尺度生产模拟方法
CN109449993B (zh) * 2018-12-26 2021-10-22 西安交通大学 一种面向多能互补电力系统的多时间尺度生产模拟方法
CN110544016A (zh) * 2019-08-09 2019-12-06 国网江苏省电力有限公司电力科学研究院 评估气象因素对电力设备故障概率影响程度的方法和设备
CN110544016B (zh) * 2019-08-09 2022-02-15 国网江苏省电力有限公司电力科学研究院 评估气象因素对电力设备故障概率影响程度的方法和设备
CN110717277A (zh) * 2019-10-14 2020-01-21 河北工业大学 一种时序风速模拟方法
CN112116127A (zh) * 2020-08-20 2020-12-22 中国农业大学 一种基于气象过程与功率波动关联的光伏功率预测方法
CN112116127B (zh) * 2020-08-20 2024-04-02 中国农业大学 一种基于气象过程与功率波动关联的光伏功率预测方法
CN115189421A (zh) * 2022-08-01 2022-10-14 长沙理工大学 基于光伏电站出力特性分析的晴空日判定方法
CN115864387A (zh) * 2022-12-09 2023-03-28 国网宁夏电力有限公司经济技术研究院 与电网负荷时序匹配的风力发电功率品质划分方法及系统
CN119515018A (zh) * 2025-01-21 2025-02-25 广东华成电力能源股份有限公司 一种动态柔性光伏箱变的运行成本评估系统及方法

Also Published As

Publication number Publication date
US20180240048A1 (en) 2018-08-23
US20180240200A1 (en) 2018-08-23
US10290066B2 (en) 2019-05-14
CN106557828A (zh) 2017-04-05

Similar Documents

Publication Publication Date Title
WO2017054537A1 (zh) 一种长时间尺度光伏出力时间序列建模方法及装置
Liu et al. Forecasting power output of photovoltaic system using a BP network method
Liu et al. PV generation forecasting with missing input data: A super-resolution perception approach
Tabone et al. Modeling variability and uncertainty of photovoltaic generation: A hidden state spatial statistical approach
CN103927695A (zh) 基于自学习复合数据源的风电功率超短期预测方法
CN103473322A (zh) 基于时间序列模型的光伏发电功率超短期预测方法
CN113159102B (zh) 一种多时间尺度的光伏功率预测方法及系统
CN114819374B (zh) 区域新能源超短期功率预测方法及系统
CN105139080A (zh) 一种基于马尔可夫链的改进光伏功率序列预测方法
CN103530473A (zh) 一种含大规模光伏电站的电力系统随机生产模拟方法
CN116109018B (zh) 一种光伏电站功率预测方法、装置及相关设备
CN111797132B (zh) 考虑时空相关性的多可再生能源电站功率场景生成方法
CN105790261A (zh) 一种随机谐波潮流计算方法
CN111709112A (zh) 一种海上风电运行模拟方法、装置及存储介质
CN103996079A (zh) 一种基于条件概率的风电功率加权预测方法
CN106058863A (zh) 一种基于随机响应面法的随机最优潮流计算方法
CN103927597A (zh) 基于自回归滑动平均模型的风电功率超短期预测方法
CN105701556A (zh) 一种光伏电站超短时功率预测修正方法
Guo et al. Wind power assessment based on a WRF wind simulation with developed power curve modeling methods
CN104091041A (zh) 基于高阶矩的发电功率估算方法和系统
Reddy et al. Hybrid approach for short term wind power forecasting
CN104462839A (zh) 基于风功率波动强度瞬时模型的风电不确定性估计方法
CN108710966B (zh) 一种基于多簇esn神经网络的光伏发电功率预测方法
CN109993335A (zh) 一种用于风电功率的概率预报方法及系统
CN103927596A (zh) 基于复合数据源自回归模型的风电功率超短期预测方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16850161

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 15751471

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16850161

Country of ref document: EP

Kind code of ref document: A1