A method and system for generating photovoltaic output time series data set

Abstract

The present invention discloses a method and system for generating a photovoltaic output time series data set, the method comprising: performing data preprocessing and simulation parameter setting, transferring and sampling daily meteorological conditions within the simulation time length through a first-order Markov chain, introducing a local cloud layer impact model to simulate the all-day irradiance of different daily meteorological conditions, and calculating the daily temperature based on the all-day simulated irradiance, considering the influence of the daily temperature and the all-day simulated irradiance on the power generation, calculating the daily output power of the photovoltaic system, and using a double screening method to screen out photovoltaic time series data sets that meet the daily power generation and the total power generation. It is possible to select appropriate time granularity and time scale according to needs, and then quickly simulate the influence of meteorological changes on the output of the photovoltaic system, and generate a data set close to the actual situation, which can be used in the fields of performance evaluation, machine learning, and integrated photovoltaic storage management and control of photovoltaic systems.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic power generation system simulation, and in particular relates to a method and system for generating a photovoltaic output time series data set. Background Art

[0002] Constructing reasonable long-term operation scenarios is crucial to dealing with the uncertainty of new energy. The volatility and intermittency of new energy output pose challenges to scenario generation. Currently, two main methods are used to solve this problem: statistical methods and time series methods.

[0003] Scenario generation methods are usually divided into two categories: statistical methods and time series methods. Statistical methods fit the probability distribution of historical data and then generate future scenarios by random sampling. The advantages of this type of method are low computational complexity, suitable for generating a large number of scenarios, and easy to implement. However, statistical methods often ignore the internal correlation of time series data, which may lead to non-compliance with the laws of meteorological changes.

[0004] The time series method focuses on using the time dependency of historical data to predict future trends through models. The time series method usually takes unsupervised learning as the core, and automatically learns the complex characteristics of renewable energy output through deep learning techniques such as generative adversarial networks (GANs) and variational autoencoders (VAEs). These models can effectively capture the hidden correlations in the scene through in-depth mining of data distribution, and then generate scenes that conform to the actual situation.

[0005] Both methods have their own advantages and disadvantages: the statistical method has high computational efficiency, but lacks temporal correlation; the time series method can better preserve temporal characteristics, but the model has high computational complexity and relies on large-scale data and computing resources, which leads to higher computational costs and implementation difficulties in practical applications. In practical applications, how to combine the advantages of these two methods to improve the quality of generated scenes is still an important direction of scene generation research.

[0006] In addition, considering that the current power analysis business has different requirements for the quality of forecast data required for different application scenarios, the method of generating photovoltaic output data should be able to consider extreme scenarios. When conducting grid stability analysis, energy storage charging and discharging strategy and capacity design, and responding to extreme weather changes, extreme photovoltaic output conditions cannot be ignored. In analysis business scenarios such as daily scheduling operations and long-term trend forecasting, the impact of extreme conditions is relatively small, and considering extreme scenarios will complicate the analysis.

[0007] Therefore, in the context of new power systems, how to combine the advantages of statistical methods and time series methods, taking into account the scenario requirements of different power analysis services, and further improve the accuracy and pertinence of photovoltaic scenario generation is the focus of current research. Summary of the invention

[0008] The present invention provides a method and system for generating a photovoltaic output time series data set, which are used to solve the technical problems of low accuracy and poor pertinence in photovoltaic scene generation.

[0009] In a first aspect, the present invention provides a method for generating a photovoltaic output time series data set, comprising:

[0010] Determine the basic parameters of the photovoltaic array to be simulated, generate the standard irradiance data of the photovoltaic array and the sunrise and sunset times of each day during the simulation through the hour angle, the distance between the sun and the earth, and the solar altitude angle, and construct the Markov meteorological transition probability matrix based on historical weather forecasts or local meteorological bureau data ;

[0011] According to the current simulation cycle month , using a first-order Markov chain, according to the Markov meteorological transition probability matrix Generate monthly daily weather types;

[0012] The local cloud impact model is introduced to simulate the all-day irradiance of different monthly daily meteorological conditions, and the daily temperature is calculated based on the all-day irradiance;

[0013] Considering the impact of daily temperature and all-day simulated irradiance on power generation, the daily output power of the photovoltaic system is calculated, and the double screening method is used to screen out the photovoltaic time series data sets that meet the daily power generation and the total power generation.

[0014] In a second aspect, the present invention provides a photovoltaic output time series data set generation system, comprising:

[0015] The construction module is configured to determine the basic parameters of the photovoltaic array required for simulation, generate the standard irradiance data of the photovoltaic array and the sunrise and sunset times of each day during the simulation time through the hour angle, the distance between the sun and the earth, and the solar altitude angle, and construct the Markov meteorological transition probability matrix based on historical weather forecasts or local meteorological bureau data ;

[0016] Generate a module, configured to be based on the current simulation cycle month , using a first-order Markov chain, according to the Markov meteorological transition probability matrix Generate monthly daily weather types;

[0017] A calculation module is configured to introduce a local cloud impact model to simulate the all-day irradiance of different monthly daily meteorological conditions, and calculate the daily temperature based on the all-day irradiance;

[0018] The screening module is configured to consider the impact of daily temperature and all-day simulated irradiance on power generation, calculate the daily output power of the photovoltaic system, and use a double screening method to screen out photovoltaic time series data sets that meet the daily power generation and the total power generation.

[0019] In a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the method for generating a photovoltaic output time series dataset of any embodiment of the present invention.

[0020] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the steps of the method for generating a photovoltaic output time series data set of any embodiment of the present invention.

[0021] The photovoltaic output time series data set generation method and system of the present application performs data preprocessing and simulation parameter setting, transfers and samples the daily meteorological conditions within the simulation time through a first-order Markov chain, introduces a local cloud impact model to simulate the all-day irradiance of different daily meteorological conditions, and calculates the daily temperature based on the all-day simulated irradiance. Considering the impact of the daily temperature and the all-day simulated irradiance on the power generation, the daily output power of the photovoltaic system is calculated, and the double screening method is used to screen out the photovoltaic time series data set that meets the daily power generation and the total power generation. It can select the appropriate time granularity and time scale according to the needs, and then quickly simulate the impact of meteorological changes on the output of the photovoltaic system, and generate a data set that is close to the actual situation, which can be used in the fields of performance evaluation of photovoltaic systems, machine learning, and integrated photovoltaic storage management. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 A flowchart of a method for generating a photovoltaic output time series data set provided by an embodiment of the present invention;

[0024] Figure 2A schematic diagram of a Markov meteorological transition probability matrix provided by an embodiment of the present invention;

[0025] Figure 3 A result diagram of a photovoltaic output data set provided by an embodiment of the present invention;

[0026] Figure 4 A structural block diagram of a photovoltaic output time series data set generation system provided by an embodiment of the present invention;

[0027] Figure 5 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] See also Figure 1 , which shows a flow chart of a method for generating a photovoltaic output time series data set of the present application.

[0030] like Figure 1 As shown in FIG. 1 , the method for generating a photovoltaic output time series data set specifically includes the following steps:

[0031] Step S101, determine the basic parameters of the photovoltaic array to be simulated, generate the standard irradiance data of the photovoltaic array and the sunrise and sunset times of each day during the simulation time through the hour angle, the distance between the sun and the earth and the solar altitude angle, and construct the Markov meteorological transition probability matrix based on the historical weather forecast or the data of the local meteorological bureau .

[0032] In this step, the basic parameters of the photovoltaic array to be simulated are determined, including photovoltaic area, photovoltaic panel efficiency, simulation area latitude, simulation duration, and simulation time granularity. The standard irradiance data of the photovoltaic array and the sunrise and sunset times of each day within the simulation duration are generated through the hour angle, the distance between the sun and the earth, and the solar altitude angle.

[0033] The sunrise time and sunset time of each day during the simulation duration are calculated by the hour angle, and the expression is:

[0034] ,

[0035] ,

[0036] ,

[0037] In the formula, is the sunrise time, It is sunset time, is the hour angle, which is negative at sunrise and positive at sunset. Every 15° corresponds to one hour. is the latitude of the photovoltaic array, is the solar declination angle;

[0038] The expression for calculating the solar declination angle is:

[0039] ,

[0040] In the formula, The Nth day of the 365 days in a year for the i-th date in the simulation duration;

[0041] Among them, the daily standard irradiance data considering seasonal differences within the simulation time can be calculated by the distance between the sun and the earth and the solar altitude angle. The expression for calculating the standard irradiance data of the photovoltaic array is:

[0042] ,

[0043] In the formula, is the standard irradiance of the photovoltaic array, is the solar constant, take 1367W / m², is the average distance between the sun and the earth, 1.496×108m, is the current distance between the Sun and the Earth, is the atmospheric transparency coefficient, which indicates the molecular scattering, ozone absorption and certain gases (such as )'s selective absorption attenuation effect on solar radiation is between 0.7 and 0.8 depending on meteorological conditions. for air quality;

[0044] The expression for calculating the current distance between the sun and the earth is:

[0045] ,

[0046] In the formula, is the approximate true anomaly, , It is the Nth day of the whole year in the simulation cycle.

[0047] Step S102, according to the current simulation cycle month , using a first-order Markov chain, according to the Markov meteorological transition probability matrix Generate monthly daily weather types.

[0048] In this step, in order to simulate various meteorological conditions from cloudy to sunny and cloudless, a Markov meteorological transition probability matrix is ​​constructed through historical weather forecasts and local meteorological data. The historical weather forecasts or local meteorological bureau data are divided by month, and cloudy to sunny and cloudless days are divided into i-level meteorological conditions. For two consecutive days of historical weather forecasts or local meteorological bureau data, the number of two-level meteorological transitions is counted (a total of i×j possible transitions for two consecutive days); all possible meteorological transition times are normalized to obtain the probability of transition, and then the Markov meteorological transition probability matrix is ​​constructed. Figure 2 The summer Markov meteorological transition probability matrix is ​​shown , the weather on day t+1 tends to be consistent with the weather conditions on day t, that is, the probability of weather changes between the two days before and after is low.

[0049] The expression of Markov meteorological transition probability matrix is:

[0050] ,

[0051] In the formula, is the probability that the first weather condition will transfer to the second weather condition the next day. is the probability that the first weather condition will change to the second weather condition the next day. is the probability that the first weather type will change to the jth weather type the next day, is the probability that the second weather condition will change to the first weather condition the next day. is the probability that the second weather type will shift to the second weather type the next day, is the probability that the second weather type will change to the jth weather type the next day, is the probability that the i-th weather condition will transfer to the first weather condition the next day, is the probability that the i-th weather condition will change to the second weather condition the next day, is the probability that the i-th weather condition will change to the j-th weather condition on the next day.

[0052] Step S103, introducing a local cloud influence model to simulate the all-day irradiance of different monthly daily weather conditions, and calculating the daily temperature based on the all-day irradiance.

[0053] In this step, since the probability of meteorological changes within the daily time scale is very low, the diagonal items in the Markov meteorological transition probability matrix have relatively high values. The fact that meteorological conditions maintain the same trend means that the Markov meteorological transition probability matrix is ​​sparse and has many zero values ​​in the non-diagonal elements. In order to enhance the simulation effect of sunny and cloudy scenes and improve the efficiency of data generation, a local cloud layer effect model is introduced to optimize the irradiance curve. The expression of the local cloud layer effect model is:

[0054] ,

[0055] In the formula, is the impact rate of cloud cover on irradiance during period t, is the number of cloud layers, is the average transmittance of randomly sampled clouds, is the thickness of the ith cloud layer, is the indicator function. When the period t is within the range of the shading duration of cloud layer i, is 1, otherwise it is 0;

[0056] By setting different average cloud transmittances under different daily weather conditions, the influence of local cloud layers on transmittance can be obtained through random sampling, and the expression of all-day irradiance is:

[0057] ,

[0058] In the formula, is the standard irradiance of the photovoltaic array, To consider the irradiance received by the photovoltaic power plant during period t after cloud cover;

[0059] Another key factor affecting photovoltaic power generation is temperature. Considering the hysteresis effect of temperature on radiation, a differential equation for the effect of solar radiation on temperature is established, which is expressed as:

[0060] ,

[0061] In the formula, is the influence coefficient of radiation on temperature, is the regression coefficient of air temperature to ambient temperature, is the temperature during period t, is the average ambient temperature, is the simulation time granularity;

[0062] Constructing the difference objective function , input historical temperature and historical solar radiation, and use nonlinear optimization algorithm to minimize the sum of square errors between model prediction and actual data to find the optimal decision variable. The difference objective function is:

[0063]

[0064] in, is the historical actual temperature at time t+1, is the historical actual temperature at time t, is the historical solar irradiance at time t, It is the ratio of the historical data duration to the historical time granularity.

[0065] Step S104, considering the impact of daily temperature and all-day simulated irradiance on power generation, the daily output power of the photovoltaic system is calculated, and the photovoltaic time series data set that meets the daily power generation and the total power generation is screened out using a double screening method.

[0066] In this step, the expression for calculating the daily output power of the photovoltaic system is:

[0067] ,

[0068] In the formula, is the photovoltaic power generation in period t, To consider the irradiance received by the photovoltaic power plant during period t after cloud cover, is the photovoltaic area, is the photovoltaic efficiency, is the temperature coefficient of the photovoltaic module, is the temperature, is the nominal operating temperature of the photovoltaic module, is the reference temperature, is the irradiance coefficient of the photovoltaic module.

[0069] It should be noted that when screening daily power generation, the daily photovoltaic power generation data that meets the upper and lower limits of daily power generation will be stored, and the daily photovoltaic power generation data that meet the upper and lower limits of daily power generation for 30 days will be spliced ​​to obtain the monthly power generation data; when the total photovoltaic power generation of the month meets the upper and lower limits of the monthly power generation, the generated continuous single-month data is considered to be qualified, otherwise the daily meteorology is generated again through the Markov chain and the daily power generation is calculated, and the double screening is repeated until the photovoltaic output time series data that meets the quantity requirements is generated.

[0070] In order to determine the upper and lower limits of daily power generation and the upper and lower limits of monthly power generation, and taking into account the different usage scenarios of data sets due to different uses of data sets, it is necessary to reasonably set the upper and lower limits of daily power generation and the upper and lower limits of monthly power generation. The specific calculation methods of the upper and lower limits of daily power generation and the upper and lower limits of monthly power generation are as follows: (1) Data preprocessing. The historical power generation time series data is divided into i types of meteorology, and divided into S groups of data by year (S=12, corresponding to 12 months), and then the daily power generation statistics are performed on long and short time scales by day and month. (2) The long and short time scale statistical results are normally fitted to generate the corresponding probability density distribution, and the expected daily fitting result is obtained. , Standard Deviation The expectation of the monthly fitting results , Standard Deviation (3) Determine the influencing parameters of the confidence interval based on whether to consider extreme scenarios Different problems have different requirements for the definition of "extreme". In statistics, based on the basic ideas of "low-probability events" and hypothesis testing, "low-probability events" usually refer to events with a probability of less than 5%, which is considered to be almost impossible to occur in a single test. The range of extreme scenario fluctuations is determined by a specific confidence interval. In the fields of financial risk management, weather forecasting, quality control, etc., it is usually chosen Equal to 2 or 3, when When the value is 2, it means that in a photovoltaic daily power generation time series simulation test, the daily power generation value has a 95.44% probability of falling within the confidence interval. When the value is 3, the probability increases to 99.74%. Furthermore, according to the statistical definition of low-probability events, when extreme scenarios are not considered, Take 2 and consider the extreme scenario. It can take a value of 3 or greater. The upper and lower limits of daily and monthly power generation are as follows without considering extreme scenarios:

[0071] ,

[0072] ,

[0073] Considering the upper and lower limits of daily and monthly power generation in extreme scenarios:

[0074] ,

[0075] ,

[0076] In the formula, is the expected daily power generation under the ith meteorological condition, is the standard deviation of daily power generation under the ith meteorological condition, is the daily power generation under the i-th meteorological scenario, is the expected total power generation in the sth simulation month, is the standard deviation of the total power generation in the sth month, is the total power generation in the sth month.

[0077] Take 5 minutes as the time granularity and do not consider extreme scenarios. , a one-month data set is generated for the photovoltaic array. The simulation parameter settings are shown in Table 1, and the simulation results are shown in Figure 3 .

[0078] Table 1 Simulation parameter settings:

[0079] .

[0080] In summary, the method of the present application performs data preprocessing and simulation parameter setting, transfers and samples the daily meteorological conditions within the simulation time through a first-order Markov chain, introduces a local cloud impact model to simulate the all-day irradiance of different daily meteorological conditions, and calculates the daily temperature based on the all-day simulated irradiance. Considering the impact of daily temperature and all-day simulated irradiance on power generation, the daily output power of the photovoltaic system is calculated, and the double screening method is used to screen out photovoltaic time series data sets that meet daily power generation and total power generation. It is possible to select appropriate time granularity and time scale as needed, and then quickly simulate the impact of meteorological changes on the output of the photovoltaic system, and generate a data set that is close to the actual situation, which can be used in the fields of photovoltaic system performance evaluation, machine learning, and integrated photovoltaic storage management.

[0081] See also Figure 4 , which shows a structural block diagram of a photovoltaic output time series data set generation system of the present application.

[0082] like Figure 4 As shown, the photovoltaic output time series data set generation system 200 includes a construction module 210 , a generation module 220 , a calculation module 230 and a screening module 240 .

[0083] Among them, the construction module 210 is configured to determine the basic parameters of the required simulated photovoltaic array, generate the standard irradiance data of the photovoltaic array and the sunrise time and sunset time of each day during the simulation time through the hour angle, the distance between the sun and the earth and the solar altitude angle, and construct the Markov meteorological transition probability matrix according to the historical weather forecast or the data of the local meteorological bureau ; Generate module 220, configured to be based on the current simulation cycle month , using a first-order Markov chain, according to the Markov meteorological transition probability matrix Generate monthly daily meteorological types; the calculation module 230 is configured to introduce a local cloud impact model to simulate the all-day irradiance of different monthly daily meteorological conditions, and calculate the daily temperature based on the all-day irradiance; the screening module 240 is configured to consider the impact of the daily temperature and the all-day simulated irradiance on the power generation, calculate the daily output power of the photovoltaic system, and use a double screening method to screen out the photovoltaic time series data set that meets the daily power generation and the total power generation.

[0084] It should be understood that Figure 4 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 4 The modules in it will not be described in detail here.

[0085] In some other embodiments, the embodiments of the present invention further provide a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the photovoltaic output time series data set generation method in any of the above method embodiments;

[0086] As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows:

[0087] Determine the basic parameters of the photovoltaic array to be simulated, generate the standard irradiance data of the photovoltaic array and the sunrise and sunset times of each day during the simulation through the hour angle, the distance between the sun and the earth, and the solar altitude angle, and construct the Markov meteorological transition probability matrix based on historical weather forecasts or local meteorological bureau data ;

[0088] According to the current simulation cycle month , using a first-order Markov chain, according to the Markov meteorological transition probability matrix Generate monthly daily weather types;

[0089] The local cloud impact model is introduced to simulate the all-day irradiance of different monthly daily meteorological conditions, and the daily temperature is calculated based on the all-day irradiance;

[0090] Considering the impact of daily temperature and all-day simulated irradiance on power generation, the daily output power of the photovoltaic system is calculated, and the double screening method is used to screen out the photovoltaic time series data sets that meet the daily power generation and the total power generation.

[0091] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the photovoltaic output time series data set generation system, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the photovoltaic output time series data set generation system via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0092] Figure 5 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 5As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 5 In the example, the connection through the bus is taken as an example. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the photovoltaic output time series data set generation method of the above-mentioned method embodiment is realized. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function control of the photovoltaic output time series data set generation system. The output device 340 may include display devices such as display screens.

[0093] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0094] As an implementation mode, the electronic device is applied to a photovoltaic output time series data set generation system, and is used for a client, and includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can:

[0095] Determine the basic parameters of the photovoltaic array to be simulated, generate the standard irradiance data of the photovoltaic array and the sunrise and sunset times of each day during the simulation through the hour angle, the distance between the sun and the earth, and the solar altitude angle, and construct the Markov meteorological transition probability matrix based on historical weather forecasts or local meteorological bureau data ;

[0096] According to the current simulation cycle month , using a first-order Markov chain, according to the Markov meteorological transition probability matrix Generate monthly daily weather types;

[0097] The local cloud impact model is introduced to simulate the all-day irradiance of different monthly daily meteorological conditions, and the daily temperature is calculated based on the all-day irradiance;

[0098] Considering the impact of daily temperature and all-day simulated irradiance on power generation, the daily output power of the photovoltaic system is calculated, and the double screening method is used to screen out the photovoltaic time series data sets that meet the daily power generation and the total power generation.

[0099] Through the description of the above implementation modes, those skilled in the art can clearly understand that each implementation mode can be implemented by means of software plus a necessary general hardware platform, or of course by hardware. Based on such an understanding, the above technical solution can essentially or in other words be embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating a photovoltaic output time series data set, characterized in that: include: Determine the basic parameters of the photovoltaic array to be simulated, generate the standard irradiance data of the photovoltaic array and the sunrise and sunset times of each day during the simulation through the hour angle, the distance between the sun and the earth, and the solar altitude angle, and construct the Markov meteorological transition probability matrix based on historical weather forecasts or local meteorological bureau data ; According to the current simulation cycle month , using a first-order Markov chain, according to the Markov meteorological transition probability matrix Generate monthly daily weather types; Introducing a local cloud layer influence model to simulate the all-day irradiance of different monthly daily meteorological conditions, and calculating the daily temperature based on the all-day irradiance, wherein the introducing a local cloud layer influence model to simulate the all-day irradiance of different monthly daily meteorological conditions, and calculating the daily temperature based on the all-day irradiance includes: The local cloud layer influence model is introduced to optimize the all-day irradiance curve to obtain the all-day irradiance, wherein the expression of the local cloud layer influence model is: ， In the formula, is the impact rate of cloud cover on irradiance during period t, is the number of cloud layers, is the average transmittance of randomly sampled clouds, is the thickness of the x-th cloud layer, is the indicator function. When the period t is within the range of the shading duration of the cloud layer x, is 1, otherwise it is 0; The expression of the whole day irradiance is: ， In the formula, is the standard irradiance of the photovoltaic array, To consider the irradiance received by the photovoltaic power plant during period t after cloud cover; Considering the hysteresis effect of temperature on radiation, a differential equation for the influence of solar radiation on temperature is established, which is expressed as follows: ， In the formula, is the influence coefficient of radiation on temperature, is the regression coefficient of air temperature to ambient temperature, is the temperature during period t, is the average ambient temperature, is the simulation time granularity; Constructing the difference objective function , input historical temperature and historical solar radiation, and use nonlinear optimization algorithm to minimize the sum of square errors between model prediction and actual data to find the optimal decision variable. The difference objective function is: in, is the historical actual temperature at time t+1, is the historical actual temperature at time t, is the historical solar irradiance at time t, It is the ratio of the length of historical data to the granularity of historical time; Considering the influence of daily temperature and all-day simulated irradiance on power generation, the daily output power of the photovoltaic system is calculated, and the photovoltaic time series data set that meets the daily power generation and the total power generation is screened out by the double screening method. The photovoltaic time series data set that meets the daily power generation and the total power generation is screened out by the double screening method includes: When screening daily power generation, the daily photovoltaic power generation data that meets the upper and lower limits of daily power generation are stored, and the daily photovoltaic power generation data that meets the upper and lower limits of daily power generation for 30 days are spliced ​​to obtain the monthly power generation data; When the total photovoltaic power generation in a month meets the upper and lower limits of the monthly power generation, the generated continuous single-month data is considered qualified. Otherwise, the daily meteorology is generated again through the Markov chain and the daily power generation is calculated. The double screening is repeated until the photovoltaic output time series data that meets the quantity requirements is generated. Among them, the upper and lower limits of daily and monthly power generation without considering extreme scenarios are: ， ， Considering the upper and lower limits of daily and monthly power generation in extreme scenarios: ， ， In the formula, is the expected daily power generation under the ith meteorological condition, is the standard deviation of daily power generation under the ith meteorological condition, is the daily power generation under the i-th meteorological scenario, is the expected total power generation in the sth simulation month, is the standard deviation of the total power generation in the sth month, is the total power generation in the sth month.

2. A method for generating a photovoltaic output time series data set according to claim 1, characterized in that: in, The sunrise time and sunset time of each day during the simulation duration are calculated by the hour angle, and the expression is: ， ， ， In the formula, is the sunrise time, It is sunset time, is the hour angle, which is negative at sunrise and positive at sunset. Every 15° corresponds to one hour. is the latitude of the photovoltaic array, is the solar declination angle; The expression for calculating the solar declination angle is: ， In the formula, is the number of days in the simulation time that the i-th date is in the 365 days of the whole year. sky; Among them, the expression for calculating the standard irradiance data of the photovoltaic array is: ， In the formula, is the standard irradiance of the photovoltaic array, is the solar constant, take 1367W / m², is the average distance between the sun and the earth, 1.496×108m, is the current distance between the Sun and the Earth, is the atmospheric transparency coefficient, for air quality; The expression for calculating the current distance between the sun and the earth is: ， In the formula, is the approximate true anomaly, , The simulation period is the first year of the year. sky.

3. The method for generating a photovoltaic output time series data set according to claim 1, characterized in that: The Markov meteorological transition probability matrix is ​​constructed based on historical weather forecasts or local meteorological bureau data. include: The historical weather forecast or the data of the local meteorological bureau are divided into months, and cloudy to sunny without clouds are divided into level i weather. For two consecutive days of historical weather forecast or the data of the local meteorological bureau, the number of changes between the two levels of weather is counted; All possible weather transition times are normalized to obtain the transition probability, and then the Markov weather transition probability matrix is ​​constructed, which is expressed as: ， In the formula, is the probability that the first weather condition will transfer to the second weather condition the next day. is the probability that the first weather condition will change to the second weather condition the next day. is the probability that the first weather type will change to the jth weather type the next day, is the probability that the second weather condition will change to the first weather condition the next day. is the probability that the second weather type will shift to the second weather type the next day, is the probability that the second weather type will change to the jth weather type the next day, is the probability that the i-th weather condition will transfer to the first weather condition the next day, is the probability that the i-th weather condition will change to the second weather condition the next day, is the probability that the i-th weather condition will change to the j-th weather condition on the next day.

4. The method for generating a photovoltaic output time series data set according to claim 1, characterized in that: The expression for calculating the daily output power of the photovoltaic system is: ， In the formula, is the daily output power of the photovoltaic system, To consider the irradiance received by the photovoltaic power plant during period t after cloud cover, is the photovoltaic area, is the photovoltaic efficiency, is the temperature coefficient of the photovoltaic module, is the temperature, is the nominal operating temperature of the photovoltaic module, is the reference temperature, is the irradiance coefficient of the photovoltaic module.

5. A photovoltaic output time series data set generation system, characterized in that: include: The construction module is configured to determine the basic parameters of the photovoltaic array required for simulation, generate the standard irradiance data of the photovoltaic array and the sunrise and sunset times of each day during the simulation time through the hour angle, the distance between the sun and the earth, and the solar altitude angle, and construct the Markov meteorological transition probability matrix based on historical weather forecasts or local meteorological bureau data ; Generate a module, configured to be based on the current simulation cycle month , using a first-order Markov chain, according to the Markov meteorological transition probability matrix Generate monthly daily weather types; The calculation module is configured to introduce a local cloud layer influence model to simulate the all-day irradiance of different monthly daily weather, and calculate the daily temperature based on the all-day irradiance, wherein the introduction of the local cloud layer influence model to simulate the all-day irradiance of different monthly daily weather, and calculate the daily temperature based on the all-day irradiance includes: The local cloud layer influence model is introduced to optimize the all-day irradiance curve to obtain the all-day irradiance, wherein the expression of the local cloud layer influence model is: ， In the formula, is the impact rate of cloud cover on irradiance during period t, is the number of cloud layers, is the average transmittance of randomly sampled clouds, is the thickness of the x-th cloud layer, is the indicator function. When the period t is within the range of the shading duration of the cloud layer x, is 1, otherwise it is 0; The expression of the whole day irradiance is: ， In the formula, is the standard irradiance of the photovoltaic array, To consider the irradiance received by the photovoltaic power plant during period t after cloud cover; Considering the hysteresis effect of temperature on radiation, a differential equation for the influence of solar radiation on temperature is established, which is expressed as follows: ， In the formula, is the influence coefficient of radiation on temperature, is the regression coefficient of air temperature to ambient temperature, is the temperature during period t, is the average ambient temperature, is the simulation time granularity; Constructing the difference objective function , input historical temperature and historical solar radiation, and use nonlinear optimization algorithm to minimize the sum of square errors between model prediction and actual data to find the optimal decision variable. The difference objective function is: in, is the historical actual temperature at time t+1, is the historical actual temperature at time t, is the historical solar irradiance at time t, It is the ratio of the length of historical data to the granularity of historical time; The screening module is configured to consider the influence of daily temperature and all-day simulated irradiance on power generation, calculate the daily output power of the photovoltaic system, and use a double screening method to screen out photovoltaic time series data sets that meet the daily power generation and the total power generation, wherein the photovoltaic time series data sets that meet the daily power generation and the total power generation by using the double screening method include: When screening daily power generation, the daily photovoltaic power generation data that meets the upper and lower limits of daily power generation are stored, and the daily photovoltaic power generation data that meets the upper and lower limits of daily power generation for 30 days are spliced ​​to obtain the monthly power generation data; When the total photovoltaic power generation in a month meets the upper and lower limits of the monthly power generation, the generated continuous single-month data is considered qualified. Otherwise, the daily meteorology is generated again through the Markov chain and the daily power generation is calculated. The double screening is repeated until the photovoltaic output time series data that meets the quantity requirements is generated. Among them, the upper and lower limits of daily and monthly power generation without considering extreme scenarios are: ， ， Considering the upper and lower limits of daily and monthly power generation in extreme scenarios: ， ， In the formula, is the expected daily power generation under the ith meteorological condition, is the standard deviation of daily power generation under the ith meteorological condition, is the daily power generation under the i-th meteorological scenario, is the expected total power generation in the sth simulation month, is the standard deviation of the total power generation in the sth month, is the total power generation in the sth month.

6. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images