Scene generation and reduction method in virtual power plant system

Abstract

The invention discloses a scene generation and reduction method in a virtual power plant system, and the method comprises the steps: S1, carrying out the processing of data in a historical power utilization scene of the virtual power plant system through a Monte Carlo simulation method, and forming an original scene set containing all scenes; s2, calculating Manhattan probability distances among the scenes, and obtaining a scene distance matrix; s3, respectively calculating the sum of probability distances between each scene and other residual scenes, and obtaining a probability distance matrix E; and S4, selecting two scenes with the minimum sum of the probability distances with other residual scenes to carry out reduction and combination, and continuously repeating the process until a final reduced scene is obtained. According to the scene generation and reduction method in the virtual power plant system provided by the invention, the problems of complex mathematical calculation, relatively high clustering center selection requirement and the like in the scene generation and reduction method of the virtual power plant can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plant dispatching, and in particular to a method for scenario generation and reduction in a virtual power plant system. Background Art

[0002] To cope with the increasingly severe energy crisis and environmental crisis, the construction of an energy system with a new power system as the theme has received unprecedented attention. As the basis for power system simulation research, methods for scenario generation and reduction in virtual power plants (such as: time series simulation method, typical day method, scenario clustering method, etc.) have also become a hot topic. However, the methods for scenario generation and reduction in virtual power plants face problems such as complex mathematical calculations and high requirements for the selection of clustering centers to a certain extent. Therefore, this has brought a certain impact on simulation research such as optimal dispatching of power systems.

[0003] Therefore, there is an urgent need to provide a method for scenario generation and reduction in virtual power plants that can overcome the above problems. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for scenario generation and reduction in a virtual power plant system, which can solve the problems of complex mathematical calculations and high requirements for the selection of clustering centers in the method for scenario generation and reduction in virtual power plants.

[0005] To achieve the above purpose, the present invention provides a method for scenario generation and reduction in a virtual power plant system, including:

[0006] S1. Process the data in the historical power consumption scenarios of the virtual power plant system through the Monte Carlo simulation method to form an original scenario set containing each scenario;

[0007] S2. Calculate the Manhattan probability distance between each scenario and obtain a scenario distance matrix;

[0008] S3. Calculate the sum of the probability distances between each scenario and the other remaining scenarios respectively, and obtain a probability distance matrix E;

[0009] S4. Select 2 scenarios with the smallest sum of probability distances from the other remaining scenarios for reduction and merging, and continuously repeat this process until the final reduced scenario is obtained.

[0010] Optionally, the step S1 includes:

[0011] S1.1. Process the data in the historical power consumption scenarios in combination with the output of wind turbines, the output of photovoltaic power, and electricity price factors to obtain optimized historical power consumption scenario data;

[0012] S1.2. Use the Monte Carlo simulation method to calculate and process the optimized historical power consumption scenario data, and obtain the original scenario set S(i) with equal probabilities.

[0013] Optionally, the processing of the historical power consumption scenario data in step S1.1 includes the following rules:

[0014] The relationship between the output of the wind turbine and the wind speed is:

[0015]

[0016] In the formula, υ N is the rated wind speed of the wind turbine; υ in is the cut-in wind speed of the wind turbine; υ out is the cut-out wind speed of the wind turbine; υ t is the real-time wind speed of the wind turbine; P WT is the rated output power of the wind turbine; P win is the output of the wind turbine;

[0017] The relationship between the output of the photovoltaic power generation and the solar irradiance is:

[0018]

[0019] In the formula, R sta is the solar irradiance under standard conditions, R rea is the solar irradiance in the current environment, R bou is the solar irradiance at a certain boundary, P N is the rated output power of the photovoltaic power station; P sun is the output of the photovoltaic power generation;

[0020] The uncertainty relationship of the electricity price is:

[0021]

[0022] In the formula, P t PB is the dynamic time-of-use electricity price at time t, are the peak, flat, and valley electricity prices respectively, α 1 、α 2 are the dynamic time-of-use electricity price division parameters, is the sum of the maximum and minimum wind and light outputs, is the sum of the wind and light outputs at time t.

[0023] Optionally, step S1.2 is specifically:

[0024] Using the Monte Carlo simulation method, a specific standard deviation, random numbers, and original prediction values are selected from the optimized historical power consumption scenario data. The standard deviation is multiplied by the standard normal distribution of the random numbers, and the result of the multiplication is added to the original prediction value to obtain an equiprobable set of original scenarios S(i), that is, S(i) = D(i) + sσ(i);

[0025] In the formula, D(i) represents the original prediction value, σ(i) represents the standard normal distribution of the random numbers generated for the i-th scenario, and s represents the standard deviation.

[0026] Optionally, the step S2 includes:

[0027] S2.1, calculating the Manhattan probability distance between each scenario in the set of scenarios:

[0028] In the formula, d ij is the Manhattan distance between scenario i and scenario j, N represents the total number of data in the scenario, and X it , X jt are the t-th data in scenario i and the t-th data in scenario j, respectively;

[0029] S2.2, obtaining the following scenario distance matrix d through the Manhattan probability distance between each scenario:

[0030]

[0031] In the formula, the number of scenarios is n, and each element in the distance matrix d represents the Manhattan distance between any one of the n scenarios and itself, as well as between other scenarios; the d 12 and d 21 both represent the Manhattan distance between scenario 1 and scenario 2, the d 1n and d n1 both represent the Manhattan distance between scenario n and scenario 1, and the d 2n and d n2 both represent the Manhattan distance between scenario n and scenario 2.

[0032] Optionally, the step S3 includes:

[0033] S3.1, respectively calculating the sum of the probability distances between each scenario and the remaining other scenarios;

[0034] S3.2, obtaining the probability distance matrix E based on the sum of the probability distances between each scenario and the remaining other scenarios calculated;

[0035]

[0036] Optionally, the step S3.1 is specifically:

[0037] The sum of the probability distances between the said scenario 1 and itself, and between it and the other remaining scenarios is d 1 Expressed as: The sum of the probability distances between the said scenario 2 and itself, and between it and the other remaining scenarios is d 2 Expressed as: …; And so on, the sum of the probability distances between the said scenario n and itself, and between it and the other remaining scenarios is d n Expressed as:

[0038] Optionally, the said step S4 includes:

[0039] S4.1, Select the 2 scenarios with the smallest calculation results of the sum of the probability distances from the other remaining scenarios, and perform reduction and merging on these 2 scenarios;

[0040] S4.2, Repeat step S4.1 multiple times to perform multiple reductions on the remaining scenarios until the final reduced scenarios are obtained.

[0041] Optionally, the reduction and merging method of the said step S4.1 is:

[0042] After selecting the 2 scenarios with the smallest sum of the probabilities from the other remaining scenarios, reduce and merge one of the scenarios into the other scenario, and the probability of the reduced and merged scenario is the sum of the probabilities of the two smallest scenarios before reduction and merging, that is:

[0043]

[0044] In the formula, S(1), S(2), …, S(n) respectively represent scenario 1, scenario 2, …, scenario n; S(g) and S(h) represent the 2 scenarios selected with the smallest sum of the probabilities from the other remaining scenarios; p g (t) represents the probability of scenario g at the t-th iteration; p h (t) represents the probability of scenario h at the t-th iteration; p g (t + 1) represents the probability of scenario g at the (t + 1)-th iteration.

[0045] Optionally, the final reduced scenarios obtained in the said step S4.2 and their corresponding probabilities are:

[0046]

[0047] In the formula, represents the k scenarios generated and reduced by the output of the wind turbine and the output of the photovoltaic power generation and their corresponding scenario probabilities; represents the k scenarios generated and reduced by the electricity price and their corresponding scenario probabilities; S wIndicates the scenario information of the final output of wind turbines and photovoltaic power generation; S e Indicates the scenario information of the final electricity price.

[0048] In summary, compared with the prior art, the present invention has the following beneficial effects:

[0049] A method for scenario generation and reduction in a virtual power plant system provided by the present invention selects specific standard deviations within a day and the standard normal distribution of generating random numbers by using the Monte Carlo method to generate specific original scenarios, and adjusts the scenarios by using the Manhattan probability distance, avoiding the problems of errors caused by cumbersome mathematical calculations and assumptions in traditional methods, over-reliance on the accuracy of data, and excessive requirements for the selection of clustering centers. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 Is the implementation flowchart of the scenario generation and reduction method of the present invention;

[0051] Figure 2 Is the scenario generation and load power diagram of the present invention;

[0052] Figure 3 Is the scenario reduction and load power diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0053] The following will be combined with the attached Figures 1 to 3 , and the technical content, structural features, achieved objectives and effects of the present invention will be described in detail through preferred embodiments.

[0054] It should be noted that the drawings adopt a very simplified form and all use non-precise scales, only for the purpose of conveniently and clearly assisting in explaining the purpose of the implementation manner of the present invention, and are not used to limit the limiting conditions for the implementation of the present invention. Therefore, they do not have technical substance significance. Any modification of the structure, change of the proportional relationship or adjustment of the size, without affecting the effects that the present invention can produce and the objectives that can be achieved, should still fall within the scope covered by the technical content disclosed by the present invention.

[0055] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "upper", "lower", "left", "right", "vertical", "horizontal", "inside", "outside", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. In addition, the terms "first", "second", "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0056] In the description of the present invention, it should be noted that unless otherwise clearly specified and defined, the terms "installation", "connection", and "coupling" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection; it can be a direct connection, or an indirect connection through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0057] The present invention provides a method for scenario generation and reduction in a virtual power plant system, as Figure 1 shown. The method for scenario generation and reduction in the virtual power plant system includes the following steps:

[0058] S1. Process the data in the historical power consumption scenarios of the virtual power plant system through the Monte Carlo simulation method to form an original scenario set containing each scenario;

[0059] S2. Calculate the Manhattan probability distance between each of the scenarios and obtain a scenario distance matrix;

[0060] S3. Calculate the sum of the probability distances between each scenario and the remaining other scenarios respectively, and obtain a probability distance matrix E;

[0061] S4. Select 2 scenarios with the smallest sum of the probability distances from the remaining other scenarios for reduction and merging, and continuously repeat this process until the final reduced scenarios are obtained.

[0062] Among them, the step S1 includes:

[0063] S1.1. Combine the changing factors such as the output of wind turbines, the output of photovoltaic power, and electricity prices to process the data in the historical power consumption scenarios and obtain optimized historical power consumption scenario data;

[0064] Process the data in the historical power consumption scenarios according to the following rules:

[0065] Among them, the relationship between the output of the wind turbine (P win ) and the wind speed is:

[0066]

[0067] In the formula, υ N is the rated wind speed of the wind turbine; υ in is the cut-in wind speed of the wind turbine; υ out is the cut-out wind speed of the wind turbine; υ t is the real-time wind speed of the wind turbine; P WT is the rated output power of the wind turbine.

[0068] Among them, the photovoltaic output (Psun ) The relationship with solar irradiance is as follows:

[0069]

[0070] In the formula, R sta is the solar irradiance under standard conditions, R rea is the solar irradiance in the current environment, R bou is the solar irradiance at a certain boundary, and P N is the rated output power of the photovoltaic power station.

[0071] Among them, the uncertainty relationship of the electricity price is:

[0072]

[0073] In the formula, P t PB is the dynamic time-of-use electricity price at time t, are the peak, flat, and valley electricity prices respectively, and α 1 and α 2 are the dynamic time-of-use electricity price division parameters, is the sum of the maximum and minimum wind and light outputs, is the sum of the wind and light outputs at time t.

[0074] S1.2. Use the Monte Carlo simulation method to calculate and process the optimized historical electricity consumption scenario data, and obtain an equiprobable original scenario set S(i);

[0075] Specifically, use the Monte Carlo simulation method to select a specific standard deviation, generate random numbers and original prediction values from the optimized historical electricity consumption scenario data, multiply the standard deviation by the standard normal distribution of the random numbers, and then add the multiplied result to the original prediction value to obtain an equiprobable original scenario set S(i), that is, S(i) = D(i) + sσ(i).

[0076] In the formula, D(i) represents the original prediction value, σ(i) represents the standard normal distribution of the random numbers generated in the i-th scenario, and s represents the standard deviation.

[0077] In a specific embodiment of the present invention, the selected specific standard deviation is 0.3.

[0078] Among them, the step S2 includes:

[0079] S2.1. Calculate the Manhattan probability distance between each scenario in the scenario set through the following formula:

[0080]

[0081] In the formula, d ijis the Manhattan distance between scenario i and scenario j, N represents the total number of data in the scenario, and X it , X jt are the t-th data in scenario i and the t-th data in scenario j, respectively.

[0082] S2.2. Obtain the following scenario distance matrix d through the Manhattan probability distance between each of the said scenarios:

[0083]

[0084] wherein, there are n said scenarios. Each element in the distance matrix d represents the Manhattan distance between any one of the n scenarios and itself, as well as between it and other scenarios.

[0085] In the formula, d 12 and d 21 both represent the Manhattan distance between scenario 1 and scenario 2, d 1n and d n1 both represent the Manhattan distance between scenario n and scenario 1, d 2n and d n2 both represent the Manhattan distance between scenario n and scenario 2.

[0086] wherein, the step S3 includes:

[0087] S3.1. Calculate the sum of the probability distances between each scenario and the remaining other scenarios respectively;

[0088] Specifically, the sum of the probability distances d 1 between scenario 1 and itself, as well as between it and the remaining other scenarios (2 to n) is expressed as: The sum of the probability distances d 2 between scenario 2 and itself, as well as between it and the remaining other scenarios (1, 3 to n) is expressed as: …; and so on, the sum of the probability distances d n between scenario n and itself, as well as between it and the remaining other scenarios (1 to n - 1) is expressed as:

[0089] S3.2. On the basis of calculating the sum of the probability distances between each scenario and the remaining other scenarios, obtain the probability distance matrix E:

[0090]

[0091] wherein, the step S4 includes:

[0092] S4.1. Select the 2 scenarios with the smallest calculation results of the sum of the probability distances from the remaining other scenarios, and perform reduction and merging on these 2 scenarios;

[0093] Specifically, when the two scenarios with the smallest sum of probability distances from other remaining scenarios are scenario g and scenario h, i.e., S(g) and S(h), the probability iteration of the merged scenario g and scenario h after reduction follows the following rule:

[0094]

[0095] In the formula, S(1), S(2), …, S(n) respectively represent scenario 1, scenario 2, …, scenario n; p g (t) represents the probability of scenario g at the t-th iteration; p h (t) represents the probability of scenario h at the t-th iteration; p g (t + 1) represents the probability of scenario g at the (t + 1)-th iteration.

[0096] The above formula can be understood as: after selecting the two scenarios with the smallest sum of probabilities from other remaining scenarios, one of the scenarios is reduced and merged into the other scenario, and the probability of the merged scenario after reduction is the sum of the probabilities of the two scenarios with the smallest probabilities before reduction and merger.

[0097] S4.2. Repeat step S4.1 multiple times to perform multiple reductions on the remaining scenarios until the final reduced scenarios as shown in Figure 2 and Figure 3 are obtained;

[0098] Among them, the final reduced scenarios and their corresponding probabilities can be expressed as:

[0099]

[0100] In the formula, represents k scenarios generated and reduced through the output of wind turbines and photovoltaic power generation and their corresponding scenario probabilities; represents k scenarios generated and reduced through electricity prices and their corresponding scenario probabilities; S w represents the scenario information of the final output of wind turbines and photovoltaic power generation; S e represents the scenario information of the final electricity price.

[0101] In summary, the method for scenario generation and reduction in a virtual power plant system provided by the present invention avoids the problems of errors caused by cumbersome mathematical calculations and assumptions in traditional methods, excessive dependence on the accuracy of data, and too high requirements for the selection of clustering centers.

[0102] Although the content of the present invention has been described in detail through the above preferred embodiments, it should be recognized that the above description should not be considered as a limitation of the invention. After those skilled in the art have read the above content, various modifications and alternatives to the present invention will be obvious. Therefore, the protection scope of the present invention should be defined by the appended claims.

Claims

1. A method for scene generation and reduction in a virtual power plant system, characterized in that: The following steps are included: S1, processing data in historical power consumption scenarios of the virtual power plant system by Monte Carlo simulation method to form an original scenario set including various scenarios; S2, calculating the Manhattan probability distance between each of the scenes and obtaining a scene distance matrix; S3, respectively calculate the sum of the probability distances between each scene and the remaining scenes, and obtain the probability distance matrix E; S4, select the two scenes with the smallest sum of probability distances to the other remaining scenes for reduction and merging, and repeat the process until the final reduced scene is obtained.

2. The scene generation and reduction method according to claim 1, characterized in that: The step S1 comprises: S1.1, combining wind turbine output, photovoltaic output, and electricity price factors, processing the data in the historical electricity consumption scenarios to obtain optimized historical electricity consumption scenario data; S1.2, the Monte Carlo simulation method is used to calculate and process the optimized historical electricity consumption scenario data, and obtain an original scenario set S(i) with equal probability.

3. The scene generation and reduction method according to claim 2, characterized in that: The historical electricity usage scenario data processing in step S1.1 includes the following rules: The relationship between the wind turbine output and wind speed is: In the formula, υ N is the rated wind speed of the wind turbine; in is the cut-in wind speed of the wind turbine; out is the cut-out wind speed of the wind turbine; t is the real-time wind speed of the wind turbine; P WT is the rated output power of the wind turbine; P win Provide power for wind turbines; The relationship between photovoltaic output and solar radiation is: In the formula, R sta is the solar radiation under standard environment, R rea is the solar radiation in the current environment, R bou is the solar radiation at a certain boundary, P N is the rated output power of the photovoltaic power station; P sun Produce power for photovoltaics; The uncertainty relationship of the electricity price is: Where P t PB is the dynamic time-of-use electricity price at time t, They are peak, flat and valley electricity prices respectively, α1 and α2 are dynamic time-of-use electricity price division parameters, For the maximum and minimum wind and solar output, Contribute to the scenery at moment t.

4. The scene generation and reduction method according to claim 2, characterized in that: The step S1.2 is specifically as follows: The Monte Carlo simulation method is used to select a specific standard deviation from the optimized historical electricity consumption scenario data, generate random numbers and original prediction values, and multiply the standard deviation with the standard normal distribution of the random number. The multiplication result is then superimposed on the original prediction value to obtain an original scenario set S(i) with equal probability, that is, S(i) = D(i) + sσ(i); Where D(i) represents the original prediction value, σ(i) represents the standard normal distribution of random numbers generated for the i-th scenario, and s represents the standard deviation.

5. The scene generation and reduction method according to claim 1, characterized in that: The step S2 comprises: S2.1, calculate the Manhattan probability distance between each scene in the scene set: Where, d ij is the Manhattan distance between scene i and scene j, N represents the total number of data in the scene, X it , X jt They are the t-th data in scene i and the t-th data in scene j respectively; S2.2, through the Manhattan probability distance between each of the scenes, the following scene distance matrix d is obtained: Wherein, the number of scenes is n, and each element in the distance matrix d represents the Manhattan distance between any scene in the n scenes and itself, and between any other scenes; the d 12 and d 21 Both represent the Manhattan distance between scene 1 and scene 2, d 1n and d n1 Both represent the Manhattan distance between scene n and scene 1, d 2n and d n2 Both represent the Manhattan distance between scene n and scene 2.

6. The scene generation and reduction method according to claim 1, characterized in that: The step S3 comprises: S3.1, calculate the sum of the probability distances between each scenario and the other remaining scenarios respectively; S3.2, based on the calculated sum of the probability distances between each scene and the other remaining scenes, the probability distance matrix E is obtained:

7. The scene generation and reduction method according to claim 6, characterized in that: The step S3.1 is specifically as follows: The sum d1 of the probability distances between the scene 1 and itself and other remaining scenes is expressed as: The sum d2 of the probability distances between the scene 2 and itself and other remaining scenes is expressed as: Similarly, the sum of the probability distances d between the scene n and itself and the remaining scenes is n It is expressed as:

8. The scene generation and reduction method according to claim 7, characterized in that: The step S4 comprises: S4.1, select the two scenarios with the smallest sum of probability distances from the remaining scenarios, and reduce and merge the two scenarios; S4.2, repeat step S4.1 multiple times to perform multiple cuts on the remaining scenes until the final cut scene is obtained.

9. The scene generation and reduction method according to claim 8, characterized in that: The reduction and merging method of step S4.1 is: When two scenes with the smallest sum of probabilities with other remaining scenes are selected, one of the scenes is cut and merged into another scene, and the probability of the scene after the reduction and merger is the sum of the two smallest scene probabilities before the reduction and merger, that is: Where S(1), S(2), …, S(n) represent scene 1, scene 2, …, scene n respectively; S(g) and S(h) represent the two selected scenes with the smallest sum of probabilities with the other remaining scenes; p g (t) represents the probability of scene g at the tth iteration; p h (t) represents the probability of scene h at the tth iteration; p g (t+1) represents the probability of scene g at the t+1th iteration.

10. The scene generation and reduction method according to claim 9, characterized in that: The final reduced scene obtained in step S4.2 and its corresponding probability are: In the formula, represents the k scenarios generated and reduced by wind turbine output and photovoltaic output and their corresponding scenario probabilities; represents the k scenarios generated and reduced by electricity prices and their corresponding scenario probabilities; S w Indicates the final scene information of wind turbine output and photovoltaic output; S e Scenario information indicating the final electricity price.

Images