Power grid time sequence scene reduction method and system based on confidence interval

Abstract

The invention discloses a power grid time sequence scene reduction method and system based on a confidence interval, and relates to the technical field of power system dispatching automation, and the method comprises the steps: carrying out the sampling according to the probability density distribution of new energy and load prediction errors, and generating an initial time sequence scene set; dividing the initial scene set through a confidence interval; correlation analysis is carried out on new energy and load prediction errors of the extreme scenes, and extreme scene sets are classified; screening out a high-risk scene set and a general scene set; for a normal scene set and various extreme scene sets, improved K-means clustering which comprehensively considers an Euclidean distance and a dynamic time bending distance is adopted to obtain a typical scene. According to the method, the scenes with different probability risks are subjected to differential reduction, the scene sets required for formulating the scheduling optimization scheme in the high-probability low-risk scene and the low-probability high-risk scene are obtained respectively, power grid time sequence scene reduction based on the confidence interval is achieved, and therefore the economical efficiency and reliability of scheduling optimization decision making are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system dispatching automation, and in particular to a confidence interval-based power grid timing scenario simplification method and system. Background Art

[0002] With the construction of new power systems, the installed capacity of new energy represented by wind power and photovoltaic power has grown rapidly. Affected by meteorological factors such as wind speed and light intensity, the output of new energy has significant uncertainty. At the same time, the penetration rate of new loads represented by electric vehicles continues to increase, and the uncertainty of load demand is gradually increasing. The uncertainty on both the source and load sides reduces the predictability of the future operation status of the power grid, bringing new challenges to the safe and stable operation and optimized dispatch of the power system.

[0003] As an optimization scheduling method that considers the uncertainty of power grid source-load, the scenario method has an intuitive model and clear physical meaning. It can not only characterize the probability distribution of renewable energy output in time and space, but also reflect the correlation of random fluctuations in time series. According to the probability density distribution obeyed by random variables, the scenario method samples and generates a large number of scenarios to simulate their uncertain changes, and then obtains multiple typical scenarios through scenario simplification to improve the calculation efficiency. However, the scenario method uses the same method to simplify scenarios with different probabilities and risks, and cannot obtain the scenario sets required for formulating optimized scheduling plans in high probability and low risk and low probability and high risk scenarios, which is not conducive to refined decision-making. To solve this problem, on the basis of generating multiple scenarios in time series, a confidence interval is set to characterize the credibility of the net load forecast error, and the normal scenarios within the confidence interval and the extreme scenarios outside the confidence interval are simplified by direct clustering and refined classification and re-clustering, so as to provide targeted scenario sets for formulating optimized scheduling plans under different probability and risk scenarios, and improve the economy and feasibility of the optimized scheduling plan in all scenarios. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: how to characterize the credibility of the net load forecast error by setting a confidence interval, and use different methods to simplify normal scenarios within the confidence interval and extreme scenarios outside the confidence interval, so as to provide a scenario set for the formulation of optimized scheduling plans suitable for scenarios with different probabilities and risks.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for simplifying power grid time series scenarios based on confidence intervals, which comprises the following steps:

[0007] Obtain new energy and load forecast values ​​and generate an initial time series scenario set based on probability distribution sampling of new energy and load forecast errors;

[0008] Calculate the confidence interval of the net load forecast error under a given confidence level, and divide the initial time series scenario set into a normal scenario set and an extreme scenario set according to the confidence interval;

[0009] Conduct correlation analysis on the new energy and load forecast errors of each time series scenario in the extreme scenario set and classify the extreme scenario set;

[0010] According to the system spare capacity and new energy ramp-up rate, high-risk scenario sets and general scenario sets are selected from various extreme scenario sets;

[0011] An improved K-means that comprehensively considers Euclidean distance and dynamic time warp distance is used to simplify various scene sets to obtain typical scenes.

[0012] As a preferred solution of the power grid time series scenario simplification method based on confidence interval described in the present invention, wherein: the confidence interval of the net load prediction error under a given confidence level is calculated, and the initial time series scenario set is divided into a normal scenario set and an extreme scenario set according to the confidence interval, including:

[0013] Calculate the net load forecast error ε of each time series scenario i in the initial time series scenario set S i , i = 1, 2, ..., N, where N is the total number of time series scenes in S;

[0014] According to the given confidence level α (0<α<1), the confidence interval of the net load forecast error is obtained [C d ,C u ];

[0015] Determine each time sequence scene i in S in turn. If C d ≤ε i ≤C u , then scene i is a normal scene, otherwise it is an extreme scene, and finally the normal scene set S is generated s and extreme scenario set S r .

[0016] As a preferred solution of the power grid time series scenario simplification method based on confidence interval described in the present invention, wherein: the correlation analysis of the new energy and load forecast errors of each time series scenario in the extreme scenario set and the classification of the extreme scenario set include:

[0017] Calculate S r The correlation coefficient r between the new energy and load forecast error in each scenario k k , the expression is:

[0018]

[0019] Among them, x kj ,ykj They represent the new energy forecast and load forecast errors at the jth moment in the extreme time series scenario k, are the mean values ​​of the forecast errors of new energy and load in time series scenario k, respectively, and T represents the number of sampling moments of the time series scenario;

[0020] Sequentially S r Each scene in k classification generates a set of extreme scenes with strong positive correlation Weak positive correlation extreme scenario set Extreme scenario set with no correlation Weak negative correlation extreme scenario set and strong negative correlation extreme scenario set

[0021] When 0.7 <r k When ≤1.0, scenario k is an extreme scenario with strong positive correlation;

[0022] When 0.3 <r k When ≤0.7, scenario k is an extreme scenario with weak positive correlation;

[0023] When -0.3 <r k When ≤0.3, scenario k is an extreme scenario with no correlation;

[0024] When -0.7 <r k When ≤-0.3, scenario k is an extreme scenario with weak negative correlation;

[0025] When -1.0≤r k When ≤-0.7, scenario k is an extreme scenario with strong negative correlation.

[0026] As a preferred solution of the power grid timing scenario simplification method based on confidence interval described in the present invention, wherein: the high-risk scenario set and the general scenario set are screened out from various extreme scenario sets according to the system spare capacity and the new energy ramp rate, including:

[0027] High-risk scenario set judgment rules;

[0028] System spare capacity conditions

[0029] P L,t -P N,t >k3P Lmax or P N,t -P L,t >k4P Lmin ;

[0030] New energy ramp rate conditions

[0031] P N,t+1 -P N,t >k1(P L,t+1 -PL,t ) or P L,t+1 -P L,t >k2(P N,t+1 -P N,t );

[0032] Among them, P N,t , P N,t+1 are the true values ​​of the new energy at the sampling time t and t+1 in the time series scenario, P L,t , P L,t+1 are the true load values ​​at sampling time t and t+1 in the time series scenario, P Lmax , P Lmin are the maximum and minimum values ​​of load prediction, respectively, and k1, k2, k3, and k4 are proportional coefficients;

[0033] If any one of the judgment rules is met, it belongs to the high-risk scenario set, otherwise it belongs to the general scenario set;

[0034] The true value of the new energy is the sum of the new energy prediction value and the prediction error;

[0035] The actual load value is the sum of the load prediction value and the prediction error.

[0036] As a preferred solution of the power grid timing scenario simplification method based on confidence intervals described in the present invention, wherein: the high-risk scenario set and the general scenario set also include;

[0037] Strong positive correlation high risk scenario set Strong positive correlation general scene set Weakly positively correlated high-risk scenario set Weakly positive correlation general scenario set Unrelated high-risk scenario set Non-correlated general scene set Weak negative correlation high risk scenario set Weak negative correlation general scene set Strong negative correlation high risk scenario set and strong negative correlation general scene set

[0038] As a preferred solution of the power grid time series scenario simplification method based on confidence interval described in the present invention, wherein: the improved K-means that comprehensively considers Euclidean distance and dynamic time bending distance is used to simplify various scenario sets to obtain typical scenarios including:

[0039] Calculate the similarity distance between the time series scene and the cluster center and improve the K-means clustering. The expression is:

[0040] d(X,Y)=β1d1(X pv,Y pv )+β2d1(X w ,Y w )+β3d1(X l ,Y l )

[0041] +β4d2(X pv ,Y pv )+β5d2(X w ,Y w )+β6d2(X l ,Y l )

[0042] Among them, d(X,Y) represents the similarity distance between the time series scene X and the cluster center Y, X pv , X w , X l are the prediction error vectors of each photovoltaic station, wind farm and load node in scenario X, Y pv , Y w , Y l are the prediction error vectors of each photovoltaic station, wind farm and load node in the cluster center Y, X = (X pv ,X w ,X l ),Y=(Y pv ,Y w ,Y l ), d1(*,*) represents the Euclidean distance, d2(*,*) represents the dynamic time warping distance, and β1~β6 are weight coefficients.

[0043] As a preferred solution of the power grid time series scenario simplification method based on confidence interval described in the present invention, wherein: the use of improved K-means that comprehensively considers Euclidean distance and dynamic time warping distance to simplify various scenario sets to obtain typical scenarios also includes:

[0044] For S s Improved K-means clustering is used, and the cluster center is selected as a typical normal time series scenario;

[0045] right and Improved K-means clustering was used to select the scene with the longest similarity distance from the cluster center in each cluster as the typical scene;

[0046] The typical scene probability is obtained by dividing the number of scenes in the cluster to which the typical scene belongs by the total number of time series scenes N.

[0047] Another object of the present invention is to provide a power grid timing scenario simplification system based on confidence intervals, which can distinguish normal scenarios from extreme scenarios by setting confidence intervals for net load prediction errors, and classify and simplify scenarios in combination with correlation analysis, risk assessment and an improved K-means clustering algorithm, to provide a set of typical scenarios with both high probability and low risk characteristics and low probability and high risk characteristics, thereby solving the problem that the scenario method in the prior art cannot effectively distinguish scenarios of different risk levels and the simplification effect is insufficient.

[0048] In order to solve the above technical problems, the present invention provides the following technical solutions: A power grid time series scenario simplification system based on confidence intervals, comprising: a data acquisition module for acquiring new energy and load forecast values;

[0049] The scenario generation module is used to generate an initial time series scenario set based on the probability distribution sampling of new energy and load forecast errors;

[0050] The scenario classification module is used to calculate the confidence interval of the net load forecast error under a given confidence level, and divide the initial time series scenario set into a normal scenario set and an extreme scenario set according to the confidence interval, perform correlation analysis on the new energy and load forecast error of each time series scenario in the extreme scenario set, and classify the extreme scenario set;

[0051] A screening module is used to screen out high-risk scenario sets and general scenario sets from various extreme scenario sets according to system spare capacity and new energy ramp-up rate;

[0052] The clustering and simplification module is used to simplify various scene sets to obtain typical scenes by using an improved K-means that comprehensively considers Euclidean distance and dynamic time warp distance.

[0053] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned confidence interval-based power grid time series scenario simplification method are implemented.

[0054] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a power grid timing scenario simplification method based on confidence intervals as described above.

[0055] Beneficial effects of the present invention: The present invention performs differentiated simplification on scenarios with different probability risks, and obtains scenario sets required for formulating optimized scheduling plans under high probability and low risk and low probability and high risk scenarios respectively, thereby realizing confidence interval-based simplification of power grid timing scenarios, thereby improving the economy and reliability of optimized scheduling decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0057] Figure 1 An overall flow chart of a confidence interval-based power grid timing scenario simplification method provided for the first embodiment of the present invention. DETAILED DESCRIPTION

[0058] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0059] Example 1, reference Figure 1 According to an embodiment of the present invention, a method for simplifying power grid time series scenarios based on confidence intervals is provided, comprising:

[0060] Obtain new energy and load forecast values ​​and generate an initial time series scenario set based on the probability density distribution of new energy and load forecast errors;

[0061] Calculate the confidence interval of the net load forecast error under a given confidence level, and divide the initial time series scenario set into a normal scenario set and an extreme scenario set according to the confidence interval;

[0062] Conduct correlation analysis on the new energy and load forecast errors of each time series scenario in the extreme scenario set and classify the extreme scenario set;

[0063] In each extreme scenario set, high-risk scenario sets and general scenario sets are selected based on system spare capacity and new energy ramp-up rate;

[0064] An improved K-means that comprehensively considers Euclidean distance and dynamic time warp distance is used to simplify various scene sets to obtain typical scenes.

[0065] Furthermore, the method for generating the normal scene set and the extreme scene set in step (2) is as follows:

[0066] 1) Calculate the net load prediction error ε of each time series scenario i in the initial time series scenario set S i , i = 1, 2, ..., N, where N is the total number of time series scenes in S;

[0067] 2) According to the given confidence level α (0<α<1), the confidence interval of the net load forecast error is obtained [C d ,C u ];

[0068] 3) Discriminate each time sequence scene i in S in turn. If C d ≤ε i ≤C u , then scene i is a normal scene, otherwise it is an extreme scene, and finally the normal scene set S is generated s and extreme scenario set S r .

[0069] Furthermore, in step (3), the correlation analysis of the new energy and load forecast error of each time series scenario in the extreme scenario set is performed and the extreme scenario set is classified, specifically:

[0070] 1) Calculate S according to the following formula r The correlation coefficient r between the new energy and load forecast error in each scenario k k :

[0071]

[0072] where x kj ,y kj They represent the new energy forecast and load forecast errors at the jth sampling point in the extreme time series scenario k, are the mean values ​​of the forecast errors of new energy and load in time series scenario k, respectively, and T represents the number of sampling moments of the time series scenario.

[0073] 2) Sequentially change S r Each scene in k classification generates a set of extreme scenes with strong positive correlation Weak positive correlation extreme scenario set Extreme scenario set with no correlation Weak negative correlation extreme scenario set and strong negative correlation extreme scenario set The classification method is as follows:

[0074] a) When 0.7 <r k When ≤1.0, scenario k is an extreme scenario with strong positive correlation;

[0075] b) When 0.3 <r k When ≤0.7, scenario k is an extreme scenario with weak positive correlation;

[0076] c) When -0.3 <r k When ≤0.3, scenario k is an extreme scenario with no correlation;

[0077] d) When -0.7 <r kWhen ≤-0.3, scenario k is an extreme scenario with weak negative correlation;

[0078] e) When -1.0≤r k When ≤-0.7, scenario k is an extreme scenario with strong negative correlation.

[0079] Furthermore, the method for selecting the high-risk scenario set and the general scenario set in step (4) is as follows: a scenario that meets one of the following conditions belongs to the high-risk scenario set, otherwise it belongs to the general scenario set:

[0080] System spare capacity conditions

[0081] P L,t -P N,t >k3P Lmax or P N,t -P L,t >k4P Lmin ;

[0082] New energy ramp rate conditions

[0083] P N,t+1 -P N,t >k1(P L,t+1 -P L,t ) or P L,t+1 -P L,t >k2(P N,t+1 -P N,t );

[0084] Among them, P N,t , P N,t+1 are the true values ​​of the new energy at the sampling time t and t+1 in the time series scenario, P L,t , P L,t+1 are the true load values ​​at sampling time t and t+1 in the time series scenario, P Lmax , P Lmin are the maximum and minimum values ​​of load prediction respectively, and k1, k2, k3, and k4 are proportional coefficients.

[0085] Furthermore, the high-risk scenario set and the general scenario set in step (4) include: a strong positive correlation high-risk scenario set Strong positive correlation general scene set Weakly positively correlated high-risk scenario set Weakly positive correlation general scenario set Unrelated high-risk scenario set Non-correlated general scene set Weak negative correlation high risk scenario set Weak negative correlation general scene set Strong negative correlation high risk scenario set and strong negative correlation general scene set

[0086] Furthermore, in step (5), an improved K-means that comprehensively considers Euclidean distance and dynamic time warping distance is used to simplify various scene sets to obtain typical scenes, specifically:

[0087] 1) Calculate the similarity distance between the time series scene and the cluster center according to the following formula to improve K-means clustering:

[0088] d(X,Y)=β1d1(X pv ,Y pv )+β2d1(X w ,Y w )+β3d1(X l ,Y l )

[0089] +β4d2(X pv ,Y pv )+β5d2(X w ,Y w )+β6d2(X l ,Y l )

[0090] Among them, d(X,Y) represents the similarity distance between the time series scene X and the cluster center Y, X pv , X w , X l are the prediction error vectors of each photovoltaic station, wind farm and load node in scenario X, Y pv , Y w , Y l are the prediction error vectors of each photovoltaic station, wind farm and load node in the cluster center Y, X = (X pv ,X w ,X l ),Y=(Y pv ,Y w ,Y l ), d1(*,*) represents the Euclidean distance, d2(*,*) represents the dynamic time warping distance, and β1~β6 are weight coefficients;

[0091] 2) For S s Improved K-means clustering is used, and the cluster center is selected as a typical normal time series scenario;

[0092] 3) Yes and Improved K-means clustering was used respectively, and the scene with the longest similarity distance to the cluster center in each clustering result was selected as the typical scene.

[0093] 4) The probability of a typical scene is obtained by dividing the number of scenes in the class to which the typical scene belongs by the total number of time series scenes N.

[0094] Embodiment 2 is an embodiment of the present invention, and provides a system for a power grid time series scenario simplification method based on confidence intervals, including:

[0095] Data acquisition module, used to obtain new energy and load forecast values;

[0096] The scenario generation module is used to generate an initial time series scenario set based on the probability distribution sampling of new energy and load forecast errors;

[0097] The scenario classification module is used to calculate the confidence interval of the net load forecast error under a given confidence level, and divide the initial time series scenario set into a normal scenario set and an extreme scenario set according to the confidence interval, perform correlation analysis on the new energy and load forecast error of each time series scenario in the extreme scenario set, and classify the extreme scenario set;

[0098] A screening module is used to screen out high-risk scenario sets and general scenario sets from various extreme scenario sets according to system spare capacity and new energy ramp-up rate;

[0099] The clustering and simplification module is used to simplify various scene sets to obtain typical scenes by using an improved K-means that comprehensively considers Euclidean distance and dynamic time warp distance.

[0100] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0101] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0102] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0103] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0104] 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 preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A power grid time series scenario simplification method based on confidence interval, characterized in that: include: Obtain new energy and load forecast values ​​and generate an initial time series scenario set based on probability distribution sampling of new energy and load forecast errors; Calculate the confidence interval of the net load forecast error under a given confidence level, and divide the initial time series scenario set into a normal scenario set and an extreme scenario set according to the confidence interval; Conduct correlation analysis on the new energy and load forecast errors of each time series scenario in the extreme scenario set and classify the extreme scenario set; According to the system spare capacity and new energy ramp-up rate, high-risk scenario sets and general scenario sets are selected from various extreme scenario sets; An improved K-means that comprehensively considers Euclidean distance and dynamic time warp distance is used to simplify various scene sets to obtain typical scenes.

2. The method for simplifying power grid time series scenarios based on confidence intervals according to claim 1, characterized in that: The calculation of the confidence interval of the net load forecast error under a given confidence level and the division of the initial time series scenario set into a normal scenario set and an extreme scenario set according to the confidence interval include: Calculate the net load forecast error ε of each time series scenario i in the initial time series scenario set S i , i = 1, 2, ..., N, where N is the total number of time series scenes in S; According to the given confidence level α (0<α<1), the confidence interval of the net load forecast error is obtained [C d ,C u ]; Determine each time sequence scene i in S in turn. If C d ≤ε i ≤C u , then scene i is a normal scene, otherwise it is an extreme scene, and finally the normal scene set S is generated s and extreme scenario set S r .

3. The method for simplifying power grid time series scenarios based on confidence intervals according to claim 2, characterized in that: The correlation analysis of the new energy and load forecast error of each time series scenario in the extreme scenario set is performed and the extreme scenario set is classified into: Calculate S r The correlation coefficient r between the new energy and load forecast error in each scenario k k , the expression is: Among them, x kj ,y kj They represent the new energy forecast and load forecast errors at the jth moment in the extreme time series scenario k, are the mean values ​​of the forecast errors of new energy and load in time series scenario k, respectively, and T represents the number of sampling moments of the time series scenario; Sequentially S r Each scene in k classification generates a set of extreme scenes with strong positive correlation Weak positive correlation extreme scenario set Extreme scenario set with no correlation Weak negative correlation extreme scenario set and strong negative correlation extreme scenario set When 0.7 <r k When ≤1.0, scenario k is an extreme scenario with strong positive correlation; When 0.3 <r k When ≤0.7, scenario k is an extreme scenario with weak positive correlation; When -0.3 <r k When ≤0.3, scenario k is an extreme scenario with no correlation; When -0.7 <r k When ≤-0.3, scenario k is an extreme scenario with weak negative correlation; When -1.0≤r k When ≤-0.7, scenario k is an extreme scenario with strong negative correlation.

4. The method for simplifying power grid time series scenarios based on confidence intervals according to claim 3, characterized in that: The high-risk scenario set and the general scenario set are selected from various extreme scenario sets according to the system spare capacity and the new energy ramp-up rate, including: High-risk scenario set judgment rules; System spare capacity conditions P L,t -P N,t >k3P Lmax or P N,t -P L,t >k4P Lmin ; New energy ramp rate conditions P N,t+1 -P N,t >k1(P L,t+1 -P L,t ) or P L,t+1 -P L,t >k2(P N,t+1 -P N,t ); Among them, P N,t , P N,t+1 are the true values ​​of the new energy at the sampling time t and t+1 in the time series scenario, P L,t , P L,t+1 are the true load values ​​at sampling time t and t+1 in the time series scenario, P Lmax , P Lmin are the maximum and minimum values ​​of load prediction, respectively, and k1, k2, k3, and k4 are proportional coefficients; If any one of the judgment rules is met, it belongs to the high-risk scenario set, otherwise it belongs to the general scenario set; The true value of the new energy is the sum of the predicted value of the new energy and the prediction error; The actual load value is the sum of the load prediction value and the prediction error.

5. The method for simplifying power grid time series scenarios based on confidence intervals according to claim 4, characterized in that: The high-risk scenario set and general scenario set also include: Strong positive correlation high risk scenario set Strong positive correlation general scene set Weakly positively correlated high-risk scenario set Weakly positive correlation general scenario set Unrelated high-risk scenario set Non-correlated general scene set Weak negative correlation high risk scenario set Weakly negatively correlated general scenario set Strong negative correlation high risk scenario set and strong negative correlation general scene set 6. A method for simplifying power grid time series scenarios based on confidence intervals according to claim 5, characterized in that: The improved K-means method that comprehensively considers Euclidean distance and dynamic time warping distance is used to simplify various scene sets to obtain typical scenes including: Calculate the similarity distance between the time series scene and the cluster center and improve the K-means clustering. The expression is: d(X,Y)=β1d1(X pv ,Y pv )+β2d1(X w ,Y w )+β3d1(X l ,Y l )+β4d2(X pv ,Y pv )+β5d2(X w ,Y w )+β6d2(X l ,Y l ) Among them, d(X,Y) represents the similarity distance between the time series scene X and the cluster center Y, X pv , X w , X l are the prediction error vectors of each photovoltaic station, wind farm and load node in scenario X, Y pv , Y w , Y l are the prediction error vectors of each photovoltaic station, wind farm and load node in the cluster center Y, X = (X pv ,X w ,X l ),Y=(Y pv ,Y w ,Y l ), d1(*,*) represents the Euclidean distance, d2(*,*) represents the dynamic time warping distance, and β1~β6 are weight coefficients.

7. A method for simplifying power grid time series scenarios based on confidence intervals according to claim 6, characterized in that: The improved K-means method that comprehensively considers Euclidean distance and dynamic time warping distance is used to simplify various scene sets to obtain typical scenes, including: For S s Improved K-means clustering is used, and the cluster center is selected as a typical normal time series scenario; right and Improved K-means clustering was used to select the scene with the longest similarity distance from the cluster center in each cluster as the typical scene; The typical scene probability is obtained by dividing the number of scenes in the cluster to which the typical scene belongs by the total number of time series scenes N.

8. A system using a power grid time sequence scenario simplification method based on confidence intervals as claimed in any one of claims 1 to 7, characterized in that: include Data acquisition module, used to obtain new energy and load forecast values; The scenario generation module is used to generate an initial time series scenario set based on the probability distribution sampling of new energy and load forecast errors; The scenario classification module is used to calculate the confidence interval of the net load forecast error under a given confidence level, and divide the initial time series scenario set into a normal scenario set and an extreme scenario set according to the confidence interval, perform correlation analysis on the new energy and load forecast error of each time series scenario in the extreme scenario set, and classify the extreme scenario set; A screening module is used to screen out high-risk scenario sets and general scenario sets from various extreme scenario sets according to system spare capacity and new energy ramp-up rate; The clustering and simplification module is used to simplify various scene sets to obtain typical scenes by using an improved K-means that comprehensively considers Euclidean distance and dynamic time warp distance.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a confidence interval-based power grid timing scenario simplification method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a confidence interval-based power grid timing scenario simplification method described in any one of claims 1 to 7 are implemented.

Images