Power supply risk early warning method and device based on wind and light output safety domain

Abstract

The invention discloses a power supply risk early warning method and device based on a wind and light output safety domain. The method comprises the following steps: carrying out clustering analysis on historical data of a wind and light output scene by adopting a deep embedding clustering method, and determining a clustering result; determining a typical scene set of a wind and light output scene according to a clustering result, and determining a wind and light output extreme value scene according to historical data; performing risk quantitative evaluation on the wind and light output extreme value scene according to a pre-established new energy power supply risk quantitative evaluation model, and determining a risk quantitative result of the wind and light output extreme value scene; when the risk quantification result is a dangerous scene, searching a scene adjacent to the dangerous scene in the typical scene set, and evaluating an early warning result of the wind and light output scene according to the adjacent scene; and when the risk quantification result is a dangerous scene, determining that the wind and light output scene has no operation risk.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid operation, and more specifically, to a power supply risk early warning method and device based on a wind and solar power output safety domain. Background Art

[0002] In recent years, the construction of wind and solar power stations has developed towards large-scale and large-scale development, and the penetration rate of wind and solar power generation in the power grid has increased rapidly. Its randomness and volatility threaten the stable operation of the power grid. In view of the problem of large uncertainty in wind and solar power output in high-proportion new energy power systems, how to carry out research on power supply risk warning methods that take into account both risk warning efficiency and comprehensive wind and solar scenarios has become an urgent requirement for further deepening theoretical research related to new energy. At present, there are the following deficiencies in power supply risk warning: First, the accuracy of wind and solar power output prediction needs to be improved, and the method for determining the boundary of the wind and solar power output safety domain still has errors at the critical point, and a more accurate method is needed to define the safety domain; second, the traditional risk assessment method is difficult to meet the needs of real-time monitoring and early warning of new power systems in terms of computational efficiency, and more efficient algorithms and models are needed to achieve rapid risk assessment. Summary of the invention

[0003] In view of the deficiencies in the prior art, the present invention provides a power supply risk warning method and device based on wind and solar power output safety domain.

[0004] According to one aspect of the present invention, a power supply risk warning method based on wind and solar power output safety domain is provided, comprising:

[0005] A deep embedding clustering method is used to cluster the historical data of wind and solar power output scenarios and determine the clustering results;

[0006] Determine the typical scenario set of wind and solar power output scenarios based on the clustering results, and determine the wind and solar power output extreme scenario based on historical data;

[0007] Conduct risk quantitative assessment on wind and solar power output extreme value scenarios based on the pre-established risk quantitative assessment model for new energy power supply, and determine the risk quantification results of wind and solar power output extreme value scenarios;

[0008] When the risk quantification result is a dangerous scenario, find the adjacent scenarios of the dangerous scenario in the typical scenario set, and evaluate the warning results of the wind and solar power output scenario based on the adjacent scenarios;

[0009] When the risk quantification result is a dangerous scenario, it is determined that the wind and solar power output scenario has no operating risk.

[0010] Optionally, the deep embedding clustering method includes: stacking autoencoders and an improved K-means algorithm, and using the deep embedding clustering method to perform clustering analysis on historical data of wind and solar power output scenarios to determine clustering results, including:

[0011] Encoding the historical data using a stacked autoencoder to determine a low-dimensional feature representation of the historical data;

[0012] The improved K-means algorithm is used to perform cluster analysis on the low-dimensional feature representation to determine the clustering results.

[0013] Optionally, the stacked autoencoder is trained using a layered pre-training method, and the specific steps are:

[0014] Step 1: Initially train the bottom layer of the encoder and the top layer of the decoder using a single-layer autoencoder;

[0015] Step 2: Use the output of the initially trained encoder as input to train the next pair of encoders and decoders, repeat 1 to complete all training processes, and obtain a stacked autoencoder.

[0016] Optionally, the network cost function for the stacked autoencoder is:

[0017]

[0018] In the formula, f(x i ) represents the hidden layer encoding vector, g(f(x i )) is the output vector of the DEC network, l is the reconstruction loss function; Ms i For cluster class s i The cluster center of , λ is the L2 regularization factor.

[0019] Optionally, the improved K-means algorithm is obtained by improving the K-means algorithm based on the density clustering principle, and the specific training steps are:

[0020] The elbow method is used to determine the optimal clustering number k of the sample set;

[0021] In a given sample set X = {x1, x2, ..., x n}, calculate the Euclidean distance d(x i , x j ), the expression is:

[0022]

[0023] Set the distance threshold d t , and according to the Euclidean distance d(x i , x j ) Calculate the distance threshold d t The local density ρ i , the expression is:

[0024]

[0025] Select the data with the largest local density as the first initial cluster center m1, select the data with the farthest distance from m1 and the highest density as the second initial cluster center m2 according to the Euclidean distance, select the data with the relatively farthest distance from m1 and the highest density as the third initial cluster center m3, and so on, and you can get k initial cluster centers (m1, m2, ..., m k ).

[0026] Optionally, the core indicator of the elbow method is the sum of squared errors S SE , the error sum of squares S SE The calculation formula is:

[0027]

[0028] In the formula, C i is the i-th cluster; p is C i Sample points in m i It is C i The cluster center of .

[0029] Optionally, a typical scene set of wind and solar power output scenes is determined according to the clustering results, including:

[0030] Taking the cluster center of the clustering result as the benchmark, the sample with the closest distance to the cluster center in each cluster sample set is selected as the representative scene of the cluster, which is used as the typical scene set.

[0031] Optionally, the objective function of the new energy power supply risk quantitative assessment model is:

[0032]

[0033] Where t is the simulation time step, P w (t,n) is the wind power generation power in region n during period t, P pv (t,n) is the power generation of photovoltaic power generation in area n during period t, e L,Curt (t,n) is the insufficient power supply power of area n at time t, and λ is the insufficient power supply penalty term;

[0034] The system power balance constraints are as follows:

[0035]

[0036] Where P G (t,n) is the total output of the n thermal power units in the region at time t, P H (t,n) is the total output of regional n hydropower units at time t, P CX (t,n) is the planned output of pumped storage unit n in region at time t, P CN (t,n) is the planned output of energy storage in area n at time t, Li (t) is the transmission power of the i-th section transmission line at time t;

[0037] The system backup requirement constraints are as follows:

[0038]

[0039] Where P re (t,n) is the positive reserve of area n at time t, N re (t,n) is the negative reserve of region n at time t;

[0040] The thermal power output range and climbing constraints are as follows:

[0041] P G,min (t,n)≤P G (t,n)≤P G,max (t,n)

[0042] P G,down ≤P G (t,n)-P G (t-1,n)≤P G,up

[0043] Where P G,min (t,n) is the total minimum technical output of n thermal power units in the region at time t, P G,max (t,n) is the total maximum technical output of n thermal power units in the region at time t; P G,down is the downward climbing amount of the thermal power unit, P G,up is the upward climbing amount of the thermal power unit;

[0044] The output constraints of new energy are as follows:

[0045]

[0046] In the formula, The predicted output of wind power at the moment, The predicted output of photovoltaic power generation at any moment;

[0047] The inter-area line transmission constraints are as follows:

[0048] L i,min ≤L i (t)≤L i,max

[0049] Where, L i,max , L i,min are the upper and lower limits of the transmission capacity of the ith transmission line, respectively. The reference direction of power is set as follows: the inflow area is the positive direction, and the outflow area is the negative direction.

[0050] Optionally, a risk quantitative assessment is performed on the wind and solar power output extreme value scenario according to a pre-established new energy power supply risk quantitative assessment model to determine the risk quantification result of the wind and solar power output extreme value scenario, including:

[0051] According to the quantitative assessment model of new energy power supply risk, power shortage, wind and solar power abandonment and cross-section transmission power are calculated for the wind and solar power extreme output scenario, and the amplitude and probability of power shortage, the amplitude and probability of wind and solar power abandonment, and the amplitude and probability of cross-section over-limit in the wind and solar power extreme output scenario are counted;

[0052] The risk quantification results of the wind and solar power output extreme scenario are determined based on the amplitude and probability of power supply shortage, the amplitude and probability of wind and solar power abandonment, and the amplitude and probability of section exceeding the limit.

[0053] Optionally, the magnitude of the underpowered and probability The calculation formula is:

[0054]

[0055] In the formula, The upper and lower limits of the conventional unit output, PD t , P l,t For loads and tie lines, To reserve for backup, Contribute to the new energy in this scenario;

[0056] The magnitude of curtailed wind and solar power and probability The calculation formula is:

[0057]

[0058] In the formula, N is the number of scenarios where insufficient backup occurs. s is the total number of scenes;

[0059] The calculation formula for the amplitude and probability of cross-section exceeding the limit is:

[0060]

[0061] In the formula, and is the maximum over-limit amplitude caused by partition i to the section between partitions i and j at time t, is the planned net output power of the section between partition i and other partitions except partition j at time t; is the number of scenarios where the cross-section between partitions i and j exceeds the limit at time t, N s is the total number of scenes; They are respectively the probability of the section between partitions i and j exceeding the limit when the positive and negative reserves in partition i are insufficient at time t.

[0062] Optionally, the risk quantification results of the wind and solar power output extreme value scenario are determined according to the magnitude and probability of power shortage, the magnitude and probability of wind and solar power abandonment, and the magnitude and probability of section over-limit, including:

[0063] Determine the power supply shortage risk based on the magnitude and probability of power supply shortage and power supply shortage risk criteria;

[0064] Determine the risk of wind and solar power abandonment based on the magnitude and probability of wind and solar power abandonment and the risk criterion of wind and solar power abandonment;

[0065] Determine the cross-section risk based on the cross-section amplitude and probability and cross-section risk criterion;

[0066] Based on the risks of insufficient power supply, wind and solar power abandonment, and section over-limit risks, the risk quantification results of the wind and solar power output extreme scenarios are determined.

[0067] Optionally, the warning result of the wind and solar power output scenario is evaluated based on adjacent scenarios, including:

[0068] Perform risk assessment calculations on adjacent scenes that have not been assessed. If a new dangerous scene is found, record the category cluster number i of the clustering result where the new dangerous scene is located, perform risk quantitative assessment on all wind and solar power output scenes in the i-th cluster, and further search for dangerous scenes in the i-th cluster;

[0069] If no new dangerous scenarios are found, the unevaluated wind-solar output scenarios are regarded as safe scenarios, and the unevaluated wind-solar output scenarios and wind-solar output safety scenarios constitute the wind-solar output safety domain;

[0070] Conduct statistical analysis on the wind and solar power output safety domain, calculate the risk level, and determine the early warning results.

[0071] According to another aspect of the present invention, a power supply risk warning device based on wind and solar power output safety domain is provided, comprising:

[0072] The cluster analysis module is used to perform cluster analysis on the historical data of wind and solar power output scenarios using a deep embedding clustering method to determine the clustering results;

[0073] A determination module is used to determine a typical scene set of wind and solar power output scenes according to the clustering results, and to determine the extreme scene of wind and solar power output according to historical data;

[0074] The first assessment module is used to perform risk quantitative assessment on the wind and solar power output extreme value scenario according to a pre-established new energy power supply risk quantitative assessment model, and determine the risk quantification result of the wind and solar power output extreme value scenario;

[0075] The second evaluation module is used to find adjacent scenes of the dangerous scene in the typical scene set when the risk quantification result is a dangerous scene, and evaluate the warning result of the wind and solar power output scene according to the adjacent scenes;

[0076] The determination module is used to determine that the wind and solar power output scenario has no operating risk when the risk quantification result is a dangerous scenario.

[0077] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the storage medium stores a computer program, and the computer program is used to execute the method described in any one of the above aspects of the present invention.

[0078] According to another aspect of the present invention, an electronic device is provided, comprising: a processor; a memory for storing instructions executable by the processor; the processor is configured to read the executable instructions from the memory and execute the instructions to implement the method described in any one of the above aspects of the present invention.

[0079] Therefore, the present invention provides a power supply risk warning method, which constructs a wind / solar output safety domain based on the extreme scenario risk assessment results, determines whether it is a risk scenario through the positional relationship between the typical wind / solar output scenario and the wind / solar output safety domain, and improves the warning efficiency by reducing the risk scenario search range, providing a rapid warning for the safe operation of the power grid. Through the precise selection and accurate traversal of wind / solar output extreme scenarios and wind / solar output typical scenarios, redundant calculations of a large number of safety scenarios are eliminated, and the division of wind / solar output safety domains and rapid warning of power supply risks are achieved at the same time; the power supply risk warning method provided by the present invention can comprehensively cover wind / solar output dangerous scenarios, and the warning results have a more significant reference value. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] A more complete understanding of exemplary embodiments of the present invention may be obtained by referring to the following drawings:

[0081] Figure 1 It is a flow chart of a power supply risk early warning method based on wind and solar power output safety domain provided by an exemplary embodiment of the present invention;

[0082] Figure 2 is a schematic diagram of a deep embedding clustering model architecture provided by an exemplary embodiment of the present invention;

[0083] Figure 3 is a schematic diagram of a risk assessment method flow chart provided by an exemplary embodiment of the present invention;

[0084] Figure 4 It is a structural schematic diagram of a power supply risk warning device based on wind and solar power output safety domain provided by an exemplary embodiment of the present invention;

[0085] Figure 5 This is a structure of an electronic device provided by an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0086] Below, the exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention, and it should be understood that the present invention is not limited to the exemplary embodiments described here.

[0087] It should be noted that the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

[0088] Those skilled in the art can understand that the terms "first" and "second" in the embodiments of the present invention are only used to distinguish different steps, devices or modules, etc., and neither represent any specific technical meaning nor indicate the necessary logical order between them.

[0089] It should also be understood that, in the embodiments of the present invention, “plurality” may refer to two or more than two, and “at least one” may refer to one, two or more than two.

[0090] It should also be understood that any component, data or structure mentioned in the embodiments of the present invention can generally be understood as one or more, unless explicitly limited or otherwise indicated in the context.

[0091] In addition, the term "and / or" in the present invention is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the present invention generally indicates that the associated objects before and after are in an "or" relationship.

[0092] It should also be understood that the description of the various embodiments of the present invention focuses on the differences between the various embodiments, and the same or similar aspects thereof can be referenced to each other, and for the sake of brevity, they will not be described one by one.

[0093] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0094] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0095] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0096] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0097] Embodiments of the present invention can be applied to electronic devices such as terminal devices, computer systems, servers, etc., which can operate with many other general or special computing system environments or configurations. Examples of well-known terminal devices, computing systems, environments and / or configurations suitable for use with electronic devices such as terminal devices, computer systems, servers, etc. include, but are not limited to: personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network personal computers, small computer systems, large computer systems, and distributed cloud computing technology environments including any of the above systems, etc.

[0098] Electronic devices such as terminal devices, computer systems, servers, etc. can be described in the general context of computer system executable instructions (such as program modules) executed by computer systems. Generally, program modules can include routines, programs, object programs, components, logic, data structures, etc., which perform specific tasks or implement specific abstract data types. Computer systems / servers can be implemented in a distributed cloud computing environment, where tasks are performed by remote processing devices linked through a communication network. In a distributed cloud computing environment, program modules can be located on local or remote computing system storage media including storage devices.

[0099] Exemplary Methods

[0100] Figure 1 is a flow chart of a power supply risk warning method based on wind and solar power output safety domain provided by an exemplary embodiment of the present invention. This embodiment can be applied to electronic devices, such as Figure 1 As shown, the power supply risk warning method 100 based on the wind and solar power output safety domain includes the following steps:

[0101] Step 101, using a deep embedding clustering method to perform cluster analysis on historical data of wind and solar power output scenarios to determine clustering results;

[0102] Step 102, determining a typical scene set of wind and solar power output scenes according to the clustering results, and determining a wind and solar power output extreme value scene according to historical data;

[0103] Step 103, performing risk quantitative assessment on the wind and solar power output extreme value scenario according to a pre-established new energy power supply risk quantitative assessment model, and determining the risk quantification result of the wind and solar power output extreme value scenario;

[0104] Step 104, when the risk quantification result is a dangerous scene, search for adjacent scenes of the dangerous scene in the typical scene set, and evaluate the warning result of the wind and solar power output scene according to the adjacent scenes;

[0105] Step 105: When the risk quantification result is a dangerous scenario, it is determined that the wind and solar power output scenario has no operating risk.

[0106] Specifically, the present invention proposes a power supply risk warning method based on wind and solar power output safety domain, which solves the problem of rapid warning of system power supply risk while realizing the division of wind and solar power output safety domain. The method constructs a wind / solar power output safety domain based on the extreme value scenario risk assessment results, determines whether it is a risk scenario based on the positional relationship between the typical wind and solar power output scenario and the wind and solar power output safety domain, and improves the warning efficiency by reducing the search range of risk scenarios, providing rapid warning for the safe operation of the power grid.

[0107] The principle of the method for early warning of power supply risks for a power system including large-scale wind / solar grid connection in the present invention is as follows:

[0108] (1) Input wind / solar output scenario data, cluster and extract typical scenarios, and calculate wind / solar output extreme scenario.

[0109] First, the DEC (deep embedding clustering) method combining stacked autoencoders and improved K-means algorithms is used to cluster wind and solar output scenarios. DEC first encodes high-dimensional input data to achieve low-dimensional feature representation of wind and solar joint output data, and then clusters the extracted low-dimensional features, that is, K-means iteration is performed on the output data of the hidden layer in the stacked autoencoder. The loss in this process consists of reconstruction loss and clustering cost, and the Adam (AdaptiveMomentQ Estimation) optimizer is used to achieve joint optimization of the overall network loss, and finally the division of typical wind and solar output scenarios is achieved. The DEC model architecture is as follows: Figure 2 .

[0110] The network cost function is:

[0111]

[0112] In the formula, f(x i ) represents the hidden layer encoding vector, g(f(x i )) is the output vector of the DEC network, l is the reconstruction loss function; Ms i For cluster class s i The cluster center of , λ is the L2 regularization factor to prevent the model from overfitting.

[0113] For the unsupervised DEC network, the layered pre-training method is used here to train the stacked autoencoder, so as to initialize the network parameters (W, Z) so that the DEC network can extract effective features. The layered pre-training process is as follows:

[0114] 1) Use a single-layer autoencoder to train the bottom layer of the encoder and the top layer of the decoder.

[0115] 2) Use the output of the trained encoder as input to train the next pair of encoder and decoder, and repeat 1) to complete all the training processes.

[0116] The pre-trained weights are loaded, a high-dimensional output data set is input to train the network, and K-means iteration is performed on the output data of the hidden layer in the stacked autoencoder to achieve deep embedding clustering of the wind-solar joint output data.

[0117] Secondly, the K-means clustering algorithm is improved based on the density clustering principle, and the elbow method is used to quantitatively evaluate the clustering effect to determine the optimal number of clusters. The principle of density clustering is that the cluster center has a high density and is relatively far away from other high-density cluster centers. The K-means clustering algorithm randomly selects the initial cluster center, which is easy to fall into the local optimum and lead to clustering failure. Now the selection method of the cluster center is improved in combination with the density clustering algorithm. The principles and steps are as follows:

[0118] 1) Given a sample set X = {x1, x2, ..., x n}, the Euclidean distance between two data points in D-dimensional space is calculated as:

[0119]

[0120] Set the threshold distance d t , and calculate the local density ρ within the distance threshold i , the expression is:

[0121]

[0122] 2) Select the data with the largest density as the first initial cluster center m1, select the data with the largest density and the farthest distance from m1 as the second initial cluster center m2 according to the Euclidean distance, and select the data with the largest density and the farthest distance from m1 and m2 as the third initial cluster center m3. And so on, we can get k initial cluster centers (m1, m2, ..., m k ).

[0123] The traditional K-means clustering algorithm is difficult to directly determine the optimal number of clusters. The present invention uses the elbow method to determine the optimal number of clusters. Its core indicator is SSE (sum of squared errors). The SSE calculation formula is shown in formula (4):

[0124]

[0125] Where: C i is the i-th cluster; p is C i Sample points in m i It is C i The cluster center of .

[0126] SSE is an important indicator to measure the quality of clustering effect. It represents the sum of squares of each sample to the center of the category. The smaller the SSE, the better the clustering effect. Therefore, for the number of clusters K, when K is less than the actual number of clusters, as the K value increases, the degree of aggregation within each cluster increases significantly, so SSE shows a significant downward trend. When the K value reaches the actual number of clusters, as the K value continues to increase, the return of the degree of aggregation between clusters rapidly decreases, so SSE shows a gentle downward trend. Therefore, the relationship between SSE and the number of clusters K is approximately in the shape of an elbow, and the K value corresponding to the elbow is the actual number of clusters of the data.

[0127] Secondly, typical scenarios are extracted for various wind and solar power output scenarios. According to the principle of the improved K-means clustering algorithm, the cluster center of each cluster is the mean of all samples in the cluster, so the cluster center is not an actual output sample. If the cluster center is directly used as a typical scenario, on the one hand, the sample will lose its authenticity, and on the other hand, the aggregation effect will weaken the volatility of the original output, making the subsequent optimization scheduling research lose its practical significance. Therefore, based on the DEC results, this paper takes the cluster center as the benchmark, selects the sample closest to the cluster center in the cluster sample set as the representative scenario of the cluster, and selects the typical scenario in this way.

[0128] Finally, the wind and solar power output extreme value scenario is calculated. The extreme value scenario is representative and can characterize the wind power output scenarios in different directions outside the wind and solar power output safety domain. It is defined here as a scenario containing any wind power output extreme value.

[0129] (2) Establish a quantitative assessment model for the risk of renewable energy power supply to assess the system supply guarantee and renewable energy consumption gap. The model uses a set of short-term wind / solar output scenarios as input, combines the system startup plan, and conducts multi-scenario power balance optimization calculations to conduct risk assessment. The model comprehensively considers constraints such as system and partition power balance, backup demand, conventional unit output range, and inter-regional line transmission. The objective function is shown in Equation (5), and the constraints are shown in Equations (6)-(11).

[0130]

[0131] Where: t is the simulation time step, P w (t,n) is the wind power generation power in region n during period t, P pv(t,n) is the power generation of photovoltaic power generation in area n during period t, e L,Curt (t,n) is the insufficient power supply power of area n at time t, and λ is the insufficient power supply penalty term.

[0132] The system power balance constraints are as follows:

[0133]

[0134] Where: P G (t,n) is the total output of the n thermal power units in the region at time t, P H (t,n) is the total output of regional n hydropower units at time t, P CX (t,n) is the planned output of pumped storage unit n in region at time t, P CN (t,n) is the planned output of energy storage in area n at time t, L i (t) is the transmission power of the i-th section transmission line at time t.

[0135] The system backup requirement constraints are as follows:

[0136]

[0137] Where: P re (t,n) is the positive reserve of area n at time t, N re (t,n) is the negative reserve of region n at time t.

[0138] The thermal power output range and climbing constraints are as follows:

[0139] P G,min (t,n)≤P G (t,n)≤P G,max (t,n) (8)

[0140] P G,down ≤P G (t,n)-P G (t-1,n)≤P G,up (9)

[0141] Where: P G,min (t,n) is the total minimum technical output of n thermal power units in the region at time t, P G,max (t,n) is the total maximum technical output of n thermal power units in the region at time t; P G,down is the downward climbing amount of the thermal power unit, P G,up is the upward climbing amount of the thermal power unit.

[0142] The output constraints of new energy are as follows:

[0143]

[0144] Where: The predicted output of wind power at the moment, Predicted photovoltaic power generation output at any given moment.

[0145] The inter-area line transmission constraints are as follows:

[0146] L i,min ≤L i (t)≤L i,max (11)

[0147] Where: L i,max , L i,min are the upper and lower limits of the transmission capacity of the ith transmission line, respectively. The reference direction of power is set as follows: the inflow area is the positive direction, and the outflow area is the negative direction.

[0148] By using the cplex solver, the power shortage and power abandonment under different wind power / photovoltaic output scenarios can be obtained by solving the optimization model. The amplitude and probability of power shortage and power abandonment risks at each moment can be obtained through statistics.

[0149] (3) Based on the model optimization results, the power shortage, wind and solar power curtailment and cross-section transmission power are calculated for the extreme scenarios, the amplitude and probability of power shortage, wind and solar power curtailment and cross-section over-limit are counted, and the system risk situation at each moment is quantitatively analyzed. Wind and solar power curtailment magnitude The calculation formula is as follows:

[0150]

[0151] in, The upper and lower limits of the conventional unit output, PD t , P l,t For loads and tie lines, To reserve for backup, Contribute to the new energy in this scenario;

[0152] Probability of insufficient power supply at each moment Probability of wind and solar power curtailment The calculation formula is as follows:

[0153]

[0154] in, N is the number of scenarios where insufficient backup occurs. s is the total number of scenes.

[0155] The calculation formula for the system cross-section over-limit value in each scenario is as follows:

[0156]

[0157] in, and is the maximum over-limit amplitude caused by partition i to the section between partitions i and j at time t, It is the planned net output power of the section between partition i and other partitions except partition j at time t.

[0158] The calculation formula of the cross-limit probability of each section at each time is as follows:

[0159]

[0160] in, is the number of scenarios where the cross-section between partitions i and j exceeds the limit at time t, N s is the total number of scenes.

[0161] The risk criterion is formulated based on the magnitude and probability of risk occurrence and is calculated using the following formula:

[0162] L = round(1-μ)L P +μL S

[0163] Where, L p is the probability (likelihood) of risk occurrence, i.e. the probability of insufficient power supply Probability of wind and solar power curtailment Or the probability of cross-section exceeding the limit, L s is the risk severity (amplitude), i.e. the power supply shortage amplitude Wind and solar power curtailment magnitude Or the cross-section limit amplitude, μ is the weight coefficient with a value interval of (0,1), and round() means rounding to the nearest integer. If μ is 1, it is a traditional deterministic safety assessment method, and risk assessment should take into account the possibility and consequences of risk occurrence, so μ is set to the interval (0,1). When μ is close to 1, the risk level result focuses more on the consequences of risk occurrence; when μ is close to 0, the risk level result focuses more on the possibility of risk occurrence. Here, based on the actual system operation and combined with the experience of professionals, considering that power companies generally focus more on the consequences of risk occurrence, and it is an extreme scenario with a low probability event, μ is set to 0.6. If the risk criterion result of any one of insufficient power supply, wind and solar power abandonment, and cross-section limit is a risk, then the scenario is judged to be a risk scenario.

[0164] (4) If no dangerous scenario is found in the risk assessment calculation of the wind / solar output extreme scenario, the system is considered to have no operating risk, the assessment result is output and the system exits; if a dangerous scenario is found, execute step 5;

[0165] (5) Search for scenes adjacent to dangerous scenes in typical wind / solar output scenes, and perform risk assessment calculations on scenes that have not been evaluated. If a new dangerous scene is found, record the category cluster number i of the wind / solar output scene, execute step (6), and repeat step 5; if no new dangerous scene is found, execute step 7.

[0166] Among them, by calculating the spatial distance, the adjacent scenes that are closer to the dangerous scene can be obtained. When performing risk assessment on these adjacent scenes, if a new dangerous scene is found, the risk assessment of the adjacent scenes of the new dangerous scene will continue until no new dangerous scene appears.

[0167] (6) Execute steps 1) to 5) for all wind / solar output scenarios in the i-th cluster, and further search for dangerous scenarios in the i-th cluster;

[0168] (7) The unevaluated wind / solar output scenarios are regarded as safe scenarios. The unevaluated wind / solar output scenarios and the wind / solar output safety scenarios constitute the wind / solar output safety domain. The risk assessment results are statistically analyzed, such as risk category, risk amplitude, risk probability, etc., the risk level is calculated, and the warning results are output.

[0169] like Figure 3 As shown in the figure, the power supply risk warning method uses the wind / solar output extreme value scenario as the initial scenario for risk assessment. If the wind / solar output extreme value scenario is a dangerous scenario, it takes these wind / solar output extreme value scenarios as the center, continuously searches for new wind / solar output typical scenarios in the surrounding area, and executes the command of evaluating the wind / solar output extreme value scenario and the wind / solar output typical scenario again in the cluster where the dangerous typical scenario is located, until no new dangerous scenario appears, so as to traverse all dangerous scenarios in the wind / solar power generation space. Finally, the unevaluated wind / solar output scenarios are regarded as safe scenarios, and the results are statistically analyzed, and the results of risk assessment are finally given, thereby eliminating the redundant calculation of a large number of safe scenarios, realizing the division of wind / solar output safety domains and rapid warning of power supply risks.

[0170] Therefore, the present invention provides a power supply risk warning method, which eliminates the redundant calculation of a large number of safety scenarios through the precise selection and accurate traversal of wind / solar output extreme value scenarios and wind / solar output typical scenarios, and simultaneously realizes the division of wind / solar output safety domains and rapid warning of power supply risks; the power supply risk warning method provided by the present invention can comprehensively cover wind / solar output dangerous scenarios, and the warning results have a more significant reference value.

[0171] Exemplary Devices

[0172] Figure 4 1 is a schematic diagram of a power supply risk warning device based on a wind and solar power output safety domain provided by an exemplary embodiment of the present invention. Figure 4 As shown, the device 400 includes:

[0173] The cluster analysis module 410 is used to perform cluster analysis on the historical data of the wind and solar power output scene using a deep embedding clustering method to determine the clustering results;

[0174] A determination module 420 is used to determine a typical scene set of wind and solar power output scenes according to the clustering results, and to determine a wind and solar power output extreme value scene according to historical data;

[0175] The first assessment module 430 is used to perform risk quantification assessment on the wind and solar power output extreme value scenario according to a pre-established new energy power supply risk quantification assessment model, and determine the risk quantification result of the wind and solar power output extreme value scenario;

[0176] The second evaluation module 440 is used to find adjacent scenes of the dangerous scene in the typical scene set when the risk quantification result is a dangerous scene, and evaluate the warning result of the wind and solar power output scene according to the adjacent scenes;

[0177] The determination module 450 is used to determine that the wind and solar power output scenario has no operating risk when the risk quantification result is a dangerous scenario.

[0178] Optionally, the deep embedding clustering method includes: stacking autoencoders and improved K-means algorithm, and the clustering analysis module 410 includes:

[0179] An encoding submodule, for encoding the historical data using a stacked autoencoder to determine a low-dimensional feature representation of the historical data;

[0180] The analysis submodule is used to perform cluster analysis on low-dimensional feature representation using an improved K-means algorithm to determine the clustering results.

[0181] Optionally, the stacked autoencoder is trained using a layered pre-training method, and the specific steps are:

[0182] Step 1: Initially train the bottom layer of the encoder and the top layer of the decoder using a single-layer autoencoder;

[0183] Step 2: Use the output of the initially trained encoder as input to train the next pair of encoders and decoders, repeat 1 to complete all training processes, and obtain a stacked autoencoder.

[0184] Optionally, the network cost function for the stacked autoencoder is:

[0185]

[0186] In the formula, f(x i ) represents the hidden layer encoding vector, g(f(x i)) is the output vector of the DEC network, l is the reconstruction loss function; Ms i For cluster class s i The cluster center of , λ is the L2 regularization factor.

[0187] Optionally, the improved K-means algorithm is obtained by improving the K-means algorithm based on the density clustering principle, and the specific training steps are:

[0188] The elbow method is used to determine the optimal clustering number k of the sample set;

[0189] In a given sample set X = {x1, x2, ..., x n}, calculate the Euclidean distance d(x i , x j ), the expression is:

[0190]

[0191] Set the distance threshold d t , and according to the Euclidean distance d(x i , x j ) Calculate the distance threshold d t The local density ρ i , the expression is:

[0192]

[0193] Select the data with the largest local density as the first initial cluster center m1, select the data with the farthest distance from m1 and the highest density as the second initial cluster center m2 according to the Euclidean distance, select the data with the relatively farthest distance from m1 and the highest density as the third initial cluster center m3, and so on, and you can get k initial cluster centers (m1, m2, ..., m k ).

[0194] Optionally, the core indicator of the elbow method is the sum of squared errors S SE , the error sum of squares S SE The calculation formula is:

[0195]

[0196] In the formula, C i is the i-th cluster; p is C i Sample points in m i It is C i The cluster center of .

[0197] Optionally, the determination module 420 determines a typical scene set of wind and solar power output scenes according to the clustering result, including:

[0198] Taking the cluster center of the clustering result as the benchmark, the sample with the closest distance to the cluster center in each cluster sample set is selected as the representative scene of the cluster, which is used as the typical scene set.

[0199] Optionally, the objective function of the new energy power supply risk quantitative assessment model is:

[0200]

[0201] Where t is the simulation time step, P w (t,n) is the wind power generation power in region n during period t, P pv (t,n) is the power generation of photovoltaic power generation in area n during period t, e L,Curt (t,n) is the insufficient power supply power of area n at time t, and λ is the insufficient power supply penalty term;

[0202] The system power balance constraints are as follows:

[0203]

[0204] Where P G (t,n) is the total output of the n thermal power units in the region at time t, P H (t,n) is the total output of regional n hydropower units at time t, P CX (t,n) is the planned output of pumped storage unit n in region at time t, P CN (t,n) is the planned output of energy storage in area n at time t, L i (t) is the transmission power of the i-th section transmission line at time t;

[0205] The system backup requirement constraints are as follows:

[0206]

[0207] Where P re (t,n) is the positive reserve of area n at time t, N re (t,n) is the negative reserve of region n at time t;

[0208] The thermal power output range and climbing constraints are as follows:

[0209] P G,min (t,n)≤P G (t,n)≤P G,max (t,n)

[0210] P G,down ≤P G (t,n)-P G (t-1,n)≤P G,up

[0211] Where PG,min (t,n) is the total minimum technical output of n thermal power units in the region at time t, P G,max (t,n) is the total maximum technical output of n thermal power units in the region at time t; P G,down is the downward climbing amount of the thermal power unit, P G,up is the upward climbing amount of the thermal power unit;

[0212] The output constraints of new energy are as follows:

[0213]

[0214] In the formula, The predicted output of wind power at the moment, The predicted output of photovoltaic power generation at any moment;

[0215] The inter-area line transmission constraints are as follows:

[0216] L i,min ≤L i (t)≤L i,max

[0217] Where, L i,max , L i,min are the upper and lower limits of the transmission capacity of the ith transmission line, respectively. The reference direction of power is set as follows: the inflow area is the positive direction, and the outflow area is the negative direction.

[0218] Optionally, the first evaluation module 430 includes:

[0219] The statistical submodule is used to calculate the power supply shortage, wind and solar power abandonment and cross-section transmission power for the wind and solar power extreme output scenario according to the new energy power supply risk quantitative assessment model, and to calculate the amplitude and probability of power supply shortage, the amplitude and probability of wind and solar power abandonment, and the amplitude and probability of cross-section over-limit for the wind and solar power extreme output scenario;

[0220] The determination submodule is used to determine the risk quantification results of the wind and solar power output extreme scenario based on the amplitude and probability of power shortage, the amplitude and probability of wind and solar power abandonment, and the amplitude and probability of section exceeding the limit.

[0221] Optionally, the magnitude of the underpowered and probability The calculation formula is:

[0222]

[0223] In the formula, The upper and lower limits of the conventional unit output, PD t , P l,t For loads and tie lines, To reserve for backup, Contribute to the new energy in this scenario;

[0224] The magnitude of curtailed wind and solar power and probability The calculation formula is:

[0225]

[0226] In the formula, N is the number of scenarios where insufficient backup occurs. s is the total number of scenes;

[0227] The calculation formula for the amplitude and probability of cross-section exceeding the limit is:

[0228]

[0229] In the formula, and is the maximum over-limit amplitude caused by partition i to the section between partitions i and j at time t, is the planned net output power of the section between partition i and other partitions except partition j at time t; is the number of scenarios where the cross-section between partitions i and j exceeds the limit at time t, N s is the total number of scenes; They are respectively the probability of the section between partitions i and j exceeding the limit when the positive and negative reserves in partition i are insufficient at time t.

[0230] Optionally, submodules are determined, including:

[0231] Determine the power supply shortage risk based on the magnitude and probability of power supply shortage and power supply shortage risk criteria;

[0232] Determine the risk of wind and solar power abandonment based on the magnitude and probability of wind and solar power abandonment and the risk criterion of wind and solar power abandonment;

[0233] Determine the cross-section risk based on the cross-section amplitude and probability and cross-section risk criterion;

[0234] Based on the risks of insufficient power supply, wind and solar power abandonment, and section over-limit risks, the risk quantification results of the wind and solar power output extreme scenarios are determined.

[0235] Optionally, the warning result of the wind and solar power output scenario is evaluated based on adjacent scenarios, including:

[0236] Perform risk assessment calculations on adjacent scenes that have not been assessed. If a new dangerous scene is found, record the category cluster number i of the clustering result where the new dangerous scene is located, perform risk quantitative assessment on all wind and solar power output scenes in the i-th cluster, and further search for dangerous scenes in the i-th cluster;

[0237] If no new dangerous scenarios are found, the unevaluated wind-solar output scenarios are regarded as safe scenarios, and the unevaluated wind-solar output scenarios and wind-solar output safety scenarios constitute the wind-solar output safety domain;

[0238] Conduct statistical analysis on the wind and solar power output safety domain, calculate the risk level, and determine the early warning results.

[0239] Exemplary Electronic Devices

[0240] Figure 5 This is a structure of an electronic device provided by an exemplary embodiment of the present invention. Figure 5 As shown, the electronic device 50 includes one or more processors 51 and a memory 52 .

[0241] The processor 51 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0242] The memory 52 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 51 may run the program instructions to implement the methods of the software programs of the various embodiments of the present invention described above and / or other desired functions. In one example, the electronic device may also include: an input device 53 and an output device 54, which are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0243] In addition, the input device 53 may also include, for example, a keyboard, a mouse, etc.

[0244] The output device 54 can output various information to the outside. The output device 54 can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto.

[0245] Of course, to simplify, Figure 5 Only some of the components related to the present invention in the electronic device are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, the electronic device may further include any other appropriate components according to specific application conditions.

[0246] Exemplary computer program products and computer-readable storage media

[0247] In addition to the above-mentioned methods and devices, an embodiment of the present invention may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the method according to various embodiments of the present invention described in the above-mentioned "Exemplary Method" section of this specification.

[0248] The computer program product may be written in any combination of one or more programming languages ​​to write program code for performing the operations of the embodiments of the present invention, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0249] In addition, an embodiment of the present invention may also be a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, enable the processor to execute the steps of the method according to various embodiments of the present invention described in the above “Exemplary Method” section of this specification.

[0250] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, system or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0251] The basic principle of the present invention is described above in conjunction with specific embodiments. However, it should be pointed out that the advantages, strengths, effects, etc. mentioned in the present invention are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. must be possessed by each embodiment of the present invention. In addition, the specific details disclosed above are only for the purpose of illustration and facilitation of understanding, rather than limitation, and the above details do not limit the present invention to being implemented by adopting the above specific details.

[0252] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0253] The block diagrams of the devices, systems, equipment, and systems involved in the present invention are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, systems, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The word "such as" used here refers to the phrase "such as but not limited to", and can be used interchangeably with it.

[0254] The method and system of the present invention may be implemented in many ways. For example, the method and system of the present invention may be implemented by software, hardware, firmware or any combination of software, hardware, firmware. The above order of steps for the method is only for illustration, and the steps of the method of the present invention are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present invention may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present invention. Thus, the present invention also covers a recording medium storing a program for executing the method according to the present invention.

[0255] It should also be noted that in the system, device and method of the present invention, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. The above description of the disclosed aspects is provided to enable any technician in the field to make or use the present invention. Various modifications to these aspects are very obvious to those skilled in the art, and the general principles defined here can be applied to other aspects without departing from the scope of the present invention. Therefore, the present invention is not intended to be limited to the aspects shown here, but in accordance with the widest range consistent with the principles and novel features disclosed here.

[0256] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present invention to the forms disclosed herein. Although a number of example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A power supply risk warning method based on wind and solar power output safety domain, characterized in that: include: A deep embedding clustering method is used to cluster the historical data of wind and solar power output scenarios and determine the clustering results; Determine a typical scenario set of the wind and solar power output scenario according to the clustering result, and determine a wind and solar power output extreme value scenario according to the historical data; Performing a risk quantitative assessment on the wind and solar power output extreme value scenario according to a pre-established new energy power supply risk quantitative assessment model, and determining a risk quantification result of the wind and solar power output extreme value scenario; When the risk quantification result is a dangerous scene, searching for adjacent scenes of the dangerous scene in the typical scene set, and evaluating the early warning result of the wind and solar power output scene according to the adjacent scenes; When the risk quantification result is a dangerous scenario, it is determined that the wind and solar power output scenario has no operating risk.

2. The method according to claim 1, characterized in that The deep embedding clustering method includes: stacking autoencoders and improved K-means algorithm, and using the deep embedding clustering method to perform clustering analysis on historical data of wind and solar power output scenes to determine clustering results, including: Encoding the historical data using the stacked autoencoder to determine a low-dimensional feature representation of the historical data; The improved K-means algorithm is used to perform cluster analysis on the low-dimensional feature representation to determine the clustering result.

3. The method according to claim 2, characterized in that The stacked autoencoder is trained using a layered pre-training method, and the specific steps are as follows: Step 1: Initially train the bottom layer of the encoder and the top layer of the decoder using a single-layer autoencoder; Step 2: Use the output of the initially trained encoder as input to train the next pair of encoders and decoders, repeat 1 to complete all training processes, and obtain the stacked autoencoder.

4. The method according to claim 2, characterized in that: The network cost function of the stacked autoencoder is: In the formula, f(x i ) represents the hidden layer encoding vector, g(f(x i )) is the output vector of the DEC network, l is the reconstruction loss function; Ms i For cluster class s i The cluster center of , λ is the L2 regularization factor.

5. The method according to claim 2, characterized in that: The improved K-means algorithm is obtained by improving the K-means algorithm based on the density clustering principle. The specific training steps are: The elbow method is used to determine the optimal clustering number k of the sample set; In a given sample set X = {x1, x2, ..., x n }, calculate the Euclidean distance d(x i , x j ), the expression is: Set the distance threshold d t , and according to the Euclidean distance d(x i , x j ) Calculate the distance threshold d t The local density ρ i , the expression is: Select the data with the largest local density as the first initial cluster center m1, select the data with the farthest distance from m1 and the highest density as the second initial cluster center m2 according to the Euclidean distance, select the data with the relatively farthest distance from m1 and the highest density as the third initial cluster center m3, and so on, and you can get k initial cluster centers (m1, m2, ..., m k ).

6. The method according to claim 5, characterized in that The core indicator of the elbow method is the error square sum S SE , the error sum of squares S SE The calculation formula is: In the formula, C i is the i-th cluster; p is C i Sample points in m i It is C i The cluster center of .

7. The method according to claim 1, characterized in that Determining a typical scene set of the wind and solar power output scene according to the clustering result includes: Taking the cluster center of the clustering result as a benchmark, the sample in each cluster sample set that is closest to the cluster center is selected as the representative scene of the cluster, which is used as the typical scene set.

8. The method according to claim 1, characterized in that: The objective function of the quantitative assessment model for new energy power supply risk is: Where t is the simulation time step, P w (t,n) is the wind power generation power in region n during period t, P pv (t,n) is the power generation of photovoltaic power generation in area n during period t, e L,Curt (t,n) is the insufficient power supply power of area n at time t, and λ is the insufficient power supply penalty term; The system power balance constraints are as follows: Where P G (t,n) is the total output of the n thermal power units in the region at time t, P H (t,n) is the total output of regional n hydropower units at time t, P CX (t,n) is the planned output of pumped storage unit n in region at time t, P CN (t,n) is the planned output of energy storage in area n at time t, L i (t) is the transmission power of the i-th section transmission line at time t; The system backup requirement constraints are as follows: Where P re (t,n) is the positive reserve of area n at time t, N re (t,n) is the negative reserve of region n at time t; The thermal power output range and climbing constraints are as follows: P G,min (t,n)≤P G (t,n)≤P G,max (t,n) P G,down ≤P G (t,n)-P G (t-1,n)≤P G,up Where P G，min (t,n) is the total minimum technical output of n thermal power units in the region at time t, P G，max (t,n) is the total maximum technical output of n thermal power units in the region at time t; P G,down is the downward climbing amount of the thermal power unit, P G,up is the upward climbing amount of the thermal power unit; The output constraints of new energy are as follows: In the formula, The predicted output of wind power at the moment, The predicted output of photovoltaic power generation at any moment; The inter-area line transmission constraints are as follows: Where, L i,max , L i,min are the upper and lower limits of the transmission capacity of the ith transmission line, respectively. The reference direction of power is set as follows: the inflow area is the positive direction, and the outflow area is the negative direction.

9. The method according to claim 1, characterized in that: A risk quantitative assessment is performed on the wind and solar power output extreme value scenario according to a pre-established new energy power supply risk quantitative assessment model to determine the risk quantification result of the wind and solar power output extreme value scenario, including: According to the new energy power supply risk quantitative assessment model, power shortage, wind and solar power abandonment and cross-section transmission power are calculated for the wind and solar power output extreme value scenario, and the amplitude and probability of power shortage, the amplitude and probability of wind and solar power abandonment and the amplitude and probability of cross-section over-limit of the wind and solar power output extreme value scenario are counted; The risk quantification result of the wind and solar power output extreme scenario is determined based on the amplitude and probability of power shortage, the amplitude and probability of wind and solar power abandonment, and the amplitude and probability of section exceeding the limit.

10. The method according to claim 9, characterized in that The magnitude of the underpowered and probability The calculation formula is: In the formula, The upper and lower limits of the conventional unit output, PD t , P l,t For loads and tie lines, To reserve for backup, Contribute to the new energy in this scenario; The magnitude of the abandoned wind and solar power and probability The calculation formula is: In the formula, N is the number of scenarios where insufficient backup occurs. s is the total number of scenes; The calculation formula for the amplitude and probability of the cross-section exceeding the limit is: In the formula, and is the maximum over-limit value caused by partition i to the section between partitions i and j at time t, is the planned net output power of the section between partition i and other partitions except partition j at time t; is the number of scenarios where the cross-section between partitions i and j exceeds the limit at time t, N s is the total number of scenes; They are respectively the probability of the section between partitions i and j exceeding the limit when the positive and negative reserves in partition i are insufficient at time t.

11. The method according to claim 9, characterized in that According to the magnitude and probability of insufficient power supply, the magnitude and probability of wind and solar power abandonment, and the magnitude and probability of cross-section limit violation, the risk quantification result of the wind and solar power output extreme value scenario is determined, including: Determining the power supply shortage risk according to the magnitude and probability of the power supply shortage and the power supply shortage risk criterion; Determine the risk of wind and solar power abandonment according to the magnitude and probability of wind and solar power abandonment and wind and solar power abandonment risk criterion; Determine the cross-section risk according to the cross-section amplitude and probability and cross-section risk criterion; The risk quantification result of the wind and solar power output extreme scenario is determined based on the power supply shortage risk, the wind and solar power abandonment risk and the section over-limit risk.

12. The method according to claim 1, characterized in that The early warning result of the wind and solar power output scenario is evaluated according to the adjacent scenario, including: Perform risk assessment calculations on the adjacent scenes that have not been assessed. If a new dangerous scene is found, record the category cluster number i of the clustering result where the new dangerous scene is located, perform risk quantitative assessment on all wind and solar power output scenes in the i-th cluster, and further search for dangerous scenes in the i-th cluster; If no new dangerous scenarios are found, the unevaluated wind-solar output scenarios are regarded as safe scenarios, and the unevaluated wind-solar output scenarios and wind-solar output safety scenarios constitute the wind-solar output safety domain; Perform statistical analysis on the wind and solar power output safety domain, calculate the risk level, and determine the warning result.

13. A power supply risk warning device based on wind and solar power output safety domain, characterized in that: include: The cluster analysis module is used to perform cluster analysis on the historical data of wind and solar power output scenarios using a deep embedding clustering method to determine the clustering results; A determination module, used to determine a typical scene set of the wind and solar power output scene according to the clustering result, and determine a wind and solar power output extreme value scene according to the historical data; A first assessment module is used to perform a risk quantitative assessment on the wind and solar power output extreme value scenario according to a pre-established new energy power supply risk quantitative assessment model, and determine a risk quantification result of the wind and solar power output extreme value scenario; A second evaluation module is used for searching adjacent scenes of the dangerous scene in the typical scene set when the risk quantification result is a dangerous scene, and evaluating the early warning result of the wind and solar power output scene according to the adjacent scenes; A determination module is used to determine that the wind and solar power output scenario has no operating risk when the risk quantification result is a dangerous scenario.

14. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and the computer program is used to execute the method according to any one of claims 1 to 12.

15. An electronic device, characterized in that: The electronic device comprises: processor; a memory for storing instructions executable by the processor; The processor is used to read the executable instructions from the memory and execute the instructions to implement the method described in any one of claims 1-12.

Images