Line breakage probability prediction and future state scheduling effect analysis method and system based on graph neural network, and product

Abstract

The invention discloses a line breakage probability prediction and future state scheduling effect analysis method, system and product based on a graph neural network, and belongs to the technical field of power line systems, and the method comprises the steps: obtaining historical region climate data and first tower body feature data; constructing a tower local climate data prediction model; predicting tower local climate data by using the tower local climate data prediction model; second tower body feature data and tower operation and maintenance information are obtained, a tower falling probability prediction model is constructed, and the probability of falling of each tower is obtained based on the tower falling probability prediction model; determining the line breakage probability of the whole line; and constructing a digital effect analysis model of the future state scheduling of the power grid in the extreme weather, and analyzing the future state scheduling effect of the power grid in the extreme weather by analyzing the loss of the future state scheduling. According to the method, the fault prediction capability of the power system under the extreme climate can be improved, and the blank of future state scheduling effect analysis of the power grid is filled.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and particularly to a method, system and product for predicting the probability of line breakage and analyzing the effect of future-state scheduling based on a graph neural network. Background Art

[0002] With the frequent occurrence of extreme climate events, disastrous weather such as typhoons and heavy rains poses a serious threat to the safe operation of power systems. In the existing power grid operation and scheduling systems, risk assessment usually relies on regional meteorological data provided by meteorological departments, and a small number of meteorological monitoring devices deployed at key transmission towers are used for local environmental perception and response. However, due to the wide geographical distribution of transmission towers and insufficient sensor coverage, it is often difficult to obtain the microclimate data around each transmission tower in real time, resulting in a large deviation in the risk judgment of the scheduling model. The existing technology estimates the probability of a transmission line failure based on regional meteorological data and historical fault statistics information, uses the regional meteorological data as input, and outputs the probability of line breakage of the entire line. It does not consider the spatial difference of meteorological conditions between transmission towers, and the prediction accuracy is limited, making it impossible to finely control the risk source. After completing the fault risk prediction, the power grid can perform advance scheduling to achieve "future-state scheduling", but currently there is a lack of a systematic method to evaluate the effect of digital measures for future-state scheduling, and it is impossible to quantitatively analyze the improvement of scheduling strategies and the reduction of losses before and after the implementation of future-state scheduling, which restricts the engineering promotion and optimized application of the scheme. Summary of the Invention

[0003] The purpose of this application is to overcome the defects of the existing technology and provide a method, system and product for predicting the probability of line breakage and analyzing the effect of future-state scheduling based on a graph neural network, so as to solve the problems of insufficient prediction accuracy of the probability of transmission line breakage under extreme climate conditions and the inability to quantitatively evaluate the effect of future-state scheduling, and improve the ability of the power grid to cope with extreme climate.

[0004] In the first aspect, this application provides a method for predicting the probability of line breakage and analyzing the effect of future-state scheduling based on a graph neural network, including the following steps: Obtain historical regional climate data and first transmission tower body characteristic data; Construct a transmission tower local climate data prediction model according to the historical regional climate data and the first transmission tower body characteristic data; Use the transmission tower local climate data prediction model to predict the key meteorological indicators of the transmission tower and obtain the transmission tower local climate data; Obtain second transmission tower body characteristic data and transmission tower operation and maintenance information, construct a transmission tower collapse probability prediction model according to the second transmission tower body characteristic data, the transmission tower operation and maintenance information, and the transmission tower local climate data, and obtain the probability of each transmission tower collapsing based on the transmission tower collapse probability prediction model; Based on the probability of tower collapse for each tower, determine the probability of conductor breakage for the entire transmission line through a probability combination model; Construct a digital effect analysis model for future-state dispatching of the power grid under extreme climate, and analyze the losses under future-state dispatching based on the digital effect analysis model for future-state dispatching of the power grid under extreme climate, so as to realize the analysis of the effect of future-state dispatching of the power grid under extreme weather.

[0005] Optionally, construct a tower local climate data prediction model according to the historical regional climate data and the first tower body characteristic data, including: Obtain the actual local climate data of the tower, set the actual local climate data of the tower as the first target output data, combine the historical regional climate data and the first tower body characteristic data to obtain the first input data; Preprocess the first input data and the first target output data; Convert the preprocessed first input data into a graph structure; Construct a first graph neural network; Use the first input data in graph structure and the preprocessed first target output data to train the first graph neural network, and use the first loss function to optimize and update the parameters to obtain the tower local climate data prediction model.

[0006] Optionally, use the tower local climate data prediction model to predict the key meteorological indicators of the tower, and the predicted value of the local climate data of the tower at time t The expression is: Where is the adjacency matrix, is the first input feature matrix, represents the tower local climate data prediction model.

[0007] Optionally, obtain the second tower body characteristic data and the tower operation and maintenance information, construct a tower collapse probability prediction model according to the second tower body characteristic data, the tower operation and maintenance information, and the tower local climate data, and obtain the probability of tower collapse for each tower based on the tower collapse probability prediction model, including: Obtain the second tower body characteristic data and the tower operation and maintenance information; Combine the second tower body characteristic data, the tower operation and maintenance information, and the tower local climate data to obtain a second input data set; Preprocess the second input data set; Convert the preprocessed second input data set into a graph structure; Construct a second graph neural network; Train the second graph neural network using the second input data set with a graph structure, and optimize and update the parameters using the second loss function to obtain a tower collapse probability prediction model for power transmission towers.

[0008] Optionally, based on the probability of each power transmission tower collapsing, determine the line break probability of the entire power transmission line through a probability combination model. The total line break probability of the power transmission line can be expressed as: where represents t the predicted value of the tower collapse probability of the i th tower at time

[0009] Optionally, construct a digital effect analysis model for the future state dispatching of the power grid under extreme climate, and analyze the losses under the future state dispatching based on the digital effect analysis model for the future state dispatching of the power grid under extreme climate, so as to realize the analysis of the effect of the future state dispatching of the power grid under extreme weather, including: Construct a digital effect analysis model for the future state dispatching of the power grid under extreme climate. The digital effect analysis model for the future state dispatching of the power grid under extreme climate includes: a line break scenario generation module, an input module, a dispatching analysis module, and a dispatching effect analysis module; Input the line break probabilities of each power transmission line in the power grid into the line break scenario generation module to obtain the line break scenarios corresponding to each power transmission line in the power grid; Send the line break scenarios corresponding to all power transmission lines in the power grid and the data in the input module to the dispatching analysis module. The dispatching analysis module includes: a generator unit combination model before extreme climate occurs, a traditional dispatching model under extreme climate, and a future state dispatching model under extreme climate; Obtain the unit dispatching plan before extreme climate occurs through the generator unit combination model before extreme climate occurs in the dispatching analysis module; Obtain the traditional dispatching plan under extreme climate through the traditional dispatching model under extreme climate in the dispatching analysis module; Obtain the future state dispatching plan under extreme climate through the future state dispatching model under extreme climate in the dispatching analysis module; Input the unit dispatching plan before extreme climate occurs, the traditional dispatching plan under extreme climate, and the future state dispatching plan under extreme climate into the dispatching effect analysis module for effect analysis, and determine the power grid loss and social loss under the future state dispatching based on the line break probability of the power transmission line and the number of disasters occurring, so as to obtain the effect of the future state dispatching of the power grid under extreme climate.

[0010] Optionally, obtain the unit scheduling plan before the occurrence of extreme climate through the unit combination model before the occurrence of extreme climate in the scheduling analysis module, including: analyzing through the unit combination model before the occurrence of extreme climate in the scheduling analysis module and optimizing using the first objective function to obtain the unit scheduling plan before the occurrence of extreme climate. The first objective function is expressed as follows: where is the output bid function of the i th unit; is the start-up cost of the i th unit; is the no-load operation cost of the i th unit; is the power generation cost of the ne th new energy unit; is whether the i th unit t starts in the t period. If it starts in the period, it is 1; otherwise, it is 0; i is whether the t th unit t shuts down in the period. If it shuts down in the i period, it is 1; otherwise, it is 0; t is whether the th unit is put into operation in the i period. If it is put into operation, it is 1; if not, it is 0; t is the output of the th unit in the t period; is the actual scheduled output of the new energy unit in the t period; d is the th load value in the s th section in the t period; is the maximum transmission capacity of the line with the head and tail nodes of nm ; is the vector form of the minimum output of the i th unit; is the vector form of the output of the i th unit in the t period; is the vector form of the output of the i th unit in the t-1 period; is the iThe vector form of the maximum output of each unit; is the ne vector form of the output of the t th new energy unit during the time period; ne is the t vector form of the maximum predicted available output of the th new energy unit during the time period; i is the set of conventional generating units; is the downward ramping rate limit of the i th unit; is the upward ramping rate limit of the n th unit; is the transmission line where the th node is located; n is the set of all transmission lines in the power grid; t is the node voltage phase angle of the th node during the m time period; t is the node voltage phase angle of the th node during the t time period; nm is the line with the start and end nodes of T during the time period; i is the starting time period of the th unit; i is the shutdown time period of the NE th unit; is the set of new energy generating units; n is the new energy generating unit of the D th node;

[0011] Optionally, the traditional scheduling plan under extreme climate is obtained through the traditional scheduling model under extreme climate in the scheduling analysis module, including: analyzing through the traditional scheduling model under extreme climate in the scheduling analysis module and optimizing with the second objective function to obtain the traditional scheduling plan under extreme climate, and the second objective function , and the expression is as follows: where is the second constraint condition when the l th line fails; is the output bidding function of the i th unit; is the j th unit with short-term rapid supply capacity during the tPeriod output quotation function; is the power generation cost of the ne th new energy unit; is the output of the j th unit with short-term rapid supply capacity in the t period; is the output of the i th unit in the t period; is the actual dispatching output of the new energy unit in the t period; is the t th load value in the d period; is the s th section's required transmitted or received power in the t period; is the vector form of the minimum output of the i th unit; is the vector form of the output of the i th unit in the t period; is the vector form of the output of the i th unit in the t-1 period; is the vector form of the maximum output of the i th unit; is the vector form of the output of the ne th new energy unit in the t period; is the vector form of the maximum predicted output of the ne th new energy unit in the t period; is the maximum transmission capacity of the line with the head and tail nodes of nm ; is the set of conventional generating units; is the set of standby capacity units with short call time; is the n th set of conventional generating units at the th node; i is the downward ramp rate limit of the th unit; i is the upward ramp rate limit of the th unit; n is the node voltage phase angle of the t th node in the period; m is the node voltage phase angle of the t th node in the is the t period with the head and tail nodes of the line beingnm circuit; T is the total period of the traditional scheduling plan under extreme climate; is the l th line set excluding the l th line after the

[0012] Optionally, obtain the future state scheduling plan under extreme climate through the future state scheduling model under extreme climate in the scheduling analysis module, including: analyze through the future state scheduling model under extreme climate in the scheduling analysis module, and optimize using the third objective function to obtain the future state scheduling plan under extreme climate, and the third objective function has the following expression: where is the output bidding function of the i th unit; is the output bidding function of the j th unit with short-term rapid supply capacity at t period; is the power generation cost of the ne th new energy unit; is the output bidding function of the k th slow standby unit at t period; is the output of the i th unit at t period; is the output of the j th unit with short-term rapid supply capacity at t period; is the output of the k th slow standby unit at t period; is the actual scheduled output of the new energy unit at t period; is t period, the d th load value; is the s th section at t period, the required power to be sent or received; is the vector form of the minimum output of the i th unit; is the vector form of the output of the i th unit at t period; is the vector form of the output of the i th unit at t-1 period; is thei The vector form of the maximum output of each unit; is the ne vector form of the output of the t th new energy unit during the is the ne vector form of the maximum predicted available output of the t th new energy unit during the is the ne minimum value of the actual and predicted historical output of the is the ne true value of the historical output of the is the ne predicted value of the historical output of the is the maximum transmission capacity of the line with the head and tail nodes of nm ; is the i downward ramping rate limit of the is the i upward ramping rate limit of the is the n phase angle of the node voltage at the t th node during the is the m phase angle of the node voltage at the t th node during the is t the line with the head and tail nodes of nm during the T is the total time period of the future state scheduling plan under extreme climate; NE is the set of new energy generating units; is the n new energy generating unit at the is the l set of all other lines except the l th line after the line fault; is the set of conventional generating units; is the set of spare capacity units with short call time; is the set of slow spare capacity units with longer call time; is the n set of conventional generating units at the G is G 1 、 G 2 、 G 3 the intersection of the three sets; D is the set of node loads in the power grid; Dn is the load of the n th node.

[0013] Optionally, the power grid loss under the future state scheduling expression is: where F is the annual occurrence times of extreme climate; N is the total number of fault types under a certain extreme climate; T is the total time period of the traditional scheduling scheme under extreme climate; a is the fault type under a certain extreme climate; is the a th fault probability of the fault type; is the standby unit cost function; is the a th required standby unit output under the t time period for the th fault type, given by the scheduling analysis module; a is the t th power supply shortage of the power system within the time period for the th fault type; is the standby capacity of the power grid; is the unit power outage compensation coefficient; where F is the annual occurrence times of extreme climate; N is the total number of fault types under a certain extreme climate; T is the total time period of the traditional scheduling scheme under extreme climate; a is the fault type under a certain extreme climate; is the a th fault probability of the fault type; is the a th required standby unit output under the t time period for the th fault type, given by the scheduling analysis module; a is the t th power supply shortage of the power system within the time period for the

[0014] In a second aspect, the present application further provides a system for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network, which is used to execute the method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network as described in any one of the first aspects, including: one or more processors; a storage device for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network as described in any one of the first aspects.

[0015] In a third aspect, the present application further provides a computer program product, including a computer program, which when executed by a processor, implements the method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network as described in any one of the first aspects.

[0016] The present application provides a method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network. By constructing a prediction model for the local climate data of the pole tower, key meteorological indicators are accurately estimated, making the meteorological influencing factors quantifiable, and accurately depicting the microclimate environment where a single pole tower is located, providing high-precision input data for subsequent calculation of the tower collapse probability; by constructing a prediction model for the tower collapse probability of the pole tower, the limitation of a single factor is broken through, making the calculation of the tower collapse probability of the pole tower more in line with the actual situation; through the probability combination model, the overall line disconnection probability is deduced from the individual probabilities of the pole towers, which can capture the individual differences in the structure and maintenance status of a single tower, and realize a scientific evaluation from the micro to the macro; by constructing a prediction model for the tower collapse probability of the pole tower, it is possible to output the disconnection probability of the entire line with high precision only by inputting regional-level meteorological data, taking into account both practicality and prediction accuracy, and significantly improving the fault prediction ability of the power transmission system under extreme climates; by constructing a digital effect analysis model for the future state scheduling of the power grid under extreme climates, the advantages of the future state scheduling in terms of emergency response efficiency and economic cost are quantified, filling the gap in the prior art for the lack of analysis of the future state scheduling effect of the power grid under extreme climate conditions, quantifying the impact of the future state scheduling on power grid losses and social losses, and providing a scientific decision-making basis for whether the power grid invests in future state scheduling, promoting the power grid scheduling under extreme climates to shift from passive response to active optimization. The method of the present application systematically improves the practicality and accuracy of power grid risk prediction and scheduling decision-making under extreme climates through hierarchical modeling and digital evaluation, and realizes a closed-loop from disconnection risk prediction to scheduling effect evaluation through data-driven and model-supported means, providing a strong basis for power grid operation and maintenance personnel to formulate protection strategies and optimize scheduling plans in advance, improving the power grid's ability to cope with extreme climates, and effectively reducing the harm of extreme weather to the operation of the power grid.

[0017] To make the above features and advantages of the invention more obvious and understandable, specific embodiments are hereinafter given and described in detail in conjunction with the accompanying drawings as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 It is a flowchart of a method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network provided in an embodiment of the present application.

[0020] Figure 2 It is a flowchart of step S2 in the method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network provided in an embodiment of the present application.

[0021] Figure 3 It is a flowchart of step S4 in the method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network provided in an embodiment of the present application.

[0022] Figure 4 It is a flowchart of step S6 in the method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network provided in an embodiment of the present application.

[0023] Figure 5 It is a structural diagram of a digital effect analysis model for future state scheduling of a power grid under extreme climate in the method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network provided in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] To make the objectives and technical solutions of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions of the embodiments of the present application with reference to the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0025] In one embodiment, please refer to Figure 1 , the present application provides a method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network. The method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network may include the following steps: step S1 to step S6.

[0026] Step S1: Obtain historical regional climate data and the first tower body characteristic data.

[0027] Step S2: Construct a prediction model for the local climate data of the tower based on the historical regional climate data and the first tower body characteristic data.

[0028] Step S3: Use the prediction model for the local climate data of the tower to predict the key meteorological indicators of the tower and obtain the local climate data of the tower.

[0029] Step S4: Obtain the second tower body characteristic data and the tower operation and maintenance information, construct a prediction model for the tower collapse probability based on the second tower body characteristic data, the tower operation and maintenance information, and the local climate data of the tower, and obtain the probability of each tower collapsing based on the prediction model for the tower collapse probability.

[0030] Step S5: Based on the probability of each tower collapsing, determine the probability of line breakage of the entire transmission line through a probability combination model.

[0031] Step S6: Construct a digital effect analysis model for the future state dispatching of the power grid under extreme climates, and analyze the losses under the future state dispatching based on the digital effect analysis model for the future state dispatching of the power grid under extreme climates, so as to realize the analysis of the effect of the future state dispatching of the power grid under extreme weather.

[0032] In the method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on graph neural network of the present application, by constructing a prediction model of local climate data of the pole tower, the key meteorological indicators are accurately predicted, the meteorological influence factors can be quantified, and the microclimate environment where a single pole tower is located is accurately characterized, providing high-precision input data for subsequent calculation of the tower collapse probability; by constructing a prediction model of the tower collapse probability of the pole tower, the limitation of a single factor is broken through, and the calculation of the tower collapse probability of the pole tower is more in line with the actual situation; through the probability combination model, the overall line disconnection probability is deduced from the individual probabilities of the pole towers, which can capture the individual differences in the structure and maintenance status of a single tower, and realize the scientific evaluation from micro to macro; by constructing a prediction model of the tower collapse probability of the pole tower, it is possible to output the disconnection probability of the entire line with high precision only by inputting regional-level meteorological data, taking into account both practicality and prediction accuracy, and significantly improving the fault prediction ability of the power transmission system under extreme climate; by constructing a digital effect analysis model of the future state scheduling of the power grid under extreme climate, the advantages of the future state scheduling in emergency response efficiency and economic cost are quantified, filling the gap in the prior art that lacks the analysis of the future state scheduling effect of the power grid under extreme climate conditions, quantifying the impact of the future state scheduling on the power grid loss and social loss, providing a scientific decision-making basis for whether the power grid invests in the future state scheduling, and promoting the power grid scheduling under extreme climate to shift from passive response to active optimization. The method of the present application systematically improves the practicality and accuracy of power grid risk prediction and scheduling decision-making under extreme climate through hierarchical modeling and digital evaluation, and realizes a closed-loop from disconnection risk prediction to scheduling effect evaluation through data-driven and model support, providing a strong basis for power grid operation and maintenance personnel to formulate protection strategies and optimize scheduling plans in advance, and effectively reducing the harm of extreme weather to the operation of the power grid.

[0033] In step S1, please refer to Figure 1 step S1 in

[0034] As an example, the historical regional climate data of the area where the power grid is located can be obtained from the database or data interface of the meteorological department , where , t is time, d is the dimension of regional climate characteristics, g represents near the g th area.

[0035] As an example, the historical regional climate data may include: regional wind speed, regional wind direction, regional rainfall, regional temperature, regional humidity, and date and time, etc.

[0036] As an example, the first pole tower body characteristic data can be obtained through geographic information system (GIS) technology combined with the design documents and construction records during the construction of the pole tower , where , m is the dimensionality of the characteristic data of the pole tower body, i represents the i th pole tower.

[0037] As an example, the first pole tower body characteristic data represents the characteristics of the area where the pole tower is located, the pole tower structure, etc., and may include: information such as the location of the pole tower, the height of the pole tower, and the altitude of the pole tower.

[0038] In step S2, please refer to Figure 1 in step S2, and construct a pole tower local climate data prediction model according to the historical regional climate data and the first pole tower body characteristic data.

[0039] As an example, please refer to Figure 2 , step S2 may include the following steps: step S21 to step S25.

[0040] Step S21: Obtain the actual local climate data of the pole tower, set the actual local climate data of the pole tower as the first target output data, and combine the historical regional climate data and the first pole tower body characteristic data to obtain the first input data.

[0041] Step S22: Preprocess the first input data and the first target output data.

[0042] Step S23: Convert the preprocessed first input data into a graph structure.

[0043] Step S24: Construct the first graph neural network.

[0044] Step S25: Use the first input data in the graph structure and the preprocessed first target output data to train the first graph neural network, and use the first loss function to optimize and update the parameters to obtain the pole tower local climate data prediction model.

[0045] As an example, in step S21, obtain the actual local climate data of the pole tower through sensors , where represents t at the i th pole tower's actual local climate condition data, n is the dimensionality of the local climate characteristics.

[0046] As an example, the actual local climate data of the pole tower may include: the actual local wind speed of the pole tower, the actual local wind direction of the pole tower, the actual local rainfall of the pole tower, the actual local temperature of the pole tower, the actual local humidity of the pole tower, etc.

[0047] Further, the actual local climate data of the pole tower is set in matrix form to obtain the first target output data matrix , and the expression is: where represents t the actual local climate condition data of the first pole tower at time represents t the actual local climate condition data of the second pole tower at time represents t the actual local climate condition data of the Nth pole tower at time N represents the number of pole towers in a certain area, n and

[0048] is the local climate feature dimension. Further, the historical regional climate data and the first pole tower body feature data are combined. That is, the historical regional climate data i is copied to each pole tower, and the input feature vector of the th pole tower is m , where d is the feature dimension of the pole tower body feature data,

[0049] and is the regional climate feature dimension. Further, the input feature vectors of all pole towers N are combined into matrix form to obtain the first input data m , where d represents the number of pole towers in a certain area, is the feature dimension of the pole tower body feature data,

[0050] As an example, in step S22, the first input data and the first target output data are normalized, aiming to eliminate the influence of different feature scale differences on model training by standardizing or normalizing the input data, so that each feature is processed at the same scale, and the efficiency and stability of model training are improved.

[0051] As an example, the min-max normalization can be used to scale the data to a fixed interval. It can be set to compress the data into the interval [0, 1].

[0052] As an example, the min-max normalization operation is performed on the first input data set X, and the expression is as follows: Among them, is the i th input feature data after normalization, is the i th data point in the original input data set, is the data set X the minimum value in; is the data set X the maximum value in.

[0053] Furthermore, each of the normalized input feature data is combined to obtain the first input matrix , where N represents the number of poles and towers in a certain area, m is the feature dimension of the pole and tower body feature data, d is the regional climate feature dimension.

[0054] As an example, the specific method for normalizing the first target output data can refer to the specific method for normalizing the first input data in step S22, which will not be elaborated here.

[0055] As an example, in step S23, due to the spatial correlation in the geographical location between poles and towers, based on this, the poles and towers can be regarded as nodes, and the two poles and towers with a connection distance less than the threshold are used as edges to obtain an undirected graph G=(V,E) , where is the node set, representing the set of poles and towers; E is the edge set, representing the set of connections between two poles and towers with a distance less than the threshold .

[0056] As an example, the expression of the edge set E is as follows: Among them, i represents the i th pole and tower, j represents the j th pole and tower, represents the i th pole and tower and the j th pole and tower distance between.

[0057] As an example, let the coordinates of the pole and tower i be p i =(x i ,y i ) , the coordinates of the pole and tower j bep j =(x j ,y j ) , then the pole tower i and the pole tower j The Euclidean distance between them is: .

[0058] Furthermore, according to the edge set E Calculate the adjacency matrix , the expression is: Furthermore, add the identity matrix A to the adjacency matrix I to obtain a new adjacency matrix to ensure that each node can utilize its own information. The new adjacency matrix after adding self-loops, the expression is: Furthermore, normalize the new adjacency matrix after adding self-loops into the symmetric Laplacian matrix , the expression is: where is the node degree matrix of the graph, which is a diagonal matrix, and its element represents the degree of the i -th node (i.e., the number of edges connected to the i -th node). The normalized adjacency matrix can adjust the intensity of node information propagation to make the propagation more balanced.

[0059] As an example, in step S24, construct a first graph neural network, and use graph convolution to aggregate the feature information of each node and its neighbor nodes. The first graph neural network may include an input layer, a graph convolution layer, and an output layer.

[0060] As an example, it can be set that the first graph neural network has a total of L layers, and the input layer can be expressed as , where represents the node feature matrix of the input layer, is the input feature matrix.

[0061] As an example, the node feature matrix of the l -th layer is , then the node feature matrix l of the +1-th layer can be obtained through the following formula: Among them, is the node feature matrix of the l th layer; is the weight matrix of the l th layer, representing the parameters of the linear transformation, which need to be learned through training; is the normalized adjacency matrix, which is used to aggregate the information of neighboring nodes to the current node; σ is a non-linear activation function.

[0062] As an example, the non-linear activation function σ can adopt the ReLU activation function to enhance the non-linear expression ability.

[0063] As an example, stacking multiple graph convolutional layers constitutes the backbone structure of the first graph neural network. Each graph convolutional layer will expand the receptive field of information, that is, the final representation of each node can integrate the information of the adjacency matrix in a wider range. For example, the first layer can integrate the information of direct neighbors, and the second layer can integrate the information of second-order neighbors. Through multi-layer stacking, the first graph neural network can capture the broader context relationship of nodes.

[0064] As an example, the output layer can be expressed as: Among them, represents t the predicted value of the local climate of the pole tower at time , L is the node feature matrix of the th layer, that is, the node feature matrix of the last layer;

[0065] As an example, in step S25, the first graph neural network is trained using the input data of the graph structure, and the weight matrix of each layer is continuously updated to make the predicted value of the local climate of the pole tower as close as possible to the true value in the first target output, and the optimal weight matrix is determined.

[0066] Specifically, the first input data of the graph structure and the first target output data are constructed into a training sample. Among them, the first input data of the graph structure can include: the normalized adjacency matrix of the graph structure , the input feature matrix. The target output data can include: the actual local climate data matrix of the pole tower

[0067] Furthermore, the first graph neural network is trained using training samples. First, the parameters of the first graph neural network are initialized, and the first input data of the graph structure is fed into the constructed first graph neural network. The first input data of the graph structure is passed layer by layer according to the structure of the first graph neural network to extract relevant features, and the predicted value of the local climate of the pole tower is obtained. Then, the predicted value of the local climate of the pole tower is compared with the first target output data, and the loss value is calculated through the first loss function to evaluate the deviation degree between the predicted value of the local climate of the pole tower obtained by the current first graph neural network and the actual local climate of the pole tower in the first target output data. Then, it is checked whether the current loss value reaches the training target. If it reaches, it means that the prediction accuracy of the first graph neural network reaches the optimal, and the training ends; if it does not reach, it continues to judge whether the preset maximum number of training times is reached. If it reaches, the training ends; if it does not reach, backpropagation is performed. According to the obtained loss value, using the backpropagation algorithm, the gradient is calculated backward from the output layer to the input layer to determine the parameters that need to be adjusted in the first graph neural network. According to the gradient calculated by backpropagation, a suitable optimization algorithm is used to update the parameters of the first graph neural network, and then forward propagation is continued. The above training process is repeated to optimize the parameters of the first graph neural network in the direction of reducing the loss value until the training target is reached or the preset maximum number of training times is reached, and the training ends to obtain the final prediction model of the local climate data of the pole tower.

[0068] As an example, the first loss function can adopt the mean square error loss function (MSE), and the expression is: Among them, N represents the number of pole towers in a certain area, is the actual local climate data of the i th pole tower, is the predicted value of the local climate data of the i th pole tower, l represents the l th layer.

[0069] As an example, the training target is to continuously reduce the loss value of the first loss function until it approaches zero or stabilizes.

[0070] As an example, the update rule of gradient descent is: Among them, is the learning rate, is the weight matrix of the l th layer to be updated, is the weight matrix of the l th layer, is the loss value of the first loss function.

[0071] As an example, the optimization algorithm can adopt the gradient descent method, or Adaptive Moment Estimation (Adam), Stochastic Gradient Descent (SGD), etc.

[0072] As an example, the maximum number of training times can be set according to the performance of the hardware device.

[0073] It should be noted that in each iteration, the forward propagation is calculated using the current parameters (i.e., the local climate prediction value of the pole tower is obtained based on the historical regional climate data), and the loss is calculated , and the weight parameters of each layer are updated using the gradient. The iteration is repeated until the loss value converges (i.e., tends to be stable), and the model training is completed.

[0074] In step S3, please refer to Figure 1 step S3 in, and use the pole tower local climate data prediction model to predict the key meteorological indicators of the pole tower to obtain the pole tower local climate data.

[0075] As an example, using the pole tower local climate data prediction model to predict the key meteorological indicators of the pole tower to obtain the pole tower local climate data, realizing the prediction of the local key meteorological indicators of each pole tower, and providing a highly reliable data basis for the subsequent pole tower risk assessment.

[0076] As an example, the predicted value of the local climate data of the pole tower at time t output by the pole tower local climate data prediction model can be expressed as: Among them, is the adjacency matrix, is the first input feature matrix, represents the pole tower local climate data prediction model.

[0077] In step S4, please refer to Figure 1 step S4 in, obtain the second pole tower body feature data and the pole tower operation and maintenance information, construct a pole tower collapse probability prediction model based on the second pole tower body feature data, the pole tower operation and maintenance information, and the pole tower local climate data, and obtain the probability of collapse of each pole tower based on the pole tower collapse probability prediction model.

[0078] As an example, please refer to Figure 3 , step S4 may include the following steps: steps S41 to S46.

[0079] Step S41: Obtain the second pole tower body feature data and the pole tower operation and maintenance information.

[0080] Step S42: Combine the second tower body feature data, the tower operation and maintenance information, and the local climate data of the tower to obtain a second input data set.

[0081] Step S43: Preprocess the second input data set.

[0082] Step S44: Convert the preprocessed second input data set into a graph structure.

[0083] Step S45: Construct a second graph neural network.

[0084] Step S46: Use the second input data set in graph structure to train the second graph neural network, optimize and update the parameters using the second loss function, and obtain a tower collapse probability prediction model.

[0085] As an example, in step S41, the second tower body feature data can be obtained by combining geographic information system (GIS) technology with the design documents and construction records during tower construction , where , is the body feature data of the i th tower, represents the feature dimension of the tower body feature data.

[0086] As an example, the second tower body feature data may include: tower height, maximum operating life of the tower, tower structure material category, etc.

[0087] As an example, the tower operation and maintenance information can be obtained through the operation and maintenance management system of the power enterprise , where , is the operation and maintenance status data of the i th tower, represents the feature dimension of the tower operation and maintenance status data.

[0088] As an example, the tower operation and maintenance information may include: corrosion grade of the tower, time since the last maintenance, years of use of the tower, etc.

[0089] As an example, in step S42, the local climate data of the tower is obtained through the local climate data prediction model of the tower , where represents t the local climate data of the i th tower at time represents the feature dimension of the local climate data of the tower.

[0090] As an example, the local climate data of the tower It may include: local wind speed of the pole tower, local wind direction of the pole tower, local rainfall of the pole tower, local temperature of the pole tower, local humidity of the pole tower, etc.

[0091] Further, the local climate data of the pole tower , the second pole tower body feature data , and the pole tower operation and maintenance information are combined to obtain the input features of each pole tower. The expression is: where is t the input feature of the i th pole tower at time ; d 1 , d 2 , d 3 represent the union of

[0092] Further, the input features of all pole towers are combined to obtain the second input data set , and the expression is: where represents t the input feature of the first pole tower at time , t the input feature of the second pole tower at time , t the input feature of the Nth pole tower at time N , represents the number of pole towers in a certain area, and

[0093] As an example, in step S43, the second input data set is normalized to eliminate the influence of different feature scale differences on model training, so that each feature is processed at the same scale, improving the efficiency and stability of model training.

[0094] As an example, min-max normalization can be used to scale the data to a fixed interval. It can be set to compress the data into the interval [0, 1].

[0095] As an example, the min-max normalization operation is performed on the second input data set , and the expression is as follows: where is the iInput data, is the first i data points, For the second input data set The minimum value in For the second input data set The maximum value in .

[0096] As an example, the data of each feature dimension in the second input data set is normalized by minimum-maximum normalization to obtain normalized input data, and the normalized input data is combined into a second input matrix ,in, N Indicates the number of towers in a certain area. is the feature dimension of the input features of each tower.

[0097] As an example, in step S44, the specific method of converting the preprocessed second input data set into a graph structure is consistent with the specific method of converting the preprocessed first input data into a graph structure in step S23, and will not be repeated here.

[0098] As an example, in step S45, a second graph neural network model is constructed, and feature information of each node and its neighboring nodes is aggregated using graph convolution. The second graph neural network may include an input layer, a graph convolution layer, and an output layer.

[0099] As an example, the second graph neural network model can be set to have a total of L2 layers, and the input layer can be expressed as ,in, Represents the node feature matrix of the input layer of the second graph neural network model, is the second input matrix.

[0100] As an example, in the second graph neural network model l The node feature matrix of the layer is , then in the second graph neural network model l +1 layer node feature matrix It can be obtained by the following formula: in, is the first l The node feature matrix of the layer; is the first l The weight matrix of the layer represents the parameters of the linear transformation and needs to be learned through training; is the normalized adjacency matrix, which is used to aggregate the information of neighboring nodes to the current node; σ is a non-linear activation function.

[0101] As an example, a non-linear activation function σ can be adopted ReLU as the activation function to enhance the non-linear expression ability.

[0102] As an example, multiple graph convolutional layers are stacked to form the backbone structure of the second graph neural network model. Each graph convolutional layer will expand the receptive field of information, that is, the final representation of each node can integrate the information of the adjacency matrix within a wider range. For example, the first layer can integrate the information of direct neighbors, and the second layer can integrate the information of second-order neighbors. Through multi-layer stacking, the second graph neural network model can capture the broader context relationship of nodes.

[0103] As an example, the output layer of the second graph neural network model can be set to be fully connected, and the expression is: where represents t the predicted value of the tower collapse probability of the tower at time is the node feature matrix of the L 2 th layer in the second graph neural network model, that is, the node feature matrix of the last layer in the second graph neural network model; is the weight matrix of the output layer in the second graph neural network model.

[0104] As an example, in step S46, the actual tower collapse label of the tower is set according to the actual tower collapse situation of the tower, and the actual tower collapse label of the tower is used as the second target output , where is t the actual tower collapse label of the first tower at time is t the actual tower collapse label of the second tower at time is t the actual tower collapse label of the Nth tower at time.

[0105] As an example, for the actual tower collapse label i of the th tower, the value can be 0 or 1, that is .

[0106] Furthermore, the second graph neural network model is trained using the second input data set with a graph structure, and the weight matrix of each layer is continuously updated to make the predicted value of the tower collapse label as close as possible to the true value in the second target output, and the optimal weight matrix is determined.

[0107] Specifically, the second input data set and the second target output of the graph structure are constructed into training samples, where the second input data set of the graph structure is the input data in the training samples, and the second target output is the output data in the training samples. The input data in the training samples includes: the normalized adjacency matrix of the graph structure , the second input matrix . The second target output may include: the actual tower collapse label matrix of the tower .

[0108] Furthermore, the second graph neural network model is trained using the training samples. First, the parameters of the second graph neural network model are initialized, and the second input data set of the graph structure is fed into the constructed second graph neural network model. The second input data set of the graph structure is passed layer by layer according to the structure of the second graph neural network model to extract relevant features, and the predicted value of the tower collapse label is obtained. Then, the predicted value of the tower collapse label is compared with the second target output label, and the loss value is calculated through the second loss function to evaluate the deviation degree between the predicted value of the tower collapse label obtained by the current second graph neural network model and the actual tower collapse label in the second target output. Then, it is checked whether the current loss value reaches the training target. If it reaches, it means that the prediction accuracy of the second graph neural network model reaches the optimal, and the training ends; if it does not reach, it is continued to judge whether the preset maximum number of training times is reached. If it reaches, the training ends; if it does not reach, then backpropagation is performed. According to the obtained loss value, using the backpropagation algorithm, the gradient is calculated reversely from the output layer to the input layer to determine the parameters that need to be adjusted in the second graph neural network model. According to the gradient calculated by backpropagation, a suitable optimization algorithm is used to update the parameters of the second graph neural network model, and then forward propagation is continued. The above training process is repeated to optimize the parameters of the second graph neural network model in the direction of reducing the loss value until the training target is reached or the preset maximum number of training times is reached, and the training ends to obtain the final tower collapse probability prediction model.

[0109] As an example, the second loss function can adopt the binary cross-entropy loss function, and the expression is: where N represents the number of towers in a certain area, is the actual tower collapse label of the i th tower, is the predicted value of the tower collapse label of the i th tower.

[0110] As an example, the training target is to continuously reduce the loss value of the second loss function until it approaches zero or stabilizes.

[0111] As an example, the update rule of gradient descent is: Among them, is the learning rate of the neural network model in the second figure, is the weight matrix of the l th updated layer in the neural network model in the second figure, is the weight matrix of the l th layer in the neural network model in the second figure, is the loss value of the second loss function.

[0112] As an example, the optimization algorithm can adopt the gradient descent method, or adaptive moment estimation (Adam), stochastic gradient descent (SGD), etc.

[0113] As an example, the maximum number of training times can be set according to the performance of the hardware device.

[0114] Furthermore, the tower collapse probability prediction model is used to predict the tower collapse probabilities of all towers in the area, obtain the probability of each tower collapsing under extreme climate scenarios, realize the prediction of the local key meteorological indicators of each tower, and provide data support for the subsequent calculation of the line break probability.

[0115] As an example, the predicted tower collapse probability value of the tower at time t output by the tower collapse probability prediction model can be expressed as: is the adjacency matrix, is the second input matrix, represents the tower collapse probability prediction model.

[0116] In step S5, please refer to Figure 1 step S5 in

[0117] Based on the probability of each tower collapsing, determine the line break probability of the entire transmission line through the probability combination model.

[0118] As an example, each tower collapse event can be regarded as an independent event, then the probability that the entire transmission line does not have a line break is the joint probability that each tower in the entire transmission line does not collapse.

[0119] As an example, a probability combination model can be constructed through a probability modeling method, and based on the probability combination model, determine the overall line break probability of the entire transmission line, integrate the local node failure risks into a system-level transmission line line break risk index, and finally obtain the line break probabilities of each transmission line in the power grid.

[0119] As an example, the total line break probability line of the transmission line can be expressed as: Among them, represents t the predicted value of the tower collapse probability of the i nth tower at the moment.

[0120] As an example, the transmission line line includes at least one tower collapse.

[0121] In step S6, refer to Figure 1 step S6 therein, construct a digital effect analysis model for the future state dispatching of the power grid under extreme climate, analyze the losses under the future state dispatching based on the digital effect analysis model for the future state dispatching of the power grid under extreme climate, and realize the analysis of the future state dispatching effect of the power grid under extreme weather.

[0122] As an example, refer to Figure 4 , step S6 may include the following steps: step S61 to step S67.

[0123] Step S61: Construct a digital effect analysis model for the future state dispatching of the power grid under extreme climate, and the digital effect analysis model for the future state dispatching of the power grid under extreme climate includes: a broken wire scenario generation module, an input module, a dispatching analysis module, and a dispatching effect analysis module.

[0124] Step S62: Input the broken wire probabilities of each transmission line in the power grid into the broken wire scenario generation module to obtain the broken wire scenarios corresponding to each transmission line in the power grid.

[0125] Step S63: Send the broken wire scenarios corresponding to all transmission lines in the power grid and the data in the input module to the dispatching analysis module, and the dispatching analysis module includes: a generator unit combination model before extreme climate occurs, a traditional dispatching model under extreme climate, and a future state dispatching model under extreme climate.

[0126] Step S64: Obtain the unit dispatching plan before extreme climate occurs through the generator unit combination model before extreme climate occurs in the dispatching analysis module.

[0127] Step S65: Obtain the traditional dispatching plan under extreme climate through the traditional dispatching model under extreme climate in the dispatching analysis module.

[0128] Step S66: Obtain the future state dispatching plan under extreme climate through the future state dispatching model under extreme climate in the dispatching analysis module.

[0129] Step S67: Input the unit scheduling plan before the occurrence of extreme climate, the traditional scheduling plan under extreme climate, and the future-state scheduling plan under extreme climate into the scheduling effect analysis module for effect analysis. Determine the power grid loss and social loss under the future-state scheduling based on the transmission line disconnection probability and the number of disasters occurred, so as to obtain the effect of the future-state scheduling of the power grid under extreme climate.

[0130] As an example, in step S61, to construct a digital effect analysis model for the future-state scheduling of the power grid under extreme climate, please refer to Figure 5 , the digital effect analysis model for the future-state scheduling of the power grid under extreme climate includes: a disconnection scenario generation module, an input module, a scheduling analysis module, and a scheduling effect analysis module. The output end of the input module is connected to the first input end of the scheduling analysis module, the output end of the disconnection scenario generation module is connected to the second input end of the scheduling analysis module, and the output end of the scheduling analysis module is connected to the input end of the scheduling effect analysis module.

[0131] As an example, the input module is used to obtain and organize the input data, and the input data may include the power grid network structure, power grid line parameters, power grid node load power, generator set parameters, unit reserve capacity purchase cost, etc.

[0132] As an example, the disconnection scenario generation module is used to judge the disconnection probability of each transmission line in the power grid according to the disconnection probability of each transmission line in the power grid obtained from the tower collapse probability prediction model, so as to obtain the disconnection scenario corresponding to each transmission line in the power grid.

[0133] As an example, the scheduling analysis module includes: a generator unit combination model before the occurrence of extreme climate, a traditional scheduling model under extreme climate, and a future-state scheduling model under extreme climate, which are used to input the disconnection scenario corresponding to each transmission line in the power grid and the input data, and respectively analyze them by using the generator unit combination model before the occurrence of extreme climate, the traditional scheduling model under extreme climate, and the future-state scheduling model under extreme climate, so as to obtain the unit scheduling plan before the occurrence of extreme climate, the traditional scheduling plan under extreme climate, and the future-state scheduling plan under extreme climate.

[0134] As an example, the scheduling effect analysis module is used to analyze the decision-making differences of the unit scheduling plan before the occurrence of extreme climate, the traditional scheduling plan under extreme climate, and the future-state scheduling plan under extreme climate, determine the power grid loss and social loss under the future-state scheduling, and obtain the effect of the future-state scheduling of the power grid under extreme climate.

[0135] As an example, in step S62, based on the disconnection probability of each transmission line in the power grid obtained from the tower collapse probability prediction model, the disconnection probability of each transmission line is compared with a set value through the scenario generation module θ If the disconnection probability of the transmission line is greater than the set value θ, a broken-line scenario corresponding to the transmission line is generated, indicating that future-state scheduling will be performed on the transmission line; if the probability of the transmission line being broken is not greater than the set value θ , no broken-line scenario is generated, indicating that future-state scheduling will not be performed on the transmission line. All transmission lines in the power grid are judged, and the broken-line scenarios corresponding to all transmission lines are output.

[0136] As an example, in step S63, the broken-line scenarios corresponding to all transmission lines in the power grid and the data in the input module are sent to the scheduling analysis module, and the scheduling analysis module includes: a generator unit combination model before the occurrence of extreme climate, a traditional scheduling model under extreme climate, and a future-state scheduling model under extreme climate.

[0137] As an example, in step S64, in the period before the occurrence of extreme climate, since the power grid does not conduct future-state early warning, it is impossible to know whether a transmission line will fail or the output of a unit will be limited. The power grid will be scheduled according to the original grid structure. Analyze through the generator unit combination model before the occurrence of extreme climate in the scheduling analysis module, and optimize it using the first objective function to obtain the unit scheduling plan before the occurrence of extreme climate.

[0138] As an example, the first objective function can be based on the power generation cost, start-up cost, no-load operation cost of conventional units, and the power generation cost of new energy units, aiming to minimize the power generation cost, start-up cost, no-load operation cost of conventional units, and the power generation cost of new energy units. Specifically, according to the upper and lower limits of the output of conventional units, the first constraint con1.1 can be added; according to the upper limit of the output of new energy units such as wind power generation and photovoltaic power generation, the second constraint con1.2 can be added; according to the limit of the change range of the output of conventional units in adjacent time periods, the third constraint con1.3 and the fourth constraint con1.4 can be added, and the third constraint and the fourth constraint can be ramp constraints; according to the unit start-stop and state transition logic, the fifth constraint con1.5 can be added; according to the requirement that conventional units need to meet the minimum continuous operation time or shutdown time, the sixth constraint con1.6 and the seventh constraint con1.7 can be added; to ensure that the start-stop and operation states are binary integer variables, the eighth constraint con1.8 can be added; according to the power supply-demand balance of the power grid, the ninth constraint con1.9 can be added; according to the capacity limit for the safe operation of transmission lines, the tenth constraint con1.10 and the eleventh constraint con1.11 can be added; since in the actual power system, there are almost no independent power grid areas, and the power grid areas are connected to other external power grids based on sections, the section connection power can be added to the ninth constraint con1.9 and the tenth constraint con1.10. Suppose a natural disaster occurs at time t, the first objective function is expressed as follows: Among them, is thei Output quotation function of the th i unit; th i No-load operation cost of the th ne unit; th i unit t Whether to start at time period t ; if starting at th i unit t Whether to stop at time period t ; if stopping at th i unit t at time period th i unit t Output at time period Actual dispatching output of the new energy unit at t time period Is t The d th load value at time period th s section t Electricity quantity to be sent or received at time period The maximum transmission capacity of the line with the head and tail nodes of nm ; th i Vector form of the minimum output of the th i unit t at time period th i unit t-1 Output vector form at time period th i unit th ne New energy unit t Output vector form at time period th ne New energy unit t Predicted maximum available output vector form at time period Set of conventional generating units; th iDownward ramp rate limit of each unit; is the i upward ramp rate limit of each unit; is the n transmission line where the th node is located; is the n th node in t time period; is the m th node in t time period; is t time period; the line with the head and tail nodes as nm ; T is the total time period of the unit scheduling plan before the occurrence of extreme climate; is the i start time period of the th unit; i is the NE shutdown time period of the th unit; n is the new energy generating unit of the D th node;

[0139] As an example, in step S65, after an extreme climate disaster occurs, the transmission lines affected by the extreme climate disaster may fail, resulting in power flow blockage in some transmission lines or limited output of wind and light units. At this time, the power grid can be rescheduled according to the post-fault grid structure or unit information. Through the traditional scheduling model under extreme climate in the scheduling analysis module for analysis and using the second objective function for optimization, a traditional scheduling plan under extreme climate is obtained.

[0140] As an example, since the current capacity after the fault may not be sufficient to support the power supply capacity and safe operation requirements of the power grid, the power grid needs to urgently activate the spare capacity with fast ramp-up ability. Since the power grid is scheduled emergently, the power generation resources that can be purchased and mobilized by the power grid in a short time only include the spare capacity with a short call time. The spare capacity with a short call time can include: spinning reserve capacity (call time within 10 minutes), fast reserve capacity (call time within 30 minutes). Therefore, when a transmission line fails, the objective function of the traditional scheduling model increases the spinning reserve capacity cost and fast reserve capacity cost compared with the non-fault situation. The output constraint and ramp constraint of the spare units can be added to the constraint conditions of the first objective function to obtain the second objective function , and the expression is as follows: Among them, The second constraint condition when the l th line fails; is the output bidding function of the i th unit; is the output bidding function of the j th unit with short - term rapid supply capacity during the t period; is the power generation cost of the ne th new - energy unit; is the output of the j th unit with short - term rapid supply capacity during the t period; is the output of the i th unit during the t period; is the actual dispatched output of the new - energy unit during the t period; is the t th load value during the d period; is the power quantity to be sent or received by the s th section during the t period; is the vector form of the minimum output of the i th unit; is the vector form of the output of the i th unit during the t period; is the vector form of the output of the i th unit during the t-1 period; is the vector form of the maximum output of the i th unit; is the vector form of the output of the ne th new - energy unit during the t period; is the vector form of the predicted maximum available output of the ne th new - energy unit during the t period; is the maximum transmission capacity of the line with the head and tail nodes of nm ; is the set of conventional generating units; is the set of reserve capacity units with short call - up time; is the set of conventional generating units at the n th node; is the downward ramp rate limit of the i th unit; is the upward ramp rate limit of the i th unit; is then The node voltage phase angle of the t node at time period is the m node voltage phase angle of the t node at time period is t the line with the head and tail nodes of nm at time period T is the total time period of the traditional scheduling scheme under extreme climate; is the l set of all other lines except the l line after the

[0141]

[0142] As an example, when analyzing the typhoon scenario, transmission line faults may occur, as well as output fluctuations of wind turbines and photovoltaic units due to typhoons. During power grid scheduling, it is necessary to schedule according to the network structure after the fault, and at the same time, constrain the upper output limits of wind turbines and photovoltaic units. When analyzing the wildfire scenario, the situation of transmission line faults caused by wildfires can be considered, and scheduling can be carried out according to the network structure after the line fault.

[0143] As an example, in step S66, in the scheduling analysis module, the power grid will formulate a scheduling plan for typical extreme climate scenarios in advance and reserve sufficient reserve capacity. Analyze through the future-state scheduling model under extreme climate in the scheduling analysis module, and optimize using the third objective function to obtain the future-state scheduling plan under extreme climate.

[0144] l As an example, the future-state scheduling model under extreme climate has more preparation time than the traditional scheduling model, so more resources can be selected for purchase. The variables of the third objective function can include reserve capacity resources with longer call times on the basis of the traditional real-time scheduling model. The reserve capacity resources with longer call times can include: slow reserve units (call time over 30 minutes) to provide more choices. l As an example, when the l th transmission line fails, the third objective function of the future-state scheduling model under extreme climate increases the slow reserve capacity cost, and the constraints increase the constraints related to slow reserve units. Considering the prediction error of the output of wind and light units, in order to ensure the reliability of the scheduling plan in the fault state, in this embodiment, the upper output limit of new energy units can be set to the predicted value multiplied by the historical minimum actual prediction ratio. Since the future-state scheduling model under extreme climate increases the analysis of the line fault probability and the output limitation of wind and light generating units under extreme climate, the power grid can consider the impact of extreme climate on the power system and make an advanced plan, so that more alternative units can be added for scheduling in the scheduling, reducing the scheduling economic cost.The third objective function in case of a line fault is expressed as follows: where, is the output bid function of the i th unit; is the output bid function of the j th unit with short-term rapid supply capacity during the t period; is the power generation cost of the ne th new energy unit; is the output bid function of the k th slow reserve unit during the t period; is the output of the i th unit during the t period; is the output of the j th unit with short-term rapid supply capacity during the t period; is the output of the k th slow reserve unit during the t period; is the actual scheduled output of the new energy unit during the t period; is the t th load value during the d period; is the amount of electricity to be sent or received by the s th section during the t period; is the vector form of the minimum output of the i th unit; is the vector form of the output of the i th unit during the t period; is the vector form of the output of the i th unit during the t-1 period; is the vector form of the maximum output of the i th unit; is the vector form of the output of the ne th new energy unit during the t period; is the vector form of the maximum predicted output of the ne th new energy unit during the t period; is the minimum of the actual and predicted historical outputs of the ne th new energy unit; is the neThe true historical output value of a new energy unit; is the ne predicted historical output value of the th new energy unit; nm is the maximum transmission capacity of the line with the head and tail nodes of is the i downward ramp rate limit of the th unit; i is the upward ramp rate limit of the n th unit; t is the node voltage phase angle of the th node at the m time period; t is the node voltage phase angle of the th node at the t time period; nm is the line with the head and tail nodes of T at the NE is the total time period of the future state scheduling plan under extreme climate; is the set of new energy generating units; n is the new energy generating unit of the th node; l is the set of all other lines except the l th line after the th line fails; is the set of standby capacity units with short call time; is the set of slow standby capacity units with long call time; is the n th set of conventional generating units at the G is the G 1 and G 2 and G 3 intersection of the three sets; D is the set of node loads in the power grid; D n is the n th node load.

[0145] As an example, in step S67, the traditional scheduling results and future state scheduling results of the power grid under different extreme climates can be obtained through the scheduling analysis module. According to the scheduling results under different extreme climates, the increased temporary output demand of the power grid is obtained, and the effects of the future state scheduling measures invested under each extreme climate are obtained.

[0146] As an example, by inputting the unit dispatch plan before extreme climate occurs, the traditional dispatch plan under extreme climate, and the future dispatch plan under extreme climate into the dispatch effect analysis module for analysis, it can be obtained that the traditional dispatch plan under extreme climate can only start emergency reserve capacity such as rotating reserve after a fault occurs. After the fault, it can be re-dispatched according to the grid information after the fault, and can call on reserve capacity resources with a longer time; the future dispatch plan under extreme climate puts the re-dispatching process in front, and quickly adjusts based on the predicted results when a fault occurs, thereby improving the efficiency of the initial emergency response.

[0147] As an example, the results for each period can be averaged and grid and social losses can be estimated based on the probability of transmission line outages and the number of disasters.

[0148] For example, when an extreme weather emergency occurs, the power grid urgently activates the backup capacity. If the backup capacity is sufficient to meet the power supply shortage of the system, the power grid company's loss is only the purchase cost of the backup capacity; if the backup capacity is not enough to meet the power supply shortage, some users will face the risk of power outages. At this time, the power grid loss can be the purchase cost of the backup capacity and the power outage compensation cost given to users who have power outages due to faults. Power grid losses in extreme weather The expression is as follows: in, F is the annual occurrence frequency of extreme climate; N is the total number of failure types under a certain extreme climate; T is the total time period of the traditional dispatch scheme under extreme climate; a The type of failure under certain extreme climate conditions; For the a The failure probability of each failure type; is the cost function of the standby unit; For the a In the following fault types t The output of the standby units required for the time period is given by the scheduling analysis module; For the a Under the fault type t The power supply shortage of the power system during the period; The reserve capacity of the power grid; It is the power outage compensation coefficient per unit of electricity.

[0149] For example, when a power system fails, if there is a power shortage in the power system causing a power outage, it will lead to industrial shutdown, commercial interruption and resident life disruption, causing serious social losses and reduced social efficiency. The power shortage can be selected as an indicator to measure social losses based on the amount of power outages for users. The social losses caused by emergencies in extreme climates The expression is as follows: Wherein, F is the annual occurrence times of extreme climate; N is the total number of fault types under a certain extreme climate; T is the total time period of the traditional scheduling scheme under extreme climate; a is the fault type under a certain extreme climate; is the a th fault probability of the fault type; is the a th required output of the standby unit at the t time period under the th fault type, which is given by the scheduling analysis module; a is the power supply deficit of the power system within the t time period under the th fault type;

[0150] As an example, by analyzing the grid loss and social loss of the future-state scheduling scheme under extreme climate, the impact on the scheduling effect before and after the implementation of the future-state scheduling is quantitatively analyzed, and the effect analysis of the future-state scheduling measures under each extreme climate is realized.

[0151] In the method for predicting the line disconnection probability and analyzing the future-state scheduling effect based on the graph neural network of the present application, by constructing a graph neural network model, the mapping from regional-scale climate information to node-scale microclimate characteristics is realized, and high-precision prediction of local climate data near each tower can be carried out in the case of missing real-time observations; the accuracy and fineness of the line disconnection probability prediction are improved through two-stage modeling. In the first stage, the graph neural network model is trained using regional-level meteorological data to predict the local meteorological characteristics of the location of each tower. In the second stage, based on the predicted local meteorological data of the tower and the tower structure attributes, the tower collapse probability model is trained, and further the disconnection probability of the entire transmission line is calculated through combinatorial logic. It can realize that only by inputting regional-level meteorological data, the disconnection probability of the entire line can be output with high precision, taking into account both practicality and prediction accuracy, and significantly improving the fault prediction ability of the power transmission system under extreme climate; through the digital effect analysis model of the future-state scheduling of the power grid under extreme climate, the gap in the lack of analysis of the future-state scheduling effect of the power grid under extreme climate conditions in the existing technology is filled. By detailed analysis of the power grid scheduling under typical extreme climate conditions, the impact of the future-state scheduling scheme on the grid loss and social loss is quantified, providing a scientific decision-making basis for whether the power grid conducts future-state scheduling investment.

[0152] It should be understood that although the steps in the flowchart of the accompanying drawings are shown in sequence according to the indication of the arrows, these steps are not necessarily executed in the sequence indicated by the arrows. Unless there is a clear description in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the accompanying drawings may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or sub-steps or stages of other steps.

[0153] In another embodiment, the present application also provides a system for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network. The system for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network includes: one or more processors; a storage device for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network as described in any of the above.

[0154] In another embodiment, the present application also provides a computer program product. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer is capable of executing each step of the method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network provided in the above embodiments.

[0155] The computer-executable instructions for implementing the method of the present application can be written in any combination of one or more programming languages. These computer-executable instructions can be provided to the processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, such that when the computer-executable instructions are executed by the processor or controller, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer-executable instructions can be executed entirely on the machine, partially on the machine, executed partially on the machine as an independent software package and partially on a remote machine, or executed entirely on a remote machine or electronic device.

[0156] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback or tactile feedback); and input from the user can be received in any form, including: acoustic input, voice input, or tactile input.

[0157] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data electronic device), or a computing system including middleware components (e.g., an application electronic device), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0158] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0159] Although this application has been disclosed above by way of embodiments, it is not intended to limit this application. Any person with ordinary knowledge in the technical field to which this application pertains can make some modifications and refinements without departing from the spirit and scope of this application. Therefore, the scope of protection of this application shall be subject to that defined by the appended patent application scope.

Claims

1. A method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network, characterized in that, Including the following steps: Obtain historical regional climate data and first pole tower body characteristic data; Construct a pole tower local climate data prediction model according to the historical regional climate data and the first pole tower body characteristic data; Use the pole tower local climate data prediction model to predict the key meteorological indicators of the pole tower to obtain pole tower local climate data; Obtain second pole tower body characteristic data and pole tower operation and maintenance information, construct a pole tower collapse probability prediction model according to the second pole tower body characteristic data, the pole tower operation and maintenance information, and the pole tower local climate data, and obtain the collapse probability of each pole tower based on the pole tower collapse probability prediction model; Based on the collapse probability of each pole tower, determine the wire break probability of the entire transmission line through a probability combination model; Construct a digital effect analysis model for future-state dispatching of the power grid under extreme climate, and analyze the losses under future-state dispatching based on the digital effect analysis model for future-state dispatching of the power grid under extreme climate to realize the analysis of the future-state dispatching effect of the power grid under extreme weather; 2. The method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on the graph neural network according to claim 1, wherein, Constructing a pole tower local climate data prediction model according to the historical regional climate data and the first pole tower body characteristic data includes: Obtain the actual local climate data of the pole tower, set the actual local climate data of the pole tower as the first target output data, and combine the historical regional climate data and the first pole tower body characteristic data to obtain the first input data; Preprocess the first input data and the first target output data; Convert the preprocessed first input data into a graph structure; Construct a first graph neural network; Use the first input data in graph structure and the preprocessed first target output data to train the first graph neural network, and use the first loss function to optimize and update the parameters to obtain the pole tower local climate data prediction model; 3. The method for predicting the line disconnection probability and analyzing the future state scheduling effect based on the graph neural network according to claim 1, wherein Predict the key meteorological indicators of the pole tower using the local climate data prediction model of the pole tower, and the predicted value of the local climate data of the pole tower at time t The expression is: Among them, is the adjacency matrix, is the first input feature matrix, represents the prediction model of the local climate data of the pole tower.

4. The method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on the graph neural network according to claim 1, wherein Obtain second pole tower body characteristic data and pole tower operation and maintenance information, construct a pole tower collapse probability prediction model according to the second pole tower body characteristic data, the pole tower operation and maintenance information, and the pole tower local climate data, and obtain the collapse probability of each pole tower based on the pole tower collapse probability prediction model, including: Obtain second pole tower body characteristic data and pole tower operation and maintenance information; Combine the second pole tower body characteristic data, the pole tower operation and maintenance information, and the pole tower local climate data to obtain a second input data set; Preprocess the second input data set; Convert the preprocessed second input data set into a graph structure; Construct a second graph neural network; Use the second input data set in graph structure to train the second graph neural network, and use the second loss function to optimize and update the parameters to obtain the pole tower collapse probability prediction model; 5. The method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on the graph neural network according to claim 1, wherein Based on the probability of each tower collapsing, the probability of conductor breakage for the entire transmission line is determined through a probability combination model, and the total probability of conductor breakage for the transmission line can be expressed as: Among them, represents t the predicted value of the tower collapse probability of the i nth tower at a certain moment.

6. The method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on the graph neural network according to claim 1, wherein Constructing a digital effect analysis model for future-state dispatching of the power grid under extreme climate, and analyzing the losses under future-state dispatching based on the digital effect analysis model for future-state dispatching of the power grid under extreme climate to realize the analysis of the future-state dispatching effect of the power grid under extreme weather, including: Construct a digital effect analysis model for the future-state dispatching of the power grid under extreme climate. The digital effect analysis model for the future-state dispatching of the power grid under extreme climate includes: a line break scenario generation module, an input module, a dispatching analysis module, and a dispatching effect analysis module; Input the line break probabilities of each transmission line in the power grid into the line break scenario generation module to obtain the line break scenarios corresponding to each transmission line in the power grid; Send the line break scenarios corresponding to all transmission lines in the power grid and the data in the input module to the dispatching analysis module. The dispatching analysis module includes: a generator unit combination model before extreme climate occurs, a traditional dispatching model under extreme climate, and a future-state dispatching model under extreme climate; Obtain the unit dispatching plan before extreme climate occurs through the generator unit combination model before extreme climate occurs in the dispatching analysis module; Obtain the traditional dispatching plan under extreme climate through the traditional dispatching model under extreme climate in the dispatching analysis module; Obtain the future-state dispatching plan under extreme climate through the future-state dispatching model under extreme climate in the dispatching analysis module; Input the unit dispatching plan before extreme climate occurs, the traditional dispatching plan under extreme climate, and the future-state dispatching plan under extreme climate into the dispatching effect analysis module for effect analysis, and determine the power grid loss and social loss under future-state dispatching based on the line break probability of the transmission line and the number of disasters occurring, so as to obtain the effect of the future-state dispatching of the power grid under extreme climate.

7. The method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on the graph neural network according to claim 6, wherein Obtain the unit scheduling plan before the occurrence of extreme climate through the unit combination model before the occurrence of extreme climate in the said scheduling analysis module, including: analyze through the unit combination model before the occurrence of extreme climate in the scheduling analysis module, and optimize it using the first objective function to obtain the unit scheduling plan before the occurrence of extreme climate, the first objective function The expression of which is as follows: wherein, is the output bidding function of the i th unit; is the start-up cost of the i th unit; is the no-load operation cost of the i th unit; is the power generation cost of the ne th new energy unit; is whether the i th unit t starts in the t period. If it starts in the t period, it is 1; otherwise, it is 0; is whether the i th unit t stops in the t period. If it stops in the t period, it is 1; otherwise, it is 0; is whether the i th unit operates in the t period. If it operates, it is 1; if it does not operate, it is 0; is the output of the i th unit in the t period; is the actual dispatched output of the new energy unit in the t period; is t the d th load value in the period; s is the amount of power to be sent or received by the t th section in the period; nm is the maximum transmission capacity of the line with the head and tail nodes of the line being nm ; is the vector form of the minimum output of the i th unit; is the vector form of the output of the i th unit in the t period; is the vector form of the output of the i th unit in the t-1 period; is the vector form of the maximum output of the i th unit; is the vector form of the output of the ne th new energy unit in the t period; is the vector form of the maximum predicted available output of the ne th new energy unit in the t period; is the set of conventional generating units; is the i Downward ramp rate limit of each unit; For the i The upward climbing rate limit of each unit; For the n The transmission line where the node is located; It is the collection of all transmission lines in the power grid; For the n Nodes in t The node voltage phase angle of the time period; For the m Nodes in t The node voltage phase angle of the time period; for t The first and last nodes of the time period line are nm The line; T The total time period of the unit dispatch plan before the extreme weather occurs; For the i The start-up period of each unit; For the i The downtime period of each unit; NE Assemble for new energy generators; For the n Nodes of new energy generators; D is the node load collection in the power grid.

8. The method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on the graph neural network according to claim 6, wherein Obtain the traditional scheduling plan under extreme climate through the traditional scheduling model in the scheduling analysis module, including: analyze through the traditional scheduling model under extreme climate in the scheduling analysis module, and optimize using the second objective function to obtain the traditional scheduling plan under extreme climate, the second objective function , and the expression is as follows: Among them, is the second constraint condition when the l th line fails; is the output bid function of the i th unit; is the output bid function of the j th unit with short-term rapid supply capacity during the t period; is the power generation cost of the ne th new energy unit; is the output of the j th unit with short-term rapid supply capacity during the t period; is the output of the i th unit during the t period; is the actual dispatched output of the new energy unit during the t period; is the t th load value during the d th period; is the power quantity to be sent or received by the s th section during the t period; is the vector form of the minimum output of the i th unit; is the vector form of the output of the i th unit during the t period; is the vector form of the output of the i th unit during the t-1 period; is the vector form of the maximum output of the i th unit; is the vector form of the output of the ne th new energy unit during the t period; is the vector form of the predicted maximum available output of the ne th new energy unit during the t period; is the maximum transmission capacity of the line with the head and tail nodes of nm ; is the set of conventional generating units; is the set of reserve capacity units with short call time; is the set of conventional generating units at the n th node; is the downward ramp rate limit of the i th unit; is the upward ramp rate limit of the i th unit; is the node voltage phase angle of the n th node at t time period; is the node voltage phase angle of the m th node at t time period; is the line with the head and tail nodes of t at nm time period; T is the total time period of the traditional scheduling scheme under extreme climate; is the set of all other lines except the l th line after the l th line fails.

9. The method for predicting the line disconnection probability and analyzing the future state scheduling effect based on the graph neural network according to claim 6, wherein Obtain the future-state scheduling plan under extreme climate through the future-state scheduling model under extreme climate in the scheduling analysis module, including: analyze through the future-state scheduling model under extreme climate in the scheduling analysis module, and optimize using the third objective function to obtain the future-state scheduling plan under extreme climate, and the third objective function has the following expression: Among them, is the output bid function of the i th unit; is the output bid function of the j th unit with short-term rapid supply capacity at t time period; is the power generation cost of the ne th new energy unit; is the output bid function of the k th slow reserve unit at t time period; is the output of the i th unit at t time period; is the output of the j th unit with short-term rapid supply capacity at t time period; is the output of the k th slow reserve unit at t time period; is the actual scheduled output of the new energy unit at t time period; is t the d th load value at is the power quantity to be sent out or received by the s th section at t time period; is the vector form of the minimum output of the i th unit; is the vector form of the output of the i th unit at t time period; is the vector form of the output of the i th unit at t-1 time period; is the vector form of the maximum output of the i th unit; is the vector form of the output of the ne th new energy unit at t time period; is the vector form of the maximum predicted output of the ne th new energy unit at t time period; is the minimum value of the historical actual output and predicted output of the ne th new energy unit; is the historical actual output value of the ne th new energy unit; is the historical predicted output value of the ne th new energy unit; is that the head and tail nodes of the line are nm The maximum transmission capacity of the line; is the i downward ramp rate limit of the th i unit; is the n upward ramp rate limit of the t th unit; m is the t phase angle of the voltage at the th t node during the nm period; T is the total period of the future-state scheduling plan under extreme climate; NE is the set of new energy generating units; is the n new energy generating unit at the th l node; l is the set of all lines except the th line after the th line fails; n is the set of conventional generating units at the G th G 1 , G 2 , G 3 the intersection of the three sets; D is the set of node loads in the power grid; D n is the n load at the th node.

10. The method for predicting the probability of line disconnection and analyzing the future state scheduling effect based on the graph neural network according to claim 6, wherein Grid losses under the future-state scheduling The expression is: Among them, F is the annual occurrence times of extreme climate; N is the total number of fault types under a certain extreme climate; T is the total time period of the traditional scheduling scheme under extreme climate; a is a fault type under a certain extreme climate; is the a fault probability of the is the standby unit cost function; is the a output of the standby unit required in the t time period under the fault type, given by the scheduling analysis module; a is the t power supply shortage of the power system within the time period under the is the spare capacity of the power grid; is the unit power outage compensation coefficient; The social loss under the future-state scheduling The expression is: Among them, F is the annual occurrence times of extreme climate; N is the total number of fault types under a certain extreme climate; T is the total time period of the traditional scheduling scheme under extreme climate; a is the fault type under a certain extreme climate; is the a fault probability of the th fault type; a is the output of the standby unit required in the t time period under the th fault type, given by the scheduling analysis module; a is the power supply shortage of the power system within the t time period under the th fault type; is the reserve capacity of the power grid.

11. A system for predicting the probability of line disconnection and analyzing the future state scheduling effect based on a graph neural network, characterized in that, The system for predicting the line break probability based on a graph neural network and analyzing the future-state dispatching effect includes: one or more processors; a storage device for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the method for predicting the line break probability based on a graph neural network and analyzing the future-state dispatching effect as described in any one of claims 1 to 10.

12. A computer program product comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method for predicting the line break probability based on a graph neural network and analyzing the future-state dispatching effect as described in any one of claims 1 to 10.

Images