CN117422260A - State estimation method, device, equipment and medium based on graph neural network - Google Patents
State estimation method, device, equipment and medium based on graph neural network Download PDFInfo
- Publication number
- CN117422260A CN117422260A CN202311427159.7A CN202311427159A CN117422260A CN 117422260 A CN117422260 A CN 117422260A CN 202311427159 A CN202311427159 A CN 202311427159A CN 117422260 A CN117422260 A CN 117422260A
- Authority
- CN
- China
- Prior art keywords
- bus
- equipment
- node
- active power
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 96
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000010586 diagram Methods 0.000 claims abstract description 70
- 238000005259 measurement Methods 0.000 claims abstract description 48
- 238000004590 computer program Methods 0.000 claims description 16
- 238000003860 storage Methods 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 12
- 230000005540 biological transmission Effects 0.000 claims description 10
- 239000003990 capacitor Substances 0.000 claims description 9
- 238000004891 communication Methods 0.000 description 9
- 238000009826 distribution Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 238000012545 processing Methods 0.000 description 8
- 230000008859 change Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000012546 transfer Methods 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 210000004027 cell Anatomy 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004931 aggregating effect Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/042—Knowledge-based neural networks; Logical representations of neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
Abstract
The invention discloses a state estimation method, device, equipment and medium based on a graph neural network. The method comprises the following steps: constructing a bus branch diagram of the power system, and calculating the active power unbalance of at least one bus node; inputting the bus branch diagram and the active power unbalance of each bus node into a graph neural network, and determining the corresponding equivalent unit configuration result of each bus; based on the configuration result of the equivalent unit, configuring the equivalent unit for the bus and acquiring the equipment power in the electric power system; and carrying out state estimation on equipment in the power system according to the equipment power, and calculating an equipment voltage estimated value of the equipment. The technical scheme of the embodiment of the invention improves the configuration efficiency of the equivalent unit, globally solves the problem of unbalanced active measurement and ensures the convergence of state estimation.
Description
Technical Field
The present invention relates to the field of artificial intelligence technologies, and in particular, to a state estimation method, apparatus, device, and medium based on a graph neural network.
Background
In power system state estimation, measurement data errors often have a severe impact on the convergence of the state estimation. To solve this problem, in the debugging of the state estimation, the convergence of the state estimation is often improved by adding an equivalent unit to the bus with unbalanced active measurements.
At present, it is usually determined on which buses equivalent units are added and how to optimally configure the equivalent units only according to experience of a debugger.
The existing equivalent unit configuration mode has low configuration efficiency, can only solve the local problem of state estimation convergence, can not globally solve the problem of unbalanced active measurement in a power system, and still has difficult guarantee of state estimation convergence.
Disclosure of Invention
The invention provides a state estimation method, device, equipment and medium based on a graph neural network, which improves the configuration efficiency of an equivalent unit, globally solves the problem of unbalanced active measurement and ensures the convergence of state estimation.
According to an aspect of the present invention, there is provided a state estimation method based on a graph neural network, the method including:
constructing a bus branch diagram of the power system, and calculating the active power unbalance of at least one bus node;
Inputting the bus branch diagram and the active power unbalance of each bus node into a graph neural network, and determining the corresponding equivalent unit configuration result of each bus;
based on the configuration result of the equivalent unit, configuring the equivalent unit for the bus and acquiring the equipment power in the electric power system;
and carrying out state estimation on equipment in the power system according to the equipment power, and calculating an equipment voltage estimated value of the equipment.
According to another aspect of the present invention, there is provided a state estimation apparatus based on a graph neural network, the apparatus including:
the bus branch diagram construction module is used for constructing a bus branch diagram of the power system and calculating the active power unbalance of at least one bus node;
the equivalent unit configuration result determining module is used for inputting the bus branch diagram and the active power unbalance of each bus node into the graphic neural network and determining the corresponding equivalent unit configuration result of each bus;
the equipment power acquisition module is used for configuring the equivalent unit for the bus based on the configuration result of the equivalent unit and acquiring equipment power in the electric power system;
and the state estimation module is used for carrying out state estimation on equipment in the power system according to the equipment power and calculating an equipment voltage estimated value of the equipment.
According to another aspect of the present invention, there is provided an electronic device including:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the state estimation method based on the graph neural network of any one of the embodiments of the present invention.
According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to implement the state estimation method based on a graph neural network of any embodiment of the present invention when executed.
According to the technical scheme, the bus branch diagram of the power system is constructed, the active power unbalance of at least one bus node is calculated, the bus branch diagram and the active power unbalance of each bus node are input into the graph neural network, the corresponding equivalent unit configuration result of each bus is determined, the equivalent unit is configured on the bus based on the equivalent unit configuration result, the equipment power in the power system is obtained, the equipment in the power system is subjected to state estimation according to the equipment power, the equipment voltage estimation value of the equipment is calculated, the problems that the existing equivalent unit configuration mode has low configuration efficiency and can only solve the local problem of state estimation convergence, the problem that the active power measurement unbalance in the power system is still difficult to guarantee can be solved, the configuration efficiency of the equivalent unit is improved, the problem of the active power measurement unbalance is solved globally, and the convergence of the state estimation is guaranteed.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a state estimation method based on a neural network according to a first embodiment of the present invention;
FIG. 2 is a diagram of an information delivery process of the neural network according to a first embodiment of the present invention;
fig. 3 is a flowchart of a state estimation method based on a neural network according to a second embodiment of the present invention;
fig. 4 is a bus bar diagram of a power system according to a second embodiment of the present invention;
FIG. 5 is a schematic diagram of a neural network of the graph determining equivalent unit configuration results for each busbar according to a second embodiment of the present invention;
Fig. 6 is a schematic structural diagram of a state estimation device based on a neural network according to a third embodiment of the present invention;
fig. 7 is a schematic structural diagram of an electronic device implementing a state estimation method based on a neural network according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
Fig. 1 is a flowchart of a state estimation method based on a graph neural network according to an embodiment of the present invention. The embodiment of the invention is applicable to the condition of carrying out state estimation on the power systemization based on the graph neural network, the method can be executed by a state estimation device based on the graph neural network, the state estimation device based on the graph neural network can be realized in a form of hardware and/or software, and the state estimation device based on the graph neural network can be configured in an electronic device carrying a state estimation function based on the graph neural network.
Referring to fig. 1, the state estimation method based on the graph neural network includes:
s110, constructing a bus branch diagram of the power system, and calculating the active power unbalance of at least one bus node.
The bus bar branching diagram may be used to characterize the device connection relationship between the buses in the power system. The bus bar diagram may embody a distribution network structure in a power system. Alternatively, the bus bar branching diagram may be used to characterize in real time the device connection relationship between the buses in the power system. The bus bar branching diagram may include bus bar nodes and branches. In the state estimation of the power system, each bus node is required to maintain active power balance, and it is understood that each bus is required to maintain active power balance. Bus active power balance is understood to mean that the difference between the device active power output by the bus and the device active power input to the bus node is zero. The bus active power is unbalanced, and it is understood that the difference between the device active power output by the bus and the device active power input to the bus is not zero. The amount of active power imbalance of the bus node may be a difference between the device active power input to the bus and the device active power output by the bus.
Specifically, the device connection relationship in the power system can be obtained in real time. According to the equipment connection relation, each bus in the power system and the equipment connection relation among the buses can be determined, and a bus branch diagram of the power system is generated. And acquiring the equipment active power of the input bus and the equipment active power output by the bus, and calculating the difference between the equipment active power of the input bus and the equipment output power output by the bus to obtain the active power unbalance of each bus node.
S120, inputting the bus branch diagram and the active power unbalance of each bus node into a graph neural network, and determining the equivalent unit configuration result of each corresponding bus.
The graph neural network can be used for determining whether the bus in the power system is provided with an equivalent unit. The graph neural network may be pre-trained. Compared with the traditional neural network (such as a convolutional neural network), the graph neural network can process measurement data of real-time change in a power system and can also consider a complex power distribution network structure. The configuration result of the equivalent unit can be whether the bus needs to be configured with the equivalent unit or not. For example, the equivalent unit configuration result may include configuration of an equivalent unit or non-configuration of an equivalent unit.
Specifically, the active power unbalance of each bus node can be used as attribute information of each bus node in the bus branch diagram. And inputting the bus branch diagram and the active power unbalance of each bus node into a graph neural network, and outputting the corresponding equivalent unit configuration result of each bus.
S130, configuring an equivalent unit for the bus based on the equivalent unit configuration result, and acquiring the equipment power in the electric power system.
The equivalent machine set can be used for realizing active power balance of the corresponding bus. Optionally, the configuration result of the equivalent unit may further include a type of the equivalent unit and an active power of the equivalent unit. When the active power unbalance of the bus is greater than zero, the type of the equivalent unit can be a load unit; when the active power unbalance of the bus is smaller than zero, the type of the equivalent unit can be a generator unit. Correspondingly, the active power of the equivalent unit can be determined according to the active power unbalance of each bus. The device power may be a power measurement of a device in the power system. For example, the device power may include device active power and device reactive power. Compared with the situation that the bus is not provided with the equivalent unit, the bus may have unbalanced active power, and the measured value of the device power is also inaccurate. After the equivalent unit is configured on the bus, the equipment power in the power system is obtained, and the measured value of the equipment power is more accurate. Correspondingly, when the state estimation of the power system is carried out, the equipment voltage estimation value is more accurate.
Specifically, if the configuration result of the equivalent unit is that the equivalent unit is configured, the equivalent unit is configured for the corresponding bus; if the equivalent unit configuration result is that the equivalent unit is not configured, the equivalent unit is not configured for the corresponding bus. And after the buses in the power system complete the configuration of the equivalent unit, acquiring the equipment power of the power system.
And S140, carrying out state estimation on equipment in the power system according to the equipment power, and calculating an equipment voltage estimated value of the equipment.
The device voltage estimate may be an estimate of the device voltage at active power balance of the bus in the power system. Because of the fact that the measured data in the power system have errors or repetition, the accuracy of the measured value of the equipment voltage is low, and the accuracy of the obtained estimated value of the equipment voltage is higher than that of the measured value of the equipment voltage through state estimation of equipment in the power system to determine the estimated value of the equipment voltage, and the subsequent analysis of the power system is facilitated.
Specifically, state estimation can be performed on equipment in the power system according to equipment power, and an equipment voltage estimated value of the equipment is calculated.
In the state estimation of a power system, bad data and network topology connection errors often have a serious influence on the convergence of the state estimation. Wherein the bad data may comprise measurement data errors or measurement data repetitions. In order to solve this problem, in the debugging of the state estimation, a mode of adding an equivalent unit to a bus with unbalanced active power measurement (i.e., unbalanced active power) may be adopted to improve the convergence of the state estimation. However, how to optimally configure the equivalent units by adding the equivalent units on the buses often only depends on the experience of debugging personnel, and a systematic and efficient optimization scheme is lacking. However, the state estimation method for debugging according to the experience of the debugger has low configuration efficiency of the equivalent unit, and the setting of the equivalent unit can only solve the local problem of convergence of state estimation, and can not solve the problem of unbalance of active measurement in the power system globally. Furthermore, deep learning methods employing artificial intelligence are a general solution to such complex problems. However, the conventional Convolutional Neural Network (CNN) cannot process measurement data of real-time variation in the power system, and cannot consider a complex power distribution network structure.
According to the technical scheme, the bus branch diagram of the power system is constructed, the active power unbalance of at least one bus in the power system is calculated, the bus branch diagram and the active power unbalance of each bus are input into the graph neural network, the equivalent unit configuration result of each bus is determined, the equivalent unit is configured for the bus based on the equivalent unit configuration result, the equipment power in the power system is obtained, the equipment in the power system is subjected to state estimation according to the equipment power, the equipment voltage estimation value of the equipment is calculated, the characteristics of real-time data change and complex structures in the power system can be adapted by using the graph neural network, the equivalent unit configuration result of the power system is determined, the problem that the existing equivalent unit configuration mode is low in configuration efficiency, the problem that the state estimation is converged can only be solved, the problem that the active power measurement in the system is unbalanced can not be solved globally is solved, the problem that the convergence of the state estimation is still difficult to ensure, the configuration efficiency of the equivalent unit is improved, the problem that the active power measurement is unbalanced is solved globally, and the convergence of the state estimation is ensured.
In an optional embodiment of the present invention, inputting the bus branch diagram and the active power unbalance of each bus node into the graph neural network, determining the equivalent unit configuration result of each corresponding bus, including: inputting a bus node branch diagram and active power unbalance of each bus node into a graph neural network; through a first convolution layer of the graph neural network, for each busbar node, carrying out weighted average on the active power unbalance of the busbar node and the active power unbalance of other busbar nodes connected with the busbar node to obtain a weighted average result of the first convolution layer; calculating the product between the weighted average result of the first convolution layer and the weight matrix coefficient of the first convolution layer through the first convolution layer of the graph neural network to obtain characteristic data output by the busbar node at the first convolution layer; through other convolution layers of the graph neural network, aiming at each busbar node, carrying out weighted average on the characteristic data output by the last convolution layer by the busbar node and the characteristic data output by the last convolution layer by other busbar nodes connected with the busbar node to obtain weighted average results of other convolution layers; calculating products between weighted average results of other convolution layers and weight matrix coefficients of other convolution layers through other convolution layers of the graph neural network to obtain characteristic data output by bus nodes in the other convolution layers until the characteristic data output by the last convolution layer is obtained; comparing the characteristic data output by the last convolution layer with a preset characteristic data threshold value; when the characteristic data output by the last convolution layer is greater than or equal to a preset characteristic data threshold value, determining an equivalent unit configuration result of the corresponding bus as a configuration equivalent unit; and when the characteristic data output by the last convolution layer is smaller than a preset characteristic data threshold value, determining that the equivalent unit configuration result of the corresponding bus is not configured.
The Graph Neural Network (GNN) may include a plurality of convolutional layers (i.e., information transfer layers). Each convolution layer implements the graph convolution process through data transfer, so that the pattern features of the data are embedded into the neurons of each convolution layer, namely the busbar nodes of the busbar branch graph. The training of the graph neural network is to adjust the weight matrix coefficient of the convolution layer, so as to obtain the optimized graph neural network. The weighted average result can be used for aggregating the characteristic data of the bus node and other bus nodes connected with the bus node, so that the structure of the power distribution network can be considered, and the characteristics of the other bus nodes connected with the bus node can be learned. The weight matrix coefficients may be used for information transfer of the feature data. The characteristic data may be data output from each convolution layer. It is understood that the feature data is a feature extraction result of the convolution layer. The preset characteristic data threshold value can be used for judging the equivalent unit configuration result of the corresponding bus.
Specifically, the bus node branch diagram and the active power unbalance of each bus node can be input into the graph neural network. And carrying out weighted average on the active power unbalance of each busbar node and the active power unbalance of other busbar nodes connected with the busbar node by the first convolution layer of the graph neural network to obtain a weighted average result of the first convolution layer. The product between the weighted average result of the first convolution layer and the weight matrix coefficient of the first convolution layer can be calculated through the first convolution layer of the graph neural network, so that the characteristic data output by the busbar node at the first convolution layer can be obtained. The characteristic data output by the previous convolution layer of the busbar node and the characteristic data output by the previous convolution layer of other busbar nodes connected with the busbar node can be weighted and averaged through other convolution layers of the graph neural network, and the weighted average result of the other convolution layers is obtained. And calculating products between weighted average results of other convolution layers and weight matrix coefficients of other convolution layers through other convolution layers of the graph neural network to obtain characteristic data output by the busbar node in the other convolution layers until the characteristic data output by the last convolution layer is obtained. The characteristic data output by the last convolution layer can be compared with a preset characteristic data threshold value, and when the characteristic data output by the last convolution layer is greater than or equal to the preset characteristic data threshold value, the equivalent unit configuration result of the corresponding bus is determined to be the configuration equivalent unit; and when the characteristic data output by the last convolution layer is smaller than a preset characteristic data threshold value, determining that the equivalent unit configuration result of the corresponding bus is not configured.
Fig. 2 is an exemplary information transfer process diagram of the neural network. As shown in fig. 2, wherein bus node 1 is at the kth layer of characteristic dataCombining characteristic data of bus nodes 2,3,4 connected thereto>Characteristic data of bus node 1 in k+1 layer is formed by aggregation +.>The convolutional layer of the graph neural network learns the characteristics of data by weighted average of busbar nodes and other busbar nodes connected with the busbar nodes through weight matrix coefficients.
Taking the bus node v as an example, the following formula can be adopted to calculate the characteristic data of the bus node v in the k+1 layer
Wherein,characteristic data of the bus node v at the k+1th layer; />Characteristic data of the bus node v at the kth layer; omega is a busbar node connected with busbar node v; n (v) is a set of other bus nodes connected to bus node v; />Characteristic data of other bus nodes connected with the bus node v; aggregate (k) For characteristic data of bus bar node and other bus bar nodes connected to bus bar node +.>Is a weighted average of the results of (a). Update (k) The characteristic data of the k layer is convolved and transferred to the k+1 layer of the graphic neural network in a mode of multiplying the characteristic data by a weight matrix coefficient system of the graphic neural network.
The characteristic data of the bus node v at the k+1 layer can also be expressed by adopting the following formulaIs calculated according to the following steps:
wherein,characteristic data of the bus node v at the k+1th layer; w (W) k+1 The weight matrix coefficients transferred to the (k+1) th layer for the (k) th layer; omega being connected to busbar node vA bus node; n (v) is a set of other bus nodes connected to bus node v; />Characteristic data of other bus nodes connected with the bus node v; c ω,v Is a normalization coefficient of other bus bar nodes connected with the bus bar node.
The equivalent unit configuration result Y of each corresponding bus can be output through the output layer of the graph neural network. Taking bus node v as an example, outputting equivalent unit configuration result Y of corresponding bus output by the layer v The characteristic data of the output of the last, i.e. nth, convolution layerAnd (5) determining.
The following formula can be adopted to represent the equivalent unit configuration result Yv of the bus corresponding to the bus node v:
wherein Y is v =1 may represent that the configuration result of the equivalent unit of the bus corresponding to the bus node v is the configured equivalent unit; y is Y v =0 may represent that the equivalent unit configuration result of the bus corresponding to the bus node v is an unconfigured equivalent unit;characteristic data representing an output of the nth convolution layer; epsilon is a preset characteristic data threshold.
According to the scheme, the equivalent unit configuration result of the bus is determined through the graph neural network, so that the processing of measurement data of real-time change in the power system is realized, a complex power distribution network structure is considered, and the efficiency and accuracy of the equivalent unit configuration result are further improved.
In an optional embodiment of the present invention, when configuring the equivalent unit for the bus based on the configuration result of the equivalent unit, and obtaining the device power in the power system, the method further includes: acquiring a device voltage measurement in a power system; after performing state estimation on a device in a power system and calculating a device voltage estimated value of the device, the method comprises the following steps: calculating a device voltage difference between the device voltage estimate and the device voltage measurement; and when the voltage difference value of the equipment is greater than or equal to a preset voltage difference value threshold value, sending out measurement error information of the equipment so as to prompt that the corresponding equipment power, equipment voltage measurement value or equipment connection relation of the equipment has errors.
The device voltage measurement may be a measured voltage value of a device in the power system. The device voltage estimate may be a voltage estimate of a device in the power system. The device voltage difference may be a difference between the device voltage estimate and the device voltage measurement. The preset voltage difference threshold may be a maximum value of a preset device voltage difference. The preset voltage difference threshold may be used to characterize the accuracy of the device voltage estimate. The measurement error information may be used to indicate an error in measurement data of devices in the power system. Alternatively, the measurement error information may be in the form of speech, text, or graphics. The measurement error information may be, for example, "measurement error exists for measurement data of XX device in Power System-! "wherein the measurement data may comprise device power, device voltage measurements or device connection relationships.
According to the method, after the state estimation is carried out on the equipment in the power system, the equipment voltage difference value between the equipment voltage estimation value and the equipment voltage measurement value is calculated, and when the equipment voltage difference value is greater than or equal to the preset voltage difference value threshold, measurement error information of the equipment is sent out to prompt that errors exist in equipment power, equipment voltage measurement value or equipment connection relation corresponding to the equipment, feedback of measurement data of the equipment in the power system is achieved, correction of the equipment and equipment connection relation in the power system is further achieved, and accuracy of state estimation of the power system is further improved.
Example two
Fig. 3 is a flowchart of a state estimation method based on a graph neural network according to a second embodiment of the present invention. Based on the embodiment, the embodiment of the invention embodies a bus branch diagram for constructing a power system as acquiring the equipment connection relation of equipment in the power system; according to the equipment connection relation, determining at least one node equipment belonging to the same bus and the same bus as a bus node; determining a branch device communicated with the bus as a branch; according to the equipment connection relation, the bus nodes are connected with the communicated branches to generate a bus branch diagram' of the power system, so that real-time acquisition of equipment data in the power system is realized, the complex structure of the power system is considered, the construction efficiency and the construction accuracy of the bus branch diagram are improved, and the comprehensiveness of data relied on by state estimation of the power system is enhanced. In the embodiments of the present invention, the descriptions of other embodiments may be referred to in the portions not described in detail.
Referring to fig. 3, the state estimation method based on the graph neural network includes:
s310, acquiring the device connection relation of the devices in the power system.
The devices in the power system may include bus bars, node devices, and branch devices. The device connection relationship may be a connection relationship between a bus bar, a node device, and a tributary device. The device connection relationship may be a connection or a connection.
Specifically, the device connection relation of the devices in the power system can be obtained in real time.
S320, determining at least one node device belonging to the same bus and the same bus as a bus node according to the device connection relation.
The bus bars may be used to power devices in a power system. The bus bar may be connected to a node device or a branch device. The voltages of the node devices belonging to the same bus are the same. All node devices belonging to the same bus can be connected in parallel on the same bus. The buses can be connected through a branching device. The voltages between the different bus bars may be the same or different. For example, the node device may include at least one of a generator set, a load set, a capacitor, a reactor, and a compensator.
Specifically, according to the device connection relationship, at least one node device connected in parallel to the same bus is determined to be at least one node device belonging to the same bus. At least one node device belonging to the same bus and the same bus can be determined together as a bus node.
S330, determining the branch equipment communicated with the bus as a branch.
The branching devices may be used to connect the various bus bars. The bypass device may include a bypass device in communication with the bus bar and a bypass device disconnected from the bus bar. By way of example, the branching device may include at least one of a transmission line, a transformer, and a switch.
Specifically, the branch device that communicates with the bus may be determined as a branch in the bus branch diagram.
And S340, connecting the bus nodes with the communicated branches according to the equipment connection relation, generating a bus branch diagram of the power system, and calculating the active power unbalance of at least one bus in the power system.
Specifically, according to the equipment connection relationship, the bus node and the corresponding connected branch are connected to generate a bus branch diagram of the power system. The device active power of the input bus and the device output power of the bus output can be obtained, and the difference value between the device active power of the input bus and the device output power of the bus output is calculated, so that the active power unbalance of at least one bus in the power system is obtained.
S350, inputting the bus branch diagram and the active power unbalance of each bus node into a graph neural network, and determining the equivalent unit configuration result of each corresponding bus.
S360, configuring an equivalent unit for the bus based on the equivalent unit configuration result, and acquiring the equipment power in the electric power system.
And S370, carrying out state estimation on equipment in the power system according to the equipment power, and calculating an equipment voltage estimated value of the equipment.
Optionally, analysis may be performed on the distribution network model file in the power system. The distribution network equipment model can be built in the graph database in real time. The distribution network equipment model may be used to store measurement data of the distribution network. Node branch diagrams of equipment such as a transmission line model, a transformer model, a BUS model (BUS), a generator set model (Unit), a Load set model (Load), a capacitance model, a reactance model, a compensator model, a switch model and the like can be respectively built in a diagram database. The transmission line model may be, for example, an alternating-current transmission line model (ac line_dot). The Transformer model may be a Three-winding Transformer model (three_port_transformer) or a double-gang Transformer model (two_port_transformer). The capacitor model and the reactor model may be a shunt capacitance reactor model (c_p). The compensator model may be a series compensator model (c_s). The switch model may include a switch (break) and a knife switch (disconnect). In the graph database, a node branch graph G (V, E) of the power system may be defined by nodes (vertex, V) and edges (Edge, E). The node V may be defined by devices such as generator sets, load sets, capacitors, reactors, compensators, etc.; by defining the transmission line, the transformer, the switch, etc. as the edge E, a node branching diagram composed of the node V and the edge E can be generated. At least one node belonging to the same bus and the same bus can be determined as one bus node, and a bus branch diagram is further generated. Illustratively, fig. 4 is a bus bar branching diagram of an electrical power system. As shown in fig. 4, the bus bar branching diagram includes 5 bus bar nodes. Each bus node in the figure represents a bus, and each side in the figure represents a device connection relationship between the buses. The amount of active power imbalance of the bus may be stored as parameter information in attribute information of bus nodes and branches. The data of x1, x2, x3 … and the like represent a data set of equivalent unit configuration results of a bus needing to be determined by an input Graph Neural Network (GNN).
According to the technical scheme, the bus branch diagram for constructing the power system is embodied into the equipment connection relation for acquiring equipment in the power system, at least one node equipment belonging to the same bus and the same bus are determined to be a bus node according to the equipment connection relation, the branch equipment communicated with the bus is determined to be a branch, the bus node is connected with the communicated branch according to the equipment connection relation, the bus branch diagram of the power system is generated, the equipment connection relation in the power system is utilized to form the bus branch diagram, real-time acquisition of equipment data in the power system is realized, the complex structure of the power system is considered, the construction efficiency and the construction accuracy of the bus branch diagram are improved, and the comprehensiveness of data depending on state estimation of the power system is enhanced.
In an alternative embodiment of the invention, the node device comprises at least one of a generator set, a load set, a capacitor, a reactor and a compensator; the bypass device includes at least one of a transmission line, a transformer, and a switch.
The genset may be used to generate device active power. The generator set may input device active power to the bus. The load set may be used to consume plant active power. The load unit can receive the equipment active power output by the bus. The capacitor can be used for storing electric energy and has the characteristics of charging and discharging, and alternating current and direct current blocking. The capacitor may input device active power to the bus or receive active power output by the bus. The reactor may be used to limit short-circuit currents or to limit higher harmonics in the grid. The reactor may input active power to the bus or receive active power output from the bus. The compensator may input active power to the bus or receive active power output by the bus. The transmission line may be used to transmit electrical energy. Transformers may be used to change the voltage level in the power grid. A switch may be used to switch the line on and off.
According to the scheme, the node equipment is embodied into at least one of a generator set, a load set, a capacitor, a reactor and a compensator, the branch equipment is embodied into at least one of a transmission line, a transformer and a switch, the screening process of the node equipment and the branch equipment is simplified, and the construction efficiency of a bus branch diagram is further improved.
In an alternative embodiment of the present invention, calculating the active power imbalance amount of at least one bus node includes: acquiring equipment connection relation and equipment active power of equipment in an electric power system; determining each device connected with the bus according to the device connection relation; and calculating the equipment active power of each equipment connected with the bus, and determining the active power unbalance of the bus as the active power unbalance of the bus node.
Specifically, the device connection relation and the device active power of the devices in the power system can be obtained. Each device connected to the bus can be detected based on the device connection relationship. And calculating the equipment active power of each equipment connected with the bus, and calculating the difference between the equipment active power of the input bus and the equipment active power output by the bus to obtain the active power unbalance of the bus as the active power unbalance of the bus node.
According to the scheme, the equipment connection relation and the equipment active power of the equipment in the power system are obtained, all the equipment connected with the bus is determined according to the equipment connection relation, the equipment active power of all the equipment connected with the bus is calculated, the active power unbalance of the bus is determined, and the calculated efficiency and the calculated accuracy of the active power unbalance of the bus are further improved as the active power unbalance of the bus node.
In an alternative embodiment of the present invention, the device active power includes at least one unit device active power corresponding to each unit time in the time period to be measured; the active power unbalance amount comprises at least one unit active power unbalance amount corresponding to each unit moment in the time period to be measured.
The time period to be measured may be a time period in which a state estimation of the power system is made. Alternatively, the time period to be measured may be a time period including the current time, so as to implement real-time state estimation of the power system. The cell time may be the time at which the data sampling is performed. Alternatively, the data may be sampled at preset time intervals during the time period to be measured. For example, the time period to be measured may be 24 hours, and the preset time interval may be 15 minutes. The active power of the unit equipment can be the active power of the equipment corresponding to the unit time. The cell active power imbalance amount may be an active power imbalance amount corresponding to a cell time.
Specifically, the device connection relationship of devices in the power system and at least one active power of a unit device corresponding to each unit time in the time period to be measured can be obtained. Each device connected to the bus can be detected based on the device connection relationship. And calculating the active power of at least one unit device corresponding to each unit time in the time period to be measured of each device connected with the bus, and calculating the difference between the active power of at least one unit device corresponding to each unit time in the time period to be measured of the input bus and the active power of at least one unit device corresponding to each unit time in the time period to be measured of the bus output, thereby obtaining the imbalance of the active power of at least one unit corresponding to each unit time in the time period to be measured of the bus.
Exemplary, FIG. 5 is a schematic diagram of a neural network for determining equivalent unit configuration results for each busbar. As shown in fig. 5, the structure of the graph neural network is given. The bus branch diagram and the corresponding data sets x1, x2 and x3 … (i.e. the active power unbalance of the bus at each unit moment in the time period to be measured) can be used as the input of the graph neural network, and the output Y is the classification prediction of the equivalent unit configuration result of each corresponding bus after the processing of the hidden layer of the graph neural network (Graph Neural Network and GNN). The input data of the neural network is the active measurement imbalance of each bus node, for example, the input data set { x1, x2, x3, … } of the bus node 1 represents the active measurement imbalance of the bus node at different unit moments { t1, t2, t3, … }, respectively. The output Y of the graph neural network is a classification identifier of the bus node and is used for identifying whether the corresponding bus is provided with an equivalent unit or not.
According to the method, the device active power is embodied into the active power of the at least one unit device corresponding to each unit time in the time period to be measured, the active power unbalance is embodied into the active power unbalance of the at least one unit corresponding to each unit time in the time period to be measured, and the calculation efficiency and the calculation accuracy of the active power unbalance of the bus are further improved.
Example III
Fig. 6 is a schematic structural diagram of a state estimation device based on a neural network according to a third embodiment of the present invention. The embodiment of the invention is applicable to the condition of carrying out state estimation on the power systemization based on the graph neural network, the device can execute a state estimation method based on the graph neural network, the device can be realized in a form of hardware and/or software, and the device can be configured in an electronic device carrying a state estimation function based on the graph neural network.
Referring to fig. 6, a state estimation apparatus based on a graph neural network includes: the system comprises a bus bar diagram construction module 610, an equivalent unit configuration result determination module 620, a device power acquisition module 630 and a state estimation module 640. The bus bar diagram construction module 610 is configured to construct a bus bar diagram of the power system, and calculate an active power unbalance amount of at least one bus node; the equivalent unit configuration result determining module 620 is configured to input the bus branch diagram and the active power unbalance of each bus node into the graph neural network, and determine the corresponding equivalent unit configuration result of each bus; the device power obtaining module 630 is configured to configure an equivalent unit for a bus based on the configuration result of the equivalent unit, and obtain device power in the power system; the state estimation module 640 is configured to perform state estimation on a device in the power system according to the device power, and calculate a device voltage estimated value of the device.
According to the technical scheme, the bus branch diagram of the power system is constructed, the active power unbalance of at least one bus node is calculated, the bus branch diagram and the active power unbalance of each bus node are input into the graph neural network, the corresponding equivalent unit configuration result of each bus is determined, the equivalent unit is configured on the basis of the equivalent unit configuration result, the equipment power in the power system is obtained, the equipment in the power system is subjected to state estimation according to the equipment power, the equipment voltage estimation value of the equipment is calculated, the characteristics of real-time data change and complex structures in the power system can be adapted by using the graph neural network, the equivalent unit configuration result of the power system is determined, the problem that the existing equivalent unit configuration mode is low in configuration efficiency, the problem that the state estimation is only converged can be solved, the problem that the active power measurement in the system is unbalanced can not be solved globally is solved, the convergence of the state estimation is still difficult to ensure, the configuration efficiency of the equivalent unit is improved, the problem that the active power measurement is unbalanced is solved globally, and the convergence of the state estimation is ensured.
In an alternative embodiment of the present invention, the bus bar branching diagram construction module 610 includes: the device connection relation acquisition unit is used for acquiring the device connection relation of the devices in the power system; the same bus node determining unit is used for determining at least one node device belonging to the same bus and the same bus as a bus node according to the device connection relation; a branch determining unit configured to determine a branch apparatus communicating with the bus as a branch; and the bus branch diagram construction unit is used for connecting the bus nodes with the communicated branches according to the equipment connection relation to generate a bus branch diagram of the power system.
In an alternative embodiment of the invention, the node device comprises at least one of a generator set, a load set, a capacitor, a reactor and a compensator; the bypass device includes at least one of a transmission line, a transformer, and a switch.
In an alternative embodiment of the present invention, the bus bar branching diagram construction module 610 includes: the equipment active power acquisition unit is used for acquiring the equipment connection relation and the equipment active power of equipment in the power system; the bus connection equipment determining unit is used for determining each equipment connected with the bus according to the equipment connection relation; the active power unbalance amount calculating unit is used for calculating the device active power of each device connected with the bus and determining the active power unbalance amount of the bus as the active power unbalance amount of the bus node.
In an alternative embodiment of the present invention, the device active power includes at least one unit device active power corresponding to each unit time in the time period to be measured; the active power unbalance amount comprises at least one unit active power unbalance amount corresponding to each unit moment in the time period to be measured.
In an alternative embodiment of the present invention, the equivalent unit configuration result determining module 620 includes: the data input unit is used for inputting the bus node branch diagram and the active power unbalance of each bus node into the graphic neural network; the first convolution layer weighting unit is used for carrying out weighted average on the active power unbalance of each busbar node and the active power unbalance of other busbar nodes connected with the busbar node through the first convolution layer of the graph neural network to obtain a weighted average result of the first convolution layer; the first convolution layer data output unit is used for calculating the product between the weighted average result of the first convolution layer and the weight matrix coefficient of the first convolution layer through the first convolution layer of the graph neural network to obtain the characteristic data output by the busbar node in the first convolution layer; the other convolution layer weighting unit is used for carrying out weighted average on the characteristic data output by the previous convolution layer by the bus node and the characteristic data output by the previous convolution layer by other bus nodes connected with the bus node aiming at each bus node through other convolution layers of the graph neural network to obtain weighted average results of other convolution layers; the other convolution layer data output unit is used for calculating products between weighted average results of other convolution layers and weight matrix coefficients of other convolution layers through other convolution layers of the graph neural network to obtain characteristic data output by the busbar node in the other convolution layers until the characteristic data output by the last convolution layer are obtained; the preset characteristic data threshold value comparison unit is used for comparing the characteristic data output by the last convolution layer with a preset characteristic data threshold value; the first equivalent unit configuration result determining unit is used for determining that the equivalent unit configuration result of the corresponding bus is the configuration equivalent unit when the characteristic data output by the last convolution layer is greater than or equal to a preset characteristic data threshold value; and the second equivalent unit configuration result determining unit is used for determining that the equivalent unit configuration result of the corresponding bus is not configured as the equivalent unit when the characteristic data output by the last convolution layer is smaller than the preset characteristic data threshold value.
In an alternative embodiment of the invention, the apparatus further comprises: the equipment voltage measurement value acquisition module is used for configuring the equivalent unit for the bus based on the configuration result of the equivalent unit, acquiring equipment power in the power system and simultaneously acquiring equipment voltage measurement values in the power system; the device voltage difference calculation module is used for calculating a device voltage difference between the device voltage estimated value and the device voltage measured value after carrying out state estimation on the device in the power system and calculating the device voltage estimated value of the device; and the measurement error information prompting module is used for sending out measurement error information of the equipment when the equipment voltage difference value is greater than or equal to a preset voltage difference value threshold value so as to prompt that the equipment power, the equipment voltage measured value or the equipment connection relation corresponding to the equipment has errors.
The state estimation device based on the graph neural network provided by the embodiment of the invention can execute the state estimation method based on the graph neural network provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
In the technical scheme of the embodiment of the invention, the acquisition, storage, application and the like of the equipment power in the related power system, the equipment connection relation of the equipment in the power system, the equipment active power, the equipment voltage measured value in the power system and the like all conform to the regulations of related laws and regulations, and the public order is not violated.
Example IV
Fig. 7 shows a schematic diagram of an electronic device 700 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 7, the electronic device 700 includes at least one processor 701, and a memory, such as a Read Only Memory (ROM) 702, a Random Access Memory (RAM) 703, etc., communicatively connected to the at least one processor 701, in which the memory stores a computer program executable by the at least one processor, and the processor 701 may perform various suitable actions and processes according to the computer program stored in the Read Only Memory (ROM) 702 or the computer program loaded from the storage unit 708 into the Random Access Memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the electronic device 700 may also be stored. The processor 701, the ROM 702, and the RAM 703 are connected to each other through a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.
Various components in the electronic device 700 are connected to the I/O interface 705, including: an input unit 706 such as a keyboard, a mouse, etc.; an output unit 707 such as various types of displays, speakers, and the like; a storage unit 708 such as a magnetic disk, an optical disk, or the like; and a communication unit 709 such as a network card, modem, wireless communication transceiver, etc. The communication unit 709 allows the electronic device 700 to exchange information/data with other devices through a computer network, such as the internet, and/or various telecommunication networks.
The processor 701 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of processor 701 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 701 performs the various methods and processes described above, such as a state estimation method based on a graph neural network.
In some embodiments, the state estimation method based on the graph neural network may be implemented as a computer program, which is tangibly embodied in a computer-readable storage medium, such as the storage unit 708. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 700 via the ROM 702 and/or the communication unit 709. When the computer program is loaded into RAM 703 and executed by processor 701, one or more steps of the state estimation method based on a graph neural network described above may be performed. Alternatively, in other embodiments, the processor 701 may be configured to perform the state estimation method based on the graph neural network in any other suitable manner (e.g., by means of firmware).
Various implementations of the systems and techniques described here above can be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end 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: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical host and VPS (Virtual Private Server ) service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. A state estimation method based on a graph neural network, the method comprising:
constructing a bus branch diagram of the power system, and calculating the active power unbalance of at least one bus node;
inputting the bus branch diagram and the active power unbalance of each bus node into a graph neural network, and determining the equivalent unit configuration result of each corresponding bus;
based on the configuration result of the equivalent unit, configuring the equivalent unit for the bus and acquiring the equipment power in the electric power system;
And carrying out state estimation on equipment in the power system according to the equipment power, and calculating an equipment voltage estimated value of the equipment.
2. The method of claim 1, wherein said constructing a bus bar branching map of a power system comprises:
acquiring a device connection relation of devices in a power system;
according to the equipment connection relation, determining at least one node equipment belonging to the same bus and the same bus as a bus node;
determining a branch device communicated with the bus as a branch;
and connecting the bus node with the communicated branch according to the equipment connection relation to generate a bus branch diagram of the power system.
3. The method of claim 2, wherein the node device comprises at least one of a generator set, a load set, a capacitor, a reactor, and a compensator; the branching device includes at least one of a transmission line, a transformer, and a switch.
4. The method of claim 1, wherein said calculating an amount of active power imbalance for at least one bus node comprises:
acquiring equipment connection relation and equipment active power of equipment in an electric power system;
Determining each device connected with the bus according to the device connection relation;
and calculating the equipment active power of each equipment connected with the bus, and determining the active power unbalance of the bus as the active power unbalance of a bus node.
5. The method of claim 4, wherein the device active power comprises at least one unit device active power corresponding to each unit time instant within a time period to be measured; the active power unbalance amount comprises at least one unit active power unbalance amount corresponding to each unit moment in the time period to be measured.
6. The method according to claim 1, wherein the inputting the bus branch graph and the active power unbalance of each bus node into the graph neural network, determining the equivalent unit configuration result of each corresponding bus, includes:
inputting the bus branch diagram and the active power unbalance of each bus node into a graph neural network;
the method comprises the steps that through a first convolution layer of the graph neural network, for each busbar node, the active power unbalance of the busbar node and the active power unbalance of other busbar nodes connected with the busbar node are weighted and averaged, and a weighted average result of the first convolution layer is obtained;
Calculating the product between the weighted average result of the first convolution layer and the weight matrix coefficient of the first convolution layer through the first convolution layer of the graph neural network to obtain the characteristic data output by the busbar node at the first convolution layer;
through other convolution layers of the graph neural network, for each busbar node, carrying out weighted average on characteristic data output by the last convolution layer by the busbar node and characteristic data output by the last convolution layer by other busbar nodes connected with the busbar node, so as to obtain a weighted average result of the other convolution layers;
calculating products between weighted average results of other convolution layers and weight matrix coefficients of the other convolution layers through other convolution layers of the graph neural network to obtain characteristic data output by the busbar node at the other convolution layers until the characteristic data output by the last convolution layer is obtained;
comparing the characteristic data output by the last convolution layer with a preset characteristic data threshold value;
when the characteristic data output by the last convolution layer is greater than or equal to a preset characteristic data threshold value, determining an equivalent unit configuration result of the corresponding bus as a configuration equivalent unit;
And when the characteristic data output by the last convolution layer is smaller than a preset characteristic data threshold value, determining that the equivalent unit configuration result of the corresponding bus is an unconfigured equivalent unit.
7. The method of claim 1, wherein configuring the equivalent unit for the bus based on the equivalent unit configuration result and obtaining the device power in the power system, further comprises:
acquiring a device voltage measurement in a power system;
after performing state estimation on equipment in the power system and calculating an equipment voltage estimated value of the equipment, the method comprises the following steps:
calculating a device voltage difference between the device voltage estimate and the device voltage measurement;
and when the equipment voltage difference value is greater than or equal to a preset voltage difference value threshold value, sending measurement error information of the equipment to prompt that errors exist in the equipment power, the equipment voltage measurement value or the equipment connection relation corresponding to the equipment.
8. A state estimation device based on a graph neural network, the device comprising:
the bus branch diagram construction module is used for constructing a bus branch diagram of the power system and calculating the active power unbalance of at least one bus node;
The equivalent unit configuration result determining module is used for inputting the bus branch diagram and the active power unbalance of each bus node into a graphic neural network and determining the corresponding equivalent unit configuration result of each bus;
the equipment power acquisition module is used for configuring an equivalent unit for the bus based on the equivalent unit configuration result and acquiring equipment power in the electric power system;
and the state estimation module is used for carrying out state estimation on equipment in the power system according to the equipment power and calculating an equipment voltage estimated value of the equipment.
9. An electronic device, the electronic device comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the graph neural network based state estimation method of any one of claims 1-7.
10. A computer readable storage medium storing computer instructions for causing a processor to implement the state estimation method based on a graph neural network of any one of claims 1-7 when executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311427159.7A CN117422260A (en) | 2023-10-30 | 2023-10-30 | State estimation method, device, equipment and medium based on graph neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311427159.7A CN117422260A (en) | 2023-10-30 | 2023-10-30 | State estimation method, device, equipment and medium based on graph neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117422260A true CN117422260A (en) | 2024-01-19 |
Family
ID=89522551
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311427159.7A Pending CN117422260A (en) | 2023-10-30 | 2023-10-30 | State estimation method, device, equipment and medium based on graph neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117422260A (en) |
-
2023
- 2023-10-30 CN CN202311427159.7A patent/CN117422260A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103793585B (en) | A kind of method for improving theoretical line loss caluclation integrality and accuracy | |
| CN109613324A (en) | A kind of detection method and device of Harmonics amplification | |
| CN117368588A (en) | Method, device, equipment and storage medium for determining consistency of voltage phase sequence | |
| CN117473384A (en) | Power grid line safety constraint identification method, device, equipment and storage medium | |
| CN114996635A (en) | Power distribution network parameter determination method, device, equipment and storage medium | |
| CN117422260A (en) | State estimation method, device, equipment and medium based on graph neural network | |
| CN114784804A (en) | Power system analysis method and device, electronic equipment and storage medium | |
| CN115598530A (en) | Method and device for evaluating no-load characteristic of generator, electronic equipment and storage medium | |
| CN108092272A (en) | A kind of voltage stabilization on-line monitoring method based on the Kalman filtering that fades | |
| CN114565058A (en) | Training method, device, equipment and medium for island detection model | |
| CN116436057B (en) | Method, device, equipment and medium for determining operation strategy of energy storage station | |
| CN117131315B (en) | Out-of-tolerance electric energy meter determining method and medium based on solving multi-element quadratic function extremum | |
| CN115453271A (en) | Power distribution network problem diagnosis method, device, equipment and storage medium | |
| CN117081094A (en) | Reactive compensation method and device, electronic equipment and storage medium | |
| CN116191405A (en) | Capacitor bank trimming method and device, electronic equipment and storage medium | |
| CN116305800A (en) | Processing method, device, equipment and medium for simulation input data of power distribution network | |
| CN117851853A (en) | Method, device, equipment and storage medium for positioning electricity stealing user | |
| CN114705947A (en) | Island detection model training method, device, equipment and medium | |
| CN116885708A (en) | Method, device, equipment and storage medium for identifying household change relation of station area | |
| CN117035405A (en) | Power system risk index assessment method and device, electronic equipment and medium | |
| CN117791579A (en) | Distributed photovoltaic cluster division optimization method, electronic equipment and storage medium | |
| CN116365523A (en) | Tidal current calculation method, device, equipment and medium applied to power distribution network | |
| CN117647702A (en) | Fault determination method, device and equipment for low-voltage distribution network and storage medium | |
| CN117833208A (en) | Power distribution network state deduction method and device, electronic equipment and storage medium | |
| CN116404650A (en) | Power grid power supply area dividing method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |