A power system modeling method, system, readable storage medium and server

Abstract

A power system modeling method, system, readable storage medium and server, the method comprises: constructing a single device external characteristic model based on a feedback fusion mode; connecting each device to form a power system wiring diagram according to the power grid topology and line parameters; constructing a mechanism data fusion driven power system model based on the single device external characteristic model of each device connected by the power system wiring diagram; and performing system simulation using the fusion driven power system model and online updating according to real-time data. The present application makes full use of the rich physical knowledge accumulated in the development process of the power system and the huge historical data accumulated since the digital construction of the power grid, breaks through the contradiction between complexity and accuracy, the lack of explainability and other shortcomings of a single modeling method, and can construct a power simulation system corresponding to the physical entity through the operation data collected by the sensor, so as to realize faster and more accurate prediction and detection of the faults and problems existing in the physical system.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power system simulation, and particularly relates to a power system modeling method and system, a readable storage medium and a server. BACKGROUND

[0002] Modern power systems have undergone tremendous changes in power source composition, network scale and load characteristics compared with the past. With the access of renewable energy, the flexible use of active loads (such as electric vehicles, etc.), and large-scale regional interconnection, the power grid has evolved into a typical dynamic large system with huge dimensions, having stronger uncertainty, complexity and nonlinearity. In power system related applications, the analysis method based on physical mechanism model has proved its effectiveness. Such a method can better adapt to changing application scenarios by analyzing the problem mechanism, extracting universal rules and implementing control decisions, but the mechanism model also has problems such as the contradiction between complexity and accuracy, and is difficult to meet the operation analysis requirements of the multi-physical field coupled giant system.

[0003] With the construction of digital new infrastructure, data-driven methods have been popularized in power systems due to their low dependence on specific mathematical models and their ability to learn from data and transfer learning from source domains. Such methods help to understand the characteristics of equipment and systems in historical operation by analyzing historical operation data, and support power grid operation situation awareness, evaluation and prediction; but the performance of data-driven methods is highly dependent on data size and quality, and lacks understanding of knowledge and explanation of results, and the generalization ability of the model restricts the applicable scenarios. Therefore, in order to adapt to the actual operation and control requirements of the new generation of power systems with new energy as the main supply, a more optimal power system model construction method needs to be designed. SUMMARY

[0004] The purpose of the present application is to provide a power system modeling method, system, readable storage medium and server to solve the problems in the prior art by fusing mechanism-driven and data-driven methods, realizing the organic combination of global and local characteristics, rules and experience of the problem, and forming a more optimal joint power system modeling method.

[0005] In order to achieve the above purpose, the present application has the following technical solutions:

[0006] In a first aspect, a power system modeling method is provided, comprising:

[0007] Constructing a single device external characteristic model of mechanism data fusion based on a feedback fusion mode;

[0008] Connecting each device to form a power system wiring diagram according to the power grid topology structure and line parameters;

[0009] The single-device external characteristic model is constructed based on the power system wiring diagram connecting each device, and the mechanism data fusion driven power system model is constructed.

[0010] The system simulation is performed by using the fusion driven power system model, and online updating is performed according to real-time data.

[0011] As a preferred scheme of the power system modeling method, the step of connecting each device according to the power grid topology and line parameters to form the power system wiring diagram specifically comprises the following steps:

[0012] The devices in the system are abstracted as nodes, the devices having a direct connection relationship are connected together according to the electrical connection relationship and relative geographical position information between the devices in the actual system, and the power system wiring diagram is formed.

[0013] The impedance and admittance of each branch are calculated according to the line length and conductor type between the nodes, and the equivalent network diagram of the power system and the line parameter admittance matrix G are recorded:

[0014]

[0015] Among them,

[0016]

[0017] As a preferred scheme of the power system modeling method, the step of constructing the single-device external characteristic model based on the feedback fusion mode specifically comprises the following steps:

[0018] A group of relatively independent input physical quantities X and output physical quantities Y are selected according to the device operation mechanism e e The mechanism model Y=f e (X) is constructed according to the device operation principle;

[0019] The calculation result of the mechanism model is obtained by bringing in the historical operation and experimental data x e , and the external environment data v e of the device collected by the sensor is obtained, and the mechanism simulation value of the device operation state is obtained Among them, the influence mode of the environment data According to the different modeling devices, the original data x e , the mechanism model output prediction value , the true value y e , and the error between the mechanism model output prediction value and the true value The correction model g e (x,y,v) is trained by using the data driven method.

[0020] ​The trained correction model is used to correct the parameters or input / output data of the mechanistic model to obtain simulated values ​​of the equipment's operating state. Among them, the correction method Choose according to the different functions of the model.

[0021] As a preferred embodiment of the power system modeling method of the present invention, in the step of constructing a power system model driven by the fusion of the external characteristic model of a single device connected by a power system wiring diagram, the fusion-driven power system model is constructed using a parallel modeling method or a serial modeling method.

[0022] Furthermore, as a preferred embodiment of the power system modeling method of the present invention, the specific steps for constructing the fusion-driven power system model using a parallel modeling approach include:

[0023] Based on the isomorphic network diagram and environmental data, a mechanistic model of the system is constructed according to physical principles. s (G,x s ,v s );

[0024] The data model g is trained using historical system operation data and environmental data. s (x s ,v s The historical operating data of the system includes steady-state operating data and fault status data;

[0025] The value f is calculated using tagged historical data and mechanistic models. s (G,x s ,v s ) and data model g s (x s ,v s The weighted mixed system simulation model was trained using a data-driven approach. Weighting method It is determined based on the training process.

[0026] Furthermore, as a preferred embodiment of the power system modeling method of the present invention, the specific steps for constructing the fusion-driven power system model using a serial modeling approach include:

[0027] Based on the isomorphic network diagram and environmental data, a mechanistic model of the system is constructed according to physical principles, and historical data is used to simplify the calculation of simulation values ​​of a set of mechanistic models.

[0028] Based on mechanism simulation values and the true value y sTraining a data model g that can map the results of mechanistic calculations to true values. s (y)=g s (f s (G,x s ));

[0029] Combined with system operating environment data v s This allows us to obtain simulation values ​​of the system's operating state. Among them, environmental data impact methods Choose according to the different functions of the models.

[0030] As a preferred embodiment of the power system modeling method of the present invention, the step of performing system simulation using a fusion-driven power system model and updating it online based on real-time data specifically includes the following steps:

[0031] The power system model, which integrates and drives various devices, is connected to the inputs and outputs according to the power grid topology to form the final system simulation model. The device models calculate the operating status of the devices based on real-time operating data. Then used as input data Input system model;

[0032] The system simulation model is based on the equipment operating status. Calculate the predicted value of the system operating state If the fusion-driven power system model is constructed using a parallel modeling approach, then the predicted system operating state after substituting the equipment operating states will be... If the fusion-driven power system model is constructed using a sequential modeling approach, then the predicted system operating state value after substituting the device operating states is...

[0033] Using system operating status prediction values The true value y s and the error between the two values Update the system simulation model regularly.

[0034] Secondly, a power system modeling system is provided, including:

[0035] The device external characteristic fusion-driven modeling module is used to construct a single device external characteristic model based on feedback fusion mode and mechanism data fusion.

[0036] The wiring diagram construction module is used to connect various devices according to the power grid topology and line parameters to form a power system wiring diagram.

[0037] The power system fusion-driven modeling module is used to construct a power system model driven by the fusion of mechanism data based on the external characteristic models of individual devices connected by the power system wiring diagram.

[0038] a model updating module, configured to perform system simulation by using the fusion-driven power system model and perform online updating according to real-time data.

[0039] In a third aspect, a computer readable storage medium is provided, which stores a computer program, and the computer program, when executed by a processor, implements the power system modeling method in the first aspect.

[0040] In a fourth aspect, a server is provided, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the power system modeling method in the first aspect when executing the computer program.

[0041] Compared with the prior art, the first aspect of the present application has at least the following beneficial effects: by decomposing a complex power system into single devices and a complex power system formed by interconnection of the devices, connecting the devices according to the power grid topology and line parameters to form a power system wiring diagram, and constructing a fusion-driven power system model based on the single device external characteristic model of each device connected according to the power system wiring diagram, the modeling process of the data mechanism fusion model of the device and the system is given for different task requirements of the local and the whole, system simulation is performed by using the fusion-driven power system model, and online updating is performed according to real-time data. The power system simulation model constructed by the present application makes full use of the rich physical knowledge accumulated in the development process of the power system and the huge historical data accumulated since the digital construction of the power grid, breaks through the contradiction between complexity and accuracy, the lack of explainability and other shortcomings of a single modeling method, and can construct a power simulation system corresponding to a physical entity by using the operation data collected by a sensor, thereby providing a new way for faster and more accurate prediction and detection of faults and problems existing in the physical system.

[0042] It can be understood that the beneficial effects of the second aspect to the fourth aspect can be referred to the related description in the first aspect, which will not be repeated here. BRIEF DESCRIPTION OF DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0044] Figure 1 The power system modeling method flowchart of the present application;

[0045] Figure 2 The feedback mechanism-data fusion model construction flowchart of the present application;

[0046] Fig. 3(a) is a schematic diagram of system topology in a power system wiring diagram formed by the present application;

[0047] Fig. 3(b) is a schematic diagram of system equivalent network in a power system wiring diagram formed by the present application;

[0048] Figure 4 A parallel mechanism-data fusion model construction flowchart of the present application;

[0049] Figure 5 A serial mechanism-data fusion model construction flowchart of the present application;

[0050] Figure 6 A structural block diagram of a power system modeling system of the present application. DETAILED DESCRIPTION

[0051] In the following description, for the purpose of explanation and not limitation, specific details are set forth, such as particular system configurations, techniques, etc., in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.

[0052] In addition, in the description of the present application and the appended claims, the terms "first", "second", "third", etc. are used only to distinguish descriptions, and cannot be understood as indicating or implying relative importance.

[0053] Embodiment 1

[0054] Referring to Figure 1 , a power system modeling method according to an embodiment of the present application includes the following steps:

[0055] S1, constructing a single device external characteristic model of mechanism-data fusion based on a feedback fusion mode;

[0056] S2, connecting each device to form a power system wiring diagram according to the power grid topology structure and line parameters;

[0057] S3, constructing a mechanism-data fusion driven power system model based on the single device external characteristic model of each device connected by the power system wiring diagram;

[0058] S4, using the fusion driven power system model to perform system simulation and online updating according to real-time data.

[0059] Since the physical mechanism of the equipment is usually clear, only the parameters in the mechanism-driven model need to be corrected by the data-driven model to better fit the nonlinear or uncertain parts of the equipment characteristics that are difficult to represent. The modeling objects include generators, transformers, energy storage batteries, etc., and the feedback fusion modeling method can be used.

[0060] To distinguish from the system model variables, the following subscript "e" (equipment) refers to the equipment model-related parameters:

[0061] In one embodiment, the single equipment external characteristic model based on the feedback fusion model is constructed by the following steps:

[0062] First, a set of relatively independent input physical quantities X e (including excitation current, winding turns, voltage, current, etc.) and output physical quantities Y e (including voltage, current, power, etc.) are selected according to the operating mechanism of the equipment, and a mechanism model Y = f e (X) is constructed according to the operating principle of the equipment (for example, the electromagnetic equation of a synchronous generator);

[0063] Then, the historical operation and experimental data x e are brought in to obtain the calculation results of the mechanism model, and combined with the equipment external environment data v e (including meteorological data, geographical data, time data, etc.) to obtain the mechanism simulation value of the equipment operating state The influence of the environment data is According to the different modeling equipment, the appropriate method is selected. According to the original data x e , the mechanism model output prediction value the true value y e and the error between the two values The correction model g e (x, y, v) is trained by using the data-driven method;

[0064] Finally, the trained correction model is used to correct the parameters or input and output data of the mechanism model, so as to obtain the simulation value of the equipment operating state The correction method is According to the different effects of the model, the appropriate method is selected.

[0065] In one embodiment, the power system wiring diagram is formed by connecting each equipment according to the power grid topology structure and line parameters, which specifically includes the following steps:

[0066] First, the equipment in the system is abstracted as a node, and according to the electrical connection relationship and relative geographical position information between each equipment in the actual system, the equipment with direct connection relationship is connected together to form a power system wiring diagram.

[0067] Then, according to the information of line length between nodes, wire type and the like, the impedance and admittance of each branch are calculated, and the equivalent network diagram of the power system is drawn, and the line parameters are recorded in the admittance matrix G:

[0068]

[0069] wherein,

[0070]

[0071] On the basis of the system network diagram, the model needs to be optimized according to the operation mechanism of the entire power system. Due to the complexity of the system operation state, a large number of simplifications are made in the traditional mechanism modeling process to construct a solvable mathematical model. In order to obtain more accurate simulation values, the neglected parameters need to be considered. In order to control the model calculation amount, the parallel or serial mode fusion modeling method can be used to improve the data-driven model, and the correlation between the simplified calculation results and the true situation is explored. In order to distinguish from the device model variables, the following subscript "s" (system) refers to the system model related parameters.

[0072] In an embodiment, referring to Figure 4 , the specific steps of constructing the mechanism data fusion driven power system model in a parallel modeling mode include:

[0073] Firstly, based on the equivalent network diagram and environmental data, the mechanism model f s (G, x s , v s ) of the system is analyzed and constructed according to the physical principle (such as the power flow equation);

[0074] Then, the data model g s (x s , v s ) is trained by using the system historical operation data (including steady state operation data and fault state data) and environmental data (such as the power system stability evaluation model);

[0075] Finally, the weighted mixed system simulation model is trained by using the labeled historical data, the mechanism model calculation value f s (G, x s , v s ) and the data model output result g s (x s , v s ) in a data-driven manner. The weighting method is determined according to the training process.

[0076] In an embodiment, referring toFigure 5 The specific steps for constructing the power system model driven by the mechanistic data fusion using a sequential modeling approach include:

[0077] First, based on the iso-network diagram and environmental data, a mechanistic model of the system is constructed according to physical principles. Then, historical data is used to simplify the calculation of simulation values ​​for a set of mechanistic models.

[0078] Then, based on the mechanism simulation values and the true value y s Training a data model g that maps the results of mechanistic calculations to true values. s (y)=g s (f s (G,x s ));

[0079] Finally, combining the system operating environment data collected by the sensors... s This allows us to obtain simulation values ​​of the system's operating state. Among them, environmental data impact methods Choose the appropriate method based on the different functions of the model.

[0080] In one implementation, the process of using a fusion-driven power system model for system simulation and updating it online based on real-time data specifically includes the following steps:

[0081] The power system model, which integrates and drives various devices, is connected to the inputs and outputs according to the power grid topology to form the final system simulation model. The device models calculate the operating status of the devices based on real-time operating data. That is, the input data x of the system model s ;

[0082] The system simulation model calculates the predicted system operating status based on the equipment operating status. If the integrated-drive power system model is constructed using a parallel modeling approach, the predicted system operating status will be... If the fusion-driven power system model is constructed using a sequential modeling approach, the predicted system operating state values ​​will be...

[0083] Using system operating status prediction values The true value y s and the error between the two values Update the system simulation model regularly.

[0084] This invention utilizes accumulated physical knowledge and data assets to create one-to-one models of each physical entity in the actual power system, simulating their appearance, function, and state during actual operation. It overcomes the problems of high computational complexity, difficulty in coping with scene changes, and poor interpretability that may exist in single modeling methods, and provides a power system simulation model with higher accuracy and faster computation.

[0085] Example 2

[0086] Another embodiment of the present invention provides a power system modeling method, comprising the following steps:

[0087] S1. Constructing a single device external characteristic model based on feedback-based fusion mode for mechanistic data fusion:

[0088] See Figure 2 The feedback mechanism data fusion modeling process is shown.

[0089] ① Construct a power source model using a photovoltaic power generation system as an example.

[0090] The electrical energy generated by photovoltaic (PV) cells connected in series and parallel is transmitted to the power grid through an inverter and corresponding filters. Therefore, the mechanism part of the external characteristic model of the entire PV power generation system should include the PV cell UI characteristic model, the equivalent of series and parallel cell connections, the inverter control model, the MPPT control model, and numerical weather prediction data. Based on physical principles, the analytical expression of the PV output curve can be obtained as follows:

[0091]

[0092] Since some parameters in the expression are difficult to determine and cannot be provided by the supplier, simplified models are generally used in engineering applications:

[0093]

[0094] The series and parallel connection of batteries is summed according to the actual structure. Furthermore, when physically modeling the inverter, losses are generally not considered (i.e., it is independent of the specific structure of the inverter). The model only needs to reflect the proportional mapping relationship between input and output, and the maximum output power is ensured by solving dP / dU = 0 (the PU relationship can be obtained from the UI characteristics of the photovoltaic cells). Based on the above physical model, a preliminary physical model of the photovoltaic power station can be constructed. solar (x).

[0095] Then, since some parameters in the model are related to dynamic variables such as light intensity and incident angle, a data-driven approach can be used based on f. solar (I m ,I sc U m U oc ) and meteorological data vweather Training the correction model g solar (x, y, v) to obtain the model parameters applicable to different external conditions, and then obtain the output prediction value of the photovoltaic power station

[0096] ②Take the star connection method as an example to construct a three-phase transformer model. The basic equation of transformer no-load closing is as follows:

[0097]

[0098] After introducing the Kirchhoff law and making some approximations to the transformer leakage reactance, the equivalent inductance, core cross-sectional area, magnetic circuit length, and number of turns of the transformer are introduced. The simplified physical model f of the transformer can be obtained trans (x). As above, the parameters of the mechanism model are adjusted using a data-driven method to obtain the correction model g trans (x, y, v) and the simulation value

[0099] ③The system load has no fixed mode, and a pure data-driven method can obtain its prediction model.

[0100] S2, connect each device to form a power system wiring diagram according to the power grid topology and line parameters:

[0101] Referring to the simple power system topology and system equivalent network diagram shown in FIGS. 3(a) and 3(b), wherein G-1, G-2 are substituted into the hybrid model of the photovoltaic power station, T-1, T-2 are substituted into the star-connected three-phase transformer model, and the power flow points of nodes 2, 3, and 5 are substituted into the load data model trained by historical data. The admittance matrix of the system in the figure is as follows:

[0102]

[0103] S3, construct a mechanism data fusion driven power system model based on the wiring diagram connecting each device model:

[0104] First, based on the power system wiring diagram and G matrix and B matrix of step S2, the three-phase power flow equation of the system that meets the physical process can be written as follows:

[0105]

[0106] Here, a serial mode can be selected to further optimize the system model.

[0107] S4, use the fusion driven power system model to perform system simulation and update online according to real-time data.

[0108] In one aspect, the application adopts a power equipment mechanism data fusion modeling method based on a feedback mode. The modeling method of the application fuses the advantages and disadvantages of mechanism models and data models, and has a wider range of application. In another aspect, the application adopts a complex power system simulation model hierarchical fusion construction method based on serial and parallel modes. The hierarchical modeling mechanism of the application firstly models one-to-one for physical entities in a complex power system, and then constructs a system model in combination with equipment models, topological relations and electrical connection relations. Meanwhile, the application proposes a dynamic online updating mechanism for the fusion model. The model construction method of the application firstly constructs a primary model according to mechanism knowledge and historical data, and then dynamically updates the model according to the difference between real-time running data collected by sensors and simulation results of the model, so as to enhance the adaptability of the model.

[0109] Embodiment 3

[0110] Please refer to Figure 6 The power system modeling system of the embodiment of the application comprises a device external characteristic fusion driving modeling module 1, a wiring diagram construction module 2, a power system fusion driving modeling module 3 and a model updating module 4, wherein:

[0111] The device external characteristic fusion driving modeling module 1 is used for constructing a single device external characteristic model of mechanism data fusion based on a feedback fusion mode;

[0112] The wiring diagram construction module 2 is used for connecting devices to form a power system wiring diagram according to a power grid topological structure and line parameters;

[0113] The power system fusion driving modeling module 3 is used for constructing a fusion-driven power system model by connecting single device external characteristic models of devices based on a power system wiring diagram;

[0114] The model updating module 4 is used for performing system simulation by using the fusion-driven power system model and performing online updating according to real-time data.

[0115] In order to construct a power system model with better performance and higher efficiency, the application divides a power system into single devices and a power system composed of interconnected devices according to granularity. In order to make full use of physical knowledge and historical running data, the above models are all mechanism-data fusion models; each part of the model corresponds to a physical entity one-to-one, and the dynamic updating of the model can be realized by real-time data collected by sensors.

[0116] Embodiment 4

[0117] Another embodiment of the present application also provides a computer readable storage medium storing a computer program, and the computer program, when executed by a processor, implements the power system modeling method in embodiment 1. The computer program includes computer program code, which can be in the form of source code, object code, executable code, or some intermediate form. The computer readable storage medium can include any entity or device capable of carrying the computer program code, a medium, a U disk, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc. It should be noted that the computer readable medium can include or exclude some contents according to the requirements of legislation and patent practice in a jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals. For the convenience of description, only the parts related to the embodiments of the present application are shown above, and the specific technical details not disclosed are described in the method part of the embodiments of the present application. The computer readable storage medium is non-transitory and can be stored in a storage device formed by various electronic devices, and can realize the execution process described in the method of the embodiments of the present application.

[0118] Embodiment 5

[0119] Another embodiment of the present application also provides a server, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor, when executing the computer program, implements the power system modeling method in embodiment 1. Similarly, for the convenience of description, only the parts related to the embodiments of the present application are shown above, and the specific technical details not disclosed are described in the method part of the embodiments of the present application.

[0120] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0121] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart Figure 1 one or more flows and / or blocks. Figure 1 one or more blocks.

[0122] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart Figure 1 one or more flows and / or blocks. Figure 1 one or more blocks.

[0123] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart Figure 1 one or more flows and / or blocks. Figure 1 one or more blocks.

[0124] Finally, it should be noted that the above-mentioned embodiments are merely used to illustrate the technical solutions of the present application, but not limit the technical solutions of the present application. Although the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalent replaced without departing from the spirit and scope of the present application, and any modification or equivalent replacement should be covered in the protection scope of the claims of the present application.

Claims

1. A method of power system modeling, characterized by, The method comprises the following steps: a single-device external characteristic model of mechanism data fusion is constructed based on a feedback fusion mode; a power system wiring diagram is formed by connecting each device according to a power grid topology structure and line parameters; a power system model driven by mechanism data fusion is constructed by connecting single-device external characteristic models of each device based on the power system wiring diagram; system simulation is performed by using the power system model driven by fusion, and online updating is performed according to real-time data; the single-device external characteristic model of mechanism data fusion constructed based on the feedback fusion mode comprises the following steps: A set of relatively independent input physical quantities X is selected according to the operating mechanism of the device e and output physical quantities Y e , a mechanism model Y = f e (X) is constructed according to the operating principle of the device; Bringing historical operation and experimental data x e Get the calculation results of the mechanism model, and combine the external environment data v collected by the sensor e Obtain the mechanism simulation value of the device running state Wherein, the influence mode of environmental data According to the different modeling devices; According to the original data x e , the mechanism model output prediction value True value y e And the error between the mechanism model output prediction value and the true value Train the correction model g e (x, y, v); The trained correction model is used to correct parameters or input and output data of the mechanism model, so as to obtain a simulation value of the equipment running state The correction method is selected according to the different functions of the model The correction method is selected according to the different functions of the model the system simulation performed by using the power system model driven by fusion and the online updating performed according to real-time data comprise the following steps: The power system model of each device fusion driving is connected with input and output according to the power grid topology to form a final system simulation model, and the device model calculates the device running state according to real-time running data post as input data input system model; The system simulation model is based on the equipment operating state The system operating state prediction value is calculated If the fusion-driven power system model is constructed in a parallel modeling manner, the system operating state prediction value after substituting the equipment operating state is If the fusion-driven power system model is constructed in a serial modeling manner, the system operating state prediction value after substituting the equipment operating state is Utilizing system operating state forecast values True value y s And error between two values Periodically update system simulation model.

2. The power system modeling method of claim 1, wherein, the power system wiring diagram formed by connecting each device according to the power grid topology structure and line parameters comprises the following steps: devices in the system are abstracted as nodes, devices having a direct connection relationship are connected together according to an electrical connection relationship and relative geographical position information between devices in an actual system, and a power system wiring diagram is formed; impedance and admittance of each branch are calculated according to line length and wire type between nodes, and an equivalent network diagram of the power system is drawn, and a line parameter admittance matrix G is recorded: in the step of constructing the power system model driven by mechanism data fusion by connecting single-device external characteristic models of each device based on the power system wiring diagram, the power system model driven by fusion is constructed in a parallel modeling mode or a serial modeling mode.

3. The power system modeling method of claim 1, wherein, the power system model driven by fusion is constructed in the parallel modeling mode, and the specific steps comprise:

4. The power system modeling method of claim 3, wherein, the power system model driven by fusion is constructed in the serial modeling mode, and the specific steps comprise: Based on equivalent network graph and environment data, mechanism model f of the system is constructed by physical principle analysis s (G,x s ,v s ); Training a data model g using system historical operation data and environment data s (x s ,v s ), wherein the system historical operation data includes steady state operation data and fault state data Using tagged historical data, mechanism model calculated values f s (G, x s , v s ) and data model g s (x s , v s ), a data driven approach is used to train a weighted hybrid system simulation model Where the weighting Is determined from the training process.

5. The power system modeling method of claim 3, wherein, The method comprises the following steps: Based on equivalent network diagram and environmental data, a mechanism model of the system is constructed by physical principle analysis, and simulation values of a group of mechanism models are simplified by using historical data According to the mechanism simulation values and the real values y s , a data model g s (y) = g s (f s (G, x s )) is trained which is capable of mapping the results of the mechanism calculation to the real values Combined with system running environment data v s Thus, the simulation value of the system running state is obtained Wherein the environment data influences the system running state The selection is made according to the different models.

6. A power system modeling system characterized by, a single-device external characteristic model of mechanism data fusion is constructed based on a feedback fusion mode by the device external characteristic fusion driving modeling module; a power system wiring diagram is formed by connecting each device according to a power grid topology structure and line parameters by the wiring diagram construction module; a power system model driven by mechanism data fusion is constructed by connecting single-device external characteristic models of each device based on the power system wiring diagram by the power system fusion driving modeling module; system simulation is performed by using the power system model driven by fusion, and online updating is performed according to real-time data by the model updating module; the single-device external characteristic model of mechanism data fusion constructed based on the feedback fusion mode comprises the following steps: the system simulation performed by using the power system model driven by fusion and the online updating performed according to real-time data comprise the following steps: According to the equipment operation mechanism, a group of relatively independent input physical quantities X e and output physical quantities Y e are selected, and a mechanism model Y = f e (X) is constructed according to the equipment operation principle. Bringing historical operation and experimental data x e Get the calculation results of the mechanism model, and combine the external environment data v collected by the sensor e Obtain the mechanism simulation value of the device operating state Among them, the influence mode of environmental data According to the different modeling devices, select according to the original data x e , the mechanism model output prediction value True value y e And the error between the mechanism model output prediction value and the true value Train the correction model g by using the data-driven method e (x, y, v); The trained correction model is used to correct parameters or input and output data of the mechanism model, so as to obtain a simulation value of the equipment running state The correction method is selected according to the different functions of the model The correction method is selected according to the different functions of the model the computer program is executed by the processor to implement the power system modeling method in any one of claims 1 to 5. The power system model of each device fusion driving is connected with input and output according to the power grid topology structure to form a final system simulation model, and the device model calculates the device running state according to real-time running data post as input data input system model; The system simulation model is based on the equipment operating state The system operating state prediction value is calculated If the fusion-driven power system model is constructed in a parallel modeling manner, the system operating state prediction value after substituting the equipment operating state is If the fusion-driven power system model is constructed in a serial modeling manner, the system operating state prediction value after substituting the equipment operating state is Utilizing system operating state forecast values True value y s And error between two values Periodically update system simulation model.

7. A computer-readable storage medium storing a computer program, wherein the computer program comprises the following steps of: receiving a request for a resource from a client; determining whether the client is authorized to access the resource; and if the client is authorized to access the resource, providing the resource to the client. the processor executes the computer program to implement the power system modeling method in any one of claims 1 to 5.

8. A server comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, ​

Images