Microgrid equipment digital twin fusion modeling method and system, terminal and medium

Abstract

The invention discloses a micro-grid equipment digital twin fusion modeling method, and discloses a system with the micro-grid equipment digital twin fusion modeling method and a storage medium, and the micro-grid equipment digital twin fusion modeling method can utilize a physical embedded neural network model to combine with measurement data. According to the method, credible micro-grid equipment digital twinborn modeling is constructed, the operation and maintenance efficiency, the full-life-cycle management capability and the system stable operation capability of distributed energy in a double-high system are improved through a digital twinborn modeling mode, the model is corrected through front-end communication data of a micro-grid system, the accuracy and the real-time performance of the model are ensured, and the real-time performance of the micro-grid equipment is improved. The problem of electric energy quality deterioration caused by electric signal injection in an active method can be effectively avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment modeling, and in particular to a method, system, terminal and medium for digital twin fusion modeling of microgrid equipment. Background Art

[0002] With the massive penetration of distributed energy in the distribution system, digital twin technology combining real-time simulation and real data has become one of the necessary technologies for effectively controlling massive distributed energy equipment and improving the observability, measurability and controllability of the system. In this context, the inertia of small-capacity hydropower units plays an important role in the stable operation of the microgrid system. Therefore, the digital twin modeling method needs to make clear data derivation of the inertia of small hydropower when considering the microgrid scenario dominated by small hydropower. In actual projects, due to asset ownership and intellectual property protection, the dispatching center of some small hydropower units cannot understand the inertia of the unit, which leads to misjudgment of the dynamic response process of the system. Even if there are clear inertia parameters, such parameters will continue to change during the operation of small hydropower due to wear and tear, and need to be accurately estimated through a reliable digital twin model.

[0003] At present, the main methods for estimating the inertia parameters of small hydropower include active impact experimental methods, mathematical traversal methods and machine learning methods. The active impact experimental method injects external small disturbance current by adding additional equipment at a fixed location, and determines the inertia of small hydropower by the response results. However, objective conditions make it impossible to conduct actual experimental tests on a large number of small hydropower equipment separately; the mathematical traversal method traverses in the form of fixed ordinary differential equations to find solution combinations, but the computational task is relatively intensive; the machine learning method fits the output curve in a data-driven way to derive parameters, but it is easy to obtain unreasonable estimation results. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a digital twin fusion modeling method for microgrid equipment, which can use physical embedded neural networks combined with measurement data to build a reliable small hydropower digital twin modeling, improve the operation and maintenance efficiency, full life cycle management capabilities and system stable operation capabilities of distributed energy in the dual-high system through digital twin modeling, and correct the model through the front-end communication data of the microgrid system to ensure the accuracy and real-time performance of the model, which can effectively avoid the deterioration of power quality caused by the injection of electrical signals by active methods.

[0005] The present invention also proposes a system having the above-mentioned microgrid equipment digital twin fusion modeling method.

[0006] The method for fusion modeling of digital twins of microgrid devices according to the first embodiment of the present invention is characterized by comprising the following steps:

[0007] Obtain the operation information of microgrid equipment;

[0008] Constructing an ordinary differential equation based on the characteristics of the microgrid equipment, and constructing a fully connected physical embedded neural network model based on the ordinary differential equation;

[0009] The operation information is combined with the physical embedded neural network model to solve the inverse problem of the ordinary differential equation to obtain the necessary parameters of the physical embedded neural network model.

[0010] The digital twin fusion modeling method for microgrid equipment according to an embodiment of the present invention has at least the following beneficial effects: the method can use a physically embedded neural network combined with measurement data to construct a reliable modeling of a digital twin of microgrid equipment, and improve the operation and maintenance efficiency, full life cycle management capability and system stable operation capability of distributed energy in the dual-high system through digital twin modeling. The model is corrected through the front-end communication data of the microgrid system to ensure the accuracy and real-time performance of the model, and can effectively avoid the problem of power quality deterioration caused by the injection of electrical signals by active methods.

[0011] According to some embodiments of the present invention, the method further comprises:

[0012] The current state of the physical embedded neural network model is retained, the operation information is updated and the parameters of the physical embedded neural network model are re-solved to achieve the update of the physical embedded neural network model.

[0013] According to some embodiments of the present invention, the power flow information includes the current speed ω, the mechanical torque T m , electromagnetic torque T e .

[0014] According to some embodiments of the present invention, obtaining the electromagnetic torque includes: determining the electromagnetic power of the small hydropower equipment, and determining the electromagnetic torque according to the electromagnetic power and the current rotation speed, wherein the calculation formula of the electromagnetic power is:

[0015] P e =V a I a cos(φ a )+V b I b cos(φ b )+V c I c cos(φ c )

[0016] Among them, V a ,I a ,V b ,I b ,V c ,Ic are the three-phase voltage and current, φ a ,φ b ,φ c is the phase angle between the voltage and current of each phase.

[0017] According to some embodiments of the present invention, before obtaining the operation information of the microgrid device, it is ensured that the operation information on the outlet side of the small hydropower device satisfies the following equation:

[0018]

[0019] Where H represents the inertia coefficient of small hydropower, and pu is expressed as per unit value.

[0020] According to some embodiments of the present invention, before solving the inverse problem of the ordinary differential equation, the constraint conditions of the physical embedding neural network are determined, and the constraint conditions include physical constraints. The error expression formula of the physical constraints of the model is as follows:

[0021]

[0022] Where L f represents the error obtained by inserting H in the ordinary differential equation obtained during the training process into the equation during the f-th training. It represents that the ordinary differential equation is a function of time t, that is, the external characteristic equation.

[0023] According to some embodiments of the present invention, the relative square error between the predicted value of the physical embedding neural network model and the real data satisfies the following formula:

[0024]

[0025] Among them, L i (θ,λ; T i ) represents the loss function between the predicted value and the measured data of the physical embedding neural network model, θ represents the hidden layer parameter cluster of the current neural network model, represents the estimated value, T i A dataset representing measurement data acquisition.

[0026] According to some embodiments of the present invention, the necessary parameters of the physical embedding neural network model are expressed as The physical embedding neural network model can be expressed as:

[0027]

[0028] Among them, pu represents the per-unit value.

[0029] The microgrid device digital twin fusion modeling system according to the second aspect of the present invention is characterized by comprising:

[0030] The power flow monitoring module can obtain the operation information of microgrid equipment;

[0031] A model building module capable of building ordinary differential equations based on microgrid device characteristics, and building a fully connected physical embedded neural network model based on the ordinary differential equations;

[0032] The parameter solving module can combine the operation information with the physical embedded neural network model to solve the inverse problem of the ordinary differential equation and obtain the necessary parameters of the physical embedded neural network model.

[0033] According to some embodiments of the present invention, the system further comprises:

[0034] The real-time update module can update the current state of the model, update the operation information and re-solve the parameters of the model to achieve the update of the model.

[0035] According to some embodiments of the present invention, the microgrid device is a small hydropower device, and the operation information includes the current speed ω, the mechanical torque T m , electromagnetic torque T e .

[0036] According to some embodiments of the present invention, the flow monitoring module obtains the electromagnetic torque by: determining the electromagnetic power of the small hydropower equipment, and determining the electromagnetic torque according to the electromagnetic power and the current rotation speed, wherein the calculation formula of the electromagnetic power is:

[0037] P e =V a I a cos(φ a )+V b I b cos(φ b )+V c I c cos(φ c )

[0038] Among them, V a ,I a ,V b ,I b ,V c ,I c are the three-phase voltage and current, φ a ,φ b ,φ c is the phase angle between the voltage and current of each phase.

[0039] According to some embodiments of the present invention, the flow monitoring module ensures that the operation information of the outlet side of the small hydropower equipment satisfies the following equation before the operation information of the microgrid device:

[0040]

[0041] Where H represents the inertia coefficient of small hydropower, and pu is expressed as per unit value.

[0042] According to some embodiments of the present invention, before solving the inverse problem of the ordinary differential equation, the constraint conditions of the physical embedding neural network are determined, and the constraint conditions include physical constraints. The error expression formula of the physical constraints of the model is as follows:

[0043]

[0044] Where L f represents the error obtained by inserting H in the ordinary differential equation obtained during the training process into the equation during the f-th training. It represents that the ordinary differential equation is a function of time t, that is, the external characteristic equation.

[0045] According to some embodiments of the present invention, the relative square error between the predicted value of the physical embedding neural network model and the real data satisfies the following formula:

[0046]

[0047] Among them, L i (θ,λ; T i ) represents the loss function between the predicted value and the measured data of the physical embedding neural network model, θ represents the hidden layer parameter cluster of the current neural network model, represents the estimated value, T i A dataset representing measurement data acquisition.

[0048] According to some embodiments of the present invention, the necessary parameters of the physical embedding neural network model are expressed as The physical embedding neural network model can be expressed as:

[0049]

[0050] Among them, pu represents the per-unit value.

[0051] According to a terminal of a third-party embodiment of the present invention, the terminal includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above-mentioned microgrid device digital twin fusion modeling method is implemented.

[0052] According to the computer-readable storage medium of the fourth aspect embodiment of the present invention, the medium stores computer-executable instructions, and the computer-executable instructions are used to execute the above-mentioned microgrid equipment digital twin fusion modeling method.

[0053] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0055] Figure 1 A schematic diagram of the steps of a digital twin fusion modeling method for microgrid equipment according to an embodiment of the present invention;

[0056] Figure 2 This is a structural block diagram of the digital twin fusion modeling system for microgrid equipment in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0058] In the description of the present invention, it should be understood that descriptions involving orientations, such as up, down, front, back, left, right, etc., and orientations or positional relationships indicated are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0059] In the description of the present invention, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed" etc. are understood as not including the number itself, and "above", "below", "within" etc. are understood as including the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing the technical features, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0060] In the description of the present invention, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in the present invention based on the specific content of the technical solution.

[0061] Embodiment 1

[0062] In order to solve the problem that the existing active parameter estimation method has an impact on the deterioration of power quality of equipment and microgrids, and the passive parameter estimation method has a low solution efficiency, the present invention provides a digital twin fusion modeling method for microgrid equipment. Taking small hydropower equipment as an example, the method includes the following steps:

[0063] Step S100: Obtain operation information of microgrid equipment.

[0064] Measure and obtain the flow information on the export side of small hydropower, and upload the flow information to the data processing center.

[0065] The microgrid device is a small hydropower device, and the operation information includes the current speed ω, the mechanical torque T m , electromagnetic torque T e , where the speed sensor is used to obtain the current speed ω of the small hydropower equipment and use it to calculate the frequency Mechanical torque T m Obtained by the torque sensor, and the mechanical power P is calculated m =ω×T m , obtain the electromagnetic torque, including: determining the electromagnetic power of the small hydropower equipment, determining the electromagnetic torque according to the electromagnetic power and the current speed, the electromagnetic power P e The calculation formula is as follows:

[0066] P e =V a I a cos(φ a )+V b I b cos(φ b )+V c I c cos(φ c )(1)

[0067] Among them, V a ,I a ,V b ,I b ,V c ,I c are the three-phase voltage and current, φ a ,φ b ,φ c is the phase angle between the voltage and current of each phase.

[0068] The electromagnetic torque T can be calculated from this e as follows:

[0069]

[0070] Step S200: construct an ordinary differential equation based on the outlet side characteristics of the microgrid device, and construct a fully connected physical embedding neural network based on the ordinary differential equation.

[0071] An ordinary differential equation is constructed to characterize the external characteristics of the small hydropower outlet side, and a fully connected physical embedding neural network is constructed based on this physical constraint.

[0072] The external characteristic equation of the small hydropower export side is constructed as follows:

[0073]

[0074] Where H represents the inertia coefficient of small hydropower, and pu represents the per-unit value. Relying on this equation as a physical constraint and as a part of the feedback of the fully connected neural network, the loss function of this neural network includes two parts. The first part is the external characteristic equation of the small hydropower outlet side mentioned above, and the second part is the relative square error between the predicted value and the real data generated during the neural network training process. The loss function of the neural network is the sum of the value of the external characteristic equation of the small hydropower outlet side and the relative square error between the predicted value and the real data generated during the neural network training process.

[0075] The error expression formula of the physical constraint of the equation is as follows:

[0076]

[0077] Where L f represents the error obtained by inserting λ (H in the present invention patent) in the ordinary differential equation obtained during the training process into the equation during the f-th training. The representative ordinary differential equation is a function of time t (the external characteristic equation in the present invention), and the relative square error between the predicted value of the second part of the neural network and the real data is as follows:

[0078]

[0079] Among them, L i (θ,λ; T i ) represents the loss function of the neural network prediction value and the measured data, θ represents the hidden layer parameter cluster of the current neural network, represents the estimated value, T i A dataset representing measurement data acquisition.

[0080] Step S300: Combine the operation information with the physical embedded neural network model to solve the inverse problem of ordinary differential equations to obtain the necessary parameters of the physical embedded neural network model.

[0081] The power flow information data obtained in step S100 is combined with the network constructed in step S200 to solve the inverse problem of the ordinary differential equation to obtain reliable necessary parameters of the model.

[0082] The solution to the inverse problem of the ordinary differential equation is to minimize the neural network loss function built in step S200 by automatic differentiation gradient descent to obtain the derived parameters

[0083] The constructed small hydropower digital twin model considering inertia can be expressed as:

[0084]

[0085] Embodiment 2

[0086] Preferably, in order to meet the requirements of real-time interaction, the method is supplemented on the basis of the first embodiment. The method further includes:

[0087] Step S400, retaining the current state of the physical embedded neural network model, updating the operation information and resolving the parameters of the physical embedded neural network model to implement the update of the physical embedded neural network model.

[0088] The device-level digital twin model is continuously updated through the flow information data transmitted back from the microgrid system-level front end to meet the real-time interaction requirements.

[0089] When the next round of measurement data communication arrives at the data center, the parameters obtained in step S300 are The value of is used as the initial point, and the above steps S100 to S300 are repeated as the real-time interaction and update of the system-level and device-level digital twin models.

[0090] Embodiment 3

[0091] Another embodiment of the present application provides a microgrid device digital twin fusion modeling system, such as Figure 2 As shown, the system 20 includes:

[0092] The power flow monitoring module 201 can obtain the operation information of the microgrid equipment;

[0093] A model building module 202 is capable of building an ordinary differential equation based on the characteristics of the microgrid equipment, and building a fully connected physical embedded neural network model based on the ordinary differential equation;

[0094] The parameter solving module 203 can combine the operation information with the physical embedded neural network model to solve the inverse problem of the ordinary differential equation and obtain the necessary parameters of the physical embedded neural network model.

[0095] This embodiment derives the inertia parameters of microgrid devices by physically embedding neural networks, and models digital twins based on the inertia parameters. It relies on front-end measurement data to achieve real-time interaction and update of digital twin modeling, providing an effective solution for digital twin modeling of distributed energy sources such as small hydropower synchronous machines, and realizing the integration of device-level digital twin modeling and system. The modeling of digital twins is achieved only by data communication and data processing communication, without the need for other external devices, which can effectively avoid the deterioration of power quality caused by the injection of electrical signals by active methods. It uses the strong fitting ability of neural networks while also combining physical constraints to make the model more reliable.

[0096] Furthermore, the system also includes:

[0097] The real-time update module (not shown in the figure) can retain the existing model, update the operation information and re-solve the parameters of the model to achieve the update of the model.

[0098] Embodiment 4:

[0099] Another embodiment of the present application provides a terminal, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned application jam detection method when executing the computer program.

[0100] Specifically, the processor may be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0101] Specifically, the processor is connected to the memory via a bus, and the bus may include a path for transmitting information. The bus may be a PCI bus or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc.

[0102] The memory can be a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, or an EEPROM, CD-ROM or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to this.

[0103] Optionally, the memory is used to store the code of the computer program for executing the solution of the present application, and the execution is controlled by the processor. The processor is used to execute the application program code stored in the memory to implement Figure 2 The illustrated embodiment provides the functions of a microgrid equipment digital twin fusion modeling system.

[0104] Embodiment 5

[0105] Another embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions for executing the above-mentioned Figure 1 The digital twin fusion modeling method of microgrid equipment is shown.

[0106] The device embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0107] It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed method above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0108] The above is a specific description of the preferred implementation of the present application, but the present application is not limited to the above-mentioned implementation mode. Technical personnel familiar with the field can also make various equivalent deformations or substitutions without violating the spirit of the present application. These equivalent deformations or substitutions are all included in the scope defined by the claims of the present application.

Claims

1. A digital twin fusion modeling method for microgrid equipment, characterized in that: The following steps are involved: Obtain the operation information of microgrid equipment; Constructing an ordinary differential equation based on the characteristics of the microgrid equipment, and constructing a fully connected physical embedded neural network model based on the ordinary differential equation; The operation information is combined with the physical embedded neural network model to solve the inverse problem of the ordinary differential equation to obtain the necessary parameters of the physical embedded neural network model.

2. The method according to claim 1, characterized in that The method further includes: The current state of the physical embedded neural network model is retained, the operation information is updated and the parameters of the physical embedded neural network model are re-solved to achieve the update of the physical embedded neural network model.

3. The method according to claim 1, characterized in that The microgrid device is a small hydropower device, and the operation information includes the current speed ω, the mechanical torque T m , electromagnetic torque T e .

4. The method according to claim 3, characterized in that Obtaining the electromagnetic torque includes: determining the electromagnetic power of the small hydropower equipment, and determining the electromagnetic torque according to the electromagnetic power and the current speed, wherein the calculation formula of the electromagnetic power is: P e =V a I a cos(φ a )+V b I b cos(φ b )+V c I c cos(φ c ) Among them, V a ,I a ,V b ,I b ,V c ,I c are the three-phase voltage and current, φ a ,φ b ,φ c is the phase angle between the voltage and current of each phase.

5. The method according to claim 3, characterized in that: Before obtaining the operation information of the microgrid device, ensure that the operation information on the outlet side of the small hydropower device satisfies the following equation: Where H represents the inertia coefficient of small hydropower, and pu is expressed as per unit value.

6. The method according to claim 3, characterized in that Before solving the inverse problem of ordinary differential equations, the constraints of physical embedding neural network are determined. The constraints include physical constraints. The error expression formula of the physical constraints of the model is as follows: Where L f represents the error obtained by inserting H in the ordinary differential equation obtained during the training process into the equation during the f-th training. It represents that the ordinary differential equation is a function of time t, that is, the external characteristic equation.

7. The method according to claim 3, characterized in that The relative square error between the predicted value of the physical embedding neural network model and the real data satisfies the following formula: Among them, L i (θ,λ; T i ) represents the loss function between the predicted value and the measured data of the physical embedding neural network model, θ represents the hidden layer parameter cluster of the current neural network model, represents the estimated value, T i A dataset representing measurement data acquisition.

8. The method according to claim 3, characterized in that The necessary parameters of the physical embedding neural network model are expressed as The physical embedding neural network model can be expressed as: Among them, pu represents the per-unit value.

9. A microgrid equipment digital twin fusion modeling system, characterized in that: include: The power flow monitoring module can obtain the operation information of microgrid equipment; A model building module capable of building ordinary differential equations based on microgrid device characteristics, and building a fully connected physical embedded neural network model based on the ordinary differential equations; The parameter solving module can combine the operation information with the physical embedded neural network model to solve the inverse problem of the ordinary differential equation and obtain the necessary parameters of the physical embedded neural network model.

10. The system according to claim 9, characterized in that The system further comprises: The real-time update module can update the current state of the model, update the operation information and re-solve the parameters of the model to achieve the update of the model.

11. The system according to claim 9, characterized in that The microgrid device is a small hydropower device, and the operation information includes the current speed ω, the mechanical torque T m , electromagnetic torque T e .

12. The system according to claim 11, characterized in that The flow monitoring module obtains the electromagnetic torque by: determining the electromagnetic power of the small hydropower equipment, and determining the electromagnetic torque according to the electromagnetic power and the current rotation speed, wherein the calculation formula of the electromagnetic power is: P e =V a I a cos(φ a )+V b I b cos(φ b )+V c I c cos(φ c ) Among them, V a ,I a ,V b ,I b ,V c ,I c are the three-phase voltage and current, φ a ,φ b ,φ c is the phase angle between the voltage and current of each phase.

13. The system according to claim 11, characterized in that The flow monitoring module ensures that the operation information of the outlet side of the small hydropower equipment satisfies the following equation before the operation information of the microgrid equipment: Where H represents the inertia coefficient of small hydropower, and pu is expressed as per unit value.

14. The system according to claim 11, characterized in that Before solving the inverse problem of ordinary differential equations, the constraints of physical embedding neural network are determined. The constraints include physical constraints. The error expression formula of the physical constraints of the model is as follows: Where L f represents the error obtained by inserting H in the ordinary differential equation obtained during the training process into the equation during the f-th training. It represents that the ordinary differential equation is a function of time t, that is, the external characteristic equation.

15. The method according to claim 11, characterized in that The relative square error between the predicted value of the physical embedding neural network model and the real data satisfies the following formula: Among them, L i (θ,λ; T i ) represents the loss function between the predicted value and the measured data of the physical embedding neural network model, θ represents the hidden layer parameter cluster of the current neural network model, represents the estimated value, T i A dataset representing measurement data acquisition.

16. The method according to claim 11, characterized in that The necessary parameters of the physical embedding neural network model are expressed as The physical embedding neural network model can be expressed as: Among them, pu represents the per-unit value.

17. A terminal, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method according to any one of claims 1 to 8.

18. A computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the method according to any one of claims 1 to 8.

Images