Physical interactive simulation method and system for power grid equipment under intelligent framework
By employing a physical interactive simulation method for power grid equipment within an embodied intelligence framework and adopting a heterogeneous resource dynamic scheduling architecture, the problem of unreasonable resource allocation in traditional simulation systems is solved, achieving high efficiency and accuracy in power grid equipment simulation and improving system response efficiency.
Patent Information
- Application Number
- CN202511612827.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-06
- Publication Date
- 2025-12-05
AI Technical Summary
Traditional simulation systems cannot dynamically allocate resources according to the timeliness and computational complexity of different simulation tasks, resulting in a waste of high-speed hardware computing resources or a backlog of simulation tasks.
A physical interactive simulation method for power grid equipment under the embodied intelligence framework is adopted. Through a heterogeneous resource dynamic scheduling architecture driven by time-sensitive needs, the operating data of power grid equipment is decomposed into first input data and second input data, and simulation calculations are performed in the hardware real-time simulation resource region and the software delay resource region, respectively.
It satisfies the timeliness requirements of physical transient simulation and the accuracy requirements of physical steady-state simulation of power grid equipment, makes full use of computing resources, improves the overall response efficiency of the system, and avoids resource waste and simulation task backlog.
Smart Images

Figure CN121072197A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of data processing, in particular to a power grid equipment physical interactive simulation method and system under a body intelligent framework. BACKGROUND
[0002] In the power system, the stable operation of the power grid equipment is directly related to the safety and reliability of power supply. With the rapid development of smart grid technology, high-precision and real-time physical state simulation of power grid equipment has become a key technical means to improve the efficiency of power grid operation and prevent potential faults.
[0003] Among them, the traditional simulation system in the prior art adopts a static resource allocation mode, that is, the operation resources in the system are not classified, but the corresponding operation resources are retrieved from the overall operation resources based on the execution order or priority of the corresponding simulation operation.
[0004] It can be seen that the unified resource scheduling mode adopted by the traditional simulation platform cannot dynamically allocate resources according to the time efficiency and calculation complexity of different simulation tasks, which is easy to cause waste of hardware high-speed operation resources or simulation task backlog. SUMMARY
[0005] The present application provides a power grid equipment physical interactive simulation method and system under a body intelligent framework, which simulates different physical states of the power grid equipment through a time efficiency demand driven heterogeneous resource dynamic scheduling architecture.
[0006] In a first aspect, the present application provides a power grid equipment physical interactive simulation system under a body intelligent framework, comprising: a physical layer, a body intelligent decision layer and an interactive simulation layer; Among them, the physical layer is used for accessing a sensor cluster for collecting operation data of the power grid equipment, and transmitting the collected operation data to the body intelligent decision layer; The body intelligent decision layer decomposes the operation data into first input data and second input data; The body intelligent decision layer calls the hardware real-time simulation resource area in the interactive simulation layer to execute the simulation operation of the first input data, and calls the software delay resource area in the interactive simulation layer to execute the simulation operation of the second input data.
[0007] Optionally, the hardware real-time simulation resource area runs on a field programmable gate array and / or a digital signal processor.
[0008] Optionally, the first input data is used to simulate the physical transient state of the power grid equipment, and the physical transient state includes at least one of electronic switch transient state and electromagnetic transient state.
[0009] Optionally, the software delay resource area runs on a central processing unit or a cloud server.
[0010] Optionally, the second input data is used to simulate a physical steady state of the power grid equipment, and the physical steady state includes at least one of a device thermal state and a power flow state.
[0011] Optionally, if at least one target parameter in the simulation result determined after simulation by using the first input data is a boundary condition of a simulation process by using the second input data, the hardware real-time simulation resource area performs simulation on the first input data in a fixed-point operation mode, and returns the obtained target parameter to the software delay resource area through the embodied intelligent decision layer. If the simulation process by using the first input data and the simulation process by using the second input data are independent of each other, the hardware real-time simulation resource area performs floating-point operation.
[0012] Optionally, if at least one target parameter in the simulation result determined after simulation by using the first input data is a boundary condition of a simulation process by using the second input data, the embodied intelligent decision layer calls the hardware real-time simulation resource area in the interactive simulation layer to perform simulation operation on the first input data, and calls the software delay resource area in the interactive simulation layer to perform simulation operation on the second input data, including: The embodied intelligent decision layer calls a same preset time correction module to perform time correction on simulation boundary conditions in the hardware real-time simulation resource area and the software delay resource area, so that the hardware real-time simulation resource area and the software delay resource area are consistent in time dimension of the simulation boundary conditions when performing simulation operation.
[0013] In a second aspect, the present application provides a power grid equipment physical interactive simulation method under an embodied intelligent framework, applied to a power grid equipment physical interactive simulation system under the embodied intelligent framework, the system including a physical layer, an embodied intelligent decision layer and an interactive simulation layer; the method including: The physical layer accesses a sensor cluster collecting operation data of the power grid equipment, and transmits the collected operation data to the embodied intelligent decision layer. The embodied intelligent decision layer decomposes the operation data into first input data and second input data. The embodied intelligent decision layer calls a hardware real-time simulation resource area in the interactive simulation layer to perform simulation operation on the first input data, and calls a software delay resource area in the interactive simulation layer to perform simulation operation on the second input data.
[0014] The power grid equipment physical interactive simulation method and system under the embodied intelligent framework provided by the application, through the sensor cluster in the physical layer, collects the operation data of the power grid equipment, and transmits the collected operation data to the embodied intelligent decision layer, so that the embodied intelligent decision layer decomposes the operation data into first input data and second input data according to the time requirement of the simulation result, then calls the corresponding hardware real-time simulation resource area or software delay resource area in the interactive simulation layer to execute the corresponding simulation operation, so as to simulate the different physical states of the power grid equipment through the heterogeneous resource dynamic scheduling architecture driven by the time requirement, which can not only meet the time requirement of the power grid equipment physical transient simulation, but also meet the accuracy requirement of the power grid equipment physical steady-state simulation, and then fully utilize the operation resources and improve the overall response efficiency of the system. BRIEF DESCRIPTION OF DRAWINGS
[0015] The accompanying drawings, which are incorporated herein and constitute part of the specification, illustrate embodiments consistent with the application and, together with the description, serve to explain the principles of the application.
[0016] Figure 1 is a structural schematic diagram of the power grid equipment physical interactive simulation system under the embodied intelligent framework according to an example embodiment of the application. Figure 2 is a flow schematic diagram of the power grid equipment physical interactive simulation method under the embodied intelligent framework according to an example embodiment of the application. Figure 3 is a flow schematic diagram of the power grid equipment physical interactive simulation method under the embodied intelligent framework according to another example embodiment of the application.
[0017] Through the above-mentioned drawings, the specific embodiments of the application have been shown, and there will be more detailed description hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of the application by any means, but to illustrate the concept of the application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION
[0018] The example embodiments will be described in detail herein, and examples thereof are shown in the drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following example embodiments do not represent all implementations consistent with the application. Instead, they are merely examples of apparatuses and methods consistent with some aspects of the application as detailed in the appended claims.
[0019] Figure 1 is a structural schematic diagram of the power grid equipment physical interactive simulation system under the embodied intelligent framework according to an example embodiment of the application. As shown in the figure, the system comprises a sensor cluster, an embodied intelligent decision layer, an interactive simulation layer, and a data storage layer.Figure 1 As shown, the physical interactive simulation system of power grid equipment under the embodied intelligent framework provided by the embodiment includes a physical layer 110, an embodied intelligent decision-making layer 120, and an interactive simulation layer 130.
[0020] The interactive simulation layer 130 includes a real-time simulation resource area 131 and a software delay resource area 132. Optionally, the real-time simulation resource area 131 can be run on a field programmable gate array and / or a digital signal processor, and the software delay resource area 132 can be run on a central processing unit or a cloud server.
[0021] The physical interactive simulation system of power grid equipment under the embodied intelligent framework described above realizes the function architecture arrangement of field data acquisition, data decomposition and simulation scheduling, and real-time / delay simulation calculation by establishing the three-layer structure, and further realizes accurate modeling and simulation of the physical state of the power grid equipment at different time scales.
[0022] Figure 2 FIG. 1 is a flowchart of a physical interactive simulation method of power grid equipment under an embodied intelligent framework according to an example embodiment of the present application. As shown, Figure 2 The physical interactive simulation method of power grid equipment under the embodied intelligent framework provided by the embodiment includes: S201, the physical layer accesses a sensor cluster for collecting operation data of the power grid equipment, and transmits the collected operation data to the embodied intelligent decision-making layer.
[0023] In this step, temperature sensors, Hall current sensors, vibration sensors, and electromagnetic field sensors can be deployed at key nodes of the power grid equipment (such as transformer windings, circuit breaker contacts, bus connection points, etc.). The sampling frequency of each sensor is dynamically adjusted according to the type of the equipment, for example, microsecond-level sampling is used for power electronic devices, and second-level sampling is used for thermal elements. After the collected data is subjected to noise filtering and dimension normalization processing by the data preprocessing module of the physical layer, it is uploaded to the embodied intelligent decision-making layer through a low-delay communication link.
[0024] S202, the embodied intelligent decision-making layer decomposes the operation data into first input data and second input data.
[0025] In this step, the embodied intelligent decision-making layer decomposes the operation data into first input data and second input data, wherein the first demand timeliness of the simulation result corresponding to the first input data as a simulation input parameter is less than the second demand timeliness of the simulation result corresponding to the second input data as a simulation input parameter.
[0026] It is worth mentioning that the operation data is usually scheduled according to signal types or task priorities, while the above steps are based on the timeliness of simulation results as the basis for data decomposition, that is, not hard-cut according to signal types, but according to the timeliness constraints of target simulation output to determine the classification (the same physical quantity can participate in transient and steady state at the same time, emphasizing task-oriented data shunting), so as to realize the scheduling of high-timeliness tasks to hardware real-time resources, and the scheduling of low-timeliness large-scale / complex model tasks to software delay resources.
[0027] Specifically, after the above embodiment intelligent decision layer receives the operation data from the physical layer, the data can be preprocessed first, including data cleaning, outlier detection and correction, etc. Then, according to the simulation requirements, the operation data can be decomposed into first input data and second input data. Among them, the first input data is the data corresponding to the short timeliness requirement of simulation results, such as the transient response data of power grid equipment under extreme working conditions, which has higher real-time requirements. And the second input data is the data corresponding to the long timeliness requirement of simulation results, such as the steady-state performance data of power grid equipment in the long-term operation process, which has relatively low real-time requirements.
[0028] And for the above timeliness judgment, the embodiment intelligent decision layer can automatically judge the timeliness requirement of the simulation result corresponding to each type of data as a simulation input parameter according to the preset rules or algorithms, so as to ensure the effective use of data.
[0029] In one possible case, assuming that an abnormal current fluctuation occurs in a key device in the power grid, the physical layer sensor quickly collects this data and transmits it to the embodiment intelligent decision layer. The embodiment intelligent decision layer analyzes the characteristics of the data (such as large fluctuation amplitude and rapid change) and the simulation task requirements (such as emergency fault prediction) through the preset algorithm, and judges that the timeliness requirement of the data is urgent. Therefore, the system immediately allocates the data as first input data to the hardware real-time simulation resource area for rapid simulation analysis, so as to quickly obtain the fault prediction result and take corresponding measures. At the same time, for other steady-state performance data of the device (such as temperature, load rate, etc.), the system performs simulation analysis in the software delay resource area according to the normal speed.
[0030] It is worth mentioning that the above first input data and second input data are not completely different, that is, if it belongs to the first input data, it cannot belong to the second input data, in other words, some parameters in the power grid equipment can be used to determine the physical transient state and the physical steady state, for example, the real-time current of the power grid equipment.
[0031] For the first input data, for example, it can be the voltage and current waveforms at the moment of the circuit breaker action, the on / off process parameters of the power electronic switch, etc. For the second input data, for example, it can be the temperature, current waveform and power parameters of the circuit breaker, etc.
[0032] In S203, the embodied intelligent decision layer calls the hardware real-time simulation resource area in the interactive simulation layer to perform the simulation operation of the first input data, and calls the software delay resource area in the interactive simulation layer to perform the simulation operation of the second input data.
[0033] In this step, for the first input data, since the time efficiency of the corresponding simulation result requirement is short, the embodied intelligent decision layer calls the hardware real-time simulation resource area in the interactive simulation layer to perform the simulation operation. The area is usually run on a field programmable gate array (FPGA) and / or a digital signal processor (DSP), so as to realize high-speed and real-time data processing and simulation analysis.
[0034] For example, when simulating the transient process of the electronic switch in the power grid equipment, the hardware real-time simulation resource area can quickly capture the current and voltage changes at the moment of the switch action, and provide accurate transient response data for the system.
[0035] For the second input data, since the time efficiency of the corresponding simulation result requirement is long, the embodied intelligent decision layer calls the software delay resource area in the interactive simulation layer to perform the simulation operation.
[0036] The area is usually run on a central processing unit (CPU) or a cloud server, and can process large-scale and complex data simulation tasks.
[0037] For example, when simulating the thermal state of the device in the long-term operation of the power grid equipment, the software delay resource area can comprehensively consider various factors such as the material of the device, the environmental temperature and the load condition, calculate the temperature distribution of each component of the device through a complex thermodynamic model, and provide a basis for the steady-state performance analysis of the system.
[0038] In the embodiment, the operation data of the power grid equipment is collected by the sensor cluster in the physical layer, and the collected operation data is transmitted to the embodied intelligent decision layer, so that the embodied intelligent decision layer decomposes the operation data into the first input data and the second input data according to the requirement of the simulation result, and then calls the corresponding hardware real-time simulation resource area or software delay resource area in the interactive simulation layer to perform the corresponding simulation operation, so as to perform the simulation operation on different physical states of the power grid equipment through the heterogeneous resource dynamic scheduling architecture driven by the time requirement, which can not only meet the timeliness requirement of the physical transient simulation of the power grid equipment, but also meet the accuracy requirement of the physical steady-state simulation of the power grid equipment, thereby fully utilizing the operation resources and improving the overall response efficiency of the system.
[0039] Specifically, through dynamic resource scheduling, the waste of hardware high-speed operation resources (such as allocating too many high-speed resources to low timeliness tasks) or simulation task backlog (such as insufficient resource allocation to high timeliness tasks) caused by unreasonable resource allocation in the traditional system is avoided. That is, high timeliness tasks can be accurately given to real-time computing resources, delay is reduced, steady-state simulation occupies hardware real-time resources is avoided, and overall response efficiency is improved. Moreover, it can also ensure that each type of task runs on the most suitable resource, shorten the overall simulation period, while maintaining the accuracy of transient and steady-state simulation.
[0040] Figure 3 is a flowchart of a power grid equipment physical interactive simulation method under the embodied intelligent framework according to another example embodiment of the present application. As shown in Figure 3 The power grid equipment physical interactive simulation method under the embodied intelligent framework provided by the embodiment includes the following steps. S301, the physical layer accesses the sensor cluster for collecting operation data of the power grid equipment, and transmits the collected operation data to the embodied intelligent decision layer.
[0041] In this step, temperature sensors, Hall current sensors, vibration sensors and electromagnetic field sensors can be deployed at key nodes of the power grid equipment (such as transformer windings, circuit breaker contacts, bus connection points, etc.). Among them, the sampling frequency of each sensor is dynamically adjusted according to the device type, for example, microsecond-level sampling is used for power electronic devices, and second-level sampling is used for thermal elements. After the collected data is subjected to noise filtering and dimension normalization processing by the data preprocessing module of the physical layer, it is uploaded to the embodied intelligent decision layer through a low-delay communication link.
[0042] S302, the embodied intelligent decision layer decomposes the operation data into the first input data and the second input data.
[0043] In this step, the embodied intelligence decision layer decomposes the running data into first input data and second input data, wherein the first demand timeliness of the simulation result corresponding to the first input data as simulation input parameters is less than the second demand timeliness of the simulation result corresponding to the second input data as simulation input parameters.
[0044] In one possible case, the first input data is used to simulate the physical transient state of the power grid equipment, and the physical transient state includes at least one of an electronic switch transient state and an electromagnetic transient state.
[0045] Among them, for electronic switch transient state simulation, when the electronic switch (such as circuit breaker, relay, etc.) in the power grid equipment acts, rapid current and voltage changes will occur, and the simulation results of these changes are crucial for evaluating the performance of the switch and the stability of the system. The system collects the voltage, current waveform and other key parameters at the moment of switch action through the sensor cluster in the physical layer in real time as the first input data to the embodied intelligence decision layer. After the embodied intelligence decision layer identifies the high timeliness simulation demand corresponding to these data, it immediately allocates them to the hardware real-time simulation resource area (such as field programmable gate array and / or digital signal processor) in the interactive simulation layer for fast simulation operation. Through the high-speed, real-time data processing capability, the system can quickly capture the transient characteristics in the switch action process, thereby providing accurate fault prediction and performance evaluation basis.
[0046] And for electromagnetic transient state simulation, to simulate the complex electromagnetic transient process generated when the power grid equipment encounters lightning, short circuit and other extreme working conditions, these processes pose a serious threat to the insulation performance of the equipment and the stability of the system. The system collects the electric field, magnetic field intensity and other key parameters in the electromagnetic transient process through electromagnetic field sensors and other devices deployed at key nodes in real time as first input data for simulation analysis. Similarly, the embodied intelligence decision layer allocates these high timeliness data to the hardware real-time simulation resource area for fast processing to accurately simulate the impact of the electromagnetic transient process on the power grid equipment.
[0047] In another possible case, the second input data is used to simulate the physical steady state of the power grid equipment, and the physical steady state includes at least one of a device thermal state and a power flow state.
[0048] Wherein, the power grid equipment in the long running process, due to the heat generated by the current will cause the temperature rise of the equipment, and then affect the performance and life of the equipment. The system through the deployment of temperature sensors and other devices in the key parts of the equipment, real-time collection of equipment temperature, load rate and other steady-state performance data as the second input data. The embodied intelligent decision layer identifies the low timeliness but high accuracy simulation requirements corresponding to these data, and allocates them to the software delay resource area (such as CPU or cloud server) in the interactive simulation layer for simulation analysis. By considering various factors such as equipment material, environmental temperature, load condition, the system can accurately calculate the temperature distribution of each component of the equipment, so as to realize the thermal state simulation of the device.
[0049] And need to carry out power flow state simulation, because the power flow state reflects the distribution and flow of electric energy in the power grid, mainly used to evaluate the economy and stability of the power grid operation. The system collects the voltage, current, power and other parameters of each node of the power grid as the second input data, and carries out complex power flow calculation in the software delay resource area. These calculations take into account the topology of the power grid, line parameters, generator output and other factors, which can accurately simulate the power flow distribution of the power grid under different operating conditions, so as to optimize the dispatching and operation of the power grid.
[0050] S303, the embodied intelligent decision layer calls the same preset time correction module to correct the simulation boundary conditions in the hardware real-time simulation resource area and the software delay resource area.
[0051] In this step, if there is at least one target parameter in the simulation result determined by using the first input data as the boundary condition of the simulation process using the second input data, the embodied intelligent decision layer calls the same preset time correction module to correct the simulation boundary conditions in the hardware real-time simulation resource area and the software delay resource area, so that the consistency of the simulation boundary conditions in the time dimension when the hardware real-time simulation resource area and the software delay resource area are simulated.
[0052] It is worth noting that the hardware real-time simulation resource area and the software delay simulation resource area usually use independent clock sources (usually the default configuration in the device) because they belong to different devices, resulting in systematic deviation in the time reference during simulation. For example, there is a significant difference between the high-precision clock (e.g., error < 1 ppm) generated by the crystal oscillator of the FPGA module in the real-time simulation resource area and the Linux system clock (e.g., error up to 50 ppm) in the software delay simulation resource area. This deviation can cause a delay between the voltage peak timestamp output by the hardware simulation and the time received by the software simulation when transferring cross-domain parameters, and can also cause phase misalignment of dynamic boundary conditions (such as fault current waveforms) in software simulation, thereby causing the long-term simulation results to deviate from the actual device behavior due to accumulated errors.
[0053] Taking a transformer in a power grid device as an example, when the transformer experiences a load mutation, its voltage and current waveforms will change rapidly, and these changes in simulation results have an important impact on the transient performance of the transformer. At the same time, the long-term operating temperature of the transformer is affected by the above transient changes, which is an important boundary condition for evaluating the steady-state performance of the transformer operating temperature. Specifically, during simulation, if it is found that the voltage peak in the simulation results of the first input data is a boundary condition for temperature calculation in the simulation process of the second input data, the preset time correction module will be called by the physical intelligent decision layer to correct the simulation boundary conditions in the two areas. By adjusting the simulation start time or applying a time offset, it is ensured that the impact of load mutation on the transformer can be accurately reflected when calculating the steady-state temperature of the transformer, so as to generate accurate and consistent simulation results.
[0054] In actual simulation, the above situation can be described as follows: after simulation using the first input data (usually corresponding to high-time-efficiency physical transient simulation), there may be at least one target parameter in the simulation results that is a boundary condition for the simulation process using the second input data (usually corresponding to low-time-efficiency but high-accuracy physical steady-state simulation).
[0055] In this case, the hardware real-time simulation resource area needs to ensure both the operation timeliness and the reliable output of simulation results during simulation operation, in order to avoid excessive waiting of the subsequent simulation process due to the inability of the previous simulation process to converge or excessive iteration times. Therefore, in the cross-domain parameter transfer scenario (such as transient voltage peak as a steady-state temperature boundary condition), fixed-point operation can reduce iteration times and ensure the real-time performance of parameter transfer through a fixed calculation path.
[0056] When the simulation tasks are independent, floating-point operation can adapt to different orders of magnitude of data (such as transient current changing from kA to A level) through dynamic adjustment of the exponent, which can avoid numerical overflow caused by fixed bit number in fixed-point operation.
[0057] Therefore, for such interdependent simulation tasks (i.e. the simulation result of the first input data as the boundary condition of the simulation of the second input data), the hardware real-time simulation resource area can perform simulation on the first input data in the manner of fixed-point operation, and return the obtained target parameters to the software time-delay resource area through the embodied intelligent decision layer. If the simulation process using the first input data and the simulation process using the second input data are independent of each other, the hardware real-time simulation resource area performs floating-point operation.
[0058] In a specific application scenario, that is, in the joint simulation of the electromagnetic transient hardware real-time domain (running in the hardware real-time simulation resource area) and the thermal / power flow steady-state software time-delay domain (running in the software time-delay resource area), the key quantities (such as phase voltage, branch current, switching state duty ratio, equivalent power) output by the electromagnetic transient hardware real-time domain are often used as the boundary conditions of the thermal / power flow steady-state software time-delay domain. The thermal / power flow steady-state software time-delay domain solver is sensitive to the time determinacy of the input, specifically, the time jitter of the boundary value and the numerical uncertainty will cause the iteration step to back off, the linearization to fail or the convergence order to decrease, resulting in a significant increase in the total wall clock time. Specifically, the steady-state solver used in the above-mentioned thermal / power flow steady-state software time-delay domain is very sensitive to the time determinacy of the input boundary condition, and it usually works with a large time step or with iteration convergence as the criterion. Once the boundary value provided by the upstream electromagnetic transient hardware real-time domain has time jitter, accidental delay spikes, or numerical discontinuity, the thermal / power flow steady-state software time-delay domain will trigger step back and re-computation, and when the iteration diverges or linearization fails, it will also enter restart or step reduction until timeout, which will cause the end-to-end wall clock time to expand.
[0059] Among them, in the prior art, fixed-point or floating-point operation is usually selected based on hardware resources or local precision, and the coupling information of "whether it is a cross-domain boundary condition" is not used for strategy switching. The above steps are triggered by "whether the first simulation result is used as the second simulation boundary condition" to drive the electromagnetic transient hardware real-time domain to switch between fixed-point and floating-point values, while taking into account the accuracy of the results, it can also improve the time determinacy and timeliness of cross-domain parameter transfer, thereby reducing the waiting and failure retry probability of the thermal / power flow steady-state software time-delay domain.
[0060] S304, the embodied intelligent decision layer calls the hardware real-time simulation resource area in the interactive simulation layer to perform simulation operation on the first input data, and calls the software time-delay resource area in the interactive simulation layer to perform simulation operation on the second input data.
[0061] In this step, for the first input data, since its corresponding simulation result demand time is short, the embodied intelligent decision layer calls the hardware real-time simulation resource area in the interactive simulation layer to perform simulation operation. The area is usually run on a field programmable gate array (FPGA) and / or a digital signal processor (DSP), so as to realize high-speed and real-time data processing and simulation analysis.
[0062] Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application being indicated by the following claims.
[0063] It is to be understood that the application is not limited to the precise details of design and construction that have been described and exemplified above and that various modifications and changes can be made by those skilled in the art without departing from the scope of the application. The scope of the application is limited only by the claims appended hereto.
Claims
1. A physical interactive simulation system for power grid equipment under embodied intelligence framework, characterized in that, The method comprises the following steps: A physical layer, a body intelligence decision layer, and an interactive simulation layer are provided. The physical layer is used for accessing a sensor cluster for collecting operation data of a power grid device and transmitting the collected operation data to the body intelligence decision layer. The body intelligence decision layer decomposes the operation data into first input data and second input data. The body intelligence decision layer calls a hardware real-time simulation resource area in the interactive simulation layer to perform simulation operation of the first input data and calls a software delay resource area in the interactive simulation layer to perform simulation operation of the second input data.
2. The power grid device physical interactive simulation system under the embodied intelligence framework according to claim 1, characterized in that, The hardware real-time simulation resource area runs on a field programmable gate array and / or a digital signal processor.
3. The power grid device physical interactive simulation system under the embodied intelligence framework according to claim 2, characterized in that, The first input data is used for simulating a physical transient state of the power grid device, and the physical transient state comprises at least one of an electronic switch transient state and an electromagnetic transient state.
4. The power grid device physical interactive simulation system under the embodied intelligence framework according to claim 1, characterized in that, The software delay resource area runs on a central processing unit or a cloud server.
5. The power grid device physical interactive simulation system under the embodied intelligence framework according to claim 4, characterized in that, The second input data is used for simulating a physical steady state of the power grid device, and the physical steady state comprises at least one of a device thermal state and a power flow state.
6. The physical interactive simulation system of power grid equipment under the embodied intelligence framework according to any one of claims 1-5, characterized in that, If at least one target parameter in a simulation result determined after simulation by using the first input data is a boundary condition of a simulation process by using the second input data, the hardware real-time simulation resource area performs fixed-point operation on the first input data, and the obtained target parameter is returned to the software delay resource area through the body intelligence decision layer. If the simulation process by using the first input data and the simulation process by using the second input data are independent of each other, the hardware real-time simulation resource area performs floating-point operation.
7. The power grid device physical interactive simulation system under the embodied intelligence framework according to claim 6, characterized in that, If at least one target parameter in a simulation result determined after simulation by using the first input data is a boundary condition of a simulation process by using the second input data, the body intelligence decision layer calls the hardware real-time simulation resource area in the interactive simulation layer to perform simulation operation of the first input data and calls the software delay resource area in the interactive simulation layer to perform simulation operation of the second input data, which comprises the following steps: The body intelligence decision layer calls a same preset time correction module to perform time correction on simulation boundary conditions in the hardware real-time simulation resource area and the software delay resource area, so as to make the simulation boundary conditions consistent in time dimension when the hardware real-time simulation resource area and the software delay resource area perform simulation operation.
8. A method for physical interactive simulation of power grid equipment under embodied intelligence framework, characterized in that, The system comprises a physical layer, a body intelligence decision layer, and an interactive simulation layer. The physical layer accesses a sensor cluster for collecting operation data of a power grid device and transmits the collected operation data to the body intelligence decision layer. The body intelligence decision layer decomposes the operation data into first input data and second input data. The body intelligence decision layer calls a hardware real-time simulation resource area in the interactive simulation layer to perform simulation operation of the first input data and calls a software delay resource area in the interactive simulation layer to perform simulation operation of the second input data. The body intelligence decision layer calls a hardware real-time simulation resource area in the interactive simulation layer to perform simulation operation of the first input data, and calls a software delay resource area in the interactive simulation layer to perform simulation operation of the second input data.
9. The method of claim 8, wherein the physical interaction of the power grid device is simulated in the embodied intelligence framework. The first input data is used to simulate a physical transient state of the power grid device, and the physical transient state includes at least one of an electronic switch transient state and an electromagnetic transient state.
10. The method of claim 8, wherein the physical interaction of the power grid device is simulated in the embodied intelligence framework. The second input data is used to simulate a physical steady state of the power grid device, and the physical steady state includes at least one of a device thermal state and a power flow state.
Citation Information
Patent Citations
Transient state and steady state hybrid simulation method and system for power distribution network
CN112803392A
Hybrid model simulation real-time scheduling system
CN117234102A
Multilayer nested control loop real-time simulation control system
CN117289622A
Line loss simulation application method and system based on intelligent fault diagnosis technology
CN119249908A
Modeling simulation analysis system for photovoltaic power station
CN119623069A
