A method and system for online digital twin analysis of power equipment

Abstract

The present invention relates to the technical field of equipment simulation, and discloses an online digital twin analysis method and system for power equipment. A fast calculation model of a target power equipment is constructed through a multi-physical field numerical simulation model, and the fast calculation model is interactively responded to sensor data collected by a sensor. The minimum error objective function is optimized by comparing the error between the sensor data and the physical field calculation value of the fast calculation model, and the physical field distribution calculation result calculated by the fast calculation model is output. The multi-physical field distribution of the power equipment can be accurately characterized, so that the multi-physical field distribution of the power equipment can be quickly obtained by using the fast calculation model according to the data collected by the sensor, thereby realizing a real-time simulation effect of the equipment status.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment simulation, and in particular to an online digital twin analysis method and system for power equipment. Background Art

[0002] With the continuous advancement of the construction of new power systems, renewable energy equipment with new energy as the main body and ultra-high voltage / ultra-high voltage transmission equipment with long-distance power transmission as the main body have been further developed and applied. As an important carrier of the power grid system, the operation and maintenance of high-voltage transmission and transformation equipment are gradually tending to establish an active and advanced safety production management system and an emergency prevention system, and to build an intrinsic safety system for equipment operation.

[0003] Digital twins mainly use physical models, sensors, historical operation data, etc. to integrate multi-disciplinary, multi-physical, multi-scale, and multi-probability simulation processes, complete mapping in virtual space, and thus reflect the full life cycle status of the corresponding physical equipment. The digital twin of power transmission and transformation equipment is a virtual representation of the power transmission and transformation equipment entity in digital space. It can be used throughout the entire life cycle of product design, manufacturing, operation and maintenance, and scrap recycling. It is a technical means to realize the digitization and intelligence of equipment.

[0004] The existing digital twin technology solutions design digital twins into physical layer, perception layer, data layer, model layer and application layer, but no practical and real-time technology solutions are proposed to achieve digital twin virtual-real fusion, perception interaction, etc. In general engineering, the main technical solutions for digital twin converter stations are geometric model display, visualization of data collected by sensors, simple threshold judgment, and simulation analysis using multi-physical field simulation models on the design side.

[0005] However, existing digital twin technology solutions are difficult to interact and respond with sensors to achieve real-time simulation of device status, and do not consider the impact of physical processes such as the distribution of multiple physical field quantities. Summary of the invention

[0006] The present invention provides an online digital twin analysis method and system for power equipment, which solves the technical problems that the existing digital twin technical solutions are difficult to interact and respond with sensors to achieve real-time simulation of equipment status, and do not consider the influence of physical processes such as the distribution of multiple physical field quantities.

[0007] In view of this, the first aspect of the present invention provides an online digital twin analysis method for power equipment, comprising:

[0008] Perform multi-physics field simulation modeling on the target power equipment and establish a multi-physics field numerical simulation model of the target power equipment;

[0009] Performing a multi-physics field simulation analysis on the target power equipment according to the multi-physics field numerical simulation model, and constructing a fast calculation model of the target power equipment according to the multi-physics field simulation analysis result;

[0010] Acquire sensor data collected by multiple preset sensors on the target power equipment;

[0011] Using the fast calculation model to perform physical field calculation on the target power equipment, and determining physical field calculation values ​​corresponding to positions of the plurality of preset sensors;

[0012] Comparing the error between the sensor data and the physical field calculation value corresponding to the sensor data, and determining whether the error is greater than a preset first error threshold;

[0013] When it is determined that the error is greater than the preset first error threshold, the preset minimum error objective function is optimized and solved until the error converges, and the physical field distribution calculation result calculated by the fast calculation model is output.

[0014] Preferably, after the step of performing multi-physics field simulation analysis on the target power equipment according to the multi-physics field numerical simulation model and constructing a fast calculation model of the target power equipment according to the multi-physics field simulation analysis result, the method further includes:

[0015] Determining whether the input amount of the fast calculation model exceeds a preset input amount range threshold;

[0016] If it is determined that the input of the fast calculation model exceeds the preset input range threshold, the multi-physics field numerical simulation model and the fast calculation model are used to perform physical field calculations on the input respectively;

[0017] Comparing the second errors of two physical field calculation values ​​obtained by respectively calculating the input quantity through the multi-physical field numerical simulation model and the fast calculation model;

[0018] When a second error between the two physical field calculation values ​​is greater than a preset second error threshold, the input quantity and the physical field calculation value calculated by the multi-physical field numerical simulation model are added to the data set used to construct the fast calculation model;

[0019] The fast calculation model is reconstructed using the added data set.

[0020] Preferably, the preset minimum error objective function is constructed with the minimum error between the sensor data and the physical field calculation value corresponding to the sensor data as the target condition, and with the device simulation state parameters in the fast calculation model as the decision variables.

[0021] Preferably, the method further comprises:

[0022] Simulating, by means of the multi-physics field numerical simulation model, the multi-physics field simulation distribution results of the target power equipment under a plurality of different preset fault conditions, the operating condition data corresponding to the multi-physics field simulation distribution results, and the physical field simulation calculation values ​​at the corresponding positions of the preset sensors;

[0023] Taking the operating condition data and the physical field simulation calculation value at the corresponding position of the preset sensor as input, and taking the multi-physical field simulation distribution result as output to construct a training data set;

[0024] Training the training data set based on a machine learning algorithm to obtain a trained multi-physics field distribution calculation model;

[0025] The multi-physical field distribution of the target power equipment in the current period is obtained by inputting the operating condition data in the current period and the sensor data collected by the preset sensor into the trained multi-physical field distribution calculation model.

[0026] Preferably, the method further comprises:

[0027] When the target power equipment fails, the sensor data and operating condition data collected by the preset sensors when the target power equipment fails are input into the trained multi-physical field distribution calculation model to obtain the multi-physical field distribution of the target power equipment in the current fault period;

[0028] The fault condition of the target power device is determined according to the multi-physical field distribution of the target power device in the current fault period.

[0029] Preferably, the method further comprises:

[0030] When the target power equipment fails, extracting the physical field calculation value at the position corresponding to the preset sensor in the multi-physical field distribution of the target power equipment during the current fault period;

[0031] Comparing the sensor data collected by the preset sensor after the target power device fails with a third error of a physical field calculation value at a position corresponding to the preset sensor in a multi-physical field distribution of the target power device;

[0032] When the third error is greater than a preset third error threshold, a fault condition when the target power equipment fails is obtained;

[0033] Adding the operating condition data under the fault condition when the target power equipment fails, the physical field simulation calculation value at the corresponding position of the preset sensor, and the multi-physical field simulation distribution result to the training data set;

[0034] The trained multi-physics field distribution calculation model is retrained using the added training data set.

[0035] Preferably, when it is determined that the error is greater than the preset first error threshold, the preset minimum error objective function is optimized and solved until the error converges, and after the step of outputting the physical field distribution calculation result calculated by the fast calculation model, it also includes:

[0036] Determine the multi-physical field value at the preset key position according to the physical field distribution calculation result;

[0037] The multi-physical field value at the preset key position is compared with the preset physical field alarm threshold, and when the multi-physical field value at the preset key position exceeds the preset physical field alarm threshold, an alarm signal is issued.

[0038] In a second aspect, the present invention further provides an online digital twin analysis system for power equipment, comprising:

[0039] A simulation modeling module, used to perform multi-physics field simulation modeling on the target power equipment and establish a multi-physics field numerical simulation model of the target power equipment;

[0040] A model building module, used to perform multi-physics field simulation analysis on the target power equipment according to the multi-physics field numerical simulation model, and to build a fast calculation model of the target power equipment according to the multi-physics field simulation analysis result;

[0041] A sensor acquisition module, used to acquire sensor data collected by multiple preset sensors on the target power equipment;

[0042] A physical field calculation module, used to perform physical field calculation on the target power equipment using the fast calculation model, and determine physical field calculation values ​​at positions corresponding to the plurality of preset sensors;

[0043] An error comparison module, used to compare the error between the sensor data and the physical field calculation value corresponding to the sensor data, and determine whether the error is greater than a preset first error threshold;

[0044] The physical field optimization module is used to optimize and solve the preset minimum error objective function when it is determined that the error is greater than the preset first error threshold, until the error converges, and then output the physical field distribution calculation result calculated by the fast calculation model.

[0045] In a third aspect, the present invention also provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the online digital twin analysis method for power equipment as described above.

[0046] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the steps of the above-mentioned online digital twin analysis method for power equipment.

[0047] It can be seen from the above technical solutions that the present invention has the following advantages:

[0048] The present invention constructs a fast calculation model of the target power equipment through a multi-physical field numerical simulation model, and enables the fast calculation model to interactively respond to the sensor data collected by the sensor, and optimizes and solves the minimum error objective function by comparing the error between the sensor data and the physical field calculation value of the fast calculation model, and outputs the physical field distribution calculation result calculated by the fast calculation model, which can accurately characterize the multi-physical field distribution of the power equipment, so that the multi-physical field distribution of the power equipment can be quickly obtained by using the fast calculation model according to the data collected by the sensor, thereby realizing the real-time simulation effect of the equipment status. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A flowchart of an online digital twin analysis method for power equipment provided by an embodiment of the present invention;

[0050] Figure 2 A schematic diagram of the structure of an online digital twin analysis system for power equipment provided by an embodiment of the present invention;

[0051] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0054] The existing digital twin technology solutions design digital twins into physical layer, perception layer, data layer, model layer and application layer, but no practical and real-time technology solutions are proposed to achieve digital twin virtual-real fusion, perception interaction, etc. In general engineering, the main technical solutions for digital twin converter stations are geometric model display, visualization of data collected by sensors, simple threshold judgment, and simulation analysis using multi-physical field simulation models on the design side.

[0055] However, existing digital twin technology solutions are difficult to interact and respond with sensors to achieve real-time simulation of device status, and do not consider the impact of physical processes such as the distribution of multiple physical field quantities.

[0056] In view of this, an embodiment of the present invention provides an online digital twin analysis method for power equipment. This embodiment can be applicable to the situation of online digital twin analysis of power equipment. The method can be executed by an online digital twin analysis device for power equipment. The online digital twin analysis device for power equipment can be implemented in the form of hardware and / or software. The online digital twin analysis device for power equipment can be configured in a computer device.

[0057] like Figure 1 As shown, an online digital twin analysis method for power equipment provided by an embodiment of the present invention includes:

[0058] Step S1, performing multi-physical field simulation modeling on the target power equipment, and establishing a multi-physical field numerical simulation model of the target power equipment.

[0059] The target power equipment refers to equipment in the power system, such as transformers, etc. The target power equipment is finely modeled, and the multi-physical field distribution of the target power equipment can be determined through the established multi-physical field numerical simulation model.

[0060] Step S2: performing multi-physics field simulation analysis on the target power equipment according to the multi-physics field numerical simulation model, and constructing a fast calculation model of the target power equipment according to the multi-physics field simulation analysis results.

[0061] The fast calculation model is the Fast CAE model, which includes the data-driven FAST CAE solver and the model-driven FAST CAE solver.

[0062] Data-driven FAST CAE solver: Based on multi-physics field simulation, enumerate the multi-physics field distribution under different operating conditions and environmental data conditions (this distribution is the data of the grid nodes in the multi-physics field simulation), and use machine learning or other similar methods to build a mapping relationship between the operating conditions, environmental data and the data of the grid nodes in the multi-physics field simulation, and train the data set composed of the mapping relationship to obtain a mapping relationship model. After inputting the operating conditions and environmental data through the mapping relationship model, the data of the grid nodes in the multi-physics field simulation, that is, the multi-physics field distribution, can be quickly calculated.

[0063] Solver of model-driven method: It uses principal component analysis and other methods to compress the size of the solution equation matrix of numerical simulation calculation and accelerate the calculation process. That is, in the calculation process of finite element and finite volume, a set of stiffness matrices will be formed. The stiffness matrix can be reduced by principal component analysis and other methods to reduce the amount of calculation and speed up the calculation.

[0064] The fast calculation model in this embodiment can be constructed using the construction method in the prior art, which will not be described in detail here.

[0065] When the operating state of the target power equipment changes, the fast calculation model can be used to quickly calculate the multi-physical field distribution of the equipment under the expected operating condition data after the target power equipment state changes, and determine the target power equipment's tolerance to system changes.

[0066] For example, in a transformer overload assessment, the current is adjusted from 4kA to 4.5kA. Before the adjustment, the multi-physical field distribution inside the transformer is quickly calculated in advance to determine whether it can withstand the overload operating conditions.

[0067] When a transient process occurs in the target power equipment, a fast calculation model can be used to quickly calculate the multi-physical fields of the equipment when it is subjected to the transient process and determine whether the design margin is exceeded.

[0068] For example, when short-circuit current enters the transformer, the transformer performs a rapid calculation of electromagnetic force to analyze whether the electromagnetic force exceeds the design margin. Combined with the cumulative effect of the transformer winding (equipment status prediction function), the remaining service life of the transformer can be predicted.

[0069] In some embodiments, the fast calculation model needs to be verified and corrected, and the process includes:

[0070] Step S21, determining whether the input quantity of the rapid calculation model exceeds a preset input quantity range threshold.

[0071] Step S22: if it is determined that the input of the fast calculation model exceeds the preset input range threshold, the multi-physics field numerical simulation model and the fast calculation model are used to perform physical field calculations on the input respectively;

[0072] Step S23, comparing the second errors of the two physical field calculation values ​​obtained by respectively calculating the input quantity through the multi-physical field numerical simulation model and the fast calculation model;

[0073] Step S24: when the second error between the two physical field calculation values ​​is greater than a preset second error threshold, the input quantity and the physical field calculation value calculated by the multi-physical field numerical simulation model are added to the data set used to construct the fast calculation model;

[0074] Step S25: Reconstruct the fast calculation model using the added data set.

[0075] Among them, the fast calculation model using data-driven mode has an upper and lower bound of the input quantity (input quantity range threshold). When the new input quantity exceeds the upper and lower bounds, it is necessary to apply the multi-physics field numerical simulation model to simulate the new input quantity, and compare the simulation calculation results (multi-physics field distribution calculation results) with the simulation calculation results calculated by the fast calculation model using data-driven mode. If the error between the two simulation calculation results is not greater than a certain value, there is no need to correct the fast calculation model. If the error between the two simulation calculation results is greater than a certain value, the new input quantity and its corresponding simulation calculation results (multi-physics field distribution calculation results) are added to the data set used to build the fast calculation model, and the fast calculation model is rebuilt.

[0076] It should be noted that if the new input data exceeds the upper limit value too much, such as the upper interface is 110°C and the new input is 180°C, it is necessary to sample the sample space (that is, take the new input as the upper limit of the variable, combine it with the operating condition data, etc., and re-construct the boundary condition sampling of the fast calculation model) to form a new data set and rebuild the fast calculation model.

[0077] Step S3: Acquire sensor data collected by multiple preset sensors on the target power equipment.

[0078] Step S4: Use a fast calculation model to perform physical field calculation on the target power equipment to determine physical field calculation values ​​corresponding to positions of multiple preset sensors.

[0079] The physical field calculation values ​​corresponding to the positions of the plurality of preset sensors calculated by the fast calculation model are of the same data type as the sensor data, such as temperature value data, magnetic value data, and the like.

[0080] It should be noted that the fast calculation model and the preset sensor perform physical field calculations on the same power equipment, so the position calculated by the fast calculation model and the position of the sensor can correspond through coordinates.

[0081] Exemplarily, a local coordinate system is established at a fixed point of the power equipment to determine the position of the sensor, and then the simulation result of the position is extracted from the multi-physical field distribution calculated by the fast calculation model.

[0082] Step S5: compare the error between the sensing data and the physical field calculation value corresponding to the sensing data, and determine whether the error is greater than a preset first error threshold.

[0083] In a general example, there are usually multiple preset sensors, such as 8, 10, etc. When comparing the errors between the sensor data and the physical field calculation values ​​corresponding to the sensor data, the errors between multiple sensor data and the physical field calculation values ​​corresponding to the multiple sensor data can be compared simultaneously. When all errors are not greater than the preset first error threshold, it means that the multi-physical field distribution obtained by the fast calculation model can reliably describe the multi-physical field distribution of the target power equipment, and the multi-physical field distribution results are output through the fast calculation model. When all errors are greater than the preset first error threshold, it means that the multi-physical field distribution obtained by the fast calculation model cannot reliably describe the multi-physical field distribution of the target power equipment.

[0084] Step S6: When the error is judged to be greater than a preset first error threshold, the preset minimum error objective function is optimized and solved until the error converges, and the physical field distribution calculation result calculated by the fast calculation model is output.

[0085] Among them, the preset minimum error objective function is constructed with the minimum error between the sensor data and the physical field calculation value corresponding to the sensor data as the target condition, and with the device simulation state parameters in the fast calculation model as the decision variables.

[0086] Among them, the minimum error objective function can be characterized by a mean square error function. In the process of constructing the minimum error objective function, in order to stabilize the stability of the target model solution process, a regularization term can be added to the minimum error objective function, and the regularization term includes L1 regularization, L2 regularization, etc.

[0087] The device simulation state parameters can characterize the operating state of the target power equipment, such as the fault state or the normal operating state. The device simulation state parameters include the material parameters and boundary conditions of the equipment. The fast calculation model is constructed by the multi-physical field numerical simulation model of the target power equipment. The device simulation state parameters participate in the construction of the fast calculation model. At the same time, since the physical field calculation values ​​corresponding to multiple different device simulation state parameters can be calculated in the fast calculation model through multiple different device simulation state parameters, there is a mapping relationship between the device simulation state parameters and the physical field calculation values ​​calculated by the fast calculation model. The correlation function between the device simulation state parameters and the physical field calculation values ​​can be formed through numerical fitting.

[0088] In some embodiments, an optimization algorithm can be used to find the optimal solution for the minimum error objective function. When the error value is less than a preset first error threshold, the optimal solution is obtained to determine the optimal device simulation state parameters, and output the calculation results of the physical field distribution that can reliably describe the target power equipment.

[0089] Among them, the optimization algorithm can adopt the particle swarm algorithm. The speed update formula of the particle swarm algorithm is:

[0090]

[0091] In the formula, and is the acceleration factor (manually selected); and is a random number, and its value range is [0, 1]. Indicates the position of each particle under the local optimal solution (the device simulation state parameter with the smallest error in the last calculation), Indicates the position of the particle swarm under the overall optimal solution (the device simulation state parameter that minimized the overall error in the last calculation), represents the particle position at the kth iteration (device simulation state parameter of the device), Indicates the speed of the kth iteration (the change in the device simulation state parameter for the kth time), represents the speed of the k+1th iteration, represents the particle position at the k+1th iteration.

[0092] Among them, the velocity update formula is used to update the position of the particle.

[0093] It should be noted that the present invention constructs a fast calculation model of the target power equipment through a multi-physical field numerical simulation model, and enables the fast calculation model to interact and respond to the sensor data collected by the sensor, and optimizes and solves the minimum error objective function by comparing the error between the sensor data and the physical field calculation value of the fast calculation model, and outputs the physical field distribution calculation results calculated by the fast calculation model, which can accurately characterize the multi-physical field distribution of the power equipment, so that the multi-physical field distribution of the power equipment can be quickly obtained according to the data collected by the sensor using the fast calculation model, thereby achieving a real-time simulation effect of the equipment status.

[0094] In some embodiments of the present invention, a multi-physics field simulation distribution calculation method based on a machine learning model is also proposed. Specifically, the method further includes:

[0095] S11. Simulate the multi-physical field simulation distribution results of the target power equipment under multiple different preset fault conditions through a multi-physical field numerical simulation model, as well as the operating condition data corresponding to the multi-physical field simulation distribution results and the physical field simulation calculation values ​​at the corresponding positions of the preset sensors.

[0096] The operating condition data includes but is not limited to the voltage, current and environmental conditions of the equipment. The physical field simulation calculation value at the corresponding position of the preset sensor can be calculated by a fast calculation model or the like.

[0097] Among them, by adjusting the boundary conditions of the multi-physics field numerical simulation model, multiple different preset fault conditions can be simulated. For example, when simulating the fault of the current-carrying connection structure of the bushing, the contact resistance value of the current-carrying connection structure can be adjusted, and the operating condition data can be determined by monitoring the infrared temperature data of the outer sheath. By adjusting the resistivity between the short-circuited turns, the short-circuit fault of the transformer winding can be simulated, and the operating condition data can be determined by monitoring the temperature data of the transformer tank. By adjusting the transformer winding shape, the deformation fault of the transformer winding can be simulated, and the vibration data of the transformer tank can be monitored to determine the operating condition data. By adjusting the bolt tightening force, the loosening fault of the decomposition switch bolt can be simulated, and the vibration signal during the tap switch stroke and the tap switch external operation can be monitored to determine the operating condition data.

[0098] S12. Taking the operating condition data and the physical field simulation calculation values ​​at the corresponding positions of the preset sensors as input, and taking the multi-physical field simulation distribution results as output, a training data set is constructed.

[0099] Among them, a mapping relationship is formed between input and output.

[0100] S13. Train the training data set based on the machine learning algorithm to obtain a trained multi-physical field distribution calculation model.

[0101] Among them, the machine learning algorithm can be a deep neural network.

[0102] S14. The multi-physical field distribution of the target power equipment in the current period is obtained by inputting the operating condition data in the current period and the sensor data collected by the preset sensors into the trained multi-physical field distribution calculation model.

[0103] It can be understood that by inputting the sensor data and operating condition data collected by the sensors in the current period into the trained multi-physical field distribution calculation model, the multi-physical field distribution of the target power equipment in the previous period can be predicted.

[0104] When a target power device fails, the failure condition of the target power device can be obtained according to the sensor data at the time of the failure.

[0105] Specifically, the method further includes:

[0106] S111. When a target power device fails, the sensor data collected by the preset sensors at the time of the target power device failure and the operating condition data are input into the trained multi-physical field distribution calculation model to obtain the multi-physical field distribution of the target power device in the current fault period.

[0107] S112. Determine a fault condition of the target power equipment according to the multi-physical field distribution of the target power equipment during the current fault period.

[0108] It can be understood that, since there is a mapping relationship between the multi-physical field distribution of the target power equipment and the fault condition, the fault condition can be determined by the multi-physical field distribution of the target power equipment in the current period.

[0109] It should be noted that since the simulated fault conditions are difficult to completely cover all fault conditions, the multi-physics field distribution calculation model needs to be corrected.

[0110] In some embodiments, the method further comprises:

[0111] S121. When a target power device fails, extract the physical field calculation value at the position corresponding to the preset sensor in the multi-physical field distribution of the target power device during the current fault period.

[0112] S122: Compare the sensor data collected by the preset sensor after the target power equipment fails with a third error of the physical field calculation value at the corresponding position of the preset sensor in the multi-physical field distribution of the target power equipment.

[0113] S123. When the third error is greater than a preset third error threshold, a fault condition when a fault occurs in the target power equipment is obtained.

[0114] Among them, when the error is large, the target power equipment can be shut down for maintenance, and the fault conditions of the target power equipment and the sensor data values ​​before the power outage can be obtained during the maintenance period.

[0115] S124, adding the operating condition data under the fault condition when the target power equipment fails, the physical field simulation calculation value at the corresponding position of the preset sensor, and the multi-physical field simulation distribution result to the training data set.

[0116] S125. Retrain the trained multi-physics field distribution calculation model using the added training data set.

[0117] In some embodiments, step S6 further includes:

[0118] Step S701: determine the multi-physical field values ​​at preset key positions according to the physical field distribution calculation results.

[0119] Step S702: compare the multi-physical field value at the preset key position with the preset physical field alarm threshold, and when the multi-physical field value at the preset key position exceeds the preset physical field alarm threshold, issue an alarm signal.

[0120] In addition, the multi-physical field distribution of the target power equipment can be displayed to the operation and maintenance personnel in real time, and the full life cycle status data (historical data) of the equipment during operation can be recorded, and the future equipment operation status can be analyzed based on the historical data and the judgment data of the equipment.

[0121] For example, the transformer may be subjected to short-circuit current shocks of varying degrees throughout its life cycle. The transformer winding has a cumulative effect on short-circuit current, that is, after a single short-circuit current shock with a smaller peak value, the winding can operate normally, but after multiple shocks in the same situation, the winding may become loose, the insulation may break down, etc. (similar to mechanical fatigue). Therefore, the short-circuit shocks that the transformer has been subjected to in history are statistically analyzed, as well as the distribution of short-circuits under the shocks. By presetting the amplitude and duration of the next shock, it can be determined whether the winding can withstand it. Similarly, the remaining life of the winding can be obtained.

[0122] The above is a detailed description of an embodiment of an online digital twin analysis method for power equipment provided by the present invention. The following is a detailed description of an embodiment of an online digital twin analysis system for power equipment provided by the present invention.

[0123] like Figure 2 As shown, an embodiment of the present invention provides an online digital twin analysis system for power equipment, including:

[0124] The simulation modeling module 100 is used to perform multi-physics field simulation modeling on the target power equipment and establish a multi-physics field numerical simulation model of the target power equipment;

[0125] A model building module 200 is used to perform multi-physics field simulation analysis on a target power device according to a multi-physics field numerical simulation model, and to build a fast calculation model of the target power device according to the multi-physics field simulation analysis result;

[0126] The sensor acquisition module 300 is used to acquire sensor data collected by multiple preset sensors on the target power equipment;

[0127] The physical field calculation module 400 is used to perform physical field calculation on the target power equipment using a fast calculation model, and determine the physical field calculation values ​​corresponding to the positions of the plurality of preset sensors;

[0128] The error comparison module 500 is used to compare the error between the sensing data and the physical field calculation value corresponding to the sensing data, and determine whether the error is greater than a preset first error threshold;

[0129] The physical field optimization module 600 is used to optimize and solve the preset minimum error objective function when the judgment error is greater than the preset first error threshold, until the error converges, and output the physical field distribution calculation result calculated by the fast calculation model.

[0130] like Figure 3 As shown, an embodiment of the present invention further provides an electronic device, the electronic device 10 includes a memory 20 and a processor 30, the memory 20 stores a computer program, and when the computer program is executed by the processor 30, the processor 30 executes the steps of the online digital twin analysis method of power equipment in any of the above embodiments.

[0131] An embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed, the steps of the online digital twin analysis method of power equipment in any of the above embodiments are implemented.

[0132] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, electronic device and computer storage medium can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0133] In several embodiments provided by the present invention, it is understood that each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and a part of a module, a program segment or a code includes one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved.

[0134] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, electronic devices, computer storage media and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0135] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0136] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0137] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for executing all or part of the steps of the method described in each embodiment of the present invention through a computer device (which can be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (full name in English: Read-Only Memory, English abbreviation: ROM), random access memory (full name in English: Random Access Memory, English abbreviation: RAM), disk or optical disk and other media that can store program codes.

[0138] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An online digital twin analysis method for power equipment, characterized in that: include: Perform multi-physics field simulation modeling on the target power equipment and establish a multi-physics field numerical simulation model of the target power equipment; Performing a multi-physics field simulation analysis on the target power equipment according to the multi-physics field numerical simulation model, and constructing a fast calculation model of the target power equipment according to the multi-physics field simulation analysis result; Acquire sensor data collected by multiple preset sensors on the target power equipment; Using the fast calculation model to perform physical field calculation on the target power equipment, and determining physical field calculation values ​​corresponding to positions of the plurality of preset sensors; Comparing the error between the sensor data and the physical field calculation value corresponding to the sensor data, and determining whether the error is greater than a preset first error threshold; When it is determined that the error is greater than the preset first error threshold, the preset minimum error objective function is optimized and solved until the error converges, and the physical field distribution calculation result calculated by the fast calculation model is output; After the step of performing multi-physics field simulation analysis on the target power equipment according to the multi-physics field numerical simulation model and constructing a fast calculation model of the target power equipment according to the multi-physics field simulation analysis result, the method further includes: Determining whether the input amount of the fast calculation model exceeds a preset input amount range threshold; If it is determined that the input of the fast calculation model exceeds the preset input range threshold, the multi-physics field numerical simulation model and the fast calculation model are used to perform physical field calculations on the input respectively; Comparing the second errors of two physical field calculation values ​​obtained by respectively calculating the input quantity through the multi-physical field numerical simulation model and the fast calculation model; When a second error between the two physical field calculation values ​​is greater than a preset second error threshold, the input quantity and the physical field calculation value calculated by the multi-physical field numerical simulation model are added to the data set used to construct the fast calculation model; The fast calculation model is reconstructed using the added data set.

2. The online digital twin analysis method for power equipment according to claim 1 is characterized in that: The preset minimum error objective function is constructed with the minimum error between the sensor data and the physical field calculation value corresponding to the sensor data as the target condition, and with the device simulation state parameters in the fast calculation model as decision variables.

3. The online digital twin analysis method for power equipment according to claim 1 is characterized in that: Also includes: Simulating, by means of the multi-physics field numerical simulation model, the multi-physics field simulation distribution results of the target power equipment under a plurality of different preset fault conditions, the operating condition data corresponding to the multi-physics field simulation distribution results, and the physical field simulation calculation values ​​at the corresponding positions of the preset sensors; Taking the operating condition data and the physical field simulation calculation value at the corresponding position of the preset sensor as input, and taking the multi-physical field simulation distribution result as output to construct a training data set; Training the training data set based on a machine learning algorithm to obtain a trained multi-physics field distribution calculation model; The multi-physical field distribution of the target power equipment in the current period is obtained by inputting the operating condition data in the current period and the sensor data collected by the preset sensor into the trained multi-physical field distribution calculation model.

4. The online digital twin analysis method for power equipment according to claim 3 is characterized in that: Also includes: When the target power equipment fails, the sensor data and operating condition data collected by the preset sensors when the target power equipment fails are input into the trained multi-physical field distribution calculation model to obtain the multi-physical field distribution of the target power equipment in the current fault period; The fault condition of the target power device is determined according to the multi-physical field distribution of the target power device in the current fault period.

5. The online digital twin analysis method for power equipment according to claim 4 is characterized in that: Also includes: When the target power equipment fails, extracting the physical field calculation value at the position corresponding to the preset sensor in the multi-physical field distribution of the target power equipment during the current fault period; Comparing the sensor data collected by the preset sensor after the target power device fails with a third error of a physical field calculation value at a position corresponding to the preset sensor in a multi-physical field distribution of the target power device; When the third error is greater than a preset third error threshold, a fault condition when the target power equipment fails is obtained; Adding the operating condition data under the fault condition when the target power equipment fails, the physical field simulation calculation value at the corresponding position of the preset sensor, and the multi-physical field simulation distribution result to the training data set; The trained multi-physics field distribution calculation model is retrained using the added training data set.

6. The online digital twin analysis method for power equipment according to claim 1 is characterized in that: When it is determined that the error is greater than the preset first error threshold, the preset minimum error objective function is optimized and solved until the error converges, and after the step of outputting the physical field distribution calculation result calculated by the fast calculation model, it also includes: Determine the multi-physical field value at the preset key position according to the physical field distribution calculation result; The multi-physical field value at the preset key position is compared with the preset physical field alarm threshold, and when the multi-physical field value at the preset key position exceeds the preset physical field alarm threshold, an alarm signal is issued.

7. An online digital twin analysis system for power equipment, characterized in that: include: A simulation modeling module, used to perform multi-physics field simulation modeling on the target power equipment and establish a multi-physics field numerical simulation model of the target power equipment; A model building module, used to perform multi-physics field simulation analysis on the target power equipment according to the multi-physics field numerical simulation model, and to build a fast calculation model of the target power equipment according to the multi-physics field simulation analysis result; A sensor acquisition module, used to acquire sensor data collected by multiple preset sensors on the target power equipment; A physical field calculation module, used to perform physical field calculation on the target power equipment using the fast calculation model, and determine physical field calculation values ​​at positions corresponding to the plurality of preset sensors; An error comparison module, used to compare the error between the sensor data and the physical field calculation value corresponding to the sensor data, and determine whether the error is greater than a preset first error threshold; A physical field optimization module, for, when it is determined that the error is greater than the preset first error threshold, optimizing and solving the preset minimum error objective function until the error converges, and outputting the physical field distribution calculation result calculated by the fast calculation model; The method further comprises: performing a multi-physics field simulation analysis on the target power equipment according to the multi-physics field numerical simulation model, and constructing a fast calculation model of the target power equipment according to the multi-physics field simulation analysis result; and then: Determining whether the input amount of the fast calculation model exceeds a preset input amount range threshold; If it is determined that the input of the fast calculation model exceeds the preset input range threshold, the multi-physics field numerical simulation model and the fast calculation model are used to perform physical field calculations on the input respectively; Comparing the second errors of two physical field calculation values ​​obtained by respectively calculating the input quantity through the multi-physical field numerical simulation model and the fast calculation model; When a second error between the two physical field calculation values ​​is greater than a preset second error threshold, the input quantity and the physical field calculation value calculated by the multi-physical field numerical simulation model are added to the data set used to construct the fast calculation model; The fast calculation model is reconstructed using the added data set.

8. An electronic device, characterized in that: The electronic device includes a memory and a processor, wherein a computer program is stored in the memory. When the computer program is executed by the processor, the processor executes the steps of the online digital twin analysis method for power equipment as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the steps of the online digital twin analysis method of power equipment as described in any one of claims 1 to 6 are implemented.

Images