A method for rapid simulation and breaking strategy decision of multi-working condition electric arc magnetic fluid of environmentally friendly gas medium ring main unit

Abstract

This invention discloses a rapid simulation and interruption strategy decision-making method for multi-condition arc magnetohydrodynamic (MHD) circuit breakers in environmentally friendly gas media ring main units, relating to the field of arc extinguishing. It uses the changing state of the current curve to predict the current at future time points. Since current generates electromagnetic force, it can produce an electric arc. Based on the gas type, current curve, and tripping speed, the method also includes rapid MHD simulation of the gas velocity, gas pressure, and gas density corresponding to the gas type at future time points using the Navier-Stokes equations. Because gas pressure, gas velocity, and gas density are difficult to detect, and current affects these parameters, calculations are used to identify data that, with a small amount of data, can significantly impact the arc extinguishing operation. Physical equipment in the ring main unit is then configured according to this arc extinguishing data for interruption. Furthermore, the method utilizes environmentally friendly gas within the ring main unit for arc extinguishing, achieving significant results with minimal effort and predicting future states.

Description

Technical Field

[0001] This invention relates to the field of arc breaking and extinguishing, and more specifically, to a rapid simulation and arc breaking strategy decision-making method for multi-condition electric arc magnetofluid in environmentally friendly gas medium ring main units. Background Technology

[0002] Arc magnetohydrodynamics is a physical model for viewing and processing electric arcs. It no longer simply views the arc as a "jumping spark," but rather as a stream of high-temperature, conductive, luminous fluid, whose motion and behavior are strongly governed by electromagnetic forces. Because gas pressure, velocity, and density are difficult to measure, the Navier-Stokes equations—essentially using mathematical and physical laws—allow us to see through this extreme environment and observe the entire process with extremely high temporal and spatial resolution. The high temperature of the arc causes a surge in local pressure, while the surrounding area has lower pressure. This pressure difference (pressure gradient) acts like an invisible hand, driving the gas from the high-pressure area to the low-pressure area. This is precisely why the gas can flow at high speed during gas blowout arc extinguishing. Fluids have viscosity (like honey being more viscous than water), and gases are no exception. Viscous forces act like "friction" within the gas, attempting to bring adjacent gas layers with different velocities into a uniform state. At the edge of the arc, the high-speed flowing arc plasma, through viscous forces, drags along the surrounding relatively stationary cold gas, leading to energy exchange and heat dissipation. The Lorentz force represents the thrust exerted by a magnetic field on flowing plasma. The gas velocity, gas pressure, and gas density obtained through the Navier-Stokes equations are ultimately incorporated into the decision model to answer the most fundamental engineering question: can this gas formulation, combined with this contact structure and this tripping speed, reliably interrupt fault current?

[0003] However, during the current decay process, the zero point will shift and the current will change, which will cause the arc intensity to change as well. When extinguishing the arc, how to accurately predict the degree of impact of multiple operating conditions on the extinguishing of different arcs is a problem when making the decision on the interruption strategy. Summary of the Invention

[0004] The purpose of this invention is to provide a rapid simulation and switching strategy decision-making method for multi-condition electric arc magnetofluid in environmentally friendly gas medium ring main units, in order to solve the above-mentioned problems existing in the prior art.

[0005] This invention provides a rapid simulation and switching strategy decision-making method for multi-condition arc magnetohydrodynamic (AMD) circuit breakers in environmentally friendly gaseous media ring main units, including: Acquire arc magnetic simulation data; the arc magnetic simulation data includes current curves, gas types, and circuit breaker opening speeds; Based on the current curve, the change in current at future time points is predicted to obtain the predicted current value; Based on the predicted current value, gas type and opening speed, the gas velocity, gas pressure and gas density of the gas type at the corresponding future time point are calculated to perform rapid simulation of arc magnetofluid. Based on the current curve, gas type, opening speed, gas velocity, gas pressure, and gas density, it is determined whether the arc has been extinguished.

[0006] Optionally, the step of predicting the change in current at future time points based on the current curve to obtain the predicted current value includes: The current curve is plotted as a current curve image; By cutting two adjacent peaks in the current curve image, multiple cut current curve images are obtained; the cut current curve images represent the current change in one cycle. Based on the cutting current curve image, the distance between the zero point and the peak is determined, and the zero point time length is obtained; the zero point time length represents the length of time from the peak current time point to the current zero point. The peak current in the cut current curve image is taken as the current intensity value; Multiple cutting currents correspond to multiple zero-point time lengths, and multiple current intensity values ​​are obtained accordingly. Based on multiple cutting current curve images, zero-point time length, and current intensity values, a temporal convolutional network is used to detect the current change through the disconnector and obtain the predicted current value.

[0007] Optionally, the step of detecting the current change through the disconnector and obtaining the predicted current value based on multiple cutting current curve images, zero-point time length, and current intensity values ​​using a temporal convolutional network includes: The zero-point time length and current intensity value are input into the first neural network to extract features and obtain the first current feature vector; The image of the cutting current curve is input into a two-dimensional convolutional neural network to extract features and obtain a second current feature vector; The first current feature vector and the second current feature vector are input into the attention mechanism to distinguish the features of zero point and current intensity, and a fused current feature vector is obtained. Multiple fused current feature vectors are obtained from multiple cutting current curve images; Multiple fused current feature vectors are input into a temporal convolutional network according to time points from morning to night to determine the decay changes of the current and obtain the predicted current value.

[0008] Optionally, the step of calculating the gas velocity, gas pressure, and gas density of the gas type at future time points based on the predicted current value, gas type, and tripping speed, and performing rapid arc magnetohydrodynamic simulation, includes: Calculate the Lorentz force at future time points based on the predicted current value; Using the Navier-Stokes equations and the Lorentz force at future time points, the gas velocity, gas pressure, and gas density of the corresponding gas type at those future time points are calculated. The gas velocity represents the speed at which the environmentally friendly gas extinguishes the arc; the gas pressure represents the driving force for the environmentally friendly gas to extinguish the arc; and the gas density represents the intensity of the environmentally friendly gas's arc extinguishing. Rapid simulation of arc magnetohydrodynamics is performed based on gas velocity, gas pressure and density, predicted current value, gas type and opening speed.

[0009] Optionally, the step of determining whether to extinguish the arc based on the current curve, gas type, opening speed, gas velocity, gas pressure, and gas density includes: Acquire historical data; the historical data includes current curves, gas type, tripping speed, gas velocity, gas pressure, and gas density at historical time points; Based on the historical data, the degree of influence on the current is determined, and an associated adjustment vector is obtained; the associated adjustment vector includes six elements: sorted current curve, gas type, opening speed, gas velocity, gas pressure, and gas density; Obtain the recovery voltage; the recovery voltage represents the voltage at which reignition is possible; Based on the current curves, opening speed, gas velocity, gas pressure, gas density, and gas type at multiple time points, the breakdown voltage is obtained using Paschen's law formula. If the breakdown voltage is greater than the recovery voltage, the arc is considered extinguished. If the breakdown voltage is less than or equal to the recovery voltage, it is assumed that the arc has not been extinguished, and an associated adjustment vector is sent.

[0010] Optionally, the step of determining the degree of influence on the current based on the historical data and obtaining the associated adjustment vector includes: Train the association network based on historical data; Based on the historical data and labeled data, an adjustment ratio is obtained; the adjustment ratio represents the degree to which the historical data changes over time without external changes. By using the controlled variable method, the values ​​of historical data are adjusted according to the adjustment ratio to obtain multiple adjustment vectors; The multiple adjustment vectors are input into the correlation network to obtain multiple correlation degree values; the correlation degree values ​​represent the correlation between the current curve, gas type, opening speed, gas velocity, gas pressure, and gas density and arc extinguishing. Based on the correlation degree values ​​from largest to smallest, the current curves, gas types, opening speeds, gas speeds, gas pressures, and gas densities corresponding to the correlation degree values ​​are sorted to obtain the correlation adjustment vector.

[0011] Optionally, the training method for the association network includes: The reciprocal of the reignition time corresponding to the historical data is used as the discriminant value; the discriminant value is used as the annotation data. The current curve, gas type, tripping speed, gas velocity, gas pressure, and gas density from the historical data are input into the correlation network to determine the correlation relationship and obtain the predicted reignition state value. The loss is calculated using the predicted reignition state values ​​and labeled data, and then the association network is trained.

[0012] Optionally, obtaining the adjustment ratio based on the historical data and labeled data includes: The difference between the historical data of the second index and the historical data of the first index is obtained. The difference between the labeled data at the second time point and the labeled data at the first time point is obtained as the labeling difference. The adjustment ratio is obtained by dividing the data difference by the corresponding labeled difference and normalizing it.

[0013] Optionally, the historical data is data from multiple operating conditions; The multiple operating conditions include multiple current curves, multiple gas categories, multiple tripping speeds, and multiple gas speeds, gas pressures, and gas densities at multiple time points.

[0014] Optionally, the number of input neurons in the input layer of the temporal convolutional network is equal to the sum of the number of elements in multiple fused current feature vectors.

[0015] Compared with the prior art, the embodiments of the present invention achieve the following beneficial effects: This invention also provides a rapid simulation and switching strategy decision-making method for multi-condition arc magnetofluid in environmentally friendly gas medium ring main units.

[0016] In this embodiment, the changing state of the current curve is used to predict the current at future time points. Since current generates electromagnetic force, it can produce an electric arc. Based on the gas type, current curve, and tripping speed, the system also includes rapid magnetohydrodynamic simulation of the electric arc by calculating the gas velocity, gas pressure, and gas density for the corresponding gas type at future time points using the Navier-Stokes equations. Because gas pressure, gas velocity, and gas density are difficult to detect, and current affects these parameters, calculations are used to determine which data, with a small amount of data, can significantly impact the arc-extinguishing operation. Physical equipment in the ring main unit is then configured according to this arc-extinguishing data for disconnection. Furthermore, the use of environmentally friendly gas within the ring main unit for arc extinguishing achieves the technical effect of arc extinguishing with minimal effort and by predicting future states. Attached Figure Description

[0017] Figure 1This is a flowchart of a rapid simulation and switching strategy decision-making method for multi-condition electric arc magnetofluid in an environmentally friendly gas medium ring main unit provided by an embodiment of the present invention. Detailed Implementation

[0018] The present invention will now be described in detail with reference to the accompanying drawings. Example

[0019] like Figure 1 As shown in the figure, this invention provides a rapid simulation and switching strategy decision-making method for multi-condition arc magnetohydrodynamic (AMD) circuit breakers in environmentally friendly gaseous media ring main units. The method includes: Because gas pressure, gas velocity, and gas density are difficult to detect, and this invention predicts that the current at future points in time will affect these parameters, calculations are used to determine which data adjustments, even with small amounts, will significantly impact the arc-extinguishing operation. Physical equipment in the ring main unit is then configured according to this arc-extinguishing data for disconnection. Furthermore, since current decays over time, the current at future points in time is predicted, and elements that will enable arc extinguishing at those future points are identified, allowing for arc extinguishing even when the arc changes due to current variations.

[0020] S101: Obtain arc magnetic simulation data; the arc magnetic simulation data includes current curve, gas type and opening speed.

[0021] Wherein, the breaking current represents the current data passing through during breaking; the gas category represents the type of environmentally friendly gas used for arc extinguishing, such as C4F7N / CO2 in this embodiment; and the tripping speed represents the speed at which the disconnector breaks (the faster the better).

[0022] S102: Based on the current curve, predict the change in current at future time points to obtain the predicted current value.

[0023] The predicted current value refers to the current value at a future time point that was once at zero after the attenuation zero-point shift and the change in current intensity. It is also the current value that causes the electric arc.

[0024] S103: Based on the predicted current value, gas type, and opening speed, calculate the gas velocity, gas pressure, and gas density of the gas type at future time points to perform rapid simulation of arc magnetofluid.

[0025] Wherein, the gas velocity represents the speed at which the environmentally friendly gas extinguishes the arc; the gas pressure represents the power of the environmentally friendly gas to extinguish the arc; and the gas density represents the intensity of the environmentally friendly gas in extinguishing the arc.

[0026] S104: Based on the current curve, gas type, opening speed, gas velocity, gas pressure, and gas density, determine whether the arc has been extinguished.

[0027] Optionally, the step of predicting the change in current at future time points based on the current curve to obtain the predicted current value includes: The current curve is plotted as a current curve image.

[0028] By cutting two adjacent peaks in the current curve, multiple cut current curve images are obtained; the cut current curve images represent the current change in one cycle.

[0029] The cutting current curve image has a DC component, so the zero crossing point and current intensity change with current decay.

[0030] Based on the cutting current curve image, the distance between the zero point and the peak is determined, and the zero point time length is obtained; the zero point time length represents the length from the peak current time point to the current zero time point.

[0031] The peak current of the cut current curve image is taken as the current intensity value.

[0032] Based on multiple cutting current curve images, zero-point time length, and current intensity values, a temporal convolutional network is used to detect the current change through the disconnector and obtain the predicted current value.

[0033] Optionally, the step of detecting the current change through the disconnector and obtaining the predicted current value based on multiple cutting current curve images, zero-point time length, and current intensity values ​​using a temporal convolutional network includes: The zero-point time length and current intensity value are input into the first neural network to extract features and obtain the first current feature vector.

[0034] In this process, the number of output neurons in the first neural network is greater than the number of input neurons, with the aim of extracting hidden features of zero-point time length and current intensity value.

[0035] The first neural network is a deep neural network (DNN).

[0036] The image of the cut current curve is input into a convolutional neural network (CNN) to extract features and obtain a second current feature vector.

[0037] The first current feature vector and the second current feature vector are input into the attention mechanism to identify the features of zero points and current intensity, and a fused current feature vector is obtained.

[0038] The attention mechanism is used to detect the second current feature vector, which pays more attention to the curves related to the zero-point time length and current intensity value compared to the blank area.

[0039] Multiple fused current feature vectors are obtained from multiple cutting current curve images; Multiple fused current feature vectors are input into a temporal convolutional network (TCN) according to time points from morning to night to determine the decay changes of the current and obtain the predicted current value.

[0040] Optionally, the above-mentioned calculation of gas velocity, gas pressure, and gas density for the gas type at future time points based on predicted current value, gas type, and tripping speed, to perform rapid arc magnetohydrodynamic simulation, includes: Calculate the Lorentz force at future time points based on the predicted current value.

[0041] This is because electric current generates an electromagnetic field, which in turn generates a Lorentz force that causes the gas used for arc extinguishing to move. Therefore, the following method is adopted: Calculation, where It indicates the magnitude and direction of the current passing through a unit area. This represents the magnetic flux density vector. The magnetic flux density vector is determined by the electric current.

[0042] Using the Navier-Stokes equations and the Lorentz force at future time points, the gas velocity, gas pressure, and gas density of the corresponding gas type at those future time points are calculated. The gas velocity represents the speed at which the environmentally friendly gas extinguishes the arc; the gas pressure represents the driving force for the environmentally friendly gas to extinguish the arc; and the gas density represents the intensity of the environmentally friendly gas's arc extinguishing.

[0043] The NS equation is as follows: + = + + ; in, It represents the rate of change of gas momentum per unit volume over time. The momentum flux representing the macroscopic flow motion of the protective gas; It represents the direct force driving the gas flow; It represents the internal friction force of the fluid when protective gas flows, and is related to the gas velocity; For Lorentz force.

[0044] Using the above method, the gas velocity, gas pressure, and gas density of the gas type after decay at future time points are calculated and simulated. The influence of the structure generating the gas velocity, gas pressure, and gas density on the arc extinguishing of the protective gas is determined.

[0045] Rapid simulation of electric arc magnetohydrodynamics is performed based on gas velocity, gas pressure, and gas density.

[0046] The above method can be used to simulate the motion state of the protective gas because gas velocity, gas pressure, and gas density are motion parameters of the protective gas.

[0047] Optionally, the step of determining whether to extinguish the arc based on the current curve, gas type, opening speed, gas velocity, gas pressure, and gas density includes: A correlation network is trained based on gas velocity, gas pressure, and gas density at historical time points; the correlation degree value represents the degree of correlation between gas velocity, gas pressure, and gas density and arc extinguishing, respectively.

[0048] Based on the historical data, the degree of influence of each on the current is determined, and an associated adjustment vector is obtained. The associated adjustment vector includes six elements: sorted current curve, gas type, opening speed, gas velocity, gas pressure, and gas density.

[0049] The current curves, gas types, tripping speeds, gas speeds, gas pressures, and gas densities at the historical time points are data from multiple time points and under multiple operating conditions.

[0050] Based on the current curve, gas type, opening speed, gas velocity, gas pressure, and gas density, determine whether the arc has been extinguished.

[0051] This embodiment uses the method described in this example.

[0052] Obtain the recovery voltage; the recovery voltage represents the voltage at which reignition can occur.

[0053] Based on the current curves at multiple time points, the opening speed, gas velocity, gas pressure, gas density, and gas type, the breakdown voltage is obtained using Paschen's law formula.

[0054] If the breakdown voltage is greater than the recovery voltage, the arc is considered extinguished.

[0055] If the breakdown voltage is less than or equal to the recovery voltage, it is assumed that the arc has not been extinguished, and an associated adjustment vector is sent.

[0056] Optionally, the step of determining the degree of influence on the current based on the historical data and obtaining the associated adjustment vector includes: Based on historical data, a correlation network is trained; the correlation degree value represents the correlation between the current curve, gas type, opening speed, gas velocity, gas pressure, and gas density and arc extinguishing. Based on the historical data, an adjustment ratio is obtained; the adjustment ratio represents the degree to which the historical data changes over time without external alteration. Using the above method, because the degree and unit of change of each historical data are different, the same change in value will achieve different effects, so the adjustment ratio is calculated and adjusted accordingly.

[0057] By using the controlled variable method, the values ​​of historical data are adjusted according to the adjustment ratio to obtain multiple adjustment vectors.

[0058] The multiple adjustment vectors are input into the correlation network to obtain multiple correlation degree values; Based on the correlation degree values ​​from largest to smallest, the current curves, gas types, opening speeds, gas speeds, gas pressures, and gas densities corresponding to the correlation degree values ​​are sorted to obtain the correlation adjustment vector.

[0059] The above method uses the different effects produced by different historical data to determine the correlation.

[0060] Optionally, the training method for the association network includes: The reciprocal of the reignition time corresponding to the historical data is used as the discriminant value; the discriminant value is used as the annotation data.

[0061] If reignition is not performed, the discrimination value is set to 0.

[0062] The reason for using the above method to determine it as the reciprocal of the reignition time is that the longer the reignition time, the smaller the discriminant value.

[0063] The current curve, gas type, tripping speed, gas velocity, gas pressure, and gas density from the historical data are input into the correlation network to determine the correlation relationship and obtain the predicted reignition state value.

[0064] The loss is calculated using the predicted reignition state values ​​and labeled data, and then the association network is trained.

[0065] In this embodiment, cross-entropy is used to calculate the loss.

[0066] Optionally, obtaining the adjustment ratio based on the historical data includes: The difference between the historical data of the second index and the historical data of the first index is obtained.

[0067] Wherein, the first subscript and the second subscript are different subscripts in the historical data.

[0068] The difference between the labeled data at the second time point and the labeled data at the first time point is obtained as the labeling difference. The adjustment ratio is obtained by dividing the data difference by the corresponding labeled difference and normalizing it.

[0069] Using the method described above, the adjustment ratio is used to detect the change in each value in the historical data and the change in the corresponding result in the labeled data. The adjustment ratio is obtained by normalizing the average of multiple differences and divisions of historical data.

[0070] Optionally, the historical data is data from multiple operating conditions; The multiple operating conditions include multiple current curves, multiple gas categories, multiple tripping speeds, and multiple gas speeds, gas pressures, and gas densities at multiple time points.

[0071] Among them, multiple current curves represent currents ranging from small currents (e.g., 200A) to large currents (e.g., rated short-circuit breaking current), multiple gas categories represent multiple gas formulations (e.g., the mixing ratio of C4F7N / CO2), and multiple tripping speeds represent multiple adjusted tripping speeds.

[0072] Among them, the current curve during detection is a Class 1 current curve, and multiple current curves are used in training the association network; the gas category during detection is equivalent to a Class 1 gas category, and multiple gas categories are used in training the association network.

[0073] Optionally, the number of input neurons in the input layer of the temporal convolutional network is equal to the sum of the number of elements in multiple fused current feature vectors.

[0074] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0075] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0076] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the apparatus according to embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

Claims

1. A rapid simulation and switching strategy decision-making method for multi-condition electric arc magnetohydrodynamic (EMD) circuit breakers in environmentally friendly gaseous media ring main units, characterized in that, include: Acquire arc magnetic simulation data; the arc magnetic simulation data includes current curves, gas types, and circuit breaker opening speeds; Based on the current curve, the change in current at future time points is predicted to obtain the predicted current value; Based on the predicted current value, gas type and opening speed, the gas velocity, gas pressure and gas density of the gas type at the corresponding future time point are calculated to perform rapid simulation of arc magnetofluid. Based on the current curve, gas type, opening speed, gas velocity, gas pressure, and gas density, it is determined whether the arc has been extinguished.

2. The rapid simulation and switching strategy decision-making method for multi-condition electric arc magnetofluid in environmentally friendly gas medium ring main units according to claim 1, characterized in that, The step of predicting the change in current at future time points based on the current curve to obtain the predicted current value includes: The current curve is plotted as a current curve image; By cutting two adjacent peaks in the current curve image, multiple cut current curve images are obtained; the cut current curve images represent the current change in one cycle. Based on the cutting current curve image, the distance between the zero point and the peak is determined, and the zero point time length is obtained; the zero point time length represents the length of time from the peak current time point to the current zero point. The peak current in the cut current curve image is taken as the current intensity value; Multiple cutting currents correspond to multiple zero-point time lengths, and multiple current intensity values ​​are obtained accordingly. Based on multiple cutting current curve images, zero-point time length, and current intensity values, a temporal convolutional network is used to detect the current change through the disconnector and obtain the predicted current value.

3. The rapid simulation and switching strategy decision-making method for multi-condition electric arc magnetofluid in environmentally friendly gas medium ring main units according to claim 2, characterized in that, The method involves detecting current changes through the disconnector using a temporal convolutional network based on multiple cutting current curve images, zero-point time length, and current intensity values ​​to obtain a predicted current value. This includes: The zero-point time length and current intensity value are input into the first neural network to extract features and obtain the first current feature vector; The image of the cutting current curve is input into a two-dimensional convolutional neural network to extract features and obtain a second current feature vector; The first current feature vector and the second current feature vector are input into the attention mechanism to distinguish the features of zero point and current intensity, and a fused current feature vector is obtained. Multiple fused current feature vectors are obtained from multiple cutting current curve images; Multiple fused current feature vectors are input into a temporal convolutional network according to time points from morning to night to determine the decay changes of the current and obtain the predicted current value.

4. The rapid simulation and switching strategy decision-making method for multi-condition electric arc magnetofluid in environmentally friendly gas medium ring main units according to claim 1, characterized in that, The method involves calculating the gas velocity, gas pressure, and gas density of the gas type at future time points based on predicted current values, gas type, and tripping speed, and performing rapid simulation of arc magnetohydrodynamics, including: Calculate the Lorentz force at future time points based on the predicted current value; Using the Navier-Stokes equations and the Lorentz force at future time points, the gas velocity, gas pressure, and gas density of the corresponding gas type at those future time points are calculated. The gas velocity represents the speed at which the environmentally friendly gas extinguishes the arc; the gas pressure represents the driving force for the environmentally friendly gas to extinguish the arc; and the gas density represents the intensity of the environmentally friendly gas's arc extinguishing. Rapid simulation of arc magnetohydrodynamics is performed based on gas velocity, gas pressure and density, predicted current value, gas type and opening speed.

5. The rapid simulation and switching strategy decision-making method for multi-condition arc magnetofluid in environmentally friendly gas medium ring main units according to claim 1, characterized in that, The determination of whether to extinguish the arc based on the current curve, gas type, opening speed, gas velocity, gas pressure, and gas density includes: Acquire historical data; the historical data includes current curves, gas type, tripping speed, gas velocity, gas pressure, and gas density at historical time points; Based on the historical data, the degree of influence on the current is determined, and an associated adjustment vector is obtained; the associated adjustment vector includes six elements: sorted current curve, gas type, opening speed, gas velocity, gas pressure, and gas density; Obtain the recovery voltage; the recovery voltage represents the voltage at which reignition is possible; Based on the current curves, opening speed, gas velocity, gas pressure, gas density, and gas type at multiple time points, the breakdown voltage is obtained using Paschen's law formula. If the breakdown voltage is greater than the recovery voltage, the arc is considered extinguished. If the breakdown voltage is less than or equal to the recovery voltage, it is assumed that the arc has not been extinguished, and an associated adjustment vector is sent.

6. The rapid simulation and switching strategy decision-making method for multi-condition arc magnetofluid in environmentally friendly gas medium ring main units according to claim 5, characterized in that, The process of determining the degree of influence on the current based on the historical data and obtaining the associated adjustment vector includes: Train the association network based on historical data; Based on the historical data and labeled data, an adjustment ratio is obtained; the adjustment ratio represents the degree to which the historical data changes over time without external changes. By using the controlled variable method, the values ​​of historical data are adjusted according to the adjustment ratio to obtain multiple adjustment vectors; The multiple adjustment vectors are input into the correlation network to obtain multiple correlation degree values; the correlation degree values ​​represent the correlation between the current curve, gas type, opening speed, gas velocity, gas pressure, and gas density and arc extinguishing. Based on the correlation degree values ​​from largest to smallest, the current curves, gas types, opening speeds, gas speeds, gas pressures, and gas densities corresponding to the correlation degree values ​​are sorted to obtain the correlation adjustment vector.

7. The rapid simulation and switching strategy decision-making method for multi-condition arc magnetofluid in environmentally friendly gas medium ring main units according to claim 6, characterized in that, The training method for the aforementioned network includes: The reciprocal of the reignition time corresponding to the historical data is used as the discriminant value; the discriminant value is used as the annotation data. The current curve, gas type, tripping speed, gas velocity, gas pressure, and gas density from the historical data are input into the correlation network to determine the correlation relationship and obtain the predicted reignition state value. The loss is calculated using the predicted reignition state values ​​and labeled data, and then the association network is trained.

8. The rapid simulation and switching strategy decision-making method for multi-condition electric arc magnetofluid in environmentally friendly gas medium ring main units according to claim 7, characterized in that, The process of obtaining the adjustment ratio based on the historical data and labeled data includes: The difference between the historical data of the second index and the historical data of the first index is obtained. The difference between the labeled data at the second time point and the labeled data at the first time point is obtained as the labeling difference. The adjustment ratio is obtained by dividing the data difference by the corresponding labeled difference and normalizing it.

9. The rapid simulation and switching strategy decision-making method for multi-condition electric arc magnetofluid in environmentally friendly gas medium ring main units according to claim 5, characterized in that, The historical data is data from multiple operating conditions; The multiple operating conditions include multiple current curves, multiple gas categories, multiple tripping speeds, and multiple gas speeds, gas pressures, and gas densities at multiple time points.

10. The rapid simulation and switching strategy decision-making method for multi-condition electric arc magnetofluid in environmentally friendly gas medium ring main units according to claim 3, characterized in that, In the temporal convolutional network, the number of input neurons in the input layer is equal to the sum of the number of elements in multiple fused current feature vectors.

Images