Switch cabinet condensation margin evaluation method based on simulation analysis and order reduction reconstruction

Abstract

The application belongs to the technical field of simulation analysis, and particularly relates to a switch cabinet condensation margin evaluation method based on simulation analysis and order reduction reconstruction, steps of which comprise: establishing a simulation analysis model of the switch cabinet, adding a physical field and performing physical field coupling and mesh partitioning; solving the simulation analysis model to obtain the temperature and humidity distribution inside the switch cabinet; optimizing the simulation analysis model; replacing electromagnetic heat with loss data as a heat source input into a non-isothermal flow field; constructing a physical field snapshot library for order reduction model training; processing the physical field snapshot library based on a POD method to construct an order reduction model; training a Kriging model to reconstruct the temperature and humidity field of the switch cabinet; constructing a condensation margin evaluation cloud map; in the actual operation process of the switch cabinet, calculating an actual dynamic dew point temperature to generate an actual operation condensation margin cloud map, and completing online real-time evaluation. The application can realize efficient and accurate reconstruction of the temperature and humidity field inside the switch cabinet and efficient evaluation of the condensation margin of the switch cabinet.

Description

Technical Field

[0001] This invention belongs to the field of simulation analysis technology, specifically relating to a method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction. Background Technology

[0002] In modern power systems, high-voltage switchgear undertakes the core functions of circuit switching, control, and protection, and its operational stability directly affects power grid security. Accurately obtaining the temperature and humidity distribution within the switchgear is crucial for optimized equipment design, defect early warning, and operational status assessment. Currently, the industry widely employs multiphysics full-order simulation for analysis, which can realistically recreate the internal field distribution under electromagnetic, thermal, and humidity coupling effects, providing quantitative data for switchgear design and maintenance.

[0003] Taking the commonly used KYN28A series switchgear as an example, its internal components include busbars, contacts, transformers, insulators, and solid-sealed poles, which are compact and spatially complex. Simulation modeling requires the division of massive meshes. At the same time, the strong coupling and iteration of multiple physical fields such as current field, heat transfer, turbulent flow and moisture transport result in slow convergence, high computational resource consumption and long single simulation time, which cannot support rapid diagnosis, online inversion and real-time evaluation in engineering sites. There is an urgent need for an efficient order reduction calculation method to achieve rapid reconstruction of temperature and humidity fields. Summary of the Invention

[0004] In view of the shortcomings of the prior art, the purpose of this invention is to provide a method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction, which can achieve efficient and accurate reconstruction of the internal temperature and humidity field of switchgear and efficient evaluation of the condensation margin of switchgear based on finite boundary monitoring data.

[0005] To achieve the above objectives, this invention provides a method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction, comprising the following steps:

[0006] S1. For medium and high voltage switchgear, establish the geometric model of the switchgear and import it into the finite element simulation software to obtain the simulation analysis model, and set the material properties of each component.

[0007] S2. Add physical fields, set the computational domain and boundary conditions for each physical field, couple the physical fields, and mesh the simulation analysis model.

[0008] S3. Add transient analysis, numerically solve the simulation analysis model, post-process the solution results, and obtain the temperature and humidity distribution inside the switch cabinet.

[0009] S4. Through temperature rise test, measure the actual temperature and humidity of the switch cabinet during operation, compare it with the simulation solution results, and optimize the simulation analysis model;

[0010] S5. Calculate the electromagnetic loss distribution in the switch cabinet, use the calculated loss data to replace electromagnetic heat as the heat source input, and import it into the non-isothermal flow field.

[0011] S6. For different working conditions, complete the full-order simulation calculation for each working condition, extract the corresponding temperature and humidity field distribution data, and build a physical field snapshot library for training the reduced-order model.

[0012] S7. Based on the intrinsic orthogonal decomposition method, the physical field snapshot library is processed, the snapshot mean is calculated and centered, and then singular value decomposition is performed to extract modes and mode coefficients. Based on the preset energy threshold and the maximum number of modes, the high-dimensional physical field is projected to the low-dimensional mode coefficient space to construct a reduced-order model.

[0013] S8. Using operating condition characteristics as input and modal coefficients as output, train the Kriging model, reconstruct the temperature and humidity field of the switch cabinet, and use the leave-one-out method to evaluate the model accuracy.

[0014] S9. The dynamic dew point temperature of each spatial node is analyzed point by point using the Magnus equation, and the dynamic dew point temperature is coupled with the reconstructed temperature field to construct a cloud map for condensation margin assessment.

[0015] S10. During the actual operation of the switchgear, real-time temperature, humidity and operating parameters are collected. The Kriging model is corrected with the measured data. The global temperature and humidity field under the current operating conditions is reconstructed through the corrected model. The actual dynamic dew point temperature is calculated, and the actual condensation margin cloud map is generated to complete the online real-time evaluation.

[0016] As a preferred embodiment of the present invention, in S1, a geometric model of the original dimensions of the switchgear is constructed using SolidWorks 3D drawing software based on the actual structural parameters of the switchgear. The model includes a current-carrying busbar, a current transformer, a circuit breaker, a switchgear housing, an insulator, and a solid-sealed pole. The solid-sealed pole is obtained by encapsulating the circuit breaker vacuum interrupter and the moving / stationary contacts into one unit using an insulating housing as a carrier. The functional compartment at the bottom of the switchgear that houses the current transformer is a cable compartment.

[0017] Import the geometric model into the COMSOL finite element simulation software, specify the current-carrying busbar as copper material, specify the current transformer, insulator and solid-sealed pole as epoxy resin material, specify the circuit breaker moving and stationary contacts as copper-chromium alloy material, specify the circuit breaker vacuum interrupter bushing as aluminum oxide material, and specify the switch cabinet shell as stainless steel material.

[0018] In S2, the added physical fields include the current physical field, the solid and fluid heat transfer physical field, the turbulence physical field, and the air moisture transport physical field.

[0019] As a preferred embodiment of the present invention, in S5, the electromagnetic loss is calculated by adding contact impedance to the contact surface of the contact and the bolt fastening surface to simulate the local heating caused by contact resistance. The bolt fastening surface is the contact surface between two conductive parts clamped and fixed by bolts, and the contact resistance R is calculated by the following formula:

[0020] ;

[0021] In the formula, K is a coefficient related to the contact material; m is a coefficient related to the contact form of the two contact surfaces, which includes surface contact, line contact, and point contact; F is the contact pressure.

[0022] Concentrated heat loss caused by contact resistance directly follows Joule's law:

[0023] ;

[0024] In the formula, P is the concentrated heat loss power generated by the contact resistance; I is the working current flowing through the contact part.

[0025] The formula for calculating the volumetric heat loss of a conductor is:

[0026] ;

[0027] In the formula, The volumetric heat loss power of the conductor unit; J is the resistivity of the busbar; J is the current density. This refers to the unit volume in the simulation analysis model.

[0028] The total electromagnetic loss is P and sum.

[0029] As a preferred embodiment of the present invention, in S6, the operating parameters of each operating condition include load current, ambient temperature and relative humidity. By inputting different operating parameters, simulation calculations are performed, and snapshots of the temperature and humidity fields under each operating condition are extracted to construct a physical field snapshot library.

[0030] As a preferred embodiment of the present invention, in S7, based on a preset energy threshold and the maximum number of modes, truncation is performed, and only the top r dominant modes with a higher energy proportion are retained. The high-dimensional physical field is projected onto the low-dimensional modal coefficient space to construct a reduced-order model. The high-dimensional physical field is the three-dimensional spatial distribution data of the temperature and humidity field inside the switch cabinet obtained from full-order simulation.

[0031] As a preferred embodiment of the present invention, in S8, a Kriging model is constructed, with the input characteristics of the working condition as the independent variable and the modal coefficients of the intrinsic orthogonal decomposition method as the dependent variable. A Gaussian process regression model is trained, and the corresponding temperature and humidity field is obtained by predicting and reconstructing the modal coefficients of the new working condition. The Gaussian process regression model adopts the Matern kernel function and the number of optimization restarts is set to 15.

[0032] As a preferred embodiment of the present invention, in S8, the leave-one-out cross-validation method is used to evaluate the model accuracy: each time, a sample of a working condition is removed, and the Kriging model is retrained with the remaining samples to predict the temperature and humidity field distribution of the removed working condition and calculate the relative error of the whole field; after traversing all working conditions, the average value of the error is taken as the model generalization error; if the model generalization error is less than the set error threshold, the Kriging model is considered to meet the accuracy requirements.

[0033] As a preferred embodiment of the present invention, in S9, based on the high-dimensional grid data of the overall temperature and humidity inside the switchgear obtained from the reconstructed temperature and humidity field, the dynamic dew point temperature of each spatial node is calculated point by point using the Magnus equation; the dynamic dew point temperature is coupled with the reconstructed temperature field data of the corresponding spatial node to obtain the condensation margin value of each spatial node, which is the temperature difference between the wall temperature and the dew point temperature of the corresponding spatial node; a global condensation margin assessment cloud map is generated based on the condensation margin value of each spatial node, and the condensation risk area inside the switchgear is determined based on the condensation margin distribution results.

[0034] As a preferred embodiment of the present invention, in S10, the condensation risk area is determined based on the condensation critical node, and the confidence level of the condensation risk area is rated, specifically as follows:

[0035] Based on the intrinsic orthogonal decomposition method, the dominant modes of the temperature field and the dominant modes of the humidity field are extracted respectively, and the contribution factor of each mode to the condensation margin gradient is calculated. The contribution factor is defined as the rate of change of condensation margin caused by the change of mode coefficient.

[0036] Based on the condensation risk level of different functional compartments of the switchgear, the weighting coefficients of the temperature field mode and the humidity field mode are dynamically adjusted: for circuit breaker compartments and busbar compartments, the weight of the temperature field mode is increased and the weight of the humidity field mode is decreased; for cable compartments, the weight of the humidity field mode is increased and the weight of the temperature field mode is decreased.

[0037] The temperature and humidity field of the switch cabinet is reconstructed based on the weighted cross-physics field modes. The dynamic dew point temperature is calculated using the Magnus equation. When the difference between the wall temperature and the dew point temperature of a certain spatial node is less than 2℃, the spatial node is marked as the condensation critical node.

[0038] Spatial cluster analysis was performed on the critical nodes of condensation to identify the geometric center and influence range of the condensation risk area. The definition rule of the condensation risk area was set as follows: for each cluster of critical nodes of condensation obtained by spatial cluster analysis, if the number of nodes in the cluster is not less than 5, the cluster is defined as an independent condensation risk area; the geometric boundary of the condensation risk area is determined by the three-dimensional convex hull of all critical nodes of condensation in the cluster.

[0039] Using the prediction variance of the Kriging model, a confidence rating for condensation risk areas is output, and treatment recommendations are generated based on the confidence rating.

[0040] As a preferred embodiment of the present invention, the criteria for determining the credibility rating are as follows:

[0041] High reliability: Kriging model prediction standard deviation at the condensation critical node With condensation margin value The ratio satisfies If the number of critical condensation nodes in the condensation risk area accounts for more than 80% of the total number of nodes in the area, it is considered to be highly reliable, and it is recommended to start the heating and dehumidification device immediately.

[0042] Medium credibility: Satisfied If the number of critical condensation nodes in the condensation risk area is between 50% and 80% of the total number of nodes in the area, it is considered to be of medium confidence. It is recommended to start the heating and dehumidification device and conduct manual verification of the condensation risk area.

[0043] Low credibility: satisfies If the number of critical condensation nodes in the condensation risk area accounts for less than 50% of the total number of nodes in the area, it is considered to be of low confidence, and it is recommended not to start active dehumidification for the time being and to increase the monitoring frequency.

[0044] The beneficial effects of this invention are:

[0045] This invention constructs a high-precision temperature and humidity field snapshot library for switchgear through multi-physics field coupling simulation, covering different load currents, ambient temperatures, and relative humidity conditions. Compared with methods that rely solely on theoretical calculations or simplified empirical models, this invention can more realistically reflect the humidity distribution characteristics under complex environments within the switchgear.

[0046] This invention introduces intrinsic orthogonal decomposition (POD) to reduce the dimensionality of high-dimensional physical fields in the snapshot library, extract the main modes, and retain most of the physical information by setting a high energy threshold and a maximum number of modes. While ensuring accuracy, it significantly reduces the model dimensionality, laying the foundation for subsequent rapid prediction.

[0047] The fast computation model proposed in this invention, based on POD mode order reduction and Kriging proxy model, enables minute-level calculation of complex physical fields in switchgear, greatly improving computational efficiency and helping to cope with sudden environmental changes. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating the principle of this invention;

[0049] Figure 2 This is a schematic diagram of the geometric model of the KYN28A switchgear in Embodiment 1 of the present invention;

[0050] Figure 3 This is a schematic diagram of the simulated temperature field in Embodiment 1 of the present invention;

[0051] Figure 4 This is a schematic diagram of the reconstructed temperature field in Embodiment 1 of the present invention;

[0052] Figure 5 This is a schematic diagram of the simulated humidity field in Embodiment 1 of the present invention;

[0053] Figure 6 This is a schematic diagram of the reconstructed humidity field in Embodiment 1 of the present invention;

[0054] Figure 7 This is a cloud map for assessing condensation margin in Embodiment 1 of the present invention. Detailed Implementation

[0055] The embodiments of the present invention will be further described below with reference to the accompanying drawings:

[0056] Example 1: As Figure 1 As shown, the method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction includes the following steps:

[0057] S1. For medium and high voltage switchgear, establish the geometric model of the switchgear and import it into the finite element simulation software to obtain the simulation analysis model, and set the material properties of each component.

[0058] S2. Add physical fields, set the computational domain and boundary conditions for each physical field, couple the physical fields, and mesh the simulation analysis model.

[0059] S3. Add transient analysis, numerically solve the simulation analysis model, post-process the solution results, and obtain the temperature and humidity distribution inside the switch cabinet.

[0060] S4. Through temperature rise test, measure the actual temperature and humidity of the switch cabinet during operation, compare it with the simulation solution results, and optimize the simulation analysis model;

[0061] S5. Calculate the electromagnetic loss distribution in the switch cabinet, use the calculated loss data to replace electromagnetic heat as the heat source input, and import it into the non-isothermal flow field.

[0062] S6. For different working conditions, complete the full-order simulation calculation for each working condition, extract the corresponding temperature and humidity field distribution data, and build a physical field snapshot library for training the reduced-order model.

[0063] S7. Based on the intrinsic orthogonal decomposition method, the physical field snapshot library is processed, the snapshot mean is calculated and centered, and then singular value decomposition is performed to extract modes and mode coefficients. Based on the preset energy threshold and the maximum number of modes, the high-dimensional physical field is projected to the low-dimensional mode coefficient space to construct a reduced-order model.

[0064] S8. Using operating condition characteristics as input and modal coefficients as output, train the Kriging model, reconstruct the temperature and humidity field of the switch cabinet, and use the leave-one-out method to evaluate the model accuracy.

[0065] S9. The dynamic dew point temperature of each spatial node is analyzed point by point using the Magnus equation, and the dynamic dew point temperature is coupled with the reconstructed temperature field to construct a cloud map for condensation margin assessment.

[0066] S10. During the actual operation of the switchgear, real-time temperature, humidity and operating parameters are collected. The Kriging model is corrected using the measured data. The global temperature and humidity field under the current operating conditions is reconstructed through the corrected model (i.e. the corrected Kriging model). The actual dynamic dew point temperature is calculated, and the actual operating condensation margin cloud map is generated to complete the online real-time evaluation.

[0067] In this embodiment, the temperature and humidity field is an abbreviation for temperature field and humidity field. The simulation and temperature rise test process of steps S1 to S4 can be implemented based on existing technology.

[0068] In S1, based on the actual structural parameters of the switchgear, a geometric model of the original dimensions of the switchgear is constructed using SolidWorks 3D drawing software. This model includes the current-carrying busbar, current transformer, circuit breaker, switchgear enclosure, insulator, and solid-sealed pole. The solid-sealed pole is obtained by encapsulating the circuit breaker vacuum interrupter and moving / stationary contacts into one unit using the insulating enclosure as a carrier. The functional compartment at the bottom of the switchgear that houses the current transformer is the cable compartment.

[0069] Taking the KYN28A switchgear as an example, the geometric model is as follows: Figure 2 As shown.

[0070] Import the geometric model into the COMSOL finite element simulation software, specify the current-carrying busbar as copper material, specify the current transformer, insulator and solid-sealed pole as epoxy resin material, specify the circuit breaker moving and stationary contacts as copper-chromium alloy material, specify the circuit breaker vacuum interrupter bushing as aluminum oxide material, and specify the switch cabinet shell as stainless steel material.

[0071] In S2, within the COMSOL finite element simulation software, add a current physical field, set the current field calculation domain, and set the boundary conditions for the electric field analysis:

[0072] The current-carrying busbar and the moving and stationary contacts of the circuit breaker are defined as the current field calculation domain. The current input boundary is set by adding surface terminals, and the current outflow boundary is set by adding ground planes. The current magnitude is set to I0, which is the effective value of the current in the actual operation of the switchgear. By adding contact impedance to the contact surface and bolt fastening surface, the local heating caused by contact resistance is simulated.

[0073] Add solid and fluid heat transfer physics fields, set the temperature field computational domain, and set the boundary conditions for temperature analysis:

[0074] All regions except the vacuum medium inside the vacuum interrupter of the circuit breaker are set as the temperature field calculation domain. The solid domain includes the current-carrying busbar, the circuit breaker body, the current transformer and the switch cabinet shell, and the fluid domain is the air inside the switch cabinet. The heat exchange with the outside air is simulated by applying convective heat flux boundary conditions to the outer surface of the air chamber shell. The heat transfer coefficient is set according to the test conditions. The air chamber is the functional compartment filled with air inside the switch cabinet, and the air chamber shell is the metal wall panel of the functional compartment where each component is located.

[0075] Add based on The turbulent physical field of the model is defined, the computational domain of the flow field is set, and the boundary conditions for fluid flow analysis are set:

[0076] The air domain inside the chamber is set as the flow field computation domain. The fluid physics model is set as a compressible flow with Mach number Ma < 0.3, including gravity terms and using the reduced pressure form. The wall boundary adopts the no-slip condition, and pressure point constraints are set at positions far away from the main flow region to ensure the stability of the numerical solution.

[0077] Add a physical field for the transport of moisture in the air, set the calculation domain for the humidity field, and set the boundary conditions for the flow of humid air:

[0078] The air domain inside the chamber is set as the calculation domain for moisture transport in the air, i.e., the humidity field calculation domain, and the humid air diffusion coefficient is set to 2.6 × 10⁻⁶. -5 m 2 / s, describing the transport process of moist air through humidity boundary conditions matched to environmental conditions. This is based on the physical field of moisture transport in air simulated in COMSOL.

[0079] The physical field of electric current is coupled with the physical field of heat transfer in solids and fluids to form an electromagnetic thermal field; the physical field of heat transfer in solids and fluids is coupled with the physical field of turbulence to form a non-isothermal flow field; the physical field of turbulence is coupled with the physical field of moisture transport in air to form a moisture flow field; and the physical field of heat transfer in solids and fluids is coupled with the physical field of moisture transport in air to form a thermal-humid field.

[0080] During mesh generation, a fine mesh is used for the moving and stationary contact areas of the circuit breaker, while a conventional mesh is used for other current-carrying conductors, including the current-carrying busbars, the conductive parts of the current transformers, the conductive parts of the solid-sealed poles, and the conductive parts of the circuit breaker excluding the moving and stationary contacts. A relatively coarse mesh is used for the air domain inside the switchgear enclosure and air chamber to balance computational accuracy and solution efficiency.

[0081] In S3, a transient study was added to the COMSOL finite element simulation software, and the transient solver MUMPS was used to numerically solve the simulation analysis model. The solution results were post-processed to obtain the humidity field distribution inside the switchgear. Several humidity measurement points on the surface of the current transformer were selected, and the steady-state relative humidity of each humidity measurement point was extracted. At the same time, the inner wall temperature and relative humidity of the cable compartment were also extracted.

[0082] In S4, during the temperature rise test, temperature and humidity sensors are arranged according to the actual operating conditions of the switchgear, and the relative humidity is measured at the bottom of the current transformer; at the same time, the air hole at the bottom of the cable compartment is selected as the characteristic point for measuring the temperature and humidity of the inner wall, and the temperature and relative humidity data of the switchgear during operation are collected.

[0083] The temperature and humidity data obtained from the temperature rise test were compared and analyzed with the simulation results. When the error between the experimental data and the simulation results exceeded the error threshold of 5%, the simulation analysis model was optimized and adjusted by adjusting the material parameters, boundary conditions and mesh density.

[0084] Before obtaining a simulation snapshot, electromagnetic losses are calculated and input as a heat source into the simulation model to improve simulation efficiency. In S5, electromagnetic losses are calculated by adding contact impedance to the contact surface and bolt fastening surface to simulate localized heating caused by contact resistance. The bolt fastening surface is the contact surface between two conductive components clamped together by bolts. The Phillips head contact is the finest component inside the switch cabinet; excessive meshing resulted in an excessively large mesh size, so it was simplified, and a contact surface was set, directly adding the contact resistance to the contact surface. The contact resistance R is calculated using the following formula:

[0085] ;

[0086] In the formula, K is a coefficient related to the contact material; m is a coefficient related to the contact form of the two contact surfaces, including surface contact, line contact, and point contact. For surface contact, m is 1; for line contact, m is 1.5~2; and for point contact, m is 2~3; F is the contact pressure, in N.

[0087] Concentrated heat loss caused by contact resistance directly follows Joule's law:

[0088] ;

[0089] In the formula, P is the concentrated heat loss power generated by the contact resistance; I is the working current flowing through the contact part.

[0090] The formula for calculating the volumetric heat loss of a conductor is:

[0091] ;

[0092] In the formula, The volumetric heat loss power of the conductor unit; J is the resistivity of the busbar; J is the current density. This refers to the unit volume in the simulation analysis model.

[0093] The total electromagnetic loss is P and sum.

[0094] In S6, the operating parameters for each working condition include load current, ambient temperature, and relative humidity. Simulation calculations are performed by inputting different operating parameters, and snapshots of the temperature and humidity fields under each condition are extracted to construct a physical field snapshot library. Full-order simulation calculations construct a complete multi-physics strongly coupled model based on four fundamental physical fields within the switchgear, according to four sets of coupling relationships. Without any order reduction, all degrees of freedom of the complete three-dimensional model are directly solved, resulting in high-precision simulation calculations of the full field distribution within the switchgear, including temperature, humidity, flow field, and electromagnetic losses.

[0095] In S7, based on the preset energy threshold and the maximum number of modes, truncation is performed, retaining only the top r dominant modes with the highest energy proportion (the temperature field and humidity field are decomposed into intrinsic orthogonal modes respectively, and their respective dominant modes are extracted independently). The high-dimensional physical field is projected onto the low-dimensional modal coefficient space to construct a reduced-order model. The high-dimensional physical field is the three-dimensional spatial distribution data of the temperature and humidity field inside the switch cabinet obtained from the full-order simulation. r takes the value of an integer between 3 and 10, the energy threshold is set to 99.95%, and the maximum number of modes is set to 25.

[0096] In S8, a Kriging model is constructed, with the input features of the working condition as the independent variable and the modal coefficients of the intrinsic orthogonal decomposition method as the dependent variable. A Gaussian process regression model is trained (a Kriging sub-model can be constructed for each dominant modal coefficient, and each sub-model is trained and predicted independently). By predicting the modal coefficients of the new working condition and reconstructing them, the corresponding temperature and humidity field is obtained. The Gaussian process regression model uses the Matern kernel function and the number of optimization restarts is set to 15.

[0097] by Figure 2 Taking the KYN28A switchgear as an example, the simulated temperature field and the reconstructed temperature field are as follows: Figure 3 and Figure 4 As shown, the simulated humidity field and the reconstructed humidity field are as follows: Figure 5 and Figure 6 As shown (%RH is relative humidity). Among them, Figure 3 and Figure 5 The temperature and humidity fields are calculated using COMSOL multiphysics simulation software. Figure 4 and Figure 6 The cloud map output after reconstruction based on the method of this embodiment is consistent with the simulated temperature and humidity field format, so as to carry out quantitative comparison and error analysis.

[0098] Figure 4 and Figure 3 In comparison, the overall average relative error of the reconstructed temperature field was 1.77%, verifying the accuracy of the method in this embodiment. The maximum error occurred in the switchgear enclosure area near the busbar, which was attributed to the strong temperature gradient and boundary heat transfer coefficient variation in this region. Figure 6 and Figure 5 In comparison, the overall average relative error is 4.6%. The largest error is located in the air region between the cable compartment and the circuit breaker compartment, where the conductor temperature rises most drastically and the temperature changes greatly, resulting in significant fluctuations in relative humidity with temperature.

[0099] Figure 7 A cloud map for assessing condensation margin. From Figure 7 As can be seen, the condensation margin is low on the inner wall of the cable compartment, the surface of the current transformer, and the bottom area of ​​the solid-sealed pole. These locations have poor air circulation, low local temperatures, and are prone to moisture accumulation, making them high-risk areas for condensation. Therefore, temperature and humidity monitoring in these areas should be a key focus.

[0100] The model accuracy is evaluated using leave-one-out cross-validation: samples of one working condition are removed each time, and the Kriging model is retrained with the remaining samples to predict the temperature and humidity field distribution of the removed working condition and calculate the relative error of the whole field; after traversing all working conditions, the average error is taken as the model generalization error; if the model generalization error is less than the set error threshold, the Kriging model is considered to meet the accuracy requirements.

[0101] In S9, based on the high-dimensional grid data of the overall temperature and humidity inside the switchgear obtained from the reconstructed temperature and humidity field, the dynamic dew point temperature of each spatial node is calculated point by point using the Magnus equation; the dynamic dew point temperature is coupled with the reconstructed temperature field data of the corresponding spatial node to obtain the condensation margin value of each spatial node, which is the temperature difference between the temperature of the corresponding spatial node and the dew point temperature; a global condensation margin assessment cloud map is generated based on the condensation margin value of each spatial node, and the condensation risk area inside the switchgear is determined based on the condensation margin distribution results.

[0102] In S10, condensation risk areas are determined based on condensation critical nodes, and a credibility rating is performed on the condensation risk areas, specifically as follows:

[0103] Based on the intrinsic orthogonal decomposition method, the dominant modes of the temperature field and the dominant modes of the humidity field are extracted respectively, and the contribution factor of each mode to the condensation margin gradient is calculated. The contribution factor is defined as the rate of change of condensation margin caused by the change of mode coefficient.

[0104] The contribution factor calculated here is used to identify the mode most sensitive to changes in the condensation margin gradient, not limited to the first few energy-dominant modes, but covering all modes that may affect the local condensation risk.

[0105] Based on the condensation risk level of different functional compartments of the switchgear, the weighting coefficients of the temperature field mode and the humidity field mode are dynamically adjusted: for circuit breaker compartments and busbar compartments, the weight of the temperature field mode is increased and the weight of the humidity field mode is decreased; for cable compartments, the weight of the humidity field mode is increased and the weight of the temperature field mode is decreased.

[0106] The temperature and humidity field of the switch cabinet is reconstructed based on the weighted cross-physics field modes. The dynamic dew point temperature is calculated using the Magnus equation. When the difference between the wall temperature and the dew point temperature of a certain spatial node is less than 2℃ (i.e., the condensation margin is less than 2℃), the spatial node is marked as the critical condensation node.

[0107] Spatial cluster analysis was performed on the critical nodes of condensation to identify the geometric center and influence range of the condensation risk area. The definition rule of the condensation risk area was set as follows: for each cluster of critical nodes of condensation obtained by spatial cluster analysis, if the number of nodes in the cluster is not less than 5, the cluster is defined as an independent condensation risk area; the geometric boundary of the condensation risk area is determined by the three-dimensional convex hull of all critical nodes of condensation in the cluster, and all finite element mesh nodes covered by the interior and boundary of the three-dimensional convex hull are the total number of nodes contained in the area.

[0108] Using the prediction variance of the Kriging model, a confidence rating for condensation risk areas is output, and treatment recommendations are generated based on the confidence rating.

[0109] The criteria for credibility rating are as follows:

[0110] High reliability: Kriging model prediction standard deviation at the condensation critical node With condensation margin value The ratio satisfies If the number of critical condensation nodes in the condensation risk area accounts for more than 80% of the total number of nodes in the area, it is considered to be highly reliable, and it is recommended to start the heating and dehumidification device immediately.

[0111] Further suggestions can be provided, such as: immediately start the cable compartment heater and cabinet dehumidifier; focus on checking whether condensation has formed on the surface of the current transformer, cable connections, and insulators.

[0112] Medium credibility: Satisfied If the number of critical condensation nodes in the condensation risk area is between 50% and 80% of the total number of nodes in the area, it is considered to be of medium confidence. It is recommended to start the heating and dehumidification device and conduct manual verification of the condensation risk area.

[0113] Further suggestions for improvement can be provided, such as: start the cable compartment heater and maintain low-power continuous heating; open the cover and inspect the bottom of the cable compartment, contact box, and insulators within 1 hour; check whether the bottom seal of the cabinet and the cable hole sealing are intact.

[0114] Low credibility: satisfies If the number of critical condensation nodes in the condensation risk area accounts for less than 50% of the total number of nodes in the area, it is considered to be of low confidence, and it is recommended not to start active dehumidification for the time being and to increase the monitoring frequency.

[0115] Further suggestions for improvement can be provided, such as: temporarily not starting active dehumidification, updating the condensation margin assessment cloud map every 30 minutes; increasing the monitoring frequency, focusing on monitoring changes in risk caused by sudden changes in ambient humidity and increases in load current; and checking whether the ventilation inside the cabinet is smooth.

[0116] Example 2: Based on Example 1, when correcting the Kriging model with measured data, the reduced-order modal coefficients of POD are used as state variables to construct the reduced-order state-space equation; the temperature and humidity data of a finite number of measured points are used as observation vectors to establish the observation equation; using the ensemble Kalman filter algorithm, a prediction-update step is performed at each sampling time: the prediction step derives the prior distribution of the modal coefficients based on the Kriging model, and the update step fuses the measured data to correct the posterior distribution of the modal coefficients; the updated modal coefficients are used to reconstruct the global temperature and humidity field at the current time, realizing the dynamic assimilation of the reduced-order model and the measured data.

[0117] The reduced-order state-space equation is as follows:

[0118] ;

[0119] In the formula, , These are the modal coefficient vectors at times k and k-1, respectively; This is a nonlinear state transition function based on the Kriging model, which characterizes the evolution of modal coefficients with varying operating conditions. This is the process noise vector, representing the model uncertainty;

[0120] in, The input is Mapping operating condition changes using the Kriging model (Load current, changes in ambient temperature and humidity, etc.), output , where u is the operating condition parameter at time k-1.

[0121] The observation equation is:

[0122] ;

[0123] In the formula, Let H be the observation vector at time k, whose elements are the measured values ​​of temperature and humidity at each measurement point; H is the observation matrix, whose row vectors are composed of the POD modal basis function values ​​at the measurement point locations, realizing the mapping from the modal coefficient space to the physical measurement point space; The observation noise vector characterizes the sensor measurement error.

[0124] Elements in H It is composed of the values ​​of the POD modal basis functions at the measurement points, i.e. , For the j-th mode, Let i be the position of the i-th sensor.

[0125] By introducing ensemble Kalman filtering into the fusion of the POD reduced-order model and sparse measured data, and using modal coefficients as state variables to construct the reduced-order state space, the physical mapping from the measurement point location to the modal space is realized through the observation matrix. While maintaining physical consistency, the online quantification of prediction uncertainty is achieved, enabling the reduced-order model to have dynamic assimilation capability and real-time performance.

[0126] Example 3: A switchgear condensation margin assessment device based on simulation analysis and order reduction reconstruction, comprising:

[0127] One or more processors;

[0128] Memory, used to store one or more computer programs;

[0129] When one or more programs are executed by one or more processors, the one or more processors perform the method in Embodiment 1 or Embodiment 2.

[0130] Example 4: A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the method in Example 1 or Example 2.

Claims

1. A method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction, characterized in that... Includes the following steps: S1. For medium and high voltage switchgear, establish the geometric model of the switchgear and import it into the finite element simulation software to obtain the simulation analysis model, and set the material properties of each component. S2. Add physical fields, set the computational domain and boundary conditions for each physical field, couple the physical fields, and mesh the simulation analysis model. Among them, the solid and fluid heat transfer physical fields are coupled with the turbulence physical field to form a non-isothermal flow field. S3. Add transient analysis, numerically solve the simulation analysis model, post-process the solution results, and obtain the temperature and humidity distribution inside the switch cabinet. S4. Through temperature rise test, measure the actual temperature and humidity of the switch cabinet during operation, compare it with the simulation solution results, and optimize the simulation analysis model; S5. Calculate the electromagnetic loss distribution in the switch cabinet, use the calculated loss data to replace electromagnetic heat as the heat source input, and import it into the non-isothermal flow field. S6. For different working conditions, complete the full-order simulation calculation for each working condition, extract the corresponding temperature and humidity field distribution data, and build a physical field snapshot library for training the reduced-order model. The working parameters for each working condition include load current, ambient temperature and relative humidity. S7. Based on the intrinsic orthogonal decomposition method, the physical field snapshot library is processed, the snapshot mean is calculated and centered, and then singular value decomposition is performed to extract modes and mode coefficients. Based on the preset energy threshold and the maximum number of modes, the high-dimensional physical field is projected to the low-dimensional mode coefficient space to construct a reduced-order model. S8. Using operating parameters as input and modal coefficients as output, train the Kriging model, reconstruct the temperature and humidity field of the switchgear, and use the leave-one-out method to evaluate the model accuracy. S9. The dynamic dew point temperature of each spatial node is analyzed point by point using the Magnus equation, and the dynamic dew point temperature is coupled with the reconstructed temperature field to construct a cloud map for condensation margin assessment. S10. During the actual operation of the switchgear, real-time temperature, humidity and operating parameters are collected. The Kriging model is corrected with the measured data. The global temperature and humidity field under the current operating conditions is reconstructed through the corrected model. The actual dynamic dew point temperature is calculated, and the actual condensation margin cloud map is generated to complete the online real-time evaluation.

2. The method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction according to claim 1, characterized in that, In S1, based on the actual structural parameters of the switchgear, a geometric model of the original dimensions of the switchgear is constructed using SolidWorks 3D drawing software. This model includes the current-carrying busbar, current transformer, circuit breaker, switchgear housing, insulator, and solid-sealed pole. The solid-sealed pole is obtained by encapsulating the circuit breaker vacuum interrupter and moving / stationary contacts into one unit using the insulating housing as a carrier. The functional compartment at the bottom of the switchgear that houses the current transformer is the cable compartment. Import the geometric model into the COMSOL finite element simulation software, specify the current-carrying busbar as copper material, specify the current transformer, insulator and solid-sealed pole as epoxy resin material, specify the circuit breaker moving and stationary contacts as copper-chromium alloy material, specify the circuit breaker vacuum interrupter bushing as aluminum oxide material, and specify the switch cabinet shell as stainless steel material. In S2, the added physical fields include the current physical field, the solid and fluid heat transfer physical field, the turbulence physical field, and the air moisture transport physical field.

3. The method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction according to claim 2, characterized in that, In S5, the electromagnetic loss is calculated by adding contact impedance to the contact surface of the contact and the bolt fastening surface to simulate the localized heating caused by contact resistance. The bolt fastening surface is the contact surface between two conductive parts clamped together by bolts. The contact resistance R is calculated using the following formula: ； In the formula, K is a coefficient related to the contact material; m is a coefficient related to the contact form of the two contact surfaces, which includes surface contact, line contact, and point contact; F is the contact pressure. Concentrated heat loss caused by contact resistance directly follows Joule's law: ； In the formula, P is the concentrated heat loss power generated by the contact resistance; I is the working current flowing through the contact part. The formula for calculating the volumetric heat loss of a conductor is: ； In the formula, The volumetric heat loss power of the conductor unit; J is the resistivity of the busbar; J is the current density. This refers to the unit volume in the simulation analysis model. The total electromagnetic loss is P and sum.

4. The method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction according to claim 1, characterized in that, In S6, the operating parameters for each operating condition include load current, ambient temperature, and relative humidity. By inputting different operating parameters, simulation calculations are performed, and snapshots of the temperature and humidity fields under each operating condition are extracted to construct a physical field snapshot library.

5. The method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction according to claim 1, characterized in that, In S7, based on a preset energy threshold and the maximum number of modes, truncation is performed, retaining only the top r dominant modes with a higher energy proportion. The high-dimensional physical field is projected onto the low-dimensional modal coefficient space to construct a reduced-order model. The high-dimensional physical field is the three-dimensional spatial distribution data of the temperature and humidity field inside the switch cabinet obtained from full-order simulation.

6. The method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction according to claim 1, characterized in that, In S8, a Kriging model is constructed, with the input features of the working condition as the independent variable and the modal coefficients of the intrinsic orthogonal decomposition method as the dependent variable. A Gaussian process regression model is trained, and the corresponding temperature and humidity field is obtained by predicting and reconstructing the modal coefficients of the new working condition. The Gaussian process regression model uses the Matern kernel function and the number of optimization restarts is set to 15.

7. The method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction according to claim 1, characterized in that, In S8, the leave-one-out cross-validation method is used to evaluate the model accuracy: each time, a sample of a working condition is removed, and the Kriging model is retrained with the remaining samples to predict the temperature and humidity field distribution of the removed working condition and calculate the relative error of the whole field; after traversing all working conditions, the average value of the error is taken as the model generalization error; if the model generalization error is less than the set error threshold, the Kriging model is considered to meet the accuracy requirements.

8. The method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction according to claim 2, characterized in that, In S9, based on the high-dimensional grid data of the overall temperature and humidity inside the switchgear obtained from the reconstructed temperature and humidity field, the dynamic dew point temperature of each spatial node is calculated point by point using the Magnus equation; the dynamic dew point temperature is coupled with the reconstructed temperature field data of the corresponding spatial node to obtain the condensation margin value of each spatial node, which is the temperature difference between the wall temperature and the dew point temperature of the corresponding spatial node; a global condensation margin assessment cloud map is generated based on the condensation margin value of each spatial node, and the condensation risk area inside the switchgear is determined based on the condensation margin distribution results.

9. The method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction according to claim 8, characterized in that, In S10, the condensation risk area is determined based on the condensation critical node, and the credibility rating of the condensation risk area is performed, specifically as follows: Based on the intrinsic orthogonal decomposition method, the dominant modes of the temperature field and the dominant modes of the humidity field are extracted respectively, and the contribution factor of each mode to the condensation margin gradient is calculated. The contribution factor is defined as the rate of change of condensation margin caused by the change of mode coefficient. Based on the condensation risk level of different functional compartments of the switchgear, the weighting coefficients of the temperature field mode and the humidity field mode are dynamically adjusted: for circuit breaker compartments and busbar compartments, the weight of the temperature field mode is increased and the weight of the humidity field mode is decreased; for cable compartments, the weight of the humidity field mode is increased and the weight of the temperature field mode is decreased. The temperature and humidity field of the switch cabinet is reconstructed based on the weighted cross-physics field modes. The dynamic dew point temperature is calculated using the Magnus equation. When the difference between the wall temperature and the dew point temperature of a certain spatial node is less than 2℃, the spatial node is marked as the condensation critical node. Spatial cluster analysis was performed on the critical nodes of condensation to identify the geometric center and influence range of the condensation risk area. The definition rule of the condensation risk area was set as follows: for each cluster of critical nodes of condensation obtained by spatial cluster analysis, if the number of nodes in the cluster is not less than 5, the cluster is defined as an independent condensation risk area; the geometric boundary of the condensation risk area is determined by the three-dimensional convex hull of all critical nodes of condensation in the cluster. Using the prediction variance of the Kriging model, a confidence rating for condensation risk areas is output, and treatment recommendations are generated based on the confidence rating.

10. The method for evaluating the condensation margin of switchgear based on simulation analysis and order reduction reconstruction according to claim 9, characterized in that, The criteria for credibility rating are as follows: High reliability: Kriging model prediction standard deviation at the condensation critical node With condensation margin value The ratio satisfies If the number of critical condensation nodes in the condensation risk area accounts for more than 80% of the total number of nodes in the area, it is considered to be highly reliable, and it is recommended to start the heating and dehumidification device immediately. Medium credibility: Satisfied If the number of critical condensation nodes in the condensation risk area is between 50% and 80% of the total number of nodes in the area, it is considered to be of medium confidence. It is recommended to start the heating and dehumidification device and conduct manual verification of the condensation risk area. Low credibility: satisfies If the number of critical condensation nodes in the condensation risk area accounts for less than 50% of the total number of nodes in the area, it is considered to be of low confidence, and it is recommended not to start active dehumidification for the time being and to increase the monitoring frequency.

Images