Middle-large size induction motor multi-physical field dynamic coupling calculation method based on multiple time scales, electronic equipment and storage medium

Abstract

The invention provides a multi-time-scale-based multi-physics-field dynamic coupling calculation method for a medium-and-large-sized induction motor, electronic equipment and a storage medium, relates to multi-physics-field calculation coupling, and is suitable for solving simulation analysis of the medium-and-large-sized induction motor in an electromagnetic, thermal and fluid multi-field coupling state. According to the method, electromagnetic and fluid transient characteristics and temperature field hot spot distribution in different operation states are mainly considered, a field-domain coupling calculation method based on multiple time scales is provided, different physical fields adopt differentiated step length calculation strategies according to dynamic characteristics of the physical fields, the calculation precision is effectively improved, and the calculation efficiency is improved. And the waste of computing resources caused by multi-field synchronous solution is avoided. According to the method, millisecond-level accurate analysis can be carried out on an electromagnetic transient stage, long-time scale simulation can be completed by adopting a self-adaptive step length strategy aiming at a steady-state working condition, and multi-physical field dynamic characteristics of the motor under a complex working condition are reflected more accurately, so that performance optimization and structural design of medium and large induction motors are better guided.

Description

A method for multi-physics dynamic coupling calculation of medium and large induction motors based on multiple time scales, electronic equipment and storage medium Technical Field

[0001] This invention belongs to the field of multi-physics field computational coupling for electric motors, and relates to a multi-physics field dynamic coupling computational method, electronic equipment, and storage medium for medium and large induction motors based on multiple time scales. Background Technology

[0002] Medium and large induction motors are widely used in industrial drives, power systems, and rail transportation, and their performance directly affects energy efficiency and system reliability. As motor capacity increases, the coupling effects of electromagnetic, thermal, and fluid multiphysics fields significantly intensify. Since electromagnetic, fluid, and temperature fields are often treated as independent subsystems, a robust dynamic coupling relationship cannot be established. This leads to losses generated by electromagnetic transient processes directly affecting the temperature field distribution, while temperature changes, in turn, influence electromagnetic field calculations through material parameters, thus affecting the reliability of electromagnetic temperature rise predictions. This bidirectional coupling mechanism is often simplified or ignored in traditional methods. Furthermore, traditional single-timescale calculation methods often struggle to balance computational accuracy and efficiency, especially under limited computational resources, failing to simultaneously meet the requirements for accurate simulation of transient processes and efficient analysis of long-term steady-state operation.

[0003] Currently, the main methods for multiphysics analysis of medium and large induction motors include the fully coupled finite element method, the sequential coupling calculation method, and the simplified model method. The fully coupled finite element method can accurately capture the interactions between the various physical fields, but it involves a huge computational load, and the timescale differences between different physical fields lead to low solution efficiency. The sequential coupling calculation method reduces computational complexity by solving each physical field step by step, but it ignores real-time feedback between multiple physical fields, resulting in significant errors, especially during transient processes. The simplified model method, while computationally fast, sacrifices model accuracy and cannot accurately reflect the motor's performance under complex operating conditions. Therefore, a multi-timescale dynamic coupling calculation method is needed that can both ensure computational accuracy and improve computational efficiency. Summary of the Invention

[0004] The purpose of this invention is to address the accuracy and efficiency issues in multi-physics coupling calculations for medium and large induction motors by proposing a dynamic coupling calculation method based on multiple time scales. This method sets differentiated calculation step-size strategies according to the different dynamic characteristics of electromagnetic, fluid, and temperature fields, and achieves efficient coupling of multiple physics fields through a real-time data feedback mechanism, thus ensuring both calculation accuracy and significantly improving computational efficiency.

[0005] To achieve the above objectives, the technical solution proposed in this invention is: a multi-physics dynamic coupling calculation method for medium and large induction motors based on multiple time scales, comprising the following steps:

[0006] Step 1: Determine the topology and research problem analysis of medium and large induction motors. The topology includes the stator and rotor cores, windings, air gap between the stator and rotor, and cooling system of medium and large induction motors. The research problem analysis includes electromagnetic performance and thermal performance.

[0007] Step 2: Determine the physical fields involved in the research problem of Step 1, including electromagnetic fields, temperature fields, and fluid fields;

[0008] Step 3: Establish a dynamic mathematical model of the induction motor and calculate its dynamic electromagnetic characteristics;

[0009] Step 4: Establish a fluid field calculation model based on multiple spatial scales, and perform multi-time scale calculations on the established fluid field model;

[0010] Step 5: Establish a temperature field gradient model and perform multi-timescale calculations on the established temperature field model;

[0011] Step 6: Establish a multi-field dynamic coupling calculation for medium and large induction motors based on multiple time scales.

[0012] Furthermore, in step 3, the specific method for establishing the dynamic mathematical model of the induction motor is as follows: affected by the rotor skin effect, the motor starting torque... The starting torque calculation formula, which is related to the rotor slot shape and takes into account the rotor skin effect, is as follows:

[0013]

[0014] in, The skin effect coefficient is given for the double-cage trough. Deep groove: Convex / Knife-shaped: Flat bottom groove: Closed groove: Pear-shaped groove: ; This represents the number of phases in the rotor winding. Phase voltage, For slippage, For synchronous angular velocity, To account for the rotor resistance due to the skin effect, , These are the converted values ​​of stator leakage reactance and rotor leakage reactance, respectively.

[0015] The time it takes for the starting current to drop from its peak value to the rated current, i.e., the electromagnetic transient duration, is calculated based on the moment of inertia. The formula is as follows:

[0016]

[0017] in, The total moment of inertia, The rated angular velocity, For starting torque and load torque;

[0018] Derivation of the mean current change during the startup process based on the duration of electromagnetic transients The calculation formula is as follows:

[0019]

[0020] in, To generate peak starting current, This is the rated current of the motor;

[0021] The motor starting process is an electromagnetic dynamic process. The various electromagnetic quantities of the motor continuously change with the rotational speed. Once the rated speed is reached, the electromagnetic quantities gradually stabilize. The solution time step of the dynamic mathematical model is set differently for different electromagnetic states. The current, torque, and speed are calculated, and the derivative of the stator current I with respect to time is solved. During the electromagnetic transient phase, when When, the electromagnetic measurement step size is set to 1ms; when At that time, the electromagnetic measurement step size was set to 0.5 ms;

[0022] During the electromagnetic steady-state phase, the time step can be extended for sensitive components, calculated using a 5ms step, while for non-sensitive components, the time step is calculated using a 50ms step.

[0023] The winding and core losses are updated on a time scale. Based on the current and magnetic flux density values ​​calculated within each time step, the stator winding loss, rotor winding loss, and core loss are calculated. The loss calculation formula is as follows:

[0024] ;

[0025] in, For the quality of the iron core, This is the hysteresis loss coefficient. This is the eddy current loss coefficient. Steinmetz index: , For operating frequency, For working magnetic flux density, For the rated frequency, Rated magnetic flux density; For copper loss; This is the effective value of the current; It is a room temperature resistor; It is the temperature coefficient of resistance; For temperature rise;

[0026] Further, in step 4, the method for establishing a fluid field calculation model based on multiple spatial scales includes the following steps: Modeling the fluid domain according to the flow characteristics within a medium-to-large induction motor; the boundary layer fluid motion state change gradient between the fluid and solid within the motor is large, so the model here is set to a small spatial scale, subdivided to 0.5 mm; the wide-path fluid domain is set to a large spatial scale, subdivided to 1 mm; for the air gap ventilation channel, the fluid velocity is jointly affected by the rotor rotation and the rotor auxiliary fan; the average fluid velocity in the air gap between the stator and rotor at steady state is estimated using the following formula:

[0027]

[0028] in, The fluid velocity generated by the rotor drag force, The fluid velocity is generated by the rotation of the rotor-attached fan, where , , respectively, are the slip coefficients of the rotor and the rotor auxiliary fan, where n and D are the rotational speed and diameter of the rotor auxiliary fan;

[0029] Derivation of average flow rate based on startup time The calculation formula is as follows:

[0030]

[0031] Differentiate the calculation step size for fluid field under different fluid states: Given the air gap fluid velocity, solve for the differential of the air gap fluid velocity with respect to time. During the startup phase, when When, the solution step size of the fluid field model is set to 2s; when When the rotor speed reaches a steady state, the solution step size of the fluid field model is set to 1 second; when the rotor speed reaches a steady state, the solution step size is set to 10 seconds.

[0032] Time coupling between dynamic mathematical model and fluid field model: Based on the solution step size of the fluid field, the dynamic mathematical model outputs rotor speed information to the fluid field model;

[0033] Further, in step 5, establishing the temperature field gradient model includes the following steps: dividing the region according to the heat generation of the medium-to-large induction motor: the core heat-generating region of the medium-to-large induction motor consists of the stator and rotor core, stator windings, rotor bars, and rotor end rings; among which, the stator windings are mainly characterized by copper losses, the rotor bars by copper losses and additional losses amplified by the skin effect, and the stator and rotor cores by iron losses; estimating the transient duration of the temperature in the core heat-generating region using the following formula:

[0034]

[0035] in, Thermal inertia amplification factor: The larger the heat capacity of the core component, the larger the value should be. Heat transfer hysteresis coefficient: ; The electromagnetic transient duration estimated in step S3;

[0036] The formula for estimating the average rate of temperature change based on the transient duration of temperature is as follows:

[0037]

[0038] in, The rated maximum allowable operating temperature of the core components. The ambient temperature;

[0039] Different step sizes are set for different components under different states: T is the core region temperature, and the derivative of the core region temperature with respect to time is solved. During the temperature transient phase, when At that time, the temperature measurement step size was set to 1 minute; when The temperature measurement step size is set to 30 seconds. During the steady-state temperature phase, the step size for the core heat-generating domain is set to 5 minutes, and the step size for the secondary heat-generating domain and the passively heated domain is set to 10 minutes. Heat source input: Real-time reception of stator winding loss, rotor winding loss, and core loss data output from the dynamic mathematical model. This data is synchronized with the current temperature field calculation step size and serves as the core heat source parameter for temperature field calculation. Boundary condition input: Real-time reception of fluid velocity boundary information output from the fluid field calculation model. This data is synchronized with the current temperature field calculation step size to ensure the accuracy of heat exchange calculation between the temperature field and the fluid field.

[0040] Temperature field feedback to electromagnetic field: When the temperature change of the stator winding and rotor bar is ≥5℃, the temperature field feeds back to the electromagnetic field once, and the resistance value is corrected according to the temperature coefficient of copper resistance; when there is no obvious temperature change, feedback is given once every 10 temperature time steps.

[0041] Furthermore, in step 6, the nodal equations for the multi-field dynamic coupling calculation are expressed as follows:

[0042]

[0043] in, , , These are the nodal potential matrices for the electromagnetic field, fluid field, and temperature field, respectively. , , These are the corresponding nodal admittance matrices; , , These are the source term matrices of the inflow nodes. The beneficial effects of this invention are:

[0044] This invention, based on the different dynamic characteristics of various physical fields in medium and large-sized induction motors, employs a multi-timescale strategy for coupled calculation. It uses millisecond-level small time steps for transient electromagnetic field processes and larger time steps for steady-state temperature fields, ensuring both computational accuracy and significantly improving computational efficiency. Through a dynamic feedback mechanism, real-time coupling of the electromagnetic field, fluid field, and temperature field is achieved, accurately capturing the multi-physics interactions during motor startup and operation. This method resolves the contradiction between accuracy and efficiency inherent in traditional single-timescale calculation methods, providing an efficient and accurate computational tool for the design optimization and performance prediction of medium and large-sized induction motors. Attached Figure Description

[0045] Figure 1 is a flowchart of the multi-physics coupling calculation of a large induction motor in an embodiment of the present invention;

[0046] Figure 2 is a topology model diagram of a large induction motor in an embodiment of the present invention; Detailed Implementation

[0047] To make the purpose, technical solution, and advantages of this invention patent clearer, the technical solutions in the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings.

[0048] This invention establishes a coupled computational model based on the multiphysics characteristics of medium and large-sized induction motors using a multi-timescale strategy. This embodiment uses an 11kV / 6500kW large induction motor as an example. This motor has a 4-pole structure, 180 stator slots, 210 rotor slots, an outer diameter of 2600mm, a core length of 1100mm, and employs a radial ventilation cooling system. The motor structure includes stator and rotor cores, stator windings, rotor cast aluminum bars and end rings, frame, end covers, and internal ventilation channels. The multiphysics coupling computation is divided into three main parts: electromagnetic field, temperature field, and fluid field. These parts are dynamically coupled through data exchange.

[0049] Multiphysics coupling calculations are implemented in the following steps:

[0050] Step 1: Determine the specific structure of the medium-to-large induction motor and analyze the research problems. The specific structure of the motor in this embodiment includes stator and rotor cores, three-phase windings, rotor bars and end rings, frame and end covers, internal ventilation channels, etc.; the research problems are the electromagnetic, thermal, and fluid coupling characteristics during the starting process, with a focus on starting torque, temperature rise distribution, and cooling effect.

[0051] Step 2: Determine the physical fields involved. This embodiment involves three physical fields: electromagnetic field, temperature field, and fluid field. The electromagnetic field and temperature field are strongly coupled, the fluid field and temperature field are moderately coupled, and the electromagnetic field and fluid field are weakly coupled.

[0052] Step 3: Establish the dynamic mathematical model of the induction motor. Due to the skin effect of the rotor, the motor in this embodiment adopts a deep-slot rotor structure. The formula for calculating the starting torque considering the skin effect is as follows:

[0053]

[0054] In this embodiment, the motor adopts a deep-slot rotor structure, and the skin effect correction coefficient is [not specified]. Take 3.5; The number of phases is set to 3; based on the motor's moment of inertia. The electromagnetic transient duration was calculated to be Seconds, and solve for the differential of the stator current I with respect to time, i.e. During the electromagnetic transient phase, when When, the electromagnetic measurement step size is set to 1ms; when At that time, the electromagnetic measurement step size was set to 0.5 ms;

[0055] Step 4: Establish the fluid field calculation model. In this embodiment, the internal fluid domain of the motor includes the stator ventilation channel, rotor ventilation channel, air gap ventilation channel, and end space. Based on the flow characteristics, the 2mm boundary layer is divided into a fine grid of 0.5mm, and the main flow region is divided into a coarse grid of 1mm. The flow velocity in the air gap ventilation channel is determined by the combined effect of rotor rotation and the rotor auxiliary fan, estimating the average air gap velocity at steady state. During the start-up phase, Given the air gap fluid velocity, solve for the differential of the air gap fluid velocity with respect to time. During the startup phase, when When, the solution step size of the fluid field model is set to 2s; when When the rotor speed reaches a steady state, the solution step size of the fluid field model is set to 1 second; when the rotor speed reaches a steady state, the solution step size is set to 10 seconds; every 10 fluid field steps, the flow velocity data is fed back to the temperature field, and every 1 fluid field step, the rotor speed information is received from the electromagnetic field.

[0056] Step 5: Establish a temperature field gradient model. In this embodiment, the core heat-generating region of the motor includes the stator windings, rotor bars, and stator and rotor cores. Based on the heat capacity and power loss, estimate the transient temperature duration and the average temperature ratio change. Let T be the core region temperature, and solve for the differential of the core region temperature with respect to time. During the temperature transient phase, when At that time, the temperature measurement step size was set to 1 minute; when At that time, the temperature measurement step size was set to 30s. During the steady-state temperature phase, the step size for the core heat-generating region was set to 5min, and the step size for the secondary heat-generating region and the passively heated region was set to 10min. Temperature field feedback to electromagnetic field: When the stator winding temperature changes by ≥5℃, feedback is sent to the electromagnetic field once to correct the winding resistance value.

[0057] Step 6: Establish multi-field dynamic coupling calculation. This embodiment employs a loosely coupled iterative method. Within each global time step, electromagnetic field calculation is performed first, followed by updating the loss distribution; then, fluid field calculation is performed to obtain the cooling medium velocity distribution; finally, temperature field calculation is performed to obtain the temperature distribution and feed it back to the electromagnetic field. The coupling calculation equation is expressed as:

[0058]

[0059] in, This indicates that the electromagnetic field source term is affected by the temperature field. This indicates that the fluid field source term is affected by the electromagnetic field. This indicates that the temperature field source term is affected by both the electromagnetic field and the fluid field.

[0060] In terms of electronic equipment, this patent uses a multi-core processor and a GPU heterogeneous chip to build a multi-field network; it uses a 128GB high-speed cache to reduce the latency of multi-field data exchange; it uses a 1TB non-volatile storage unit to achieve power-off data protection; an FPGA hardware acceleration module is used to accelerate matrix solving; an industrial-grade PCIe 4.0 communication interface provides a data transmission rate of 32GB / s; and a high-precision sensor is used for real-time data acquisition.

[0061] Storage medium implementation: A 2TB SSD is used to store the mapping relationship between multiphysics coupling calculation and electromagnetic field, temperature field and fluid field; the multi-timescale adaptive step size adjustment algorithm of this patent is built in; and the calculation results can be accessed in real time and compared with historical data for analysis.

[0062] Furthermore, this patent selects the Super-Relaxed Iterative Method (SOR) for solving the problem, which incorporates a relaxation factor. Furthermore, this patent employs the magnetic potential equation for solution, which significantly increases the iterative convergence speed while ensuring computational accuracy. The following equation represents the computational component form for solving the nodal magnetic potential equation:

[0063]

[0064] In the formula, k is the number of iterations. The hyperrelaxation factor (1 < <2).

[0065] When the iteration converges or reaches the upper limit of convergence, the iteration process terminates, and the solution is completed. That is, it conforms to the following formula:

[0066]

[0067] in, To ensure accuracy, the larger the error, the worse the accuracy; therefore, the error accuracy should not be too large.

[0068] The method of this invention is not only applicable to medium and large induction motors, but can also be extended to multiphysics field analysis of various motor types such as permanent magnet synchronous motors and DC motors, and has wide applicability.

[0069] The above describes preferred embodiments of the present invention, but the scope of application of the present invention is not limited to the above embodiments. All derivative technical solutions based on the concept of the present invention fall within the protection scope of the present invention. Any equivalent modifications and refinements made by other skilled in the art without departing from the principles of the present invention are within the protection scope of the present invention and are protected by the appended claims.

Claims

1. A multi-physics dynamic coupling calculation method, electronic device, and storage medium for medium-to-large induction motors based on multiple time scales includes the following steps: Step Step 1: Determine the topology and research problem analysis of medium and large-sized induction motors. The topology includes the stator and rotor cores, windings, air gap between the stator and rotor, and cooling system. The research problem analysis focuses on electromagnetic and thermal performance. Step 2: Determine the physical fields involved in the research problem of Step 1. These physical fields include electromagnetic fields, temperature fields, and fluid fields. Step 3: Establish a dynamic mathematical model of the induction motor and calculate its dynamic electromagnetic characteristics. Step 4: Establish a fluid field calculation model based on multiple spatial scales and perform multi-time scale calculations on the established fluid field model. Step 5: Establish a temperature field gradient model and perform multi-time scale calculations on the established temperature field model. Step 6: Establish a multi-field dynamic coupling calculation for medium and large-sized induction motors based on multiple time scales.

2. The method, electronic device, and storage medium for multi-physics dynamic coupling calculation of medium and large induction motors based on multiple time scales as described in claim 1, wherein step 3 is characterized in that, The specific method for establishing a dynamic mathematical model of an induction motor is as follows: The time it takes for the starting current to drop from its peak value to the rated current, i.e., the electromagnetic transient duration, is calculated based on the moment of inertia. Simultaneously, due to the rotor skin effect, the motor's starting torque is related to the rotor slot shape; the rotor skin effect is taken into account, and the starting torque is calculated accordingly. The average current change during the starting process is derived based on the electromagnetic transient duration. The motor starting process is an electromagnetic dynamic process, where various electromagnetic quantities of the motor continuously change with the rotational speed. Once the rated speed is reached, the various electromagnetic quantities of the motor gradually stabilize. The solution time step of the dynamic mathematical model is set differently for different electromagnetic states, and the current, torque, and speed are calculated. During the electromagnetic steady-state stage, a 5ms time step is used for sensitive components, and a 50ms time step is used for non-sensitive components. The winding and core losses are updated on a time scale. The stator winding loss, rotor winding loss and core loss are calculated based on the current and magnetic flux density values ​​obtained in each time step.

3. The multi-field dynamic coupling calculation method for medium and large induction motors based on multiple time scales according to claim 1, wherein step 4 is characterized in that, The specific method for establishing a fluid field calculation model based on multiple spatial scales is as follows: Fluid domain modeling is performed based on the flow characteristics within medium and large-sized motors. The boundary layer fluid motion state changes rapidly at the contact point between the fluid and solid within the motor, so a small spatial scale is set for this area, subdivided to 0.5 mm. A large spatial scale is set for the wide-path fluid domain, subdivided to 1 mm. For the air gap ventilation channel, the fluid velocity is influenced by both rotor rotation and the rotor's auxiliary fan. The average fluid velocity in the air gap between the stator and rotor is estimated at steady state. The average velocity is derived based on the startup duration. The fluid field calculation step size is set differently for different fluid states. Once the motor rotor speed reaches a steady state, the dynamic mathematical model and the fluid field model are coupled in time. Based on the fluid field solution step size, the dynamic mathematical model outputs rotor speed information to the fluid field model.

4. The method for multi-physics dynamic coupling calculation of medium and large induction motors based on multiple time scales according to claim 1, wherein step 5 is characterized in that, The specific method for establishing the temperature field gradient model is as follows: The model is divided into domains based on the heat generation of medium and large induction motors. The core heat-generating domains of medium and large induction motors include the stator and rotor cores, stator windings, rotor bars, and rotor end rings. The stator windings are primarily affected by copper losses, the rotor bars by copper losses and additional losses amplified by the skin effect, and the stator core by iron losses. The transient duration of the temperature in the core heat-generating region is estimated. The average temperature change is estimated based on the transient duration. The motor startup process is a transient temperature process; the motor temperature rises continuously with startup, and the core temperature gradually stabilizes after reaching the rated speed. Different step sizes are set for different components under different conditions. In the steady-state temperature phase, the step size for the core heat-generating domain is set to 5 minutes, and the step size for the secondary heat-generating domain and the passively heated domain is set to 10 minutes. Temperature field feedback to the electromagnetic field: When the temperature change of the stator windings and rotor bars is ≥5℃, the temperature field feeds back to the electromagnetic field once, correcting the motor resistance value according to the copper resistance temperature coefficient. When there is no significant temperature change, feedback is performed every 10 temperature time steps.

5. The method for multi-physics dynamic coupling calculation of medium and large induction motors based on multiple time scales as described in claim 1, characterized in that, An electronic device comprising a multi-core processor and a GPU heterogeneous chip, a cache, a non-volatile memory unit, an FPGA hardware acceleration module, an industrial-grade PCIe 4.0 communication interface, and high-precision current, temperature, and flow rate sensors, for realizing real-time multi-field dynamic coupling calculation of medium and large induction motors.

6. The method for multi-physics dynamic coupling calculation of medium and large induction motors based on multiple time scales according to claim 1, wherein step 6 is characterized in that, The method introduces nodal equations for multi-field dynamic coupling calculation and employs an over-relaxation iterative method to improve calculation accuracy. It also includes a storage medium that uses a high-speed data throughput SSD loading algorithm to store the computer program of the method described in this patent.

Images