A motor temperature rise real-time estimation method and device based on an equivalent thermal network model

Abstract

The application belongs to the technical field of electric machines, and particularly relates to a motor temperature rise real-time estimation method and device based on an equivalent thermal network model. The method first constructs a motor thermal network model, then establishes a corrected thermal balance equation of each node according to the loss of each node and the equivalent thermal resistance of each node, solves the corrected thermal balance equation of each node in combination with the running time, and obtains the temperature rise of each node. The matrix form of the corrected thermal balance equation is: P is the loss of the node, theta is the temperature rise of the node, and R is the equivalent thermal resistance between the nodes. The application draws lessons from the concept of RC transient circuit in a circuit, combines the concept of a capacitor in heat transfer, adds a thermal capacity as a capacitor to the motor equivalent thermal network model, reflects the process of temperature change with time, can calculate not only the steady-state temperature rise but also the temperature rise under the transient state, and realizes real-time estimation of the motor temperature rise.

Description

Technical Field

[0001] The present invention belongs to the technical field of motors, and in particular relates to a method and device for real-time estimation of motor temperature rise based on an equivalent thermal network model. Background Art

[0002] As the power source for current new energy vehicles, the operating reliability of the drive motor is closely related to the performance of the entire vehicle. Temperature is an important indicator of motor operating status. Quickly and accurately determining the temperature rise of key components within the motor (such as stator windings, rotor magnets, and bearings) is crucial for improving motor reliability. Depending on the location of the components, motor temperature rise calculations are primarily categorized as stator and rotor. The stator temperature rise can be directly measured using wired temperature sensors, while the rotor temperature rise can also be measured using wireless temperature sensors (such as Bluetooth or infrared transmitters). However, wireless temperature sensors are expensive and have poor reliability. Furthermore, the placement of temperature sensors increases the failure rate of the motor during operation.

[0003] A Chinese invention patent, authorized with publication number CN107391884B, discloses a method for calculating the temperature rise of a dual-redundant permanent magnet synchronous motor based on an equivalent thermal network model. This method establishes a three-dimensional equivalent thermal network model to determine thermal resistance and heat sources, establishes a heat balance equation, and solves it to determine the temperature rise of each motor node. The relationship between the temperature rise of each node within the motor and the circuit model is established through the similarities between heat transfer and circuit theory. However, this method can only calculate the motor temperature rise in steady state—that is, the temperature rise after the motor reaches thermal equilibrium. It cannot reflect the transient temperature rise process within the motor in real time, making real-time calculation impossible. Furthermore, the entire model is open-loop control, resulting in low accuracy. Summary of the Invention

[0004] The present invention provides a method and device for real-time estimation of motor temperature rise based on an equivalent thermal network model, which is used to solve the problem that the prior art can only calculate the motor temperature rise in a steady state.

[0005] To solve the above technical problems, the technical solutions included in the present invention and the corresponding beneficial effects of the technical solutions are as follows:

[0006] The present invention provides a method for real-time estimation of motor temperature rise based on an equivalent thermal network model, comprising the following steps:

[0007] 1) Divide the nodes on the motor according to the characteristics of each motor component and construct the motor thermal network model;

[0008] 2) During the operation of the motor, obtain the current operating condition data of the motor; determine the loss of each node in the motor thermal network model based on the current operating condition data of the motor;

[0009] 3) determining the equivalent thermal resistance between each node according to the thermal conduction characteristics between each node;

[0010] 4) establishing a corrected thermal balance equation of each node according to the loss of each node and the equivalent thermal resistance between each node, solving the corrected thermal balance equation of each node in combination with the running time to obtain the temperature rise of each node; wherein the matrix form of the corrected thermal balance equation is:

[0011]

[0012] wherein P is the loss of the node, θ is the temperature rise of the node, R is the equivalent thermal resistance between the nodes, and C is the specific heat capacity.

[0013] The beneficial effects of the above technical solutions are that the application borrows the concept of RC transient circuit in the circuit, combines the concept of capacitor in heat transfer, adds the heat capacity as the capacitor to the equivalent thermal network model of the motor, reflects the process of temperature change with time, can calculate not only the steady-state temperature rise but also the temperature rise under the transient state, and realizes real-time estimation of the motor temperature rise.

[0014] Further, in order to introduce the feedback correction of temperature to the motor loss to improve the temperature calculation accuracy, the application further comprises the step of correcting the obtained loss of each node according to the temperature change of the motor under the current operating condition.

[0015] Further, in order to introduce the feedback correction of temperature to the copper loss to improve the temperature rise calculation accuracy, if the loss of the node is the copper loss, the corrected loss of the node is:

[0016] P cu1 = P cu0 [1+(T1-T0)*k1]

[0017] wherein P cu1 is the copper loss under the temperature T1; P cu0 is the copper loss under the temperature T0; and k1 is a set coefficient for representing the influence of temperature change on the copper loss.

[0018] Further, in order to introduce the feedback correction of temperature to the iron loss to improve the temperature rise calculation accuracy, if the loss of the node is the iron loss, the corrected loss of the node is:

[0019] P fe1 = P fe0 [1-(T1-T0)*k2] 2

[0020] wherein P fe1 is the iron loss under the temperature T1; P fe0is the iron loss at temperature T0; k2 is the set coefficient used to represent the impact of temperature change on iron loss.

[0021] Furthermore, in order to introduce temperature feedback correction on mechanical friction loss to improve the accuracy of temperature rise calculation, if the loss of the node is mechanical friction loss, the corrected node loss is:

[0022] P fw1 =P fw0 [1-(T1-T0)*k3] 2

[0023] Among them, P fw1 is the mechanical friction loss at temperature T1; P fw0 is the mechanical friction loss at temperature T0; k3 is the set coefficient used to represent the impact of temperature change on mechanical friction loss.

[0024] Furthermore, in order to facilitate the rapid calculation of the current motor loss, in step 2), the means for determining the loss of each node in the motor thermal network model based on the current operating condition data of the motor is: based on the benchmark operating condition data and the benchmark node loss corresponding to the benchmark operating condition data obtained by calibrating the benchmark operating condition, and combined with the current operating condition data, interpolation operation is performed to obtain the loss of each node corresponding to the current operating condition data.

[0025] Furthermore, in order to reflect the phenomenon that after the motor enters the weak magnetic field area, the motor loss increases because the motor stator needs to provide an additional part of the current to weaken the permanent magnet magnetic field, it also includes correcting the loss of each node in the motor thermal network model obtained according to the current motor speed: if the current motor speed is greater than the weak magnetic inflection point speed, the current motor speed and the weak magnetic coefficient are multiplied to correct the loss of each node.

[0026] Furthermore, the nodes include: a casing shell, a contact surface between the casing inner wall and the stator core, a contact surface between the casing and the rear end cover, a contact surface between the casing and the front end cover, a middle part of the winding, a front end part of the winding, a rear end part of the winding, air in front of the motor, air in back of the motor, a stator core, a rotor permanent magnet, a stator-rotor gap, a front end cover, a rear end cover, a front bearing inner ring, a front bearing outer ring, a rear bearing inner ring, a rear bearing outer ring, and a motor shaft.

[0027] Furthermore, the operating condition data includes motor speed, motor torque and cooling water temperature.

[0028] The present invention also provides a real-time estimation device for motor temperature rise based on an equivalent thermal network model, comprising a memory and a processor, wherein the processor is used to execute instructions stored in the memory to implement the above-mentioned real-time estimation method for motor temperature rise based on an equivalent thermal network model, and achieve the same beneficial effects as the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a flow chart of the method for real-time estimation of motor temperature rise based on an equivalent thermal network model of the present invention;

[0030] Figure 2 This is a schematic diagram of the temperature rise curve of the internal components of the motor of the present invention;

[0031] Figure 3 It is a schematic diagram of the transient temperature rise curve of the present invention;

[0032] Figure 4 This is a model diagram of an M file in MATLAB for correcting copper loss using temperature according to the present invention;

[0033] Figure 5 This is a model diagram of the copper loss influencing factor displayed using Simulink in MATLAB of the present invention;

[0034] Figure 6 This is a model diagram of the iron loss influencing factor displayed by Simulink in MATLAB of the present invention;

[0035] Figure 7 This is a model diagram of the mechanical friction loss influencing factor displayed using Simulink in MATLAB of the present invention;

[0036] Figure 8 This is a model diagram of a motor equivalent thermal network model after modularization and packaging displayed using Simulink in MATLAB of the present invention;

[0037] Figure 9 This is a module diagram of a housing (3×2) displayed using Simulink in MATLAB of the present invention;

[0038] Figure 10 The stator and winding module diagram of the present invention is displayed using Simulink in MATLAB;

[0039] Figure 11 This is a front-end winding loss and temperature rise module diagram displayed using Simulink in MATLAB of the present invention;

[0040] Figure 12 This is a module diagram of winding loss and temperature rise in the core displayed by Simulink in MATLAB of the present invention;

[0041] Figure 13 This is a diagram of a rotor core module displayed using Simulink in MATLAB;

[0042] Figure 14 This is a shaft module diagram displayed using Simulink in MATLAB of the present invention;

[0043] Figure 15 This is a diagram of the end cap module of the present invention displayed using Simulink in MATLAB;

[0044] Figure 16 This is a diagram of the internal front-end air module of the present invention displayed using Simulink in MATLAB;

[0045] Figure 17 This is a bearing module diagram displayed using Simulink in MATLAB of the present invention;

[0046] Figure 18 This is a schematic diagram of the heat transfer process distribution of the motor 1 / 2 model components of the present invention;

[0047] Figure 19 It is a structural diagram of the motor temperature rise real-time estimation device based on the equivalent thermal network model of the present invention. DETAILED DESCRIPTION

[0048] The basic concept of the present invention is as follows: drawing on the concept of RC transient circuits in electrical circuits and combining it with the concept of capacitance in heat transfer, the heat capacity is added as a capacitor to the equivalent thermal network model, enabling real-time analysis of the heat transfer process of the motor and real-time estimation of the temperature rise under transient conditions. Furthermore, temperature feedback on copper loss, iron loss, and mechanical friction loss is introduced into the motor equivalent thermal network model, reflecting the real-time change of the loss heat source with temperature, thereby improving the accuracy of the motor temperature rise calculation.

[0049] Method Example:

[0050] In order to achieve real-time estimation of motor temperature, it is necessary to build a motor equivalent thermal network model that can calculate the transient model. Currently, most motor temperature estimation methods based on the motor thermal network model mainly use thermal resistance and heat source to calculate the steady-state temperature rise. This method has two main problems: it cannot calculate the transient temperature rise, and the temperature rise calculation accuracy is low because the loss does not provide feedback correction for the temperature rise. To solve these two problems, the methods of the present invention are as follows:

[0051] 1. Solve the problem of "unable to calculate transient temperature rise".

[0052] The present invention draws on the principle that when the current in the RC transient circuit in the circuit flows through the node containing the capacitor, the capacitor will be charged. The concept of heat capacitance is introduced. That is, when the heat source in the thermal circuit model passes through a certain component node, it will first charge the heat capacitance at the node, which delays the temperature rise of the node to a certain extent. The temperature rise curve presents a parabolic shape, as shown in Figure 2. Figure 3 shown.

[0053] In steady state, the heat source loss P, temperature rise θ and equivalent thermal resistance R satisfy the following thermal equilibrium relationship:

[0054]

[0055] When the temperature rise has not reached a steady state, it will change over time, introducing heat capacity C. The standard definition of heat capacity is: "When a system's temperature rises by a small amount of heat dQ and rises by dT, the quantity dQ / dT is the system's heat capacity." The unit is J / K. The heat capacity is equal to the product of the object's specific heat capacity and its mass. The formula is:

[0056] C=C m *m (2)

[0057] Specific heat capacity, also known as specific heat, is one of the physical properties of a substance. It refers to the amount of heat absorbed for every 1K increase in temperature or the amount of heat released for every 1K decrease in temperature when a unit mass of the substance absorbs or releases heat, causing the temperature to rise or fall. It is usually represented by the symbol Cm and the unit is J / (kg.K).

[0058] After adding the concept of heat capacity, the heat balance equation (1) can be improved as follows:

[0059]

[0060] Where t is time. The meaning of this thermal equilibrium relationship is that under transient conditions, the heat generated by an object is equal to the heat dissipated plus the heat absorbed by the object itself, the latter causing its temperature to continue to rise. Its specific expression is:

[0061]

[0062] Analyzing the first-order differential equation shown in formula (3), we can obtain:

[0063]

[0064] Or:

[0065]

[0066] Among them, θ ∞ The steady-state temperature rise means that after a long enough time, the heat generation and heat dissipation reach a balance and the temperature rise no longer continues. ∞ =P*R; τ is the object's heating time constant, representing the object's temperature rise rate. The larger the heat capacity and the smaller the thermal conductivity, the larger the heating time constant, which means it takes more time to reach a steady state. τ = C*R; T is the object's heating time constant, and θ0 is the initial temperature rise. This formula shows that the temperature rise θ is a curve that changes exponentially with time, such as Figure 3As shown. At t = 3τ, When , it can generally be considered to have reached a steady state.

[0067] Based on the above principles, capacitors are added to the nodes of key motor components when constructing the motor thermal network model.

[0068] 2. Solve the problem of "low accuracy of temperature rise calculation due to lack of feedback correction for temperature rise due to loss".

[0069] In actual motor operation, motor losses are not constant. To maintain the same output as the temperature rises, the motor's input current increases accordingly. Considering the effects of temperature on winding resistance, the magnetic properties of the magnetic material, bearing lubrication, and friction coefficient, various motor losses will vary with temperature rise, necessitating modifications to the constructed model. The loss heat source in the equivalent thermal network model is expressed as a controllable current source, with the specific loss input source directly replaced by the current source. This invention introduces a temperature influence factor to address this issue.

[0070] 1) Copper loss influencing factors.

[0071] The copper loss factor is determined by the measured winding temperature VcoilR and the initial loss P cu2 We jointly decided to use MATLAB to show the schematic diagram of this relationship as follows Figure 4 As shown, the copper loss input is simply P cu2 Added copper loss influence factor P withTem1 as the feedback factor.

[0072] Introducing the negative feedback of temperature on copper loss, according to the thermal conductivity of copper, its temperature rises by 1 ° C, the resistance changes by 0.004, the copper loss P at a known temperature T0 is cu0 In the case of temperature T1, the copper loss P can be obtained. cu1 The calculation formula is:

[0073] P cu1 =P cu0 [1+(T1-T0)*0.004] (7)

[0074] Then, the copper loss influence factor P can be displayed using MATLAB. withTem1 The module diagram of the expression is as follows Figure 5 As shown, the input interface In1 is connected to the measured winding temperature, and the output interface Out1 is connected to the loss heat source.

[0075] 2) Iron loss influencing factors.

[0076] Introducing negative feedback of temperature on iron loss, temperature changes affect the magnetic induction intensity, unit iron loss and magnetic properties of permanent magnets of silicon steel. Analyze the no-load iron loss data of motors at different temperatures, and the iron loss P at a known temperature T0 fe0 In the case of temperature T1, the iron loss P can be obtained. fe1 The calculation formula is:

[0077] P fe1 =P fe0 [1-(T1-T0)*0.0012] 2 (8)

[0078] Then we can get the module diagram of the iron loss influence factor expression using MATLAB as follows: Figure 6 As shown, the input interface In1 is connected to the measured magnetic steel temperature, and the output interface Out1 is connected to the core loss heat source.

[0079] 3) Mechanical friction loss factor.

[0080] Introducing the negative feedback of temperature on mechanical friction loss, analyzing the motor mechanical friction loss data at different temperatures, the mechanical loss P at a known temperature T0 fw0 In the case of temperature T1, the mechanical loss P can be obtained. fw1 The calculation formula is:

[0081] P fw1 =P fw0 [1-(T1-T0)*0.003] 2 (9)

[0082] Then we can use MATLAB to show the module diagram of the mechanical friction loss factor expression as follows: Figure 7 As shown, the input interface In1 is connected to the measured bearing temperature, and the output interface Out1 is connected to the mechanical loss heat source.

[0083] After solving the above two problems, the motor equivalent thermal network model built by MATLAB is as follows Figure 8 As shown in the figure, the motor equivalent thermal network model is mainly composed of motor component modules represented by Arabic numerals (1, 2, 3...) and heat source input modules represented by Roman numerals (I, II, III...). The serial number identifiers (①, ②, ③...) represent the temperature output of each node. The specific information is:

[0084] Arabic numeral 1 represents the casing module, which mainly includes the motor water-cooled casing, cooling water channels, coolant, etc., and is in direct contact with the motor stator core and end cover; Arabic numeral 2 represents the motor stator module, which mainly includes the stator core, stator winding, etc., and is in direct contact with the motor casing, the air gap between the stator and rotor, and the end air; Arabic numeral 3 represents the motor rotor module, which includes the rotor core, magnetic steel, etc., and is in direct contact with the motor shaft, the air gap between the stator and rotor, and the end air; Arabic numeral 4 represents the motor shaft, Arabic numerals 5 and 6 represent the front and rear bearings, Arabic numerals 7 and 8 represent the front and rear end covers, and Arabic numerals 9 and 10 represent the air in the front and rear of the motor.

[0085] The Roman numerals I, II, and III represent the stator winding's center, front, and rear copper losses, respectively. The Roman numerals IV and VII represent the stator and rotor core losses, respectively. The Roman numeral VI represents the permanent magnet losses, and the Roman numeral V represents the stator-rotor air gap module and the friction losses due to rotor friction. The mechanical losses due to friction in the front and rear bearings are integrated into the front axle bearing modules 5 and 6 and are no longer shown separately.

[0086] Serial number ① outputs the temperature of the casing shell, Serial number ② outputs the temperature of the contact surface between the casing inner wall and the stator core, Serial number ③ outputs the temperature of the contact surface between the casing and the rear end cover, Serial number ④ outputs the temperature of the contact surface between the casing and the front end cover, Serial number ⑤ outputs the temperature of the middle part of the winding, Serial number ⑥ outputs the temperature of the front end of the winding, Serial number ⑦ outputs the temperature of the rear end of the winding, Serial number ⑧ outputs the temperature of the air in front of the motor, Serial number ⑨ outputs the temperature of the air in the rear of the motor, Serial number ⑩ outputs the temperature of the stator core, Serial number outputs Temperature of the rotor permanent magnet, No. Output the temperature of the air gap between the stator and rotor, serial number Output the temperature of the front cover, serial number Output the temperature of the rear end cover, serial number Output the temperature of the inner ring of the front bearing, serial number Output the temperature of the outer ring of the front bearing, serial number Output the temperature of the inner ring of the bearing, serial number Output the temperature of the outer ring of the rear bearing, serial number Output motor shaft temperature. The output is the loss value in the middle of the stator winding, which is used as an observation quantity for subsequent redundant control. Because the stator copper loss can be simply determined by Joule's law, the loss heat P is equal to the current I and the winding resistance R. Therefore, this observation can be used to determine whether there are any abnormalities in the settings of the loss heat source in the model and its negative feedback based on the node temperature. If there are any abnormalities, it can be terminated in time.

[0087] The following is a detailed introduction to the motor component module.

[0088] 1) Casing module. The heat transfer route between the motor casing and the stator core is: external environment - casing exterior - water channel - casing interior - the front and rear parts of the casing are connected in parallel with the stator core. In combination with the motor production process, the casing and the stator core are heat-shrink-fitted to achieve an interference fit. In the model, there is a certain contact thermal resistance between the casing and the stator core. Figure 9 As shown, three interfaces are used to connect the front and rear end covers and the stator core respectively, and the two signal outputs represent the temperature of the contact point between the housing and the stator core and the contact point between the housing and the end covers respectively.

[0089] 2) Motor stator module. Stator core (8×0) module such as Figure 10 The 8 interfaces are connected to the chassis, core winding copper loss, core winding temperature, winding rear end copper loss, winding front end copper loss, winding front end temperature measurement point, stator iron loss, air gap and winding rear end temperature measurement point. Figure 11 As shown. The two interfaces are connected to the stator module's interface 4 front-end winding copper loss and interface 6 front-end winding temperature measurement points respectively. One signal output interface is connected to an oscilloscope to display the front-end winding temperature change. The stator rear-end winding loss and temperature rise module is similar to the front-end and will not be described in detail. The stator middle winding loss and temperature rise (2×2) module is as follows: Figure 12 The two interfaces are connected to the stator module's interface 2 (measuring the copper loss of the central winding in the core) and interface 3 (measuring the temperature of the central winding in the core). The two signal output interfaces are connected to an oscilloscope to display the temperature changes of the central winding in the core and the changes in the copper loss of the winding.

[0090] 3) Motor rotor module. The rotor core (6×0) module is as follows Figure 13 The 6 interfaces are connected to the air gap between the stator and rotor, the internal front air, the internal rear air, the rotor iron loss signal, the magnetic steel loss signal and the shaft. Figure 14 As shown in the figure, the three interfaces are connected to the rotor core, the front bearing and the rear bearing respectively.

[0091] 4) Front and rear end covers. Front end cover (3×0) module as Figure 15 As shown in the figure, the three ports connect to the housing, internal air, and bearings. The rear cover is similar to the front cover.

[0092] 5) Air inside the motor. Internal front air (4×0) module such as Figure 16 As shown in the figure, the four interfaces connect the front end of the housing, the front cover, the front end winding, and the rotor module interface 2. The internal rear air module is similar to the front end.

[0093] 6) Front and rear bearings. Front bearing (2×2) module such as Figure 17The two interfaces are connected with the end cover and the rotating shaft respectively, and the two signal outputs are connected with the oscilloscope to represent the inner ring temperature and the outer ring temperature of the bearing. The rear bearing is similar to the front end. The mechanical loss of the motor mainly includes the loss generated by the friction between the bearing and the oil seal and the rotating shaft, and the loss generated by the friction between the rotor and the air; the mechanical loss is separated by combining the mechanical loss test data of the bearing provided by the bearing manufacturer and the mechanical loss test data with and without the oil seal; part of the mechanical loss is applied to the contact point between the bearing inner ring and the rotating shaft, and the other part of the mechanical loss is applied to the rotor loss.

[0094] After the equivalent thermal network model of the motor is constructed according to the above content, the motor temperature rise can be estimated in real time according to the steps shown in Figure 1 The equivalent thermal network model of the motor is constructed according to the above content, the motor temperature rise can be estimated in real time according to the steps shown in

[0095] S0: Preparation stage, mainly according to the specific motor to calculate the corresponding thermal resistance, heat capacity, heat source and other data to establish the equivalent thermal network model of the motor, in order to carry out real-time estimation of motor temperature, the model needs to have the ability of transient calculation and the function of closed loop feedback to temperature.

[0096] S1: After the establishment of the equivalent thermal network model of the motor, the program can be run to estimate the motor temperature rise in real time.

[0097] S2: Sampling data, mainly MCU sampling the running condition data of the motor in the current time t, the sampled running condition data includes motor speed (for convenience, the letter R is used instead of the following), torque (Q), cooling water temperature (T), sampling interval (△t), the obtained data is shown in Table 1. This is the input parameter of the model, which is mainly used to calculate the internal heat source distribution of the motor under the current working condition.

[0098] Table 1: sampled motor running condition data

[0099]

[0100]

[0101] S3: Read the running condition data sequence, judge how many groups of data are sampled in t time, and count M.

[0102] S4: Introduce the current calculation sequence N, judge the running condition data sequence: as long as M-N is greater than zero, it means that there is data not yet calculated, at this time, go to S5 to run the program for subsequent processing; if M-N is not greater than zero, it is considered that the calculation is completed, and directly go to S12 to run the program for subsequent processing.

[0103] S5: Read the detailed running condition data, including R, Q and T, which are used for subsequent program and model calculation.

[0104] S6: According to the read current R, Q, T, combined with the reference working condition data, and the reference motor loss corresponding to the reference working condition data (including the loss of each node) obtained by calibrating the reference working condition, interpolation operation is performed to obtain the current motor loss corresponding to the current R, Q, T (part of the nodes are nodes as heat sources, and their losses are not 0, and part of the nodes are passive nodes, and their losses are 0).

[0105] S7: Determine whether the current motor speed R is greater than the field weakening speed inflection point R0: if the current motor speed R is not greater than the field weakening speed inflection point R0, the current motor loss calculated by interpolation in S6 is directly used for subsequent calculation processing, that is, directly go to S8 for next step calculation; if the current motor speed R is greater than the field weakening speed inflection point R0, it is considered that the motor has entered the field weakening region at the current speed, at this time, in order to suppress the high motor speed causing the high motor back electromotive force, the motor stator needs to provide an additional part of the current to weaken the permanent magnet magnetic field, so the motor loss will increase, and the current motor loss calculated by interpolation in S6 needs to be corrected, that is, multiplying the current motor loss calculated by interpolation in S6 by a field weakening coefficient fk, the field weakening coefficient is different for different motor saliency rates and control modes, and the size is usually between 1.5-2.5. Then go to S8 for next step calculation.

[0106] S8: Assign the finally obtained current motor loss to each heat source of the motor equivalent thermal network model. And calculate the equivalent thermal resistance between nodes.

[0107] S9: Run the motor equivalent thermal network model built in MATLAB's Simulink, the running time is t, according to the loss of each node and the equivalent thermal resistance between nodes, Simulink uses the formula as shown in formula (2) to establish the corrected heat balance equation of each node in the model, combined with the running time t, the corrected heat balance equation of each node is solved, so as to obtain the voltage (i.e. temperature) of each node, and the oscilloscope is used for collection. In this process, the loss of each node will be continuously corrected according to the change of temperature, if it is copper loss, the correction formula is as shown in formula (7), if it is iron loss, the correction formula is as shown in formula (8), and if it is mechanical friction loss, the correction formula is as shown in formula (9). That is, under the current operating condition, the loss of each heat source is not fixed, but changes with temperature, that is, the loss input to the node brings heat, which will immediately consider the heat caused by the loss itself.

[0108] S10: Record the data stored in the oscilloscope after model calculation to a specific table, as shown in Table 2 below (the data in Table 2 is not complete, and all nodes are not shown).

[0109] Table 2 Temperature rise data of key components inside the motor

[0110]

[0111] S11: After calculating one operating condition data, the current calculation sequence N+1 is transferred to S4 to automatically calculate the next operating condition data.

[0112] S12: If the MN calculated in S4 changes from greater than zero to not greater than zero, it is considered that all the current working condition data have been calculated and reported to the MCU for the next step.

[0113] S13: The temperature data of the key components of the motor calculated and stored in S4 to S10 within the time t are read out uniformly and sorted out to obtain the component temperature rise curve, such as Figure 2 is shown, and then automatically goes to the next step.

[0114] S14: Clear the original sampling data, transfer to S2 to automatically perform the next round of time sampling and the next round of S4 to S10 program operation, and then automatically update the temperature rise curve obtained in S13 to realize the real-time estimation function.

[0115] In step S8, the equivalent thermal resistance between each node needs to be calculated. The following describes in detail how to calculate the equivalent thermal resistance.

[0116] The thermal resistance and thermal capacitance in the thermal circuit represent the resistance and capacitance elements in the equivalent thermal network model respectively. The resistance and thermal resistance represent the strength of the object's electrical and thermal conductivity, respectively, while the capacitance and thermal capacitance represent the object's ability to absorb and store electricity and heat, respectively. Through detailed analysis and disassembly of each component in the motor model, such as Figure 18 As shown, a schematic diagram of the motor 1 / 2 model structure is established.

[0117] The atmospheric environment serves as the boundary of the model, and the temperature is approximately assumed to remain constant. The motor casing and end cover exchange heat with the environment through the air. The transfer forms include heat radiation, heat conduction, and heat convection. The calculation formula for its transfer thermal resistance is complex, and the main heat exchange form of the motor is forced water cooling. Therefore, the empirical formula is used as a reference for this thermal resistance in the model.

[0118] The various mechanical components of the motor include the casing, end covers, shaft, bearings, stator and rotor, etc., and each has its own conductive thermal resistance and heat capacity, which can be calculated according to their respective formulas. Among them, the stator winding is an important heat source inside the motor. Although it is a whole, the effective electromagnetic part is only in the middle section, and the boundary conditions for heating and heat dissipation in the middle and end sections are different. Therefore, the conductive thermal resistance of the stator winding itself is divided into three parts; the contact thermal resistance needs to be considered when there are gaps in the fitting parts of each component due to tolerance matching. The contact thermal resistance is calculated when the heat conduction medium is air; there is an insulating layer and varnish between the motor winding and the stator, and there is filling glue between the rotor core and the magnet, so there is also contact thermal resistance.

[0119] The thermal resistance and heat capacity of each motor component, including mechanical structural components such as the casing, end cover, shaft, and bearings, as well as key electromagnetic components such as the stator core, rotor core, and magnets, are extracted. In order to better apply to platform-based motors, the sizes of the variables corresponding to the heat source, heat capacity, and thermal resistance in the Simulink model are all expressed as parameters. The values ​​of each sheet in the Excel table are read using an M file and then assigned to each parameter, as shown in Tables 3 and 4.

[0120] Table 3 Model electrical (thermal) capacitance list

[0121] entry Model variable name entry Model variable name chassis C_house rotor core C_rotor Front cover C_endcapF magnetic steel C_pm rear end cover C_endcapR shaft C_shaft Middle of winding C_coil front bearing C_bearingR Winding front end C_coilF rear bearing C_bearingF Winding rear end C_coilR Front Air C_ESR stator core C_stator rear air C_ESF

[0122] Table 4 List of electrical (thermal) resistances in the model

[0123]

[0124]

[0125] Depending on the type of heat transfer, thermal resistance can also be conductive thermal resistance, convection thermal resistance and radiation thermal resistance. Among them, conductive thermal resistance is the most basic, and the calculation formula is similar to that of resistance. Where λ is the thermal conductivity of the object, d is the thickness of the heat conducting layer, and S is the area of ​​the heat conducting layer.

[0126] The following is the calculation of the main equivalent thermal resistance in the motor model.

[0127] a) Thermal resistance between coolant and housing:

[0128]

[0129] Where d1 is the equivalent diameter of the water channel (m), S1 is the total heat exchange area of ​​the water channel (m 2 ), λ1 is the thermal conductivity of the water channel coolant (W / mK), and Nu is the Nusselt number of the coolant.

[0130] b) Case thermal resistance:

[0131]

[0132] Where d2 is the thickness of the heat transfer layer of the casing (m), S2 is the surface area of ​​the inner wall of the casing (m 2 ), λ2 is the thermal conductivity of the casing (W / mK).

[0133] c) Thermal resistance between stator and casing:

[0134]

[0135] Among them, d3 is the thickness of the heat transfer layer between the stator and the casing (the air gap is calculated as air) (i.e. the fitting clearance between the stator and the casing is generally very small, so this part can be ignored in the simple calculation), S3 is the inner wall surface area of ​​the casing (m 2 ), λ3 is the thermal conductivity of the heat transfer layer (air) (W / mK).

[0136] d) Thermal conduction resistance of the stator core yoke. The thermal conduction of the stator core yoke can be calculated based on the conduction of a hollow cylinder or a flat plate.

[0137] Method 1: When the stator core yoke is calculated as a hollow cylinder, the thermal resistance is:

[0138]

[0139] Where r1 and r2 are the inner and outer diameters of the stator core yoke (m), λ4 is the thermal conductivity of the core (W / mK), and l is the length of the core (m).

[0140] Method 2: When the stator core yoke is calculated based on flat plate conduction, the thermal resistance is:

[0141]

[0142] Where d4 is the thickness of the heat transfer layer of the stator yoke (m), S4 is the surface area of ​​the inner diameter of the core yoke (m 2 ), λ4 is the thermal conductivity of the core (W / mK).

[0143] e) Thermal conduction resistance of the stator core teeth. The thermal resistance of the stator core teeth is calculated as a hollow cylinder:

[0144]

[0145] Where r3 and r4 are the inner and outer diameters of the stator core teeth (m), λ5 is the thermal conductivity of the core (W / mK), l is the length of the core (m), and k is the ratio of the tooth area to the area of ​​the hollow cylinder.

[0146] f) Thermal conduction resistance between the winding and the core in the slot:

[0147]

[0148] Where d6 is the thickness of the insulating material in the slot (m), S6 is the surface area of ​​the slot (m 2 ), λ6 is the thermal conductivity of the insulating material (W / mK).

[0149] g) Thermal conduction resistance between the winding in the slot and the end winding:

[0150]

[0151] Where d7 is the thickness of the insulation material in the slot (m), S7 is the cross-sectional area of ​​the conductor in the slot (m 2 ), λ7 is the thermal conductivity of the conductor winding (W / mK).

[0152] h) Convective thermal resistance of the air gap. Heat transfer in an air gap occurs through similar mechanisms to liquids, including conduction, radiation, and convection. In the absence of axial airflow, the airflow is laminar at low speeds, transitioning to a turbulent state at higher speeds. In laminar flow, the airflow has no axial velocity, and heat transfer occurs solely through conduction and radiation. In turbulent flow, the airflow has axial velocity, and convection becomes the primary heat transfer mechanism.

[0153] Different states of airflow can be expressed by Taylor numbers:

[0154]

[0155] Where ω is the angular velocity (rad / s), r i is the rotor outer diameter (m), D h is the hydraulic diameter (m), ν is the kinematic viscosity ((m 2 / s). When T a When <1740, the airflow is in laminar state, Nu=2; when T a When ≥1740, the airflow is in a turbulent state.

[0156] The convection thermal resistance of the air gap is:

[0157]

[0158] Where S8 is the total heat transfer area of ​​the air gap layer (m 2 ), λ8 is the thermal conductivity of air (W / mK), and δ is the air gap length (m).

[0159] i) Thermal resistance of rotor yoke:

[0160]

[0161] Where r5 and r6 are the inner and outer diameters of the rotor core yoke (m), λ9 is the thermal conductivity of the core (W / mK), and l is the length of the core (m).

[0162] j) Thermal resistance from magnet to rotor:

[0163]

[0164] Among them, d 10 is the thickness of the air gap in the magnetic slot (m), S 10 is the surface area of ​​the magnetic groove (m 2 ), λ 10 is the thermal conductivity of the air gap (W / mK).

[0165] k) Thermal resistance of magnetic steel:

[0166]

[0167] Where, d is the thickness of the magnet (m), S is the heat transfer area of ​​the magnet (m 2 ), λ is the thermal conductivity of the magnetic steel (W / mK).

[0168] l) Thermal conduction resistance inside the rotor:

[0169]

[0170] Among them, r7 and r8 are the inner and outer diameters of the rotor (m), λ 12 is the thermal conductivity of the core (W / mK), and l is the length of the core (m).

[0171] m) Thermal conduction resistance of the shaft and rotor:

[0172]

[0173] Among them, d 13 is the thickness of the rotor shaft gap (m), S 13 is the heat transfer area of ​​the contact surface between the rotor and the shaft (m 2 ), λ 13 is the thermal conductivity of the air gap (W / mK).

[0174] n) Thermal conduction resistance of the shaft:

[0175]

[0176] Among them, d 14 is the thickness of the magnetic steel (m), S 14 is the heat transfer area of ​​the shaft cross section (m 2 ), λ 41 is the thermal conductivity of the shaft (W / mK).

[0177] o) Heat transfer resistance from shaft to bearing:

[0178]

[0179] Among them, d 15 is the thickness of the gap between the shaft and the bearing (m), S 15 is the heat transfer area of ​​the contact surface between the shaft and the bearing (m 2 ), λ 15 is the thermal conductivity of the air gap between the shaft and bearing (W / mK).

[0180] p) Thermal conduction resistance between the winding end and the air inside the motor:

[0181]

[0182]

[0183] Among them, h δ1 is the heat transfer coefficient between the winding end and the air inside the motor (W / m 2 K), S 16 is the total heat exchange area between the winding end and the air inside the motor (m 2 ),u r is the rotor linear speed (m / s). The single-side thermal resistance is 50% of the total thermal resistance.

[0184] q) Heat transfer resistance between the rotor and the air inside the motor:

[0185]

[0186]

[0187] Among them, h δ2 is the heat transfer coefficient between the rotor and the air inside the motor (W / m 2 K), S 17 is the total heat exchange area of ​​the rotor and the air inside the motor (m 2 ). The single-side thermal resistance is 50% of the total thermal resistance.

[0188] r) Thermal resistance of heat transfer from the air inside the motor to the casing:

[0189]

[0190]

[0191] Among them, h δ3 is the heat transfer coefficient (W / m 2 K), S 18 is the total heat exchange area of ​​the motor's internal air and casing (m 2 ). The single-side thermal resistance is 50% of the total thermal resistance.

[0192] s) Thermal resistance of heat transfer from the air inside the motor to the end cover:

[0193]

[0194] Among them, S 19 Total heat exchange area of ​​the motor internal air and end cover (m 2 ).

[0195] The present invention provides a method for estimating the temperature rise of a motor in real time based on an equivalent thermal network model, which can realize the sampling, calculation processing and automatic loop iteration of the motor operating condition data; by adding heat capacity to the motor equivalent network model, it can not only calculate the steady-state temperature rise, but also calculate and record the temperature rise under transient conditions, thereby realizing real-time estimation of the temperature rise. Moreover, under a certain operating condition, the loss calculated using steps S5 to S7 is equivalent to the initial value of the loss of each node under this operating condition. As the motor continues to operate under this operating condition, the temperature of each node of the motor will continue to change, and the loss of each node will also continue to change. Therefore, the present invention introduces temperature feedback on the heat source, realizing real-time changes in loss with temperature, thereby improving the accuracy of the model. However, if the operating condition changes, it is necessary to recalculate the initial value of the loss of each node using steps S5 to S7. As the motor continues to operate under this operating condition, the loss is continuously corrected using temperature.

[0196] Device Example:

[0197] An embodiment of a motor temperature rise real-time estimation device based on an equivalent thermal network model of the present invention is as follows: Figure 19 As shown, the system includes a memory, a processor, and an internal bus. The processor and memory communicate and exchange data with each other via the internal bus. The memory includes at least one software function module stored in the memory. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing a real-time motor temperature rise estimation method based on an equivalent thermal network model as described in the method embodiment of the present invention.

[0198] The processor may be a microprocessor MCU, a programmable logic device FPGA or other processing devices.

[0199] The memory can be various types of memories that use electrical energy to store information, such as RAM, ROM, etc.; it can also be various types of memories that use magnetic energy to store information, such as hard disks, floppy disks, magnetic tapes, magnetic core memories, bubble memories, USB flash drives, etc.; it can also be various types of memories that use optical methods to store information, such as CDs, DVDs, etc.; of course, it can also be other types of memories, such as quantum memories, graphene memories, etc.

Claims

1. A real-time estimation method for motor temperature rise based on an equivalent thermal network model, characterized in that: include: 1) Divide the nodes on the motor according to the characteristics of each motor component and construct the motor thermal network model; 2) obtaining the current operating condition data of the motor during the operation of the motor, where the operating condition data includes at least the motor speed; Based on the benchmark operating condition data and the benchmark motor node losses corresponding to the benchmark operating condition data obtained by calibration of the benchmark operating condition, and combined with the operating condition data at a certain moment in the current period, interpolation calculation is performed to obtain the losses of each node serving as a heat source corresponding to the operating condition data at each moment in the current period; Then, determine whether the current motor speed is greater than the field-weakening speed inflection point. If so, multiply the interpolation result by the field-weakening coefficient to correct the losses of each node serving as a heat source. Otherwise, no correction is made to the losses of each node serving as a heat source. 3) Determine the equivalent thermal resistance between each node based on the heat conduction characteristics between the nodes. 4) Based on the loss of each node and the equivalent thermal resistance between each node, a revised heat balance equation for each node is established. Combined with the running time, the revised heat balance equation for each node is solved to obtain the temperature rise of each node; wherein the matrix form of the revised heat balance equation is: P is the loss of the node, θ is the temperature rise of the node, R is the equivalent thermal resistance between nodes, C is the specific heat capacity, and t is time.

2. The method for real-time estimation of motor temperature rise based on an equivalent thermal network model according to claim 1, characterized in that: The method also includes the step of correcting the obtained losses of each node according to the temperature change of the motor under the current operating condition.

3. The method for real-time estimation of motor temperature rise based on an equivalent thermal network model according to claim 2, characterized in that: If the node loss is copper loss, the corrected node loss is: P cu1 =P cu0 [1+(T1-T0)*k1] Among them, P cu1 is the copper loss at temperature T1; P cu0 is the copper loss at temperature T0; k1 is the set coefficient used to represent the impact of temperature change on copper loss.

4. The method for real-time estimation of motor temperature rise based on an equivalent thermal network model according to claim 2, characterized in that: If the node loss is iron loss, the corrected node loss is: P fe1 =P fe0 [1-(T1-T0)*k2] 2 Among them, P fe1 is the iron loss at temperature T1; P fe0 is the iron loss at temperature T0; k2 is the set coefficient used to represent the impact of temperature change on iron loss.

5. The method for real-time estimation of motor temperature rise based on an equivalent thermal network model according to claim 2, characterized in that: If the node loss is mechanical friction loss, the corrected node loss is: P fw1 =P fw0 [1-(T1-T0)*k3] 2 Among them, P fw1 is the mechanical friction loss at temperature T1; P fw0 is the mechanical friction loss at temperature T0; k3 is the set coefficient used to represent the impact of temperature change on mechanical friction loss.

6. The method for real-time estimation of motor temperature rise based on an equivalent thermal network model according to claim 1, characterized in that: The nodes include: a casing shell, a contact surface between the casing inner wall and the stator core, a contact surface between the casing and the rear end cover, a contact surface between the casing and the front end cover, a middle part of the winding, a front end part of the winding, a rear end part of the winding, air in front of the motor, air in back of the motor, a stator core, a rotor permanent magnet, a stator-rotor gap, a front end cover, a rear end cover, a front bearing inner ring, a front bearing outer ring, a rear bearing inner ring, a rear bearing outer ring, and a motor shaft.

7. The method for real-time estimation of motor temperature rise based on an equivalent thermal network model according to claim 1, characterized in that: The operating condition data also includes motor torque and cooling water temperature.

8. A real-time estimation device for motor temperature rise based on an equivalent thermal network model, characterized in that: The method comprises a memory and a processor, wherein the processor is used to execute instructions stored in the memory to implement the real-time estimation method for motor temperature rise based on an equivalent thermal network model as claimed in any one of claims 1 to 7.

Images