A motor temperature control method and device based on a digital twin thermal model and a medium

Abstract

The application discloses a motor temperature control method and device based on a digital twin thermal model and a medium, and relates to the technical field of data processing. The method comprises the following steps: a centralized parameter thermal network model containing three nodes of windings, a stator core and a motor shell is constructed, and real-time collected motor electrical parameters, shell temperatures and wellbore fluid physical property parameters are combined to dynamically update the convective heat transfer boundary conditions and accurately calculate motor losses as heat sources, and then a winding hot spot temperature reflecting an insulation safety limit is reconstructed; on the basis, under the premise of keeping the electromagnetic torque constant, the hot spot temperature is taken as feedback to generate a current vector control instruction to drive a frequency converter to actively adjust the motor operating state. The method effectively solves the model inaccuracy caused by the fixed thermal boundary in the traditional temperature control technology, the extensive control caused by the dependence on the shell temperature, and the passive response defects of the lack of working condition adaptive adjustment capability.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, device and medium for motor temperature control based on a digital twin thermal model. Background Technology

[0002] With the expansion of heavy oil resource development, electric submersible screw pumps are widely used due to their high torque and low speed characteristics. However, their downhole submersible motors face severe heat dissipation challenges under high temperature, high pressure, and weak convection conditions.

[0003] Submersible motors (SMPs) for electric submersible screw pumps operate in high-temperature, high-pressure downhole environments, especially in heavy oil wells. Due to the high viscosity and low flow rate of crude oil, natural convection cooling is severely insufficient, easily leading to overheating and even burnout of the motor windings. Current technologies primarily use parametric thermal models to predict motor temperature, combined with frequency converters for frequency reduction or current limiting to suppress temperature rise. However, this approach has significant drawbacks, such as: fixed boundary conditions in the thermal model leading to misjudgments of cooling capacity; reliance solely on current amplitude adjustment causing a significant decrease in fluid production while controlling temperature; fixed model parameters resulting in deteriorated accuracy over long-term operation; and outputting only the average winding temperature, leading to delayed early warnings. In summary, existing temperature control technologies suffer from model inaccuracies, coarse control, and a lack of adaptive capabilities. Summary of the Invention

[0004] This application provides a method, device, and medium for motor temperature control based on a digital twin thermal model. The main purpose is to solve the problems of model inaccuracy, coarse control, and lack of adaptive capability in existing temperature control technologies.

[0005] In a first aspect, embodiments of this application provide a motor temperature control method based on a digital twin thermal model, the method comprising: constructing a lumped parameter thermal network model including three temperature nodes: windings, stator core, and motor housing;

[0006] Real-time acquisition of downhole operating data of the electric submersible screw pump motor, the downhole operating data including at least the motor electrical parameters, motor housing temperature and wellbore fluid physical properties;

[0007] The convective heat transfer boundary conditions of the thermal network model are dynamically updated based on the fluid properties of the wellbore and the temperature of the motor casing.

[0008] The motor loss is calculated based on the motor electrical parameters, and the motor loss is used as the heat source input convection heat transfer boundary condition to update the thermal network model, thereby obtaining the winding average temperature.

[0009] By combining the average winding temperature and local loss distribution information, the winding hot spot temperature is reconstructed;

[0010] While maintaining a constant output electromagnetic torque of the motor, a current vector control command is generated based on the hot spot temperature of the winding, and the frequency converter is driven to adjust the operating state of the motor based on the current vector control command in order to suppress temperature rise.

[0011] In one implementation of this application, the wellbore fluid properties include crude oil viscosity, sand concentration, fluid velocity, and fluid gas content.

[0012] The convective heat transfer boundary conditions of the thermal network model dynamically updated based on the wellbore fluid properties and the motor casing temperature include:

[0013] By inputting the physical property parameters of the well fluid and the temperature of the motor housing into a preset heat dissipation capacity mapping relationship, the heat exchange parameters between the motor housing and the well fluid in the thermal network model are obtained.

[0014] The convective heat transfer boundary conditions of the thermal network model are dynamically updated based on the heat exchange parameters.

[0015] In one implementation of this application, before reconstructing the winding hot spot temperature by combining the winding average temperature and local loss distribution information, the method includes:

[0016] The three-phase current signal in the motor electrical parameters is subjected to spectral decomposition, and the fifth and seventh harmonic amplitudes are extracted from the decomposition results.

[0017] The amplitude of the fifth harmonic is compared with the reference amplitude of the fifth harmonic under normal operating conditions, and the amplitude of the seventh harmonic is compared with the reference amplitude of the seventh harmonic under normal operating conditions.

[0018] If the amplitude of any harmonic exceeds a preset multiple of the corresponding reference amplitude, a local loss enhancement indication signal is generated. The local loss enhancement indication signal is used to characterize the trend of non-uniform heating aggravation in the winding.

[0019] Based on the local loss enhancement indication signal, the corresponding local loss distribution information is queried from the preset hot spot temperature rise mapping table.

[0020] In one implementation of this application, reconstructing the winding hot spot temperature by combining the winding average temperature and local loss distribution information includes:

[0021] The average temperature of the winding is added to the temperature rise offset of the basic hot spot to obtain the intermediate temperature corresponding to the hottest part of the winding.

[0022] The additional temperature rise compensation value is determined based on the local loss distribution information.

[0023] The additional temperature rise compensation value is added to the intermediate temperature to obtain the winding hot spot temperature.

[0024] In one implementation of this application, the step of generating a current vector control command based on the winding hot spot temperature includes:

[0025] The initial value of the q-axis current is determined based on the electromagnetic torque required for production.

[0026] Compare the winding hot spot temperature with the safe temperature threshold.

[0027] If the hot spot temperature of the winding is higher than the safe temperature threshold, then the initial value is reduced and the target value of the d-axis current is increased accordingly, so that the combined current vector of the adjusted target value and the adjusted initial value can maintain the output of the electromagnetic torque required for production.

[0028] Generate current vector control commands based on the combined current vectors;

[0029] The step of adjusting the motor operating state by driving the frequency converter based on the current vector control command includes:

[0030] The frequency converter generates voltage commands corresponding to the d-axis and q-axis respectively through the d-axis and q-axis current closed-loop regulators according to the current vector control command.

[0031] By combining the real-time position information of the motor rotor, the voltage command is transformed from the rotating coordinate system to the stationary coordinate system to obtain a two-phase voltage reference signal;

[0032] Based on the two-phase voltage reference signal, space vector pulse width modulation is performed to generate a three-phase PWM drive signal;

[0033] The motor's operating state is adjusted by regulating the three-phase current of the motor based on the three-phase PWM drive signal.

[0034] In one implementation of this application, the method includes:

[0035] The measured temperature of the motor casing is obtained based on the variable frequency drive, and the predicted motor casing temperature at the same time is obtained from the thermal network model.

[0036] Calculate the difference between the measured temperature of the housing and the predicted temperature of the motor housing;

[0037] When the absolute value of the difference exceeds the preset tolerance limit, the thermal resistance between the stator core and the motor housing in the thermal network model, as well as the convective heat transfer resistance between the motor housing and the well fluid, are adjusted according to the difference.

[0038] In one implementation of this application, the method further includes:

[0039] After each correction of the thermal conductivity resistance parameter and the convective heat transfer resistance, the first relative rate of change between the corrected thermal conductivity resistance and the original thermal conductivity resistance is calculated.

[0040] When the first relative rate of change exceeds the preset rate of change threshold, a motor insulation aging warning signal is generated and uploaded to the monitoring system.

[0041] In one implementation of this application, the method further includes:

[0042] Calculate the second relative rate of change between the corrected convective heat transfer resistance and the original convective heat transfer resistance;

[0043] The first relative rate of change and the second relative rate of change are weighted and fused to obtain the health index of the thermal network model. The health index characterizes the degree of comprehensive degradation of the internal thermal conductivity and external heat dissipation capacity of the motor.

[0044] When the health index exceeds the preset health indicator threshold, a motor maintenance warning signal is generated and uploaded to the monitoring system;

[0045] Obtain the trend of the health index changes within a preset time period;

[0046] If the health index shows an upward trend, the safe temperature threshold is dynamically lowered, and the preset rate of change threshold is reduced, so that the winding hot spot temperature comparison and insulation aging early warning judgment are performed based on the adjusted safe temperature threshold and the preset rate of change threshold.

[0047] Secondly, embodiments of this application also provide a motor temperature control device based on a digital twin thermal model. The device includes: at least one processor; and a memory communicatively connected to the at least one processor. The memory stores instructions executable by the at least one processor, which, when executed, enable the at least one processor to: construct a lumped-parameter thermal network model containing three temperature nodes: windings, stator core, and motor housing; collect downhole operating data of an electric submersible screw pump motor in real time, the downhole operating data including at least motor electrical parameters, motor housing temperature, and wellbore fluid properties; dynamically update the convective heat transfer boundary conditions of the thermal network model based on the wellbore fluid properties and the motor housing temperature; calculate motor losses based on the motor electrical parameters and input the motor losses as a heat source into the updated thermal network model to obtain the average winding temperature; reconstruct the winding hot spot temperature by combining the average winding temperature with local loss distribution information; and, while maintaining a constant output electromagnetic torque of the motor, generate a current vector control command based on the winding hot spot temperature, and drive a frequency converter driver to adjust the motor's operating state based on the current vector control command to suppress temperature rise.

[0048] Thirdly, this application also provides a non-volatile computer storage medium for motor temperature control based on a digital twin thermal model, storing computer-executable instructions. These instructions are configured to: construct a lumped-parameter thermal network model including three temperature nodes: windings, stator core, and motor casing; collect downhole operating data of the electric submersible screw pump motor in real time, the downhole operating data including at least motor electrical parameters, motor casing temperature, and wellbore fluid properties; dynamically update the convective heat transfer boundary conditions of the thermal network model based on the wellbore fluid properties and the motor casing temperature; calculate motor losses based on the motor electrical parameters and input the motor losses as a heat source into the updated thermal network model to obtain the winding average temperature; reconstruct the winding hot spot temperature by combining the winding average temperature with local loss distribution information; and, while maintaining a constant output electromagnetic torque of the motor, generate a current vector control command based on the winding hot spot temperature, and drive the frequency converter driver to adjust the motor's operating state based on the current vector control command to suppress temperature rise.

[0049] This application provides a motor temperature control method, device, and medium based on a digital twin thermal model, which has the following beneficial effects: By constructing a lumped-parameter thermal network model containing three nodes—winding, stator core, and motor casing—and combining real-time collected motor electrical parameters, casing temperature, and wellbore fluid properties, the convective heat transfer boundary conditions are dynamically updated, and motor losses are accurately calculated as the heat source, thereby reconstructing the winding hotspot temperature that reflects the insulation safety limit. Based on this, while maintaining a constant electromagnetic torque, a current vector control command is generated using the hotspot temperature as feedback to drive the frequency converter to actively adjust the motor's operating state. This effectively solves the problems of model inaccuracies caused by using fixed thermal boundaries in traditional temperature control technologies, the coarse control resulting from relying solely on casing temperature, and the passive response defects of lacking adaptive adjustment capabilities under operating conditions. It significantly improves the accuracy of downhole motor temperature rise prediction, realizing a shift from average temperature control to precise hotspot suppression. While ensuring stable production output, it effectively delays insulation aging, avoids overheating shutdowns, and improves the operational safety and reliability of electric submersible screw pump systems in complex downhole environments. Attached Figure Description

[0050] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0051] Figure 1 A flowchart of a motor temperature control method based on a digital twin thermal model is provided for an embodiment of this application;

[0052] Figure 2 This is a schematic diagram of the internal structure of a motor temperature control device based on a digital twin thermal model, provided as an embodiment of this application. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0054] This application provides a method, device, and medium for motor temperature control based on a digital twin thermal model, in order to solve the following technical problems: the problems of model inaccuracy, coarse control, and lack of adaptive capability in existing temperature control technologies.

[0055] The technical solutions proposed in the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0056] Figure 1 This document provides a flowchart of a motor temperature control method based on a digital twin thermal model, as illustrated in an embodiment of this application. Figure 1 As shown in the figure, the motor temperature control method based on a digital twin thermal model provided in this application specifically includes the following steps:

[0057] Step 101: Construct a lumped parameter thermal network model containing three temperature nodes: windings, stator core, and motor housing;

[0058] In some embodiments, based on the physical structure of the electric submersible screw pump motor, a lumped-parameter thermal network model is established, consisting of three key thermal nodes: the winding (heat source core area), the stator core (thermal conduction medium), and the motor casing (heat exchange interface with the wellbore fluid). The nodes are connected by thermal resistance (such as the thermal resistance of the insulation material between the winding and the core, and the structural thermal resistance between the core and the casing) and heat capacity (reflecting the heat storage capacity of the node), forming an equivalent thermal path. This model, based on the heat conduction equation, simplifies the complex three-dimensional thermal field into a set of low-order differential equations that can be solved in real time, balancing computational efficiency and physical interpretability. This thermal network model is a type of digital twin thermal model.

[0059] Step 102: Real-time acquisition of downhole operating data of the electric submersible screw pump motor, the downhole operating data including at least the motor electrical parameters, motor housing temperature and wellbore fluid physical properties;

[0060] In some embodiments, the following data are synchronously acquired through downhole sensors and the built-in monitoring module of the variable frequency drive: electrical parameters such as motor three-phase current, voltage, and power factor; the casing temperature measured by a temperature sensor installed on the surface of the motor casing; and wellbore fluid properties obtained through downhole multi-parameter instruments or surface inversion, including fluid density, specific heat capacity, thermal conductivity, viscosity, and flow velocity. All data are uploaded to the control unit via a downhole communication link at a frequency of not less than 1 Hz, providing real-time input for thermal model updates and control decisions.

[0061] Step 103: Dynamically update the convective heat transfer boundary conditions of the thermal network model based on the fluid properties of the wellbore and the temperature of the motor casing;

[0062] In some embodiments, the convective heat transfer coefficient between the casing and the wellbore fluid is calculated in real time using the collected wellbore fluid properties and the motor housing geometry, employing empirical correlations (such as the Dittus-Boelter formula or a Nusselt number model modified for the annular space). h ; will h Substituting the values ​​into the boundary conditions of the motor casing nodes in the thermal network model, i.e., the heat flow... q = h A ( T 外壳 T 流体 ),in A For heat exchange area, T 流体 For the wellbore fluid temperature, T 外壳 This refers to the temperature of the motor casing. This mechanism enables the thermal model to adapt to fluctuations in cooling capacity caused by changes in operating conditions such as well depth, production rate, and water cut.

[0063] Step 104: Calculate the motor loss based on the motor electrical parameters, and use the motor loss as a heat source input to update the thermal network model after convection heat transfer boundary conditions to obtain the winding average temperature;

[0064] In some embodiments, copper losses are calculated based on real-time current, voltage, and motor parameters. P Cu =3 I 2 R Considering skin effect correction and iron loss (based on voltage frequency and magnetic flux density amplitude, using the Bertotti model); the total loss is distributed to the winding nodes as a concentrated heat source input; the updated convective heat transfer boundary conditions and the heat source are substituted into the thermal network differential equation, and the steady-state or transient temperature distribution is solved by numerical integration (such as the fourth-order Runge-Kutta method), outputting the average winding temperature. T avg .

[0065] Step 105: Combine the average winding temperature and local loss distribution information to reconstruct the winding hot spot temperature;

[0066] In some embodiments, based on the distribution law of copper loss in the winding slot (such as end concentration, high temperature at the bottom of the slot, etc.) obtained in advance by motor electromagnetic design data or finite element simulation, a hot spot temperature rise correction coefficient Δ is established. T hotspot = k ( T avg T 环境 ),in k >1 is the hotspot factor; the final reconstructed winding hotspot temperature is T hot= T avg +Δ T hotspot This method avoids the engineering difficulties of directly measuring hot spots, and at the same time reflects the actual thermal stress of the insulating material better than a single average temperature.

[0067] Step 106: While maintaining the constant output electromagnetic torque of the motor, a current vector control command is generated based on the hot spot temperature of the winding, and the frequency converter is driven to adjust the operating state of the motor based on the current vector control command to suppress temperature rise.

[0068] In some embodiments, within a vector control framework, the q-axis current is maintained. i q To meet the electromagnetic torque required by the current load ( T e ∝ i q When detected T hot When the temperature approaches a preset safety threshold (e.g., 155°C), moderately increase the d-axis demagnetizing current. i d (Negative adjustment) to reduce the total effective current and copper loss; generate new ( i d , i q The command is sent to the current loop controller of the frequency converter driver; the driver adjusts the inverter PWM output accordingly to achieve adaptive operation with "constant torque and suppressed temperature rise", effectively preventing overheating faults and extending motor life.

[0069] This application provides a motor temperature control method based on a digital twin thermal model, which has the following advantages: By constructing a lumped-parameter thermal network model containing three nodes—windings, stator core, and motor casing—and combining real-time collected motor electrical parameters, casing temperature, and wellbore fluid properties, the convective heat transfer boundary conditions are dynamically updated, and motor losses are accurately calculated as the heat source, thereby reconstructing the winding hotspot temperature that reflects the insulation safety limit. Based on this, while maintaining a constant electromagnetic torque, a current vector control command is generated using the hotspot temperature as feedback to drive the frequency converter to actively adjust the motor's operating state. This effectively solves the problems of model inaccuracies caused by using fixed thermal boundaries in traditional temperature control technologies, the coarse control resulting from relying solely on casing temperature, and the passive response defects of lacking adaptive adjustment capabilities under operating conditions. It significantly improves the accuracy of downhole motor temperature rise prediction, realizing a shift from average temperature control to precise hotspot suppression. While ensuring stable production output, it effectively delays insulation aging, avoids overheating shutdowns, and improves the operational safety and reliability of electric submersible screw pump systems in complex downhole environments.

[0070] As a refinement of the above embodiments, the wellbore fluid properties include crude oil viscosity, sand concentration, fluid velocity, and gas content. When performing step 103, which involves dynamically updating the convective heat transfer boundary conditions of the thermal network model based on the wellbore fluid properties and the motor casing temperature, the following implementation methods can also be used, but are not limited to: inputting the wellbore fluid properties and the motor casing temperature into a preset heat dissipation capacity mapping relationship to obtain the heat exchange parameters between the motor casing and the well fluid in the thermal network model; and dynamically updating the convective heat transfer boundary conditions of the thermal network model based on the heat exchange parameters. The well fluid is the wellbore fluid.

[0071] In some embodiments, real-time acquired wellbore fluid properties are input to a preset heat dissipation capacity mapping relationship. These wellbore fluid properties include crude oil viscosity (e.g., 50 mPa·s), sand concentration (e.g., 2% volume fraction), fluid velocity (e.g., 0.8 m / s), and fluid gas content (e.g., 5% volume fraction). This mapping relationship can be established through offline CFD multiphase flow simulation or downhole measured data calibration, and is expressed as a convective heat transfer coefficient. h = f ( μ , C s , v , G ),in μ For crude oil viscosity, C s The sand concentration, v For fluid velocity, G This refers to the gas content of the fluid. For example, as crude oil viscosity increases, fluid flowability decreases, leading to weakened convective heat transfer; while increased sand concentration can improve the effective thermal conductivity of the fluid, it may weaken the heat transfer effect due to particle deposition or increased flow resistance; increased fluid velocity significantly enhances forced convective heat transfer; and increased gas content will significantly reduce the overall heat transfer efficiency because the thermal conductivity of gas is much lower than that of liquid. The above-mentioned multi-factor coupling effects are uniformly quantified through the mapping function. Based on this function, the heat exchange parameters under the current operating conditions are calculated. h (like h =95W / (m 2 K). Simultaneously, combining the motor installation depth (e.g., 1500 meters) and the well temperature gradient model (e.g., a geothermal gradient of 3°C / 100m), the wellbore fluid temperature at the motor location is determined. T 井液 (e.g., 65°C); combined with the real-time collected motor casing temperature T 外壳 (e.g., 85°C), will h Substituting the temperature difference into the convective boundary conditions of the motor casing nodes in the thermal network model, i.e., the heat flow equation...q = hA ( T 外壳 T 井液 This method dynamically updates the convective heat transfer boundary conditions. By simultaneously considering fluid properties (gas content, sand content, viscosity, and flow velocity) and the spatially related well temperature distribution, the thermal model can accurately reflect changes in complex downhole cooling environments such as high viscosity, sand content, gas content, and low flow velocity. This provides a high-fidelity digital twin foundation for winding hotspot temperature reconstruction and temperature control strategies. Furthermore, it enables dynamic updates of the convective heat transfer boundary conditions, allowing the thermal model to accurately reflect changes in complex downhole cooling environments such as high viscosity, sand content, and low flow velocity, providing a reliable basis for subsequent temperature rise prediction and control.

[0072] As a refinement of the above embodiments, before reconstructing the winding hot spot temperature by combining the winding average temperature and local loss distribution information, the method may also adopt, but is not limited to, the following implementation methods, for example: performing spectral decomposition on the three-phase current signal in the motor electrical parameters, and extracting the fifth harmonic amplitude and seventh harmonic amplitude from the decomposition result; comparing the fifth harmonic amplitude with the fifth harmonic reference amplitude under normal operating conditions, and comparing the seventh harmonic amplitude with the seventh harmonic reference amplitude under normal operating conditions; if any harmonic amplitude exceeds a preset multiple of the corresponding reference amplitude, a local loss enhancement indication signal is generated, the local loss enhancement indication signal being used to characterize the trend of non-uniform heating aggravation in the winding; according to the local loss enhancement indication signal, querying the corresponding local loss distribution information from a preset hot spot temperature rise mapping table.

[0073] To facilitate understanding of the above embodiments, this application provides an exemplary description, including: During downhole operation, the system performs a Fast Fourier Transform (FFT) on the acquired three-phase current signal, extracting a fifth harmonic amplitude of 1.8A and a seventh harmonic amplitude of 1.2A; while the reference amplitudes of the fifth and seventh harmonics of the motor under normal operating conditions are 0.5A and 0.4A, respectively (obtained through statistical analysis of historical steady-state operating data). Setting a preset multiplier of 3, the fifth harmonic threshold is 1.5A, and the seventh harmonic threshold is 1.2A. Since the fifth harmonic amplitude (1.8A) exceeds 1.5A, the system determines that there is an aggravated harmonic distortion phenomenon and generates a "local loss enhancement indication signal." This signal indicates that due to the skin effect of harmonic current or uneven end leakage flux in the winding, copper losses in certain areas within the slot may significantly increase. Subsequently, based on the indication signal, the control system queries the corresponding local loss distribution pattern from the preset hot spot temperature rise mapping table. For example, when the fifth harmonic dominates, hot spots are more likely to appear at the winding end near the lead area, and the corresponding hot spot factor increases from the conventional 1.15 to 1.25. Finally, this corrected local loss distribution information is used for hot spot temperature reconstruction in step 105, enabling the temperature control strategy to respond in advance to potential local overheating risks and improving the foresight and accuracy of insulation protection.

[0074] The hotspot temperature rise mapping table is pre-constructed through a combination of offline multiphysics simulation and experimental calibration: First, based on the actual structure and material parameters of the motor, a high-fidelity electromagnetic-thermal coupling model is established using software such as ANSYS Maxwell / Fluent or JMAG; then, typical harmonic currents of different amplitudes and combinations, such as the fifth and seventh harmonics, are applied in the simulation to calculate the spatial distribution of winding copper losses and solve the steady-state temperature field to extract the hotspot temperature rise (i.e., the difference ΔT between the hotspot temperature and the average temperature). hot =T hot T av9 Furthermore, the harmonic type (such as "5th dominant" or "7th + 5th superposition") and its multiple relative to normal operating conditions are used as input indices, corresponding to the hot spot temperature rise or hot spot factor (k=T). hot / T av9 As output, a structured mapping table is constructed; finally, key operating conditions are verified and corrected through prototype temperature rise tests (such as fiber optic temperature measurement) to ensure engineering reliability. During operation, when the system detects harmonic anomalies and generates a "local loss enhancement indication signal", the mapping table can be queried in real time to obtain the matching local loss distribution pattern and temperature rise compensation amount. Thus, the hot spot temperature can be reconstructed with high accuracy based on only the average winding temperature. This transforms the complex three-dimensional non-uniform heating problem into a lightweight, table-based online model, significantly improving the adaptability, accuracy, and insulation safety of the temperature control system, and effectively preventing local overheating failures caused by harmonics.

[0075] As a refinement of the above embodiments, when reconstructing the winding hot spot temperature by combining the winding average temperature and local loss distribution information in step 105, the following implementation methods can also be adopted, for example: adding the winding average temperature to the basic hot spot temperature rise offset to obtain the intermediate temperature corresponding to the hottest part of the winding; determining an additional temperature rise compensation value based on the local loss distribution information; adding the additional temperature rise compensation value to the intermediate temperature to obtain the winding hot spot temperature.

[0076] For example, the current average winding temperature is calculated to be 95°C using a thermal network model. Based on motor design data or historical steady-state operating data, a baseline hot spot temperature rise offset of 10°C is set (reflecting the inherent temperature difference caused by uneven structure and heat dissipation under normal load). Therefore, the initial intermediate temperature is 95°C + 10°C = 105°C. Further, if the amplitude of the fifth harmonic current exceeds four times the baseline value during the process, the system generates a "local loss enhancement indication signal" and retrieves the corresponding additional temperature rise compensation value of 8°C (reflecting the additional non-uniform copper loss caused by harmonics) from the hot spot temperature rise mapping table. Finally, this additional compensation value is superimposed on the intermediate temperature, resulting in a reconstructed winding hot spot temperature of 105°C + 8°C = 113°C. This result more realistically reflects the thermal stress state of the weakest insulation region, providing a reliable basis for subsequent vector control based on hot spot temperature and effectively avoiding temperature control lag or overheating risks caused by ignoring the influence of dynamic harmonics. It should be understood that the above parameters are only for better illustrating the execution process and do not constitute a limitation of this disclosure.

[0077] As a refinement of the above embodiment, when executing step 106 to generate the current vector control command based on the winding hot spot temperature, the following implementation methods can also be adopted, but are not limited to: determining the initial value of the q-axis current according to the electromagnetic torque required for production; comparing the winding hot spot temperature with a safe temperature threshold; if the winding hot spot temperature is higher than the safe temperature threshold, decreasing the initial value and correspondingly increasing the target value of the d-axis current, so that the combined current vector of the adjusted target value and the adjusted initial value can maintain the output of the electromagnetic torque required for production; generating a current vector control command based on the combined current vector; the step of driving the frequency converter to adjust the motor operating state based on the current vector control command includes: the frequency converter generating voltage commands corresponding to the d-axis and q-axis respectively through the d-axis and q-axis current closed-loop regulators according to the current vector control command; combining the real-time position information of the motor rotor, converting the voltage commands from the rotating coordinate system to the stationary coordinate system to obtain a two-phase voltage reference signal; performing space vector pulse width modulation based on the two-phase voltage reference signal to generate a three-phase PWM drive signal; adjusting the three-phase current of the motor operation based on the three-phase PWM drive signal to adjust the motor operating state.

[0078] For example, current production requirements dictate that the motor output electromagnetic torque is 120 N·m. Based on the motor parameters and vector control relationships, the initial value of the q-axis current is preliminarily determined. i q0 =40A, while the d-axis current is initially set to 0A (normal maximum torque / current control mode); at this time, the winding hot spot temperature obtained by system reconfiguration is 162°C, exceeding the preset safe temperature threshold (such as 155°C for Class F insulation); in order to suppress the temperature rise while maintaining constant torque, the controller adjusts the current vector according to the following strategy: i q The current is reduced from 40A to 38A, while a negative d-axis current (demagnetizing direction) is introduced. i d = 6A; because the electromagnetic torque is approximately proportional to i q The product of the air gap flux and the air gap flux, which varies with | i d The increase in current vectors slightly decreases the overall current vector, therefore, accurate motor modeling or table lookup compensation is required to ensure the adjusted combined current vector is within acceptable limits. i d = 6A, i qThe actual torque generated by the 38A current vector remained stable at 120 N·m. Ultimately, this combined current vector was encapsulated into a current vector control command and sent to the inner loop controller of the inverter driver to drive the inverter to adjust the PWM output. Although the total effective current increased slightly, due to copper losses and i 2 With the q-axis current being proportional to the torque and the dq current being the dominant current, a reasonable allocation of the dq current can significantly reduce the winding thermal load without sacrificing production efficiency, thus achieving adaptive operation with "constant torque and controlled temperature rise".

[0079] In some embodiments, the method may also be implemented in, but is not limited to, the following ways: obtaining the measured temperature of the motor casing based on the variable frequency drive, and obtaining the predicted motor casing temperature at the same time from the thermal network model; calculating the difference between the measured casing temperature and the predicted motor casing temperature; when the absolute value of the difference exceeds a preset tolerance limit, adjusting the thermal resistance between the stator core and the motor casing in the thermal network model, and the convective heat transfer resistance between the motor casing and the well fluid, according to the difference.

[0080] For example, in actual operation, the measured temperature of the motor casing is obtained through the temperature acquisition module built into the frequency converter driver, and the predicted casing temperature at the same time is extracted from the thermal network model. For instance, under a certain operating condition, the measured casing temperature is 82°C, while the predicted value of the thermal network model is 76°C, a difference of +6°C. If the preset tolerance limit is ±3°C, this deviation exceeds the allowable range, indicating that the parameters of the current thermal network model (such as thermal conductivity or convective heat transfer conditions) are mismatched with the actual downhole environment. At this time, the system activates the online correction mechanism: based on the positive deviation (measured > predicted), it is determined that the actual heat dissipation capacity is weaker than the model expectation, which may be due to a decrease in the well fluid flow rate or scaling leading to deterioration of heat transfer; therefore, the controller increases the convective heat transfer resistance between the motor casing and the well fluid proportionally (i.e., effectively reduces the convective heat transfer coefficient), while moderately increasing the thermal conductivity resistance between the stator core and the casing (reflecting that the structural contact thermal resistance may increase due to thermal expansion or deposits); the adjusted thermal resistance parameters are updated to the thermal network model and used for subsequent temperature extrapolation. After 1–2 control cycles, the predicted shell temperature gradually converges to the measured value (e.g., from 76°C to 81°C), and the model accuracy is restored, thus ensuring the reliability of the winding hot spot temperature reconstruction and forming a "sensing, correction, prediction" closed loop, which significantly improves the adaptability and robustness of the thermal management system.

[0081] In some embodiments, the method may also be implemented in, but is not limited to, the following ways: after each correction of the thermal conductivity resistance parameter and the convective heat transfer resistance, a first relative rate of change between the corrected thermal conductivity resistance and the original thermal conductivity resistance is calculated; when the first relative rate of change exceeds a preset rate of change threshold, a motor insulation aging warning signal is generated and uploaded to the monitoring system.

[0082] For example, to achieve proactive assessment of the motor's health status, aging trend judgment can be performed after each online correction of the thermal network model parameters. For instance, during thermal resistance correction, the thermal resistance between the stator core and the motor housing is updated from the original value of 0.15 K / W to 0.21 K / W; its first relative change rate is calculated as |(0.21) / W. 0.15) / 0.15∣=40%; if the preset change rate threshold is 30%, then this change rate has exceeded the limit. Since a significant increase in thermal resistance usually reflects aging phenomena such as insulation material deterioration, slot wedge loosening, or air gap carbon buildup inside the motor, hindering the effective conduction of heat from the windings to the casing, the system immediately generates a "motor insulation aging warning signal" and uploads it along with current operating condition information (such as running time, cumulative temperature rise, and thermal resistance change trend) to the ground equipment health management platform. Maintenance personnel can then arrange preventative maintenance based on this, such as early inspection or adjustment of load strategies, to avoid sudden shutdowns caused by accelerated insulation failure due to heat accumulation, thereby upgrading the temperature control system from passive heat dissipation to an active health sensing and early warning mechanism.

[0083] In some embodiments, the method may also employ, but is not limited to, the following implementations: calculating a second relative rate of change between the corrected convective heat transfer resistance and the original convective heat transfer resistance; performing a weighted fusion calculation on the first relative rate of change and the second relative rate of change to obtain a health index of the thermal network model, wherein the health index characterizes the comprehensive degradation degree of the motor's internal thermal conductivity and external heat dissipation capacity; when the health index exceeds a preset health indicator threshold, generating a motor maintenance warning signal and uploading it to the monitoring system; acquiring the trend of the health index change within a preset time period; if the health index shows an upward trend, dynamically lowering the safety temperature threshold and reducing the preset rate of change threshold, so as to perform the winding hot spot temperature comparison and insulation aging warning judgment based on the adjusted safety temperature threshold and the preset rate of change threshold.

[0084] For example, multi-dimensional thermal resistance variation information is further integrated to achieve a more refined health assessment. For instance, during downhole operation, after model correction, the thermal resistance between the stator core and the casing increases from 0.15 K / W to 0.21 K / W (first relative change rate of 40%), while the convective heat transfer resistance between the motor casing and the well fluid increases from 0.30 K / W to 0.42 K / W (second relative change rate of 40%). Considering that internal thermal conductivity degradation has a more critical impact on insulation life, the system performs a weighted fusion of the two (e.g., weights of 0.7 and 0.3 respectively), calculating the health index of the thermal network model as 0.7 × 40% + 0.3 × 40% = 40%. This index comprehensively characterizes the degree of synergistic degradation between the decline in internal thermal conductivity and the weakening of external heat dissipation capacity of the motor. If the preset health index threshold is 35%, the current health index has exceeded the limit, and the system generates a "motor maintenance warning signal" and uploads it to the monitoring platform. Furthermore, the system analyzed the trend of the health index over the past 7 days and found that it was continuously rising (from 25% to 32% to 40%), indicating that the motor's thermal performance was in an accelerated degradation phase. Subsequently, the safe temperature threshold was dynamically lowered (e.g., from 155°C to 148°C) and the preset rate of change threshold was reduced (e.g., from 30% to 25%). Afterward, the winding hot spot temperature comparison and insulation aging warning were both executed based on the adjusted, more stringent thresholds, thereby triggering protection actions earlier and improving the system's operational safety and warning sensitivity under aging conditions.

[0085] In some embodiments, the method of this application may also be implemented in, but is not limited to, the following ways: constructing a lumped-parameter thermal network model containing three heat capacity nodes: winding, stator core, and motor housing, and initializing the model parameters; real-time acquisition of downhole operating data, including motor three-phase current, DC bus voltage, motor housing temperature, wellbore fluid viscosity, sand concentration, and fluid velocity; dynamically calculating the convective heat transfer coefficient between the motor housing and the well fluid based on the wellbore fluid viscosity, sand concentration, and fluid velocity, and updating the boundary conditions of the thermal network model; calculating copper loss and iron loss based on the three-phase current and DC bus voltage, and inputting them as heat source terms into the thermal network model to solve for the average winding temperature, stator core temperature, and motor housing temperature; and modifying the model based on the average winding temperature and local copper loss distribution by adjusting the hot spot temperature. The model reconstructs the winding hot spot temperature, and the local copper loss distribution is obtained by current harmonic characteristic identification. With the constraint of maintaining constant electromagnetic torque, a multi-objective optimization problem is constructed to minimize copper loss and thermal stress accumulation, and the optimal d-axis and q-axis current commands are obtained by solving the problem. These optimal d-axis and q-axis current commands are sent to the vector frequency converter to adjust the current vector output of the submersible motor, thereby actively suppressing the winding hot spot temperature. Based on the residual between the measured and model-predicted motor casing temperatures, an extended Kalman filter algorithm is used to identify the convective heat transfer resistance online every 5 minutes, and a recursive least squares algorithm is used to identify the winding-core thermal conductivity resistance and winding thermal capacity online every 24 hours. The thermal model health index is calculated based on the updated thermal resistance parameters. When the thermal model health index exceeds a preset threshold, a maintenance warning signal is generated.

[0086] The above are embodiments of the method proposed in this application. Based on the same inventive concept, embodiments of this application also provide a motor temperature control device based on a digital twin thermal model, the structure of which is as follows: Figure 2 As shown.

[0087] Figure 2 This is a schematic diagram of the internal structure of a motor temperature control device based on a digital twin thermal model, provided as an embodiment of this application. Figure 2 As shown, the device includes:

[0088] At least one processor 201;

[0089] And a memory 202 that is communicatively connected to at least one processor;

[0090] The memory 202 stores instructions executable by at least one processor 201, which in turn executes the instructions to enable the processor 201 to: construct a lumped-parameter thermal network model comprising three temperature nodes: windings, stator core, and motor housing; acquire downhole operating data of the electric submersible screw pump motor in real time, the downhole operating data including at least motor electrical parameters, motor housing temperature, and wellbore fluid properties; dynamically update the convective heat transfer boundary conditions of the thermal network model based on the wellbore fluid properties and the motor housing temperature; calculate motor losses based on the motor electrical parameters and input the motor losses as a heat source into the updated thermal network model to obtain the winding average temperature; reconstruct the winding hot spot temperature by combining the winding average temperature with local loss distribution information; and, while maintaining a constant output electromagnetic torque of the motor, generate a current vector control command based on the winding hot spot temperature, and drive the frequency converter driver to adjust the motor operating state based on the current vector control command to suppress temperature rise.

[0091] Some embodiments of this application provide corresponding to Figure 1 A non-volatile computer storage medium for motor temperature control based on a digital twin thermal model stores computer-executable instructions. These instructions are configured to: construct a lumped-parameter thermal network model comprising three temperature nodes: windings, stator core, and motor casing; collect downhole operating data of an electric submersible screw pump motor in real time, including at least motor electrical parameters, motor casing temperature, and wellbore fluid properties; dynamically update the convective heat transfer boundary conditions of the thermal network model based on the wellbore fluid properties and motor casing temperature; calculate motor losses based on the motor electrical parameters and input the motor losses as a heat source into the updated thermal network model to obtain the winding average temperature; reconstruct the winding hot spot temperature by combining the winding average temperature with local loss distribution information; and, while maintaining a constant output electromagnetic torque of the motor, generate a current vector control command based on the winding hot spot temperature, and drive a frequency converter driver to adjust the motor's operating state to suppress temperature rise.

[0092] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments for IoT devices and media are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0093] The systems, media, and methods provided in this application are one-to-one correspondences. Therefore, the systems and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the systems and media will not be repeated here.

[0094] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0095] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0096] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0097] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0098] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0099] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0100] Computer-readable media include both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0101] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0102] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A motor temperature control method based on a digital twin thermal model, characterized in that, The method includes: Construct a lumped parameter thermal network model that includes three temperature nodes: windings, stator core, and motor housing; Real-time acquisition of downhole operating data of the electric submersible screw pump motor, the downhole operating data including at least the motor electrical parameters, motor housing temperature and wellbore fluid physical properties; The convective heat transfer boundary conditions of the thermal network model are dynamically updated based on the fluid properties of the wellbore and the temperature of the motor casing. The motor loss is calculated based on the motor electrical parameters, and the motor loss is used as the heat source input convection heat transfer boundary condition to update the thermal network model, thereby obtaining the winding average temperature. By combining the average winding temperature and local loss distribution information, the winding hot spot temperature is reconstructed; While maintaining a constant output electromagnetic torque of the motor, a current vector control command is generated based on the hot spot temperature of the winding, and the frequency converter is driven to adjust the operating state of the motor based on the current vector control command in order to suppress temperature rise. Before reconstructing the winding hot spot temperature by combining the winding average temperature and local loss distribution information, the method includes: The three-phase current signal in the motor electrical parameters is subjected to spectral decomposition, and the fifth and seventh harmonic amplitudes are extracted from the decomposition results. The amplitude of the fifth harmonic is compared with the reference amplitude of the fifth harmonic under normal operating conditions, and the amplitude of the seventh harmonic is compared with the reference amplitude of the seventh harmonic under normal operating conditions. If the amplitude of any harmonic exceeds a preset multiple of the corresponding reference amplitude, a local loss enhancement indication signal is generated. The local loss enhancement indication signal is used to characterize the trend of non-uniform heating aggravation in the winding. Based on the local loss enhancement indication signal, the corresponding local loss distribution information is queried from the preset hot spot temperature rise mapping table; The process of reconstructing the winding hot spot temperature by combining the average winding temperature and local loss distribution information includes: The average temperature of the winding is added to the temperature rise offset of the basic hot spot to obtain the intermediate temperature corresponding to the hottest part of the winding. The additional temperature rise compensation value is determined based on the local loss distribution information. The additional temperature rise compensation value is added to the intermediate temperature to obtain the winding hot spot temperature.

2. The motor temperature control method based on a digital twin thermal model according to claim 1, characterized in that, The fluid properties in the wellbore include crude oil viscosity, sand concentration, fluid velocity, and gas content. The convective heat transfer boundary conditions of the thermal network model dynamically updated based on the wellbore fluid properties and the motor casing temperature include: By inputting the physical property parameters of the well fluid and the temperature of the motor housing into a preset heat dissipation capacity mapping relationship, the heat exchange parameters between the motor housing and the well fluid in the thermal network model are obtained. The convective heat transfer boundary conditions of the thermal network model are dynamically updated based on the heat exchange parameters.

3. The motor temperature control method based on a digital twin thermal model according to claim 1, characterized in that, The current vector control command generated based on the winding hot spot temperature includes: The initial value of the q-axis current is determined based on the electromagnetic torque required for production. Compare the winding hot spot temperature with the safe temperature threshold. If the hot spot temperature of the winding is higher than the safe temperature threshold, then the initial value is reduced and the target value of the d-axis current is increased accordingly, so that the combined current vector of the adjusted target value and the adjusted initial value can maintain the output of the electromagnetic torque required for production. Generate current vector control commands based on the combined current vectors; The step of adjusting the motor operating state by driving the frequency converter based on the current vector control command includes: The frequency converter generates voltage commands corresponding to the d-axis and q-axis respectively through the d-axis and q-axis current closed-loop regulators according to the current vector control command. By combining the real-time position information of the motor rotor, the voltage command is transformed from the rotating coordinate system to the stationary coordinate system to obtain a two-phase voltage reference signal; Based on the two-phase voltage reference signal, space vector pulse width modulation is performed to generate a three-phase PWM drive signal; The motor's operating state is adjusted by regulating the three-phase current of the motor based on the three-phase PWM drive signal.

4. A motor temperature control method based on a digital twin thermal model according to any one of claims 1-3, characterized in that, The method includes: The measured temperature of the motor casing is obtained based on the variable frequency drive, and the predicted motor casing temperature at the same time is obtained from the thermal network model. Calculate the difference between the measured temperature of the housing and the predicted temperature of the motor housing; When the absolute value of the difference exceeds the preset tolerance limit, the thermal resistance between the stator core and the motor housing in the thermal network model, as well as the convective heat transfer resistance between the motor housing and the well fluid, are adjusted according to the difference.

5. A motor temperature control method based on a digital twin thermal model according to claim 4, characterized in that, The method further includes: After each correction of the thermal conductivity resistance parameter and the convective heat transfer resistance, the first relative rate of change between the corrected thermal conductivity resistance and the original thermal conductivity resistance is calculated. When the first relative rate of change exceeds the preset rate of change threshold, a motor insulation aging warning signal is generated and uploaded to the monitoring system.

6. The motor temperature control method based on a digital twin thermal model according to claim 5, characterized in that, The method further includes: Calculate the second relative rate of change between the corrected convective heat transfer resistance and the original convective heat transfer resistance; The first relative rate of change and the second relative rate of change are weighted and fused to obtain the health index of the thermal network model. The health index characterizes the degree of comprehensive degradation of the internal thermal conductivity and external heat dissipation capacity of the motor. When the health index exceeds the preset health indicator threshold, a motor maintenance warning signal is generated and uploaded to the monitoring system; Obtain the trend of the health index changes within a preset time period; If the health index shows an upward trend, the safe temperature threshold is dynamically lowered, and the preset rate of change threshold is reduced, so that the winding hot spot temperature comparison and insulation aging early warning judgment are performed based on the adjusted safe temperature threshold and the preset rate of change threshold.

7. A motor temperature control device based on a digital twin thermal model, characterized in that, The device includes: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform a motor temperature control method based on a digital twin thermal model as described in any one of claims 1-6.

8. A computer storage medium storing computer-executable instructions, characterized in that, When the computer-executable instructions are executed, they implement a motor temperature control method based on a digital twin thermal model as described in any one of claims 1-6.

Images