A parameter identification based speed regulation method and system

Abstract

This invention discloses a speed regulation method and system based on parameter identification, belonging to the field of motor parameter identification technology. It aims to solve the problem that conventional online parameter identification mechanisms use fixed convergence gains, which deviate from underlying thermodynamic laws, leading to oscillating and unstable speed estimation. The method includes: using the thermodynamic law of conservation of energy to determine the highest possible physical rate of change of the motor stator temperature; setting the physical speed limit boundary of the online parameter identification mechanism; obtaining a dynamic convergence gain with physical boundary awareness; constructing a sliding mode observer with an adaptive identification interface; performing stator resistance parameter identification through the received dynamic convergence gain; and resolving the actual rotor speed of the synchronous motor from the obtained stator resistance identification value to adjust the output power of the circulating water pump. This invention suppresses the oscillation and instability of the stator resistance identification value, ensuring the accuracy of speed analysis of the synchronous motor and the stability of the circulating water pump output power under high-temperature alternating conditions.

Description

Technical Field

[0001] This invention relates to the field of motor parameter identification technology. More specifically, this invention relates to a speed control method and system based on parameter identification. Background Technology

[0002] In the operation system of smart heating pipe networks, circulating water pumps are the core physical hubs driving heat energy transmission. In order to achieve energy saving and consumption reduction and precise regulation of heating flow, industrial control systems need to acquire the rotor speed of synchronous motors in real time and then implement speed control. However, placing sensors in the high-temperature and high-pressure pump cavity is not only costly in terms of hardware, but also prone to physical damage under long-term harsh operating conditions. Therefore, the industry usually adopts a sensorless vector control architecture. Among them, the speed analysis technology based on sliding mode observers is mature. This technology collects the stator voltage and stator current of the synchronous motor at the electrical port, constructs an equivalent mathematical observation equation inside the control chip, and uses preset basic electrical parameters such as stator resistance to perform reverse deduction to estimate the rotor speed. This technology can maintain a good steady-state operation under constant ambient temperature, and its underlying logic relies on the constant stator resistance parameter.

[0003] When the aforementioned electrical control technology is directly applied to the actual physical conditions of a smart heating network, the circulating water pump needs to continuously deliver high-temperature fluid during peak shaving and daily operation. The fluid temperature often fluctuates drastically between 40 and 90 degrees Celsius. This drastically alternating high-temperature heat energy will generate a strong heat conduction effect directly from the pump body to the synchronous motor casing and internal components. According to Joule's law and the objective laws of the physical properties of metallic conductors, the actual stator resistance of the synchronous motor will drift and increase as the conduction temperature rises. At this time, if the sliding mode observer still calls the factory-preset fixed stator resistance parameters, it will cause severe phase distortion in the rotor speed signal output by the equivalent mathematical observation equation, which will then cause severe vibration of the circulating water pump or even mechanical damage to the drive shaft.

[0004] To compensate for the stator resistance drift caused by heat conduction, existing technologies have introduced an online parameter identification mechanism. A mathematical integrator is added outside the sliding mode observer for correction: the error between the estimated and actual stator current is multiplied by a fixed convergence gain to obtain the compensation amount finally added to the original stator resistance, which is used to correct the stator resistance value in real time. When the fluid temperature of the smart heating network changes abruptly due to changes in supply and demand load, the fixed convergence gain deviates from the underlying thermodynamic laws, causing overcorrection or tracking lag in the numerical iteration space of the stator resistance. This leads to violent high-frequency oscillations in the stator resistance identification value, which not only fails to accurately complete parameter tracking and smooth speed regulation, but also induces low-frequency resonance of the synchronous motor or even stall and shutdown. Summary of the Invention

[0005] To address the technical problem that the conventional online parameter identification mechanism uses a fixed convergence gain, which deviates from the underlying thermodynamic laws and leads to oscillations and instability in speed estimation, this invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a speed regulation method based on parameter identification, applied to driving a circulating water pump, comprising: constructing a sliding mode observer with an adaptive identification interface for receiving dynamic convergence gain; the method for obtaining the dynamic convergence gain is as follows: based on the real-time fluid temperature of the circulating water pump operating in the heating network and the stator current of the synchronous motor, the highest possible physical rate of change of the motor stator temperature is determined using the thermodynamic energy conservation law; the physical speed limit boundary of the parameter online identification mechanism is set according to the highest physical rate of change, and a dynamic convergence gain with physical boundary awareness is obtained; the stator resistance parameter is updated periodically using the dynamic convergence gain to obtain the stator resistance identification value; the corresponding matrix element in the sliding mode observer is updated using the stator resistance identification value and the actual rotor speed of the synchronous motor is parsed out, and a closed-loop control command is generated to adjust the output power of the circulating water pump.

[0007] This invention establishes an objective thermodynamic boundary by introducing real-time fluid temperature and uses the highest physical rate of change of the motor stator temperature as the physical speed limit benchmark for online identification. This ensures that the adjustment logic of the dynamic convergence gain conforms to the actual heating process of the stator winding. This physical sensing mechanism effectively solves the problem of excessive numerical correction caused by conventional identification logic when facing fluid temperature change shocks, thereby suppressing the oscillation and instability of the stator resistance identification value and ensuring the accuracy of speed analysis of the synchronous motor and the stability of the output power of the circulating water pump under high-temperature alternating conditions.

[0008] Preferably, the formula for calculating the highest possible rate of physical change in the motor stator temperature is: In the formula, Indicates the sequence number of the previous sampling period; This represents the physical heat capacity of the motor stator; Represents stator current. This represents the stator resistance identification value from the previous sampling period; This represents the estimated stator temperature from the previous sampling period; Indicates the real-time temperature of the fluid; It represents the equivalent physical thermal resistance between the motor stator and the fluid.

[0009] This invention fully considers the dynamic balance between the Joule heat generated during motor operation and the heat dissipated to the external fluid. By quantifying the highest physical change rate caused by energy imbalance within a unit sampling period, it provides a clear physical inertial boundary for subsequent parameter identification. This operation ensures that the update step size of the resistance parameter is always controlled by the actual physical limitations of material heating, avoiding speed estimation deviations caused by ignoring changes in the heat dissipation environment, and significantly improving the control robustness of the speed regulation system under scenarios of drastic fluctuations in heating load.

[0010] Preferably, the formula for calculating the dynamic convergence gain is: In the formula, Indicates the sequence number of the current sampling period; Indicates the sequence number of the previous sampling period; This represents the dynamic convergence gain for the current sampling period; Indicates the underlying calibration coefficient; This represents the temperature coefficient of resistance of the stator winding material; This represents the physical heat capacity of the motor stator; Represents stator current. This represents the stator resistance identification value from the previous sampling period; This represents the estimated stator temperature from the previous sampling period; Indicates the real-time temperature of the fluid; This represents the equivalent physical thermal resistance between the motor stator and the fluid. This indicates taking the absolute value.

[0011] This invention successfully transforms the extreme value of thermodynamic evolution into a dynamic convergence gain that can be directly involved in digital control by integrating the highest physical change rate, the inherent properties of the material, and the underlying calibration coefficients. This transformation process enables the online identification mechanism to have an adaptive adjustment capability for the actual heating properties of the material, ensuring that the correction step size always operates within the physical safety envelope, and effectively preventing the risk of low-frequency resonance of the motor induced by an excessively large calculation step size.

[0012] Preferably, constructing a sliding mode observer with an adaptive identification interface includes: removing the fixed constant constraint from the internal stator resistance of the sliding mode observer and configuring the stator resistance as a dynamic variable to be updated with an external data injection interface.

[0013] This invention breaks away from the fixed parameter architecture of traditional sliding mode observers. By reconfiguring the stator resistance into a dynamic variable with an external injection interface, it gives the mathematical integrator the flexibility to evolve with the physical environment. This reconfiguration action eliminates the state matrix tilting problem caused by parameter locking in the original technology, and provides an execution architecture for subsequent real-time resistance compensation using dynamic convergence gain, ensuring the sliding mode observer's ability to continuously track actual impedance shifts.

[0014] Preferably, the step of updating the stator resistance parameters periodically using the dynamic convergence gain to obtain the stator resistance identification value includes: In the formula, Indicates the sequence number of the current sampling period; Indicates the sequence number of the previous sampling period; The stator resistance identification value for the current sampling period; This represents the stator resistance identification value from the previous sampling period; This represents the dynamic convergence gain for the current sampling period; and The stator current in the current sampling period is respectively shaft and The estimated error value of the shaft; and These are the current sampling periods. shaft and Estimated stator current under the shaft.

[0015] Preferably, the stator current is in shaft and The method for obtaining the estimation error value of the shaft includes: performing a Clark orthogonal transformation to project the stator voltage and stator current in the three-phase stationary coordinate system to a two-phase stationary orthogonal coordinate system, wherein the orthogonal coordinate system includes... shaft and The extracted stator voltage components under the orthogonal axis are substituted into the sliding mode observer with an adaptive identification interface. The state differential equation is iteratively solved using a discrete numerical integration algorithm, and the estimated stator current under the orthogonal axis for the current sampling period is output. The estimated stator current under the orthogonal axis is compared with the actual collected stator current components under the orthogonal axis to extract the values ​​respectively. Shaft estimation error value and Shaft estimation error value.

[0016] Preferably, the equivalent physical thermal resistance is obtained by the control system executing an offline temperature rise calibration program in advance. The calibration program includes: acquiring the real-time temperature of the fluid in thermal equilibrium at the moment of cold start-up of the equipment and recording it as the initial fluid temperature; controlling the inverter to inject a weak DC test signal into the synchronous motor and extracting the basic cold resistance; controlling the inverter to inject a calibration DC current with a constant amplitude into the synchronous motor; continuously monitoring the stator voltage reference command until the command value no longer drifts; synchronously recording the current stator steady-state current and the final steady-state voltage command; calculating the steady-state stator resistance by dividing the steady-state voltage command by the stator steady-state current; deriving the stator steady-state temperature using the initial fluid temperature, the steady-state stator resistance, and the basic cold resistance through the hot resistance method; calculating the absolute heating power by the square of the stator steady-state current and the steady-state stator resistance; and calculating the equivalent physical thermal resistance by dividing the difference between the stator steady-state temperature and the initial fluid temperature by the absolute heating power.

[0017] Preferably, the calibration steps for the bottom calibration coefficient include: controlling the circulating water pump to operate at a constant rated speed in a normal temperature water environment to achieve a dynamic physical steady state; injecting a fixed proportion of bias error into the current stator resistance identification value in the sliding mode observer to generate a stator current estimation error; gradually increasing the bottom calibration coefficient test value from 0, and simultaneously monitoring the convergence trajectory of the stator resistance identification value; when the stator resistance identification value exhibits constant amplitude high-frequency oscillation, recording the current bottom calibration coefficient test value as the critical oscillation coefficient; and multiplying the critical oscillation coefficient by a safety attenuation factor to obtain the bottom calibration coefficient.

[0018] In a second aspect, the present invention provides a speed regulation system based on parameter identification, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the aforementioned speed regulation method based on parameter identification is implemented.

[0019] By adopting the above technical solution, a speed regulation method based on parameter identification is generated into a computer program and stored in a memory for loading and execution by a processor. This allows for the creation of a terminal device based on the memory and processor, facilitating its use.

[0020] The beneficial effects of this invention are as follows: This invention establishes an objective thermodynamic boundary by introducing real-time fluid temperature and uses the highest physical rate of change of the motor stator temperature as the physical speed limit benchmark for online identification. This ensures that the adjustment logic of the dynamic convergence gain conforms to the actual heating process of the stator winding. This physical sensing mechanism effectively solves the problem of excessive numerical correction caused by conventional identification logic when facing fluid temperature change shocks, thereby suppressing the oscillation and instability of the stator resistance identification value and ensuring the accuracy of speed analysis of the synchronous motor and the stability of the output power of the circulating water pump under high-temperature alternating conditions. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating a speed regulation method based on parameter identification in this invention. Figure 2 This is a schematic diagram illustrating the control waveform of motor speed changing over time when the conventional parameter online identification mechanism uses a fixed convergence gain; Figure 3 This is a schematic diagram illustrating the adaptive adjustment waveform of the dynamic convergence gain over time in the method of the present invention; Figure 4 This is a schematic diagram illustrating the control waveform of motor speed changing over time after applying the dynamic convergence gain of the method of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0024] This invention discloses a speed regulation method based on parameter identification, referring to... Figure 1 This includes steps S1-S4: S1: Synchronously acquire the real-time temperature of the fluid running in the circulating water pump in the heating network, as well as the stator current and stator voltage of the synchronous motor driving the pump.

[0025] To address the high-temperature-variable load interference environment of the heating system, real-time fluid temperature is introduced to integrate objective thermodynamic boundary conditions into the core control link, thereby providing a physical benchmark for the subsequent construction of a dynamic constraint mechanism. Specifically, a high-precision industrial-grade platinum resistance temperature probe is deployed on the outer wall of the main return pipe of the circulating water pump. It is connected to an external signal conditioning board through an armored shielded cable. The physical heat is converted into a 4 to 20 mA standard analog current signal through a four-wire transmitter circuit, and then input to the 12-bit analog-to-digital converter unit inside the controller. The real-time fluid temperature is then extracted by median filtering in the software.

[0026] Meanwhile, a Hall effect current transformer and a high-frequency voltage isolation transmitter module are embedded in the three-phase power supply harness of the synchronous motor. Relying on the controller's built-in control timer to trigger the dual-channel synchronous sampling mechanism, the voltage and current signals are strictly aligned within the sampling window. Finally, the high-frequency harmonics of the inverter switch are removed by the discrete Butterworth low-pass filtering algorithm, and the stator voltage and stator current are extracted.

[0027] S2: Construct a sliding mode observer with an adaptive identification interface to evaluate the estimation error of the stator current.

[0028] Based on the existing electromagnetic differential mechanism of AC synchronous motors, a known sliding mode observer is constructed. On the basis of this existing technical system, a key structural reconstruction is performed: the fixed constant restriction of the stator resistance inside the sliding mode observer is removed, and it is configured as a dynamic variable to be updated with an external data injection interface so that the parameter overwriting can be performed in the future. Thus, a sliding mode observer with an adaptive identification interface is constructed.

[0029] In addition, by using the processor to perform the Clark orthogonal transformation known in the art, the electrical variables in the three-phase stationary coordinate system, namely stator voltage and stator current, are projected onto a two-phase stationary orthogonal coordinate system, and the stator current components and stator voltage components in the orthogonal axis system are extracted. Stator current components of the shaft Stator current components of the shaft The stator voltage components of the shaft and Stator voltage component of the shaft.

[0030] Furthermore, the extracted stator voltage components in the orthogonal axis system are used as known excitation inputs and substituted into the sliding mode observer with the adaptive identification interface. The state differential equation of the observer is iteratively solved using the discrete numerical integration algorithm inside the controller to calculate and output the estimated value of the stator current in the orthogonal axis system for the current sampling period, including: The estimated value of the stator current of the shaft and The estimation error value of the axis; then, the estimated value of the stator current under the orthogonal axis system is compared with the actual collected stator current components under the orthogonal axis system, and the difference is extracted respectively. Shaft estimation error value and Shaft estimation error value This provides closed-loop data feedback support for subsequent parameter identification.

[0031] S3: Based on the thermodynamic decay law, reconstruct the dynamic convergence gain of the sliding mode observer and perform stator resistance parameter identification.

[0032] As introduced in the background technology, conventional online parameter identification mechanisms sample a fixed convergence gain. However, this fixed convergence gain deviates from the underlying thermodynamic laws, leading to overcorrection or tracking lag in the numerical iteration space of stator resistance, causing severe high-frequency oscillations in the stator resistance identification value. To solve the above problems, this invention dynamically adjusts the convergence gain based on the thermodynamic decay law. In this process, the highest possible physical rate of change of the motor stator temperature is determined by the thermodynamic energy conservation law. The physical rate limit boundary of the online parameter identification mechanism is set according to the highest physical rate of change, thereby limiting the dynamic adjustment boundary of the convergence gain in the online parameter identification mechanism, making it subject to the actual physical temperature rise rate of the motor stator, thus establishing a stable parameter tracking mechanism under thermal shock.

[0033] According to the thermodynamic law of conservation of energy, due to the thermal inertia of metal materials, their temperature cannot change instantaneously; the rate of change of winding temperature is only subject to two types of physical energy: one is the internal Joule heat generated by the current passing through the metal wire, and the other is the heat dissipated by the motor to the surrounding circulating fluid; by using sensors to obtain the actual water temperature of the pipe network and combining it with the inherent heat dissipation coefficient of the motor, the control system can calculate the highest possible rate of change of winding temperature at the current instant.

[0034] Specifically, the thermal evolution of the motor stator within a unit sampling period is constrained by the law of conservation of energy; the rate of change of heat within the stator is equal to the difference between the Joule heat actively generated by the motor and the heat dissipated to the external fluid. Furthermore, according to Joule's law, the Joule heat power generated by the motor stator windings is equal to the stator current. The square of the value and the stator resistance identification value of the previous sampling period The product of, i.e. According to Newton's law of cooling in heat transfer, the heat power dissipated by the stator to the external circulating fluid is equal to the estimated stator temperature of the previous sampling period. Real-time temperature of fluid Divide the difference by the equivalent physical thermal resistance between the motor stator and the fluid. ,Right now Finally, the physical limit boundary of stator temperature change is derived. Combined with the heat capacity formula, it can be seen that the rate of change of heat is equal to the rate of change of stator temperature. Multiplied by the physical heat capacity of the stator ,Right now .

[0035] Furthermore, based on the law of conservation of energy, the rate of change of heat inside the stator is equal to the Joule heat power generated by the stator windings minus the heat power dissipated by the stator to the external circulating fluid. Based on this, an energy balance equation can be constructed. Divide both sides of the energy balance equation by the physical heat capacity of the stator. This allows us to determine the highest possible rate of physical change in the motor stator temperature at the current instant: This characterizes the limit rate at which the real resistance may deviate.

[0036] Furthermore, after determining the highest possible rate of physical change in the motor stator temperature, a cross-dimensional parameter replacement is performed: according to common sense in materials science, the resistance of a metal is linearly related to temperature, and the proportion of resistance change for every 1 degree Celsius change in temperature is determined by the temperature coefficient of resistance of the stator winding material. Therefore, the extreme value of the actual change in resistance is equal to... .

[0037] The derived extreme values ​​of resistance change are used as the safety envelope of the online parameter identification mechanism, and underlying calibration coefficients are introduced. By adjusting it, the dynamic convergence gain is obtained, and the specific calculation formula is as follows:

[0038] In the formula, Indicates the sequence number of the current sampling period; Indicates the sequence number of the previous sampling period; This represents the dynamic convergence gain for the current sampling period; Indicates the underlying calibration coefficient; The temperature coefficient of resistance (TCR) represents the temperature coefficient of resistance of the stator winding material. In the stator winding of a motor, the TCR of the winding material determines the characteristic of the winding resistance changing with temperature. For common copper conductors, the TCR of resistance is 0.00386 at 20°C. °C -1 That is, for every 1 degree Celsius increase in temperature °C, the resistance value increases by 0.00386Ω per ohm of initial resistance; The physical heat capacity of the motor stator can be obtained by consulting the motor technical manual. Represents stator current. This represents the stator resistance identification value from the previous sampling period; This represents the estimated stator temperature from the previous sampling period; Indicates the real-time temperature of the fluid; This represents the equivalent physical thermal resistance between the motor stator and the fluid. This indicates taking the absolute value.

[0039] The formula includes a term based on Joule heating. With heat conduction and heat dissipation The physical constraint combination term is formed as follows: When a large current surges into the electromagnetic drive system due to the increase in pipeline resistance, the Joule heating term spikes. The positive amplification of the physical constraint combination term drives the dynamic convergence gain to adaptively expand, giving the sliding mode observer's mathematical integrator a wider range of numerical update permissions to match the rapid heating rate in the real physical space. Conversely, when the fluid system is in steady-state circulating heat exchange, the heat conduction and heat dissipation terms and the Joule heating term tend to dynamically cancel each other out. The contraction of the physical constraint combination term limits the dynamic convergence gain to a safe range.

[0040] It should be noted that in adaptive control, the dynamic convergence gain only serves as a step size limiter, and its value must always be non-negative. By adding an absolute value sign, this invention distinguishes between the physical rate of change and the mathematical correction direction: the physical constraint combination term in the calculation formula is only used to extract the degree of change in resistance due to thermodynamic factors, i.e., the absolute velocity, as the step size limit boundary for parameter identification; while the specific direction of the stator resistance correction, i.e., whether it increases or decreases, is determined by the subsequent two-dimensional vector inner product term. To be confirmed.

[0041] Finally, the stator resistance parameters are updated periodically using the obtained dynamic convergence gain with physical boundary awareness. The specific calculation formula is as follows:

[0042] In the formula, Indicates the sequence number of the current sampling period; Indicates the sequence number of the previous sampling period; The stator resistance identification value for the current sampling period; This represents the stator resistance identification value from the previous sampling period; This represents the dynamic convergence gain for the current sampling period; and The stator current in the current sampling period is respectively shaft and The estimated error value of the shaft; and These are the current sampling periods. shaft and Estimated stator current under the shaft.

[0043] It should be added that, for obtaining the stator temperature estimate, based on engineering practice needs, the real-time fluid temperature obtained in the previous sampling period is directly used as the stator temperature estimate for the current sampling period.

[0044] In addition, the equivalent physical thermal resistance is an inherent thermodynamic structural property of the electromagnetic drive system. In order to ensure the accuracy of the online identification benchmark of the parameters, the control system needs to perform an offline temperature rise calibration procedure in advance to obtain the specific value of the equivalent physical thermal resistance before the circulating water pump equipment is put into operation. The calibration process is as follows: (1) At the moment of cold start of the equipment, the real-time temperature of the fluid in thermal equilibrium state is obtained and recorded as the initial fluid temperature. At the same time, the inverter is controlled to inject a weak DC test signal into the synchronous motor to obtain the basic cold resistance of the stator winding. Subsequently, the inverter is controlled to inject a calibration DC current with a constant amplitude into the synchronous motor, forcing the stator winding to continuously generate Joule heat. (2) During this temperature rise process, since the actual impedance of the stator increases with the temperature rise, in order to maintain the constant calibration DC current, the stator voltage reference command output by the control system will adaptively rise. The stator voltage reference command is continuously monitored until the value of the command no longer drifts. At this time, it is determined that the heat generation inside the motor and the heat dissipation to the external fluid have reached a physical steady-state equilibrium; (3) After reaching the steady-state equilibrium state, the current stator steady-state current and the final steady-state voltage command are recorded synchronously; the steady-state stator resistance is obtained by dividing the steady-state voltage command by the stator steady-state current; combined with the temperature coefficient of resistance of the stator winding material, the stator steady-state temperature is derived by using the initial fluid temperature, steady-state stator resistance and basic cold resistance through the hot resistance method; the hot resistance method is a method of measuring temperature by using the physical characteristics of the resistance of a metal conductor changing with temperature, which is a well-known technology and will not be described in detail here; (4) The absolute heating power is calculated by the square of the stator steady-state current and the steady-state stator resistance, and the equivalent physical thermal resistance between the motor stator and the fluid is calculated by dividing the difference between the stator steady-state temperature and the initial fluid temperature by the absolute heating power; the obtained equivalent physical thermal resistance is solidified and stored in the controller non-volatile memory.

[0045] It should be added that this invention determines the dynamic convergence gain based on the highest possible physical rate of change of the motor stator temperature. The dynamic convergence gain is a parameter used to control the refresh frequency of the controller; therefore, it is necessary to calibrate the underlying calibration coefficients. To match the engineering dimension conversion between continuous physical time and discrete digital sampling period; in order to obtain the underlying calibration coefficients, during the factory commissioning phase before the circulating water pump is officially put into operation, the controller executes a critical oscillation trial program for automated offline calibration; the specific operation is as follows: the circulating water pump is controlled to operate at a constant rated speed in a normal temperature water environment to ensure that the internal heat transfer reaches a dynamic physical steady state; in the sliding mode observer inside the controller, a fixed proportion of bias error is artificially injected into the current stator resistance identification value, forcing the system to generate stator current estimation error; subsequently, the control program gradually increases the test value of the underlying calibration coefficient from 0, and simultaneously monitors the convergence trajectory of the stator resistance identification value; when the stator resistance identification value shows a constant amplitude high-frequency oscillation near the true resistance value, it indicates that the current discrete system stability physical limit has been broken, and the system immediately captures and records the test value as the critical oscillation coefficient; further, combined with the conventional stability margin criterion of control engineering, the critical oscillation coefficient is multiplied by the safety attenuation factor, and the resulting product value is used as the final underlying calibration coefficient, which is stored in the controller's non-volatile memory for subsequent retrieval.

[0046] S4: Update the corresponding matrix element in the sliding mode observer using the obtained stator resistance identification value, parse out the actual rotor speed of the synchronous motor, and generate closed-loop control commands to adjust the output power of the circulating water pump.

[0047] Specifically, the stator resistance identification value of the current sampling period is extracted, mapped, and overwritten into the sliding mode observer with the adaptive identification interface, replacing the original resistance element in the state space matrix. After the controller performs matrix multiplication and state recursion, the sliding mode observer uses its internal smoothing switching function to extract the back EMF signal. Relying on the well-known quadrature phase-locked loop, the back EMF signal is phase-decoded and frequency-locked to accurately extract the actual rotor speed of the synchronous motor. Then, the actual rotor speed is input into a conventional outer-loop speed proportional-integral regulator, and the difference and proportional-integral operation are performed with the set desired speed to generate a quadrature-axis current reference command to maintain the desired flow rate. Further, combined with the quadrature-axis current reference command, multiple complementary drive pulse signals are generated through a well-known space vector pulse width modulation algorithm to control the inverter to output three-phase AC voltage and adjust the transient drive torque of the circulating water pump.

[0048] For example, Figure 2 This schematically illustrates the control waveform of motor speed versus time when a conventional online parameter identification mechanism uses a fixed convergence gain; for example... Figure 2As shown, under the conventional online parameter identification mechanism, due to the use of a fixed convergence gain, the setting of this gain deviates from the underlying objective thermodynamic laws. When the thermal conduction state in the equipment operating environment changes, the fixed parameter update step size cannot match the thermal physical limitations of the real material, causing overcorrection in the numerical iteration space of the stator resistance, which in turn leads to obvious high-frequency distortion in the resolved rotor speed signal. This parameter resolution inaccuracy is directly reflected at the system output, causing the motor speed to exhibit continuous high-frequency oscillation and instability when approaching the target speed.

[0049] For example, Figure 3 This is a schematic diagram illustrating the adaptive adjustment waveform of the dynamic convergence gain over time in the method of the present invention; for example... Figure 3 As shown, this invention constructs an objective thermodynamic boundary by introducing real-time fluid temperature, transforming the highest possible physical rate of change of motor stator temperature into a physical speed limit benchmark for parameter identification. Under the action of this physical sensing mechanism, the dynamic convergence gain exhibits the characteristic of adaptive adjustment with thermodynamic evolution: during the rapid temperature rise phase of system startup or the initial stage of thermal shock, the value of the dynamic convergence gain is adaptively expanded to give the system a wider range of numerical update permissions; as internal heat generation and external fluid heat dissipation gradually tend towards a dynamic physical steady state, the dynamic convergence gain smoothly decays and shrinks to a safe range, matching the actual temperature rise process of the stator winding throughout the entire process.

[0050] For example, Figure 4 This schematically illustrates the control waveform diagram of motor speed changing over time after applying the dynamic convergence gain of the method of this invention; for example... Figure 4 As shown, thanks to the dynamic convergence gain with physical boundary awareness, the sliding mode observer with adaptive identification interface achieves cycle-by-cycle smooth tracking of the stator resistance parameters. This mechanism fundamentally suppresses the parameter oscillation phenomenon induced by the uncontrolled numerical iteration step size, ensuring extremely high accuracy in extracting the back EMF signal and analyzing the actual rotor speed of the synchronous motor under high-temperature alternating conditions. The resulting closed-loop control command makes the transient drive torque output of the circulating water pump extremely stable, and the motor speed can quickly and accurately converge to the rated target operating state.

[0051] This invention also discloses a speed control system based on parameter identification, including a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement a speed control method based on parameter identification according to the present invention.

[0052] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

Claims

1. A speed regulation method based on parameter identification, applied to a synchronous motor driving a circulating water pump, characterized in that, include: Construct a sliding mode observer with an adaptive identification interface to receive the dynamic convergence gain; The method for obtaining the dynamic convergence gain is as follows: based on the real-time fluid temperature of the circulating water pump in the heating network and the stator current of the synchronous motor, the highest possible physical rate of change of the motor stator temperature is determined by using the thermodynamic energy conservation law, and the physical speed limit boundary of the parameter online identification mechanism is set according to the highest physical rate of change to obtain the dynamic convergence gain with physical boundary awareness. The stator resistance parameters are updated cycle by cycle using the dynamic convergence gain to obtain the stator resistance identification value. The corresponding matrix element in the sliding mode observer is updated using the stator resistance identification value, and the actual rotor speed of the synchronous motor is analyzed to generate a closed-loop control command to adjust the output power of the circulating water pump.

2. The speed regulation method based on parameter identification according to claim 1, characterized in that, The formula for calculating the highest possible rate of physical change in the motor stator temperature is: In the formula, Indicates the sequence number of the previous sampling period; This represents the physical heat capacity of the motor stator; Represents stator current. This represents the stator resistance identification value from the previous sampling period; This represents the estimated stator temperature from the previous sampling period; Indicates the real-time temperature of the fluid; It represents the equivalent physical thermal resistance between the motor stator and the fluid.

3. The speed regulation method based on parameter identification according to claim 1, characterized in that, The formula for calculating the dynamic convergence gain is: ； In the formula, Indicates the sequence number of the current sampling period; Indicates the sequence number of the previous sampling period; This represents the dynamic convergence gain for the current sampling period; Indicates the underlying calibration coefficient; This represents the temperature coefficient of resistance of the stator winding material; This represents the physical heat capacity of the motor stator; Represents stator current. This represents the stator resistance identification value from the previous sampling period; This represents the estimated stator temperature from the previous sampling period; Indicates the real-time temperature of the fluid; This represents the equivalent physical thermal resistance between the motor stator and the fluid. This indicates taking the absolute value.

4. The speed regulation method based on parameter identification according to claim 1, characterized in that, The construction of the sliding mode observer with an adaptive identification interface includes: Remove the fixed constant constraint from the internal stator resistance of the sliding mode observer and configure the stator resistance as a dynamic variable to be updated with an external data injection interface.

5. The speed regulation method based on parameter identification according to claim 1, characterized in that, The step of updating the stator resistance parameters periodically using the dynamic convergence gain to obtain the stator resistance identification value includes: ； In the formula, Indicates the sequence number of the current sampling period; Indicates the sequence number of the previous sampling period; The stator resistance identification value for the current sampling period; This represents the stator resistance identification value from the previous sampling period; This represents the dynamic convergence gain for the current sampling period; and The stator current in the current sampling period is respectively shaft and The estimated error value of the shaft; and These are the current sampling periods. shaft and Estimated stator current under the shaft.

6. The speed regulation method based on parameter identification according to claim 5, characterized in that, The stator current is shaft and Methods for obtaining the estimated error value of the shaft include: Performing the Clark orthogonal transformation projects the stator voltage and stator current from the three-phase stationary coordinate system to a two-phase stationary orthogonal coordinate system, wherein the orthogonal coordinate system includes... shaft and axis; The extracted stator voltage components in the orthogonal axis system are substituted into the sliding mode observer with the adaptive identification interface, and the state differential equation is solved iteratively using the discrete numerical integration algorithm to output the estimated stator current in the orthogonal axis system for the current sampling period. The estimated stator current in the orthogonal axis system is compared with the actual collected stator current components in the orthogonal axis system using a difference calculation to extract the values. Shaft estimation error value and Shaft estimation error value.

7. The speed regulation method based on parameter identification according to claim 2, characterized in that, The equivalent physical thermal resistance is obtained by the control system performing an offline temperature rise calibration program in advance. The calibration program includes: The real-time temperature of the fluid in thermal equilibrium state is acquired and recorded as the initial fluid temperature at the moment of cold start-up of the equipment. The inverter is controlled to inject a weak DC test signal into the synchronous motor and extract the base cold resistance. The inverter is also controlled to inject a constant amplitude calibration DC current into the synchronous motor. The stator voltage reference command is continuously monitored until the command value no longer drifts. The current stator steady-state current and the final steady-state voltage command are recorded synchronously. The steady-state stator resistance is calculated by dividing the steady-state voltage command by the stator steady-state current. The stator steady-state temperature is derived using the initial fluid temperature, the steady-state stator resistance, and the basic cold-state resistance through the hot-state resistance method. The absolute heating power is calculated from the square of the stator steady-state current and the steady-state stator resistance; the equivalent physical thermal resistance is calculated by dividing the difference between the stator steady-state temperature and the initial fluid temperature by the absolute heating power.

8. The speed regulation method based on parameter identification according to claim 3, characterized in that, The calibration steps for the underlying calibration coefficients include: The circulating water pump is controlled to operate at a constant rated speed in a normal temperature water environment to achieve a dynamic physical steady state. A fixed-proportion bias error is injected into the current stator resistance identification value in the sliding mode observer to generate a stator current estimation error; Starting from 0, the test value of the underlying calibration coefficient is gradually increased, and the convergence trajectory of the stator resistance identification value is monitored simultaneously. When the stator resistance identification value exhibits constant-amplitude high-frequency oscillation, the current test value of the underlying calibration coefficient is recorded as the critical oscillation coefficient. The critical oscillation coefficient is multiplied by the safety attenuation factor to obtain the underlying calibration coefficient.

9. A speed control system based on parameter identification, characterized in that, include: A processor and a memory, the memory storing computer program instructions that, when executed by the processor, implement a speed regulation method based on parameter identification according to any one of claims 1-8.

Images