Optimization control method for dynamic performance of frequency converter

Abstract

The invention relates to a frequency converter dynamic performance optimization control method, and belongs to the technical field of power electronics. The method comprises the steps that frequency converter operation and device characteristic signals and external working condition correlation signals are collected in real time and integrated into a multi-dimensional operation data set, an analysis model is built, and a mapping relation is preset; extracting transient characteristics of a device switch, calculating a dynamic impedance parameter based on a mapping relation, and matching a parameter self-tuning rule base to generate a current loop proportionality coefficient, an integral time constant and a modulation wave duty ratio correction; a rotating speed and current double-loop fusion framework is built, the weight of an outer loop is dynamically adjusted according to rotating speed deviation and load sudden change, an inner loop adopts an adaptive reaching law and disturbance compensation, and a cooperative control instruction is generated through inter-loop interaction; monitoring instruction execution data to form feedback, correcting a model mapping relation and optimizing control parameters. The dynamic response precision of the frequency converter under complex working conditions is improved, the anti-disturbance capability is enhanced, and meanwhile, the device operation loss and the manual debugging cost are reduced.

Description

Technical Field

[0001] This invention belongs to the field of power electronics technology, and specifically relates to a method for dynamic performance optimization control of frequency converters. Background Technology

[0002] As a core control device in the field of power electronics, frequency converters are widely used in industrial motor drives, new energy power generation, rail transit, smart homes and other fields. Their dynamic performance directly determines the operating efficiency, stability and energy consumption of terminal equipment. With the improvement of industrial automation and the increase in demand for complex working conditions (such as frequent load changes in the metallurgical industry, grid voltage fluctuations of photovoltaic inverters, and frequent start-stop switching of elevator traction machines), the requirements for the dynamic performance of frequency converters have been upgraded from "basic control" to "high precision, high adaptability, and low loss coordination".

[0003] Current mainstream inverter dynamic performance optimization solutions mostly revolve around improving a single control algorithm or tuning static parameters. Firstly, traditional PID control remains the most widely used solution in the industrial field due to its simple structure and ease of implementation. However, this solution relies on fixed parameters tuned offline. When operating conditions change, the parameters cannot adapt in real time, which can easily lead to speed overshoot, response delay, or current oscillation. Especially when core switching devices (such as IGBTs) experience dynamic impedance changes due to aging and temperature rise, PID control struggles to compensate for the performance degradation caused by the decay of device characteristics. Secondly, while advanced algorithms such as Model Predictive Control (MPC) and Sliding Mode Control can balance multi-objective optimization and fast response, MPC requires accurate motor mathematical models and operating parameters. In complex scenarios such as load changes and grid harmonic pollution, the model is prone to inaccuracy, and the high computational complexity limits the response speed. Sliding Mode Control suffers from "chattering," and conventional approach law designs struggle to balance response speed and control stability. Furthermore, it is not correlated with the dynamic characteristics of the devices and cannot avoid the increased losses caused by high-frequency switching. Furthermore, existing solutions generally suffer from three major defects: a disconnect between device characteristics and operating parameters, insufficient coordination in multi-loop control, and a lack of closed-loop optimization mechanisms. On the one hand, most control strategies treat the parameters of core switching devices (such as on-resistance and junction capacitance) as fixed values, ignoring the dynamic impedance changes of devices under different temperatures and aging levels. Existing solutions do not establish a correlation mechanism between device characteristics and control parameters, making it impossible to dynamically correct control strategies. On the other hand, in the multi-loop control architecture of speed and current loops, conventional solutions use fixed weight allocation. When operating conditions change (such as from stable operation to sudden load changes), key performance indicators (such as speed tracking accuracy) cannot be prioritized, and information exchange between loops is lagging, resulting in poor coordinated control effects. At the same time, existing solutions are mostly open-loop control-one-time optimization modes, lacking real-time feedback and model correction of operating data. After long-term operation, dynamic performance will continue to degrade due to operating condition drift and device aging, requiring repeated manual debugging and increasing maintenance costs. Summary of the Invention

[0004] To address the aforementioned problems in the existing technology, this invention provides a method for optimizing the dynamic performance control of a frequency converter. The objective of this invention can be achieved through the following technical solutions: include: S1: Real-time acquisition of inverter operation and device characteristic signals, synchronous acquisition of external operating condition related signals, forming a multi-dimensional operation dataset; based on the multi-dimensional operation dataset, construct a device operating condition coupling analysis model, and preset the mapping relationship between dynamic impedance parameters and switching transient characteristic quantities and operating condition parameters; S2: Extract the switching transient characteristics of the device, calculate the dynamic impedance parameters under the current operating condition based on the mapping relationship; construct a parameter self-tuning rule library, preset the control parameter adjustment thresholds corresponding to different dynamic impedance ranges; generate the current loop proportional coefficient, integral time constant and modulation wave duty cycle correction amount according to the adjustment strategy in the dynamic impedance parameter matching rule library. S3: Build a speed and current integrated control architecture. The outer loop is the speed control loop, which constructs a dynamic weight allocation model and dynamically adjusts the weight ratio according to the speed deviation and the degree of load change. The inner loop is the current control loop, which presets an adaptive approach law and a grid disturbance compensation rule. Establish an inter-loop information exchange mechanism to generate collaborative control commands. S4: Monitor the operation-related data after the command is executed in real time to obtain the performance feedback dataset; input the performance feedback dataset into the device operating condition coupling analysis model to correct the mapping relationship between the dynamic impedance parameter and the operating condition parameter; optimize the parameter self-tuning rule base threshold, double-loop weight allocation coefficient and sliding mode control reaching law parameter.

[0005] As a preferred technical solution of the present invention, the specific process of S1 is as follows: First, synchronously collect the inverter operation and device characteristic signals (output current, voltage, switching transient waveforms) and external operating condition related signals (load speed, grid voltage fluctuations). After adaptive filtering and noise reduction, clock synchronization calibration, amplitude and timing alignment, the data is integrated into a multi-dimensional operating dataset. Then, based on this dataset, a framework is built according to the two dimensions of "device characteristics - operating condition parameters", incorporating physical constraints such as power electronic conversion principles and device thermal conduction characteristics to construct a device operating condition coupling analysis model. Finally, by collecting measured data of device dynamic impedance under different operating conditions, statistical analysis and data fitting are performed to preset the mapping relationship between dynamic impedance parameters and switching transient characteristic quantities and operating condition parameters, providing accurate data support and suitable analysis basis for subsequent steps.

[0006] Specifically, the process of acquiring inverter operation and device characteristic signals in real time and synchronously acquiring external operating condition related signals is as follows: synchronously capturing the current change, voltage fluctuation and switching transient waveforms of the inverter output terminal; simultaneously acquiring the real-time load rotation status and grid voltage stability and fluctuation trend; using an adaptive filtering algorithm to filter grid interference and environmental noise; using the inverter operating clock as a reference for signal synchronization calibration; after eliminating invalid interference data, performing amplitude calibration and timing alignment on all signals, and integrating them into a multi-dimensional operating dataset.

[0007] Specifically, the process of constructing the device operating condition coupling analysis model is as follows: based on the signal characteristics in the multi-dimensional operating data set, the device characteristic dimension and the operating condition parameter dimension are divided; a correlation analysis framework between the device characteristic dimension and the operating condition parameter dimension is established; according to the power electronic conversion principle and the device thermal conduction characteristics, the physical mechanism constraints of the frequency converter operation are incorporated to form a structured coupling analysis model.

[0008] Specifically, the mapping relationship between the preset dynamic impedance parameters and the switching transient characteristics and operating parameters is as follows: collect measured data of the device's dynamic impedance under different operating conditions; extract the corresponding switching transient characteristics and operating parameter ranges; classify and organize the measured data using statistical analysis methods; establish corresponding association rules through data fitting; and form a preset mapping relationship based on the association rules.

[0009] Specifically, the process of extracting the switching transient features of the device is as follows: extract the waveform of the turn-on and turn-off periods of the core switching device from the standardized device feature signal, extract the rising edge slope, falling edge slope, peak value feature and duration feature of the waveform, use feature normalization processing to eliminate the signal amplitude difference under different operating conditions, and quantize the extracted features into feature vectors.

[0010] Specifically, the process of calculating the dynamic impedance parameters under the current operating condition based on the mapping relationship is as follows: the extracted switching transient feature vector and the current operating condition parameters are substituted into the preset mapping relationship, and the calculation coefficients are corrected by combining the device static parameters and the real-time operating condition; feature matching and parameter derivation are performed through the built-in calculation logic of the model, and the dynamic impedance value corresponding to the current operating condition is output.

[0011] Specifically, the process of constructing the parameter self-tuning rule library is as follows: divide the interval boundaries according to the rated operating parameters and extreme operating conditions of the frequency converter; divide the interval gradient of dynamic impedance according to the adjustment range of the frequency converter control parameters; set the corresponding current loop proportional coefficient adjustment range, integral time constant correction range and modulation wave duty cycle adjustment range for each interval gradient, and store the intervals and adjustment rules to form a rule library.

[0012] Specifically, the process of dynamically adjusting the weight ratio is as follows: set thresholds and judgment criteria according to the performance requirements of the inverter application scenario; set speed deviation thresholds and load change judgment criteria; when the speed deviation exceeds the preset threshold or the load changes suddenly, increase the weight ratio corresponding to the speed tracking accuracy; when the speed deviation is within the preset threshold range, increase the weight ratio corresponding to the device loss control.

[0013] Specifically, the process of the preset adaptive approach law is as follows: based on the control accuracy requirements of the current loop, the dynamic response characteristics of the approach law are optimized by incorporating the current change rate feedback term, an approach law expression including a deviation feedback term and a rate adjustment term is designed, the dynamic adjustment range of the approach law parameters is set, and the parameter values ​​are matched in real time according to the magnitude of the current deviation.

[0014] Specifically, the process of establishing the inter-loop information interaction mechanism is as follows: a signal transmission channel is preset between the speed control loop and the current control loop, and a real-time communication protocol is adopted; the speed control loop transmits the weight allocation result to the current control loop in real time; the current control loop feeds back the current regulation status to the speed control loop, forming a two-way information interaction link.

[0015] Specifically, the process of correcting the mapping relationship between dynamic impedance parameters and operating parameters is as follows: the actual dynamic impedance values ​​in the performance feedback dataset are compared with the model prediction values ​​to calculate the deviation; based on the distribution law of the deviation, the coefficients are adjusted using a gradient correction method, and the feature matching coefficients in the mapping relationship are adjusted to gradually reduce the deviation between the model prediction values ​​and the actual values.

[0016] Specifically, the process for optimizing the self-tuning rule base threshold, the dual-loop weight allocation coefficient, and the sliding mode control reaching law parameters is as follows: based on the performance feedback dataset, analyze the sensitivity of each parameter to the control effect and divide the parameter optimization priority; set the parameter optimization range for different operating scenarios, use iterative optimization algorithms to adjust the parameter values, and verify the optimization effect through multiple sets of operating condition tests.

[0017] The beneficial effects of this invention are as follows: (1) By setting up a device operating condition coupling analysis model and a preset mapping relationship between dynamic impedance parameters and switching transient characteristics and operating condition parameters, combined with adaptive filtering signal processing and feature normalization transient feature extraction, the dynamic characteristic changes of core switching devices and external operating condition fluctuations can be captured in real time. Then, relying on the parameter self-tuning rule library, the control parameters can be accurately adapted. This avoids the overshoot and delay problems of traditional fixed parameter control when the device ages or the operating condition changes suddenly. Furthermore, through the real-time calculation and correction of dynamic impedance parameters, it ensures that key parameters such as the current loop proportional coefficient and integral time constant are always matched with the device characteristics and operating condition requirements, thereby improving the dynamic response accuracy and device adaptability of the frequency converter under complex operating conditions and reducing the increase in energy consumption and performance degradation caused by parameter mismatch. (2) By setting up a speed and current integrated control architecture and a two-way information interaction mechanism between loops, combined with a dynamic weight allocation model and an adaptive approach law design, and at the same time constructing a performance feedback-driven closed-loop optimization mechanism, it is possible to achieve coordinated linkage and long-term performance stability of multi-loop control. Dynamic weight allocation can flexibly adjust the control priority according to speed deviation and load change, ensuring speed tracking or loss control requirements in key scenarios. Inter-loop information interaction eliminates the information lag problem of traditional multi-loop control and improves the efficiency of coordinated control. Based on the mapping relationship correction and parameter iteration optimization of feedback data, it can not only gradually reduce the deviation between model prediction and actual operation, but also adapt to the operating condition drift and device aging in long-term operation, avoid repeated manual debugging, reduce operation and maintenance costs, and ensure the long-term stability and reliability of the inverter's dynamic performance. Attached Figure Description

[0018] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0019] Figure 1 This is a flowchart illustrating a method for optimizing the dynamic performance control of a frequency converter according to the present invention. Figure 2 This is a data flow diagram of a frequency converter dynamic performance optimization control method according to the present invention; Figure 3 This is a schematic diagram of the device operating condition coupling analysis model in this invention. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0021] Please see Figure 1-3 A method for optimizing the dynamic performance control of frequency converters; include: S1: Real-time acquisition of inverter operation and device characteristic signals, synchronous acquisition of external operating condition related signals, forming a multi-dimensional operation dataset; based on the multi-dimensional operation dataset, construct a device operating condition coupling analysis model, and preset the mapping relationship between dynamic impedance parameters and switching transient characteristic quantities and operating condition parameters; S2: Extract the switching transient characteristics of the device, calculate the dynamic impedance parameters under the current operating condition based on the mapping relationship; construct a parameter self-tuning rule library, preset the control parameter adjustment thresholds corresponding to different dynamic impedance ranges; generate the current loop proportional coefficient, integral time constant and modulation wave duty cycle correction amount according to the adjustment strategy in the dynamic impedance parameter matching rule library. S3: Build a speed and current integrated control architecture. The outer loop is the speed control loop, which constructs a dynamic weight allocation model and dynamically adjusts the weight ratio according to the speed deviation and the degree of load change. The inner loop is the current control loop, which presets an adaptive approach law and a grid disturbance compensation rule. Establish an inter-loop information exchange mechanism to generate collaborative control commands. S4: Monitor the operation-related data after the command is executed in real time to obtain the performance feedback dataset; input the performance feedback dataset into the device operating condition coupling analysis model to correct the mapping relationship between the dynamic impedance parameter and the operating condition parameter; optimize the parameter self-tuning rule base threshold, double-loop weight allocation coefficient and sliding mode control reaching law parameter.

[0022] As a preferred technical solution of the present invention, the specific process of S1 is as follows: First, synchronously collect the inverter operation and device characteristic signals (output current, voltage, switching transient waveforms) and external operating condition related signals (load speed, grid voltage fluctuations). After adaptive filtering and noise reduction, clock synchronization calibration, amplitude and timing alignment, the data is integrated into a multi-dimensional operating dataset. Then, based on this dataset, a framework is built according to the two dimensions of "device characteristics - operating condition parameters", incorporating physical constraints such as power electronic conversion principles and device thermal conduction characteristics to construct a device operating condition coupling analysis model. Finally, by collecting measured data of device dynamic impedance under different operating conditions, statistical analysis and data fitting are performed to preset the mapping relationship between dynamic impedance parameters and switching transient characteristic quantities and operating condition parameters, providing accurate data support and suitable analysis basis for subsequent steps.

[0023] Specifically, the process of acquiring inverter operation and device characteristic signals in real time and synchronously acquiring external operating condition related signals is as follows: synchronously capturing the current change, voltage fluctuation and switching transient waveforms of the inverter output terminal; simultaneously acquiring the real-time load rotation status and grid voltage stability and fluctuation trend; using an adaptive filtering algorithm to filter grid interference and environmental noise; using the inverter operating clock as a reference for signal synchronization calibration; after eliminating invalid interference data, performing amplitude calibration and timing alignment on all signals, and integrating them into a multi-dimensional operating dataset.

[0024] In this embodiment, the inverter operation and device characteristic signals refer to a set of signals that directly reflect the inverter's own operating status and the working characteristics of the core switching devices, including the current change trend at the inverter output terminal, voltage fluctuation, and transient waveforms of the core switching devices during turn-on and turn-off. The external operating condition related signals refer to external environment and load and power grid status signals related to the inverter operation, including the real-time rotation status of the load (such as speed magnitude and speed change trend), the voltage stability of the power grid, and the amplitude and trend of power grid voltage fluctuation. These signals directly affect the operating conditions of the inverter and are key inputs for constructing the coupling relationship between devices and operating conditions.

[0025] Specifically, the process of constructing the device operating condition coupling analysis model is as follows: based on the signal characteristics in the multi-dimensional operating data set, the device characteristic dimension and the operating condition parameter dimension are divided; a correlation analysis framework between the device characteristic dimension and the operating condition parameter dimension is established; according to the power electronic conversion principle and the device thermal conduction characteristics, the physical mechanism constraints of the frequency converter operation are incorporated to form a structured coupling analysis model.

[0026] Specifically, the mapping relationship between the preset dynamic impedance parameters and the switching transient characteristics and operating parameters is as follows: collect measured data of the device's dynamic impedance under different operating conditions; extract the corresponding switching transient characteristics and operating parameter ranges; classify and organize the measured data using statistical analysis methods; establish corresponding association rules through data fitting; and form a preset mapping relationship based on the association rules.

[0027] In this embodiment, the dynamic impedance parameter refers to the real-time parameters of the core switching device, such as the dynamic on-resistance and dynamic junction capacitance, which reflect the conduction and switching characteristics of the device under real-time operating conditions; the switching transient characteristic refers to the characteristic parameters extracted from the turn-on / turn-off waveform of the core switching device, including the rising slope, falling slope, peak value, and duration of the waveform, which are the direct basis for deriving the dynamic impedance parameter.

[0028] In this embodiment, the operating parameters refer to key parameters that describe the external conditions of the inverter's operation, including load speed, load change rate (such as the magnitude of load increase or decrease), grid voltage fluctuation value, grid harmonic content, and inverter operating ambient temperature.

[0029] Specifically, the process of extracting the switching transient features of the device is as follows: extract the waveform of the turn-on and turn-off periods of the core switching device from the standardized device feature signal, extract the rising edge slope, falling edge slope, peak value feature and duration feature of the waveform, use feature normalization processing to eliminate the signal amplitude difference under different operating conditions, and quantize the extracted features into feature vectors.

[0030] Specifically, the process of calculating the dynamic impedance parameters under the current operating condition based on the mapping relationship is as follows: the extracted switching transient feature vector and the current operating condition parameters are substituted into the preset mapping relationship, and the calculation coefficients are corrected by combining the device static parameters and the real-time operating condition; feature matching and parameter derivation are performed through the built-in calculation logic of the model, and the dynamic impedance value corresponding to the current operating condition is output.

[0031] In this embodiment, the control parameters corresponding to the different dynamic impedance ranges refer to the core inverter control parameters adapted to each range after dividing the range according to the numerical range of the dynamic impedance of the core switching device. These parameters include the proportional coefficient adjustment range of the current loop, the correction value of the integral time constant, and the adjustment amount of the modulation wave duty cycle. These parameters are preset through the self-tuning rule library.

[0032] In this embodiment, the device operating condition coupling analysis model consists of five levels: signal acquisition and preprocessing, feature extraction and state recognition, dynamic impedance parameter calculation, parameter mapping and decision-making, and control parameter generation and optimization. Each level is sequentially connected to form a closed-loop link: the signal acquisition and preprocessing layer is responsible for capturing and filtering inverter operation and device characteristic signals, as well as external operating condition related signals, and completing timing alignment and standardization processing; the feature extraction and state recognition layer extracts switching transient features and operating condition parameters from the preprocessed signals, and identifies the operating condition mode and device operating state; the dynamic impedance parameter calculation layer integrates the device's multi-physics field characteristics, establishes the correlation between characteristics and operating conditions, and derives dynamic impedance parameters in real time; the parameter mapping and decision-making layer matches preset mapping rules based on dynamic impedance parameters and operating condition types, dynamically allocates operating condition weights, and generates control parameter adjustment strategies; the control parameter generation and optimization layer outputs current loop parameters and modulation wave duty cycle correction, and continuously optimizes the mapping relationship and rule base based on operating feedback data to ensure that the model adapts to changes in device characteristics and complex operating condition requirements.

[0033] In this embodiment, the specific process for generating the current loop proportional coefficient, integral time constant, and modulation wave duty cycle correction is as follows: First, based on the current dynamic impedance parameter output by the device operating condition coupling analysis model, a preset parameter self-tuning rule library is queried to determine the interval gradient to which the dynamic impedance parameter belongs (e.g., the dynamic impedance is in the low, middle, or high interval). The rule library has preset corresponding control parameter adjustment logic for each interval gradient—if the dynamic impedance parameter is in the low interval (indicating good device conduction characteristics), the current loop proportional coefficient is reduced by a small amount, the integral time constant is appropriately extended, and the modulation wave duty cycle adjustment is maintained at the base value. The system is designed to maintain a certain range to avoid overshoot due to excessively fast response. If the dynamic impedance is in the high range (indicating a decrease in the device's conduction capability), the current loop proportional coefficient is increased according to the preset ratio in the rule library to compensate for the response speed, the integral time constant is shortened to accelerate steady-state convergence, and the modulation wave duty cycle adjustment is increased to adapt to changes in the device's switching characteristics. Finally, based on the range matching results and combined with the current operating parameters (such as the degree of load change and grid voltage fluctuations), the preset adjustment is fine-tuned to generate the final current loop proportional coefficient correction, integral time constant correction, and modulation wave duty cycle correction, ensuring that the correction values ​​both conform to the device's dynamic characteristics and adapt to real-time operating requirements.

[0034] In this embodiment, the specific process of building the integrated speed and current control architecture is as follows: First, the dual-loop nested logic of the architecture is clarified, with the speed control loop as the outer loop and the current control loop as the inner loop, and the two achieving data communication through a dedicated signal transmission channel; for the outer loop (speed control loop), a dynamic weight allocation model is first constructed, with the initial weight ratio of speed tracking accuracy and device loss control preset, and then a speed deviation detection module and a load change identification module are embedded, setting the speed deviation threshold and load change judgment criteria, so that the model can dynamically adjust the weight ratio according to the magnitude of the real-time detected speed deviation value and the degree of load change; for the inner loop (current control loop), a design is first designed based on the current loop control accuracy requirements, including... The adaptive approach law of the deviation feedback term and the rate adjustment term is integrated with the current change rate feedback term to optimize the dynamic response characteristics. Then, the grid disturbance compensation rule is preset, and the grid disturbance observation module captures the disturbance information such as grid voltage fluctuation and harmonic pollution in real time to generate the corresponding compensation signal. Finally, a two-way information interaction mechanism between the loops is established. The outer loop transmits the real-time weight allocation result to the inner loop as the basis for adjusting the inner loop approach law parameter and adapting the disturbance compensation intensity. The inner loop feeds back the operating status such as current regulation accuracy and current jitter amplitude to the outer loop to assist the outer loop in optimizing the weight allocation strategy. This forms a closed-loop nested architecture of outer loop decision-inner loop execution-status feedback, completing the construction of the speed and current fusion control architecture.

[0035] Specifically, the process of constructing the parameter self-tuning rule library is as follows: divide the interval boundaries according to the rated operating parameters and extreme operating conditions of the frequency converter; divide the interval gradient of dynamic impedance according to the adjustment range of the frequency converter control parameters; set the corresponding current loop proportional coefficient adjustment range, integral time constant correction range and modulation wave duty cycle adjustment range for each interval gradient, and store the intervals and adjustment rules to form a rule library.

[0036] Specifically, the process of dynamically adjusting the weight ratio is as follows: set thresholds and judgment criteria according to the performance requirements of the inverter application scenario; set speed deviation thresholds and load change judgment criteria; when the speed deviation exceeds the preset threshold or the load changes suddenly, increase the weight ratio corresponding to the speed tracking accuracy; when the speed deviation is within the preset threshold range, increase the weight ratio corresponding to the device loss control.

[0037] Specifically, the process of the preset adaptive approach law is as follows: based on the control accuracy requirements of the current loop, the dynamic response characteristics of the approach law are optimized by incorporating the current change rate feedback term, an approach law expression including a deviation feedback term and a rate adjustment term is designed, the dynamic adjustment range of the approach law parameters is set, and the parameter values ​​are matched in real time according to the magnitude of the current deviation.

[0038] Specifically, the process of establishing the inter-loop information interaction mechanism is as follows: a signal transmission channel is preset between the speed control loop and the current control loop, and a real-time communication protocol is adopted; the speed control loop transmits the weight allocation result to the current control loop in real time; the current control loop feeds back the current regulation status to the speed control loop, forming a two-way information interaction link.

[0039] Specifically, the process of correcting the mapping relationship between dynamic impedance parameters and operating parameters is as follows: the actual dynamic impedance values ​​in the performance feedback dataset are compared with the model prediction values ​​to calculate the deviation; based on the distribution law of the deviation, the coefficients are adjusted using a gradient correction method, and the feature matching coefficients in the mapping relationship are adjusted to gradually reduce the deviation between the model prediction values ​​and the actual values.

[0040] In this embodiment, the execution-related data refers to the feedback data reflecting the inverter's operating effect and working condition after the execution of the collaborative control command. This includes speed response delay time, speed overshoot, current regulation accuracy, current jitter amplitude, operating losses of core switching devices, real-time grid voltage status, and load speed stability. These data are the core basis for the closed-loop optimization model and parameters.

[0041] Specifically, the process for optimizing the self-tuning rule base threshold, the dual-loop weight allocation coefficient, and the sliding mode control reaching law parameters is as follows: based on the performance feedback dataset, analyze the sensitivity of each parameter to the control effect and divide the parameter optimization priority; set the parameter optimization range for different operating scenarios, use iterative optimization algorithms to adjust the parameter values, and verify the optimization effect through multiple sets of operating condition tests.

[0042] In this embodiment, taking the scenario of a motor-driven conveyor belt in a metallurgical plant as an example, the execution process of this solution is explained in detail: In this scenario, the load on the conveyor belt often changes abruptly from the initial value a to b (b is greater than a), the grid voltage may experience a temporary drop of magnitude c, and the dynamic impedance of the core switching device gradually decreases from the initial value d to e (e is greater than d) due to long-term high-temperature operation. The specific process is as follows: In stage S1, the current changes, voltage fluctuations, and turn-on / off transient waveforms of the core switching devices at the inverter output terminal are synchronously acquired (i.e., inverter operation and device characteristic signals). At the same time, the real-time speed of the conveyor belt (load speed signal), the stability of the grid voltage, and the fluctuation trend of the amplitude c (external operating condition related signal) are also acquired. After filtering grid interference by an adaptive filtering algorithm, the signal synchronization calibration and timing alignment are completed with the inverter operating clock as the reference, and integrated into a multi-dimensional operating dataset. Based on this dataset, a device operating condition coupling analysis model is built, incorporating the power electronic conversion principle and the thermal conduction characteristics of the device, and pre-setting the mapping relationship between dynamic impedance parameters and switching transient characteristic quantities (such as waveform rise slope and peak value) and operating condition parameters (load speed a / b, grid fluctuation c). In stage S2, the switching transient feature vector of the core switching device is extracted from the standardized signal, substituted into the preset mapping relationship, and combined with the current operating condition parameters (load a suddenly changes to b, power grid fluctuation c) to calculate the current dynamic impedance parameter e. The parameter self-tuning rule library is queried to determine the high impedance range to which e belongs. According to the preset logic of the rule library, the correction amount of the current loop proportional coefficient is increased, the correction amount of the integral time constant is shortened, and the correction amount of the modulation wave duty cycle is increased. Then, combined with the degree of load sudden change, the final correction amount of the three control parameters is generated. In the S3 phase, a fusion control architecture with nested speed and current loops is constructed: the outer loop is the speed control loop, which builds a dynamic weight allocation model and initially sets the weight ratios of speed tracking accuracy and device loss control. When the load is detected to change abruptly from a to b (speed deviation exceeds the threshold), the weight ratio of speed tracking accuracy is automatically increased, while the weight ratio of device loss control is decreased. The inner loop is the current control loop, which adopts an adaptive approach law containing deviation feedback and rate adjustment terms. At the same time, the voltage sag of magnitude c is captured by the grid disturbance observation module to generate a disturbance compensation signal. A two-way interaction mechanism between loops is established through a dedicated signal transmission channel. The outer loop transmits the weight adjustment results to the inner loop, and the inner loop feeds back the current regulation status to the outer loop to generate collaborative control commands. In the S4 stage, the speed response after command execution is monitored in real time (such as the delay time for the speed to recover from the steady state corresponding to a to the target speed corresponding to b), current overshoot, and operating losses of core switching devices (operation-related data after execution) to form a performance feedback dataset. This dataset is then input into the device operating condition coupling analysis model to correct the mapping relationship between the dynamic impedance parameter e and the operating condition parameters (load b, grid fluctuation c). At the same time, the interval threshold, double-loop weight allocation coefficient, and sliding mode control reaching law parameters of the parameter self-tuning rule library are optimized.

[0043] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for optimizing the dynamic performance control of a frequency converter, characterized in that, include: S1: Real-time acquisition of inverter operation and device characteristic signals, synchronous acquisition of external operating condition related signals, forming a multi-dimensional operation dataset; based on the multi-dimensional operation dataset, construct a device operating condition coupling analysis model, and preset the mapping relationship between dynamic impedance parameters and switching transient characteristic quantities and operating condition parameters; S2: Extract the switching transient characteristics of the device, calculate the dynamic impedance parameters under the current operating condition based on the mapping relationship; construct a parameter self-tuning rule library, preset the control parameter adjustment thresholds corresponding to different dynamic impedance ranges; generate the current loop proportional coefficient, integral time constant and modulation wave duty cycle correction amount according to the adjustment strategy in the dynamic impedance parameter matching rule library. S3: Build a speed and current integrated control architecture. The outer loop is the speed control loop, which constructs a dynamic weight allocation model and dynamically adjusts the weight ratio according to the speed deviation and the degree of load change. The inner loop is the current control loop, which presets an adaptive approach law and a grid disturbance compensation rule. Establish an inter-loop information exchange mechanism to generate collaborative control commands. S4: Monitor the operation-related data after the command is executed in real time to obtain the performance feedback dataset; input the performance feedback dataset into the device operating condition coupling analysis model to correct the mapping relationship between the dynamic impedance parameter and the operating condition parameter; The parameters for self-tuning rule base threshold, double-loop weight allocation coefficient, and sliding mode control reaching law parameters are optimized.

2. The method according to claim 1, characterized in that, The specific process of acquiring inverter operation and device characteristic signals in real time, and synchronously acquiring external operating condition related signals is as follows: Synchronously capture the current changes, voltage fluctuations, and switching transient waveforms of the inverter output terminal; Simultaneously, the real-time rotation status of the load and the stability and fluctuation trend of the grid voltage are obtained; An adaptive filtering algorithm is used to filter out grid interference and environmental noise, and signal synchronization calibration is performed based on the inverter's operating clock. After removing invalid interference data, amplitude calibration and timing alignment are performed on all signals, and they are integrated into a multi-dimensional running dataset.

3. The method according to claim 1, characterized in that, The specific process for constructing the device operating condition coupling analysis model is as follows: Based on the signal characteristics in the multi-dimensional running dataset, the device characteristic dimension and the operating condition parameter dimension are divided. Establish a correlation analysis framework between the device characteristic dimension and the operating condition parameter dimension; Based on the power electronic conversion principle and the pre-set constraints of the device's thermal conduction characteristics, and incorporating the physical mechanism constraints of the frequency converter operation, a structured coupled analysis model is formed.

4. The method according to claim 1, characterized in that, The specific process of mapping the preset dynamic impedance parameters with the switching transient characteristics and operating parameters is as follows: collect measured data of the device's dynamic impedance under different operating conditions; extract the corresponding switching transient characteristics and operating parameter ranges; classify and organize the measured data using statistical analysis methods; establish corresponding association rules through data fitting; and form a preset mapping relationship based on the association rules.

5. The method according to claim 1, characterized in that, The specific process for extracting the switching transient features of the device is as follows: extract the waveform of the turn-on and turn-off periods of the core switching device from the standardized device feature signal, extract the rising edge slope, falling edge slope, peak value features and duration features of the waveform, use feature normalization processing to eliminate the signal amplitude differences under different operating conditions, and quantize the extracted features into feature vectors.

6. The method according to claim 1, characterized in that, The specific process of calculating the dynamic impedance parameters under the current operating condition based on the mapping relationship is as follows: the extracted switching transient feature vector and the current operating condition parameters are substituted into the preset mapping relationship, and the calculation coefficients are corrected by combining the device static parameters and real-time operating conditions; feature matching and parameter derivation are performed through calculation logic, and the dynamic impedance value corresponding to the current operating condition is output.

7. The method according to claim 1, characterized in that, The specific process of constructing the parameter self-tuning rule library is as follows: divide the interval boundaries according to the rated operating parameters and extreme operating conditions of the frequency converter; divide the interval gradient of dynamic impedance according to the adjustment range of the frequency converter control parameters; set the corresponding adjustment range of current loop proportional coefficient, integral time constant correction range and modulation wave duty cycle adjustment range for each interval gradient, and store the intervals and adjustment rules to form the parameter self-tuning rule library.

8. The method according to claim 1, characterized in that, The specific process of dynamically adjusting the weight ratio is as follows: Set thresholds and judgment criteria according to the performance requirements of the inverter application scenario, and set speed deviation threshold and load change judgment criteria. When the detected speed deviation exceeds the preset threshold or a sudden change in load occurs, the weight ratio corresponding to speed tracking accuracy is increased; when the speed deviation is within the preset threshold range, the weight ratio corresponding to device loss control is increased.

9. The method according to claim 1, characterized in that, The specific process of the preset adaptive approach law is as follows: based on the control accuracy requirements of the current loop, the dynamic response characteristics of the approach law are optimized by incorporating the current change rate feedback term, an approach law expression including a deviation feedback term and a rate adjustment term is designed, the dynamic adjustment range of the approach law parameters is set, and the parameter values ​​are matched in real time according to the magnitude of the current deviation.

10. The method according to claim 1, characterized in that, The specific process of establishing the inter-loop information interaction mechanism is as follows: a signal transmission channel is preset between the speed control loop and the current control loop, and a real-time communication protocol is adopted; the speed control loop transmits the weight allocation result to the current control loop in real time. The current control loop feeds back the current regulation status to the speed control loop, forming a two-way information interaction link.

11. The method according to claim 1, characterized in that, The specific process of correcting the mapping relationship between dynamic impedance parameters and operating parameters is as follows: compare the actual dynamic impedance values ​​in the performance feedback dataset with the predicted values ​​and calculate the deviation; adjust the coefficients using a gradient correction method based on the distribution law of the deviation, and adjust the feature matching coefficients in the mapping relationship to gradually reduce the deviation between the predicted and actual values.

12. The method according to claim 1, characterized in that, The specific process for optimizing the self-tuning rule base threshold, the dual-loop weight allocation coefficient, and the sliding mode control reaching law parameters is as follows: Based on the performance feedback dataset, the sensitivity of each parameter to the control effect is analyzed, and the parameter optimization priority is determined. Parameter optimization ranges are set for different working conditions, and iterative optimization algorithms are used to adjust parameter values. The optimization effect is verified through multiple sets of working condition tests.

Images