A method and system for diagnosing and limiting overvoltage trends in the stator of a doubly-fed converter

Abstract

This invention discloses a method and system for diagnosing and limiting overvoltage trends in a doubly-fed induction generator (DFIG) stator, relating to the fields of power electronics and intelligent control technology. The method includes fitting overvoltage trend curves to the stator A-phase, B-phase, and C-phase voltages to obtain trend parameters of the three-phase voltages over time, and constructing a three-phase cross-voltage trend graph. Based on the three-phase cross-voltage trend graph, it determines phase asynchrony overvoltage trends. Based on the rotor speed range and grid-connected operating status, it executes adaptive voltage limiting control. The method of this invention, by constructing a three-phase voltage trend graph and identifying phase asynchrony overvoltage trends, combined with speed and grid-connected status, achieves trend-driven adaptive voltage limiting control and feedback correction, improving the dynamic updating capability of control parameters and the adaptability of voltage limiting strategies.

Description

Technical Field

[0001] This invention relates to the field of power electronics and intelligent control technology, specifically to a method and system for diagnosing and limiting overvoltage trends in the stator of a doubly-fed converter. Background Technology

[0002] With the large-scale grid connection of wind power systems, doubly-fed induction generators have become the mainstream structural form for medium and large wind turbine generators due to their excellent grid connection regulation capabilities and control flexibility. In the doubly-fed structure, the converter, as the core control component, is responsible for regulating the rotor current and power exchange with the grid. Especially under high dynamic conditions, the response speed and adaptability of the converter control strategy directly determine the stability of the system operation. With the stator side directly connected to the grid, the voltage is affected by grid disturbances, changes in unit status, and control strategies, often resulting in dynamic fluctuations, phase-to-phase differential evolution, and short-term overshoot. In actual operation, scenarios such as low-voltage ride-through, rapid grid connection, and reconnection are becoming increasingly frequent, and the voltage change process exhibits a significant nonlinear trend. The three-phase voltage response shows asynchronous characteristics in amplitude and time, becoming a potential risk point that needs attention in wind power systems.

[0003] Current technologies for stator voltage control typically rely on threshold settings and amplitude judgments, primarily depending on single-point sampling or voltage extreme values ​​to trigger voltage limiting operations. This approach struggles to capture trend-based evolution, especially when phase-to-phase trend changes occur at inconsistent rates, failing to promptly reflect potential overvoltage risks. While some technologies attempt to improve response capabilities by introducing model predictive control algorithms, most still rely on fixed sampling mechanisms and modulation strategies. Internal PWM modulation in converters mostly employs preset control methods, lacking dynamic matching capabilities to the current system operating state. Furthermore, the ineffective switching of modulation parameters across different operating phases results in a lack of flexible adjustment during voltage limiting.

[0004] Typically, single-wheel control logic is used, which lacks the ability to extract deviations from voltage response results and provide feedback corrections. It also lacks a sustainable adaptive closed-loop optimization mechanism. Voltage control strategies have certain limitations in terms of trend recognition accuracy, response timeliness, and control parameter update logic, making it difficult to meet the voltage control requirements of doubly-fed converters under complex grid dynamic conditions. Summary of the Invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by this invention is that existing stator voltage control methods have problems such as: inability to accurately identify phase asynchrony overvoltage trends, lack of real-time modeling means for three-phase voltage evolution trends, strong voltage limiting response and lack of adaptability to operating conditions, inability of control parameters to achieve closed-loop correction based on voltage response results, and low adaptive adjustment capability under complex dynamic operating conditions of wind power systems. The invention also addresses how to identify, determine and dynamically limit voltage linkage control of voltage trends under different operating stages.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for diagnosing and limiting overvoltage trends in a doubly-fed converter stator, comprising fitting overvoltage trend curves to the stator A-phase, B-phase, and C-phase voltages respectively, obtaining trend parameters of the three-phase voltages changing over time, and constructing a three-phase cross-voltage trend map; determining the phase asynchrony overvoltage trend based on the three-phase cross-voltage trend map; and performing adaptive voltage limiting control based on the rotor speed range and grid-connected operation status.

[0008] As a preferred embodiment of the stator overvoltage trend diagnosis and voltage limiting method for doubly-fed converters described in this invention, the overvoltage trend curve fitting includes: continuously sampling the three-phase voltage signals; using a third-order polynomial fitting model to perform least-squares fitting on the sampled data; obtaining the trend parameters of the three-phase voltage changing with time by calculating the first derivative of the polynomial at the current moment; setting a voltage anomaly point deviation threshold; and automatically removing sampled data whose voltage values ​​deviate from the voltage anomaly point deviation threshold.

[0009] As a preferred embodiment of the stator overvoltage trend diagnosis and voltage limiting method for doubly fed converters described in this invention, the three-phase cross voltage trend graph includes: after each round of trend parameter update, comparing the trend parameter differences between the three phases, marking the trend parameter with the largest parameter difference, and storing the results in chronological order.

[0010] As a preferred embodiment of the stator overvoltage trend diagnosis and voltage limiting method for doubly fed converters described in this invention, the phase asynchrony overvoltage trend includes: continuously monitoring the trend parameter update results of the three-phase voltage; if the trend parameter update rate of any phase voltage is significantly greater than that of the other two phases, and the trend parameter difference continues to exceed a set time window, the current phase is marked as an abnormal trend phase, triggering a trend warning state.

[0011] As a preferred embodiment of the doubly fed converter stator overvoltage trend diagnosis and voltage limiting method described in this invention, the adaptive voltage limiting control includes: after detecting a phase asynchrony overvoltage trend, judging the trend warning level, dynamically adjusting the voltage sampling coefficient and PWM modulation strategy, and feedback correction of the voltage limiting control parameters.

[0012] As a preferred embodiment of the stator overvoltage trend diagnosis and voltage limiting method for doubly-fed converters described in this invention, the dynamic adjustment of the voltage sampling coefficient and PWM modulation strategy includes: a sampling calibration module adjusting the stator PT voltage sampling coefficient, compressing the original sampled value by setting a multiplier, maintaining the PT coefficient stable under grid-connected operation, and recording the trend offset; receiving trend parameters to determine the warning level signal; if the trend level is level one, maintaining the original sinusoidal PWM modulation mode, limiting the amplitude; if the trend level is level two or above, switching to space vector PWM, and adjusting the upper and lower limits of the PWM modulation duty cycle.

[0013] As a preferred embodiment of the doubly fed converter stator overvoltage trend diagnosis and voltage limiting method described in this invention, the feedback correction includes recording the actual stator voltage response value and the corresponding trend parameters, calculating a correction factor based on the voltage difference before and after control and the trend change direction to update the upper and lower limits of the PWM modulation duty cycle, the voltage sampling coefficient table, and the trend identification criterion for the next round, and constructing a closed-loop self-learning control mechanism.

[0014] Another objective of this invention is to provide a stator overvoltage trend diagnosis and voltage limiting system for a doubly-fed converter. This system can achieve closed-loop logic processing from voltage trend identification to control parameter correction by constructing a functional coupling relationship between a trend identification module, a trend judgment module, and an adaptive voltage limiting control module, and by combining multi-condition information interaction with the operating status. This solves the problems of current voltage control systems under complex operating conditions, such as lag in trend identification, misjudgment of phase-to-phase asynchronous evolution, inability to coordinate modulation strategies, and lack of synchronous updates between voltage sampling and control parameters. This improves the adaptability and dynamic response capability of the voltage limiting strategy.

[0015] As a preferred embodiment of the stator overvoltage trend diagnosis and voltage limiting system for a doubly-fed converter described in this invention, the system includes: a trend identification module, a trend determination module, and an adaptive voltage limiting control module. The trend identification module continuously samples and models the stator A-phase, B-phase, and C-phase voltages to construct a three-phase cross-voltage trend graph and update the trend parameters. The trend determination module identifies phase-asynchronous overvoltage trends based on the three-phase cross-voltage trend graph, dynamically determines the trend level by analyzing the rate of change of trend parameters and the phase-to-phase difference, and triggers early warning states and control flow calls. The adaptive voltage limiting control module comprehensively selects a voltage limiting strategy based on the trend determination result, rotor speed range, and grid-connected operation status, dynamically adjusts the voltage sampling coefficient and PWM modulation strategy, and performs parameter feedback correction based on the voltage response.

[0016] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement a method for diagnosing and limiting overvoltage trends in a doubly-fed converter stator.

[0017] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a method for diagnosing and limiting overvoltage trends in a doubly-fed converter stator.

[0018] The beneficial effects of this invention are as follows: The stator overvoltage trend diagnosis and voltage limiting method for doubly-fed converters provided by this invention models the evolution trend of the three-phase stator voltage through trend identification, thereby improving the ability to predict overvoltage risks; trend determination can distinguish phase asynchrony characteristics, enhancing the identification accuracy of non-consistent overvoltage trends; adaptive voltage limiting control achieves dynamic matching between the voltage limiting control strategy and the operating state through linkage voltage sampling adjustment and PWM modulation strategy switching; the feedback correction mechanism can correct control parameters based on voltage response, realizing closed-loop self-updating of the trend judgment and control process. This invention achieves better results in terms of voltage trend identification accuracy, voltage control response flexibility, and adaptive update capability of control parameters. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is an overall flowchart of a doubly fed converter stator overvoltage trend diagnosis and voltage limiting method provided in Embodiment 1 of the present invention. Detailed Implementation

[0021] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0022] Example 1, referring to Figure 1 As an embodiment of the present invention, a method for diagnosing and limiting overvoltage trends in a doubly-fed converter stator is provided, comprising:

[0023] S1: Fit the overvoltage trend curves of the stator phases A, B, and C respectively to obtain the trend parameters of the three-phase voltage change over time, and construct the three-phase cross voltage trend map.

[0024] Furthermore, the overvoltage trend curve fitting includes continuously sampling the three-phase voltage signal, using a third-order polynomial fitting model to perform least-squares fitting on the sampled data, obtaining the trend parameters of the three-phase voltage changing with time by calculating the first derivative of the polynomial at the current moment, setting a voltage anomaly point deviation threshold, and automatically removing sampled data whose voltage values ​​deviate from the voltage anomaly point deviation threshold.

[0025] Furthermore, the three-phase cross voltage trend graph includes comparing the trend parameter differences between the three phases after each round of trend parameter updates, marking the trend parameter with the largest parameter difference, and storing the results in chronological order.

[0026] It should be noted that the automatic removal of sampled data whose voltage values ​​deviate from the voltage anomaly deviation threshold includes: continuous sampling with a period of no more than 20 milliseconds; setting a dynamic deviation tolerance based on engineering experience and simulation data, for example, setting it to ±5% of the expected voltage value; subtracting the actual voltage value of each sample point from the calculated value of the third-order polynomial fitting curve at the corresponding time point to form a voltage difference, i.e., the trend slope; sampled data whose absolute value of the trend slope difference is greater than the voltage anomaly deviation threshold will be marked as an anomaly and removed from the current round of fitting data to avoid distorting the shape of the fitting curve.

[0027] It should also be noted that by introducing a trend extraction method based on a third-order polynomial fitting model and first-order derivative calculation, and because the third-order polynomial fitting model can better describe nonlinear trends than first-order and second-order fitting, it is suitable for the complex voltage fluctuation process in wind power systems. Combined with a dynamic deviation tolerance mechanism to identify and eliminate abnormal voltage data, high-precision modeling of the stator three-phase voltage change trend over time is achieved, capturing the change trend of three-phase voltage on a microsecond time scale, and dynamically filtering outliers caused by transient disturbances or measurement errors. This effectively improves the stability and robustness of trend fitting, providing a more reliable input basis for subsequent phase-to-phase trend difference identification and trend early warning judgment.

[0028] S2: Based on the three-phase cross voltage trend graph, determine the phase asynchrony overvoltage trend.

[0029] Furthermore, the phase asynchrony overvoltage trend includes continuously monitoring the trend parameter update results of the three-phase voltage. If the trend parameter update rate of any phase voltage is significantly greater than that of the other two phases, and the trend parameter difference continues to exceed the set time window, the current phase is marked as the trend abnormal phase, triggering the trend warning state.

[0030] It should be noted that the marker of an abnormal trend is represented as follows:

[0031]

[0032] Δk i(t)=k i (t)-k i (t-δt)

[0033]

[0034] in, represents the arithmetic mean of the trend slopes of the other two phases (excluding phase i) at time t; represents the average trend of the voltage changes of the other two phases at the current time, used as the benchmark value for phase-to-phase comparison in the three-phase cross-voltage trend graph; t represents the current calculation time point, corresponding to the instantaneous time point within each trend parameter update cycle; j represents any one of the two phases different from phase i; i represents the voltage phase currently being analyzed; k j (t) represents the trend slope of the j-th phase at time t; Δk i (t) represents the change in the trend slope of the i-th phase between two adjacent sampling periods, reflecting the trend update rate (unit: V / ms). 2 ); δt represents the continuous sampling period of the three-phase voltage signal, set to within 10ms to 20ms; k i (t) represents the trend slope of the i-th phase at time t; Ψ i (t) represents the trend anomaly judgment function, which is a binary logic function (True / False) used to determine whether the i-th phase (A / B / C) of the stator is in a phase asynchrony overvoltage trend state at time t. It is 1 when the condition is met and 0 when the condition is not met. This represents the trend deviation criterion, indicating whether the deviation of the trend slope of phase i from the arithmetic mean of the slopes of the other two phases exceeds the set dynamic deviation tolerance θ1. θ1 is set based on engineering experience and simulation data; Δk i (t)>θ2 represents the trend update rate criterion, indicating whether the time update rate of the voltage trend slope of the i-th phase is higher than the update rate θ2, where θ2 is set based on engineering experience and simulation data; t0 represents the start time of the current trend anomaly identification window; Δt stable This represents the minimum duration window required to confirm an abnormal trend. Based on the sliding time window integration mechanism, when the difference in trend parameters continuously exceeds the set time window and meets the time accumulation integration condition, the i-th phase is identified as an abnormal trend phase.

[0035] It should also be noted that by introducing a joint criterion of trend slope deviation and update rate, combined with a sliding time window integration mechanism, the identification of phase asynchronous overvoltage trends can be achieved. This enables the stable extraction of abnormal behavior in the evolution of three-phase voltage trends, providing a reliable early warning basis for subsequent voltage limiting control.

[0036] S3: Based on the rotor speed range and grid-connected operation status, perform adaptive voltage limiting control.

[0037] Furthermore, the adaptive voltage limiting control includes, after detecting a phase asynchrony overvoltage trend, classifying the trend warning level, dynamically adjusting the voltage sampling coefficient and PWM modulation strategy, and providing feedback correction to the voltage limiting control parameters.

[0038] It should be noted that the dynamic adjustment of voltage sampling coefficient and PWM modulation strategy includes: the sampling calibration module adjusts the stator PT voltage sampling coefficient, compresses the original sampled value by setting a multiplier, keeps the PT coefficient stable under grid-connected operation, and records the trend offset; it receives trend parameters to determine the warning level signal. If the trend level is level one, the original sinusoidal PWM modulation mode is maintained, limited to amplitude; if the trend level is level two or above, it switches to space vector PWM and adjusts the upper and lower limits of the PWM modulation duty cycle.

[0039] It should be noted that the feedback correction includes recording the actual stator voltage response value and the corresponding trend parameters, calculating the correction factor based on the voltage difference before and after control and the direction of trend change, and using it to update the upper and lower limits of the PWM modulation duty cycle, the voltage sampling coefficient table, and the trend identification criteria for the next round, thus constructing a closed-loop self-learning control mechanism.

[0040] It should also be noted that the calculation of the correction factor includes extracting the difference between the current three-phase voltage trend parameter and the trend parameter of the previous period in each time period to obtain the trend change amplitude, collecting the actual stator voltage response value under the current control output, comparing it with the target voltage value set in the current period to obtain the voltage response error, and then weighting and synthesizing a trend and response coupled correction factor according to the trend change amplitude and the voltage response error respectively.

[0041] It should also be noted that the adaptive voltage limiting control updates the upper and lower limits of the PWM modulation duty cycle, the voltage sampling coefficient table, and the trend identification criteria for the next cycle, realizing a closed-loop correlation between trend identification, control execution, and parameter optimization. The correction process is automatically triggered after each trend warning cycle, realizing self-learning adaptive updating of the voltage limiting control parameters.

[0042] Example 2, an embodiment of the present invention, provides a stator overvoltage trend diagnosis and voltage limiting system for a doubly fed converter, including a trend identification module, a trend determination module, and an adaptive voltage limiting control module.

[0043] The trend recognition module is used to continuously sample and model the voltage trends of stator phases A, B, and C, construct a three-phase cross voltage trend map, and update the trend parameters.

[0044] The trend determination module is used to identify phase asynchronous overvoltage trends based on the three-phase cross voltage trend graph. By analyzing the rate of change of trend parameters and the phase difference, it dynamically determines the trend level and triggers early warning status and control process calls.

[0045] The adaptive voltage limiting control module is used to comprehensively select the voltage limiting strategy based on the trend judgment result, rotor speed range and grid-connected operation status, dynamically adjust the voltage sampling coefficient and PWM modulation strategy, and perform parameter feedback correction in combination with voltage response.

Claims

1. A method for diagnosing and limiting overvoltage trends in a doubly-fed converter stator, characterized in that, include: Overvoltage trend curves were fitted to the stator A-phase, B-phase, and C-phase voltages respectively to obtain the trend parameters of the three-phase voltages changing with time, and a three-phase cross voltage trend map was constructed. Based on the three-phase cross voltage trend graph, determine the phase asynchrony overvoltage trend; Based on the rotor speed range and grid-connected operation status, adaptive voltage limiting control is executed; Overpressure trend curve fitting includes, The three-phase voltage signal is continuously sampled, and the sampled data is fitted with a third-order polynomial fitting model using least squares. The trend parameters of the three-phase voltage change over time are obtained by calculating the first derivative of the polynomial at the current time. A voltage anomaly deviation threshold is set, and sampled data whose voltage values ​​deviate from the voltage anomaly deviation threshold are automatically removed. The three-phase cross voltage trend graph includes... After each round of trend parameter updates, compare the trend parameter differences among the three phases, mark the trend parameter with the largest difference, and store the results in chronological order. Phase asynchrony overvoltage trends include, The trend parameter update results of the three-phase voltage are continuously monitored. If the trend parameter update rate of any phase voltage is significantly greater than that of the other two phases and the difference in trend parameters continues to exceed the set time window, the current phase is marked as the abnormal trend phase, and the trend warning state is triggered. Performing adaptive voltage limiting control includes, After detecting a phase asynchrony overvoltage trend, the trend warning level is judged, the voltage sampling coefficient and PWM modulation strategy are dynamically adjusted, and the voltage limiting control parameters are corrected by feedback.

2. The method for diagnosing and limiting overvoltage trends in a doubly-fed converter stator as described in claim 1, characterized in that: The dynamically adjusted voltage sampling coefficient and PWM modulation strategy include, The sampling calibration module adjusts the stator PT voltage sampling coefficient, compresses the original sampled value by setting a multiplier, keeps the PT coefficient stable under grid-connected operation, and records the trend offset. The system receives trend parameters to determine the warning level. If the trend level is level one, it maintains the original sinusoidal PWM modulation mode, limiting the amplitude. If the trend level is level two or above, it switches to space vector PWM and adjusts the upper and lower limits of the PWM modulation duty cycle.

3. The method for diagnosing and limiting overvoltage trends in a doubly-fed converter stator as described in claim 2, characterized in that: The feedback correction includes, Record the actual stator voltage response value and the corresponding trend parameters. Based on the voltage difference before and after control and the direction of trend change, calculate the correction factor to update the upper and lower limits of the PWM modulation duty cycle, the voltage sampling coefficient table, and the trend identification criteria for the next round, and construct a closed-loop self-learning control mechanism.

4. A stator overvoltage trend diagnosis and voltage limiting system for a doubly-fed converter, employing the stator overvoltage trend diagnosis and voltage limiting method for a doubly-fed converter as described in any one of claims 1 to 3, characterized in that: Includes a trend recognition module, a trend determination module, and an adaptive pressure limiting control module; The trend recognition module is used to continuously sample and model the voltages of stator phases A, B, and C, construct a three-phase cross voltage trend map, and update the trend parameters. The trend determination module is used to identify phase asynchronous overvoltage trends based on the three-phase cross voltage trend graph. By analyzing the rate of change of trend parameters and the phase difference, it dynamically determines the trend level and triggers early warning status and control process calls. The adaptive voltage limiting control module is used to comprehensively select a voltage limiting strategy based on the trend judgment result, rotor speed range and grid-connected operation status, dynamically adjust the voltage sampling coefficient and PWM modulation strategy, and perform parameter feedback correction in combination with voltage response.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the doubly fed converter stator overvoltage trend diagnosis and voltage limiting method as described in any one of claims 1 to 3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the doubly fed converter stator overvoltage trend diagnosis and voltage limiting method as described in any one of claims 1 to 3.

Images