Low-temperature pressure compensation control method for yaw brake system of wind turbine in cold region

Abstract

This invention discloses a low-temperature pressure compensation control method for the yaw braking system of a wind turbine in cold regions, relating to the field of wind turbine control technology. The method includes acquiring multi-source data such as ambient temperature, hydraulic oil temperature, braking pressure, and yaw angular velocity; fusing temperature information to generate system pressure loss assessment parameters; using yaw dynamic data and a braking performance reference model to calculate the theoretical required pressure under standard operating conditions as a dynamic benchmark; comparing this theoretical pressure with the actual pressure and combining it with the assessment parameters to calculate a precise pressure compensation amount; and using an adaptive controller to drive the hydraulic unit to achieve dynamic adjustment of the braking pressure. This method achieves accurate diagnosis and adaptive dynamic compensation of low-temperature pressure loss, effectively improving the reliability and control accuracy of the yaw braking system of a wind turbine in cold regions under low-temperature environments.

Description

Technical Field

[0001] This invention belongs to the field of wind turbine generator control technology, specifically a low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine generator. Background Technology

[0002] In cold regions, the hydraulic oil in the yaw braking system of wind turbines experiences a significant increase in viscosity due to low temperatures. This increases flow resistance in the braking lines, delays system response, and causes significant pressure loss. The braking system cannot output the designed braking torque under low-temperature conditions. To address this, the hydraulic oil in the yaw braking system of wind turbines experiences a significant increase in viscosity due to low temperatures. Therefore, based on the collected ambient temperature or hydraulic oil temperature, and using a preset temperature-pressure correlation curve or lookup table, the pressure command of the braking system is statically adjusted. This method uses a single temperature variable as the sole or primary basis for compensation calculations.

[0003] Existing static compensation methods have two main drawbacks. First, the pressure target referenced for compensation is a fixed value or a finite range, failing to incorporate the dynamic load information of the unit's real-time yaw movements. Second, the pressure loss caused by low temperatures is coupled with the actual pressure demand of the yaw load; static methods cannot effectively separate these two factors, leading to a disconnect between the compensation benchmark and real-time operating conditions. Third, relying on data from a single temperature point makes it difficult to comprehensively assess the overall state of the hydraulic braking circuit affected by low temperatures. Fourth, the interaction between the ambient cold source and the oil temperature changes generated within the system due to operation is not quantitatively considered, making it impossible to accurately characterize the degree of real-time performance degradation of the system.

[0004] A control method is needed to establish a dynamic pressure reference that changes in real time with yaw load in order to accurately isolate cryogenic losses; and a characteristic parameter that integrates multi-source temperature information and real-time system pressure status needs to be constructed to accurately diagnose the degree of cryogenic degradation of the system, thereby achieving precise adaptive pressure compensation. Summary of the Invention

[0005] This invention aims to solve at least one of the technical problems existing in the prior art; Therefore, this invention proposes a low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine, comprising: During the operation of the wind turbine, a multi-source real-time monitoring data set of the yaw braking system is acquired. The multi-source real-time monitoring data set includes ambient temperature data collected by the ambient temperature sensor, hydraulic oil temperature data collected by the hydraulic oil temperature sensor, actual pressure data of the brake pipeline collected by the brake pressure sensor, and angular velocity data of the moving body collected by the yaw state sensor. Based on the ambient temperature data and the hydraulic oil temperature data, the comprehensive temperature characteristic value of the hydraulic oil is calculated, and combined with the actual pressure data of the brake line, the pressure loss assessment parameters of the brake system under the current low temperature conditions are generated. Using the angular velocity data of the moving body and the actual pressure data of the braking line, the theoretical required pressure value under standard temperature conditions is calculated through a pre-constructed braking performance reference model. By comparing the theoretical required pressure value with the actual pressure data of the brake line, and combining the pressure loss assessment parameters of the brake system, the pressure compensation amount required to compensate for the effects of low temperature is calculated. Based on the pressure compensation amount, a corresponding hydraulic system control command is generated by a pre-trained adaptive compensation controller. The hydraulic system control command is sent to the hydraulic actuator, which drives the hydraulic actuator to dynamically adjust the braking pressure of the yaw brake, thereby achieving pressure compensation control in low-temperature environments.

[0006] Furthermore, based on the ambient temperature data and the hydraulic oil temperature data, a comprehensive temperature characteristic value of the hydraulic oil is calculated. Combined with the actual pressure data of the brake lines, pressure loss assessment parameters for the braking system under current low-temperature conditions are generated, including: The ambient temperature data and the hydraulic oil temperature data are input into a temperature coupling model for data fusion. The temperature coupling model is constructed based on the heat exchange law between ambient temperature and hydraulic oil temperature. The temperature coupling model outputs a comprehensive temperature characteristic value that characterizes the overall thermal state of the hydraulic oil. The comprehensive temperature characteristic value and the actual pressure data of the brake line are input into the pressure loss evaluation function, which describes the quantitative relationship between the effect of hydraulic oil viscosity change on line pressure loss under a specific comprehensive temperature characteristic value. The pressure loss assessment function is used to calculate and output the pressure loss assessment parameters of the braking system.

[0007] Furthermore, using the angular velocity data of the moving body and the actual pressure data of the braking line, the theoretical required pressure value under standard temperature conditions is calculated through a pre-constructed braking performance reference model, including: The angular velocity data of the moving body and the actual pressure data of the braking line are input into the braking performance reference model. The braking performance reference model reflects the relationship between the braking pressure required to achieve specific yaw rate control and the measured pressure under standard temperature conditions. The braking performance reference model is based on the input current actual pressure data of the braking line and the angular velocity data of the moving body, and reversely calculates the braking pressure required to achieve the same braking performance under standard oil temperature and viscosity conditions. The braking pressure obtained by the reverse calculation is output as the theoretical required pressure value.

[0008] Furthermore, by comparing the theoretical required pressure value with the actual pressure data of the brake line, and combining this with the brake system pressure loss assessment parameters, the pressure compensation amount required to compensate for the effects of low temperature is calculated, including: Calculate the pressure difference between the theoretical required pressure value and the actual pressure data of the brake line; The pressure difference is correlated with the pressure loss assessment parameters of the braking system to distinguish the pressure loss component caused by increased viscosity at low temperature and the pressure deviation component caused by other factors. The pressure loss component caused by the increase in viscosity at low temperature is multiplied by a preset safety margin coefficient, and the product is used as the basic pressure compensation amount. Based on the sign and magnitude of the pressure deviation component caused by other factors, the basic pressure compensation amount is finely adjusted and corrected, and the pressure compensation amount is finally output.

[0009] Furthermore, based on the pressure compensation amount, a corresponding hydraulic system control command is generated through a pre-trained adaptive compensation controller, including: The hydraulic system control commands are used to adjust the output pressure and flow rate of the hydraulic pump; The pressure compensation amount is input into the input interface of the adaptive compensation controller, which internally includes a feedforward control channel and a feedback adjustment channel. In the feedforward control channel, based on the value and trend of the pressure compensation amount, a preliminary pressure control command is generated by looking up a pre-stored control quantity mapping table. The initial pressure control command is sent to the hydraulic system, and the actual pressure data of the brake line after execution is collected simultaneously and input as a feedback signal into the feedback adjustment channel. The feedback adjustment channel calculates the adjustment control command based on the deviation between the target pressure and the actual feedback pressure using a built-in proportional-integral algorithm. The initial pressure control command and the adjustment control command are superimposed to synthesize the final hydraulic system control command.

[0010] Furthermore, the hydraulic system control command is sent to the hydraulic actuator to drive the hydraulic actuator to dynamically adjust the braking pressure of the yaw brake, including: The hydraulic system control commands are transmitted to the variable frequency drive of the hydraulic pump via a fieldbus. The variable frequency drive adjusts the operating frequency of the motor according to the pressure set value in the command. The change in motor frequency drives the hydraulic pump to change its output flow and pressure, and the adjusted hydraulic oil is delivered to the brake cylinder of the yaw brake via the control valve group. A pressure sensor installed at the brake cylinder monitors changes in brake pressure in real time and transmits the monitored pressure data back to the control system. The control system compares the returned braking pressure data with the theoretical required pressure value. If the set target is not met, it reacquires multi-source real-time monitoring data and sequentially executes the pressure loss assessment, theoretical required pressure calculation, pressure compensation determination, and control command generation process to generate new hydraulic system control commands until the braking pressure reaches the expected range.

[0011] Furthermore, the method for constructing the pre-built braking performance reference model includes: In a standard temperature laboratory, angular velocity data of the moving body of the yaw braking system of the wind turbine under different braking pressures were collected to form a standard operating condition dataset. The standard working condition dataset is analyzed to establish a mathematical relationship between braking pressure and yaw deceleration effect. This mathematical relationship constitutes the core mapping relationship of the braking performance reference model. Other secondary factors affecting braking performance, including the range of variation in the brake disc friction coefficient and the degree of slight mechanical wear, are introduced as correction factors into the core mapping relationship. Through extensive experimental data, the parameters of the model after introducing the correction factor were identified and optimized, and the complete structure and parameters of the braking performance reference model were finally determined.

[0012] Furthermore, the training method for the pre-trained adaptive compensation controller includes: A semi-physical simulation test platform including a hydraulic system, sensors, and controllers was built to simulate a cold-region low-temperature environment. Under simulated low-temperature conditions, pressure compensation requirements with different amplitudes and rates of change are applied, and the hydraulic system control commands output by the adaptive compensation controller and their corresponding actual pressure regulation effects are recorded. With the optimization goal of quickly eliminating pressure deviation and ensuring a smooth control process without overshoot, machine learning algorithms are used to iteratively optimize the parameters inside the adaptive compensation controller based on the recorded input and output data. When the adaptive compensation controller's control performance on the test dataset meets the preset accuracy index, training is completed and its internal parameters are solidified.

[0013] Furthermore, the temperature coupling model is constructed based on the heat exchange law between ambient temperature and hydraulic oil temperature, including: Obtain the structural parameters of the wind turbine hydraulic system, including hydraulic tank volume, pipeline length and diameter, and material thermal conductivity; Based on the structural parameters and combined with the principles of fluid mechanics and heat transfer, a differential equation describing the dynamic heat exchange relationship between ambient temperature and hydraulic oil temperature is established. Historical operating data is used to identify and calibrate the thermal resistance and heat capacity parameters in the differential equation to ensure that the model output matches the measured oil temperature data. The real-time collected ambient temperature data is used as the model input to solve the differential equation and obtain the dynamic response of hydraulic oil temperature change. The dynamic response is weighted and averaged to output the comprehensive temperature characteristic value that characterizes the overall thermal state of the hydraulic oil.

[0014] Furthermore, the standard operating condition dataset is analyzed to establish a mathematical relationship between braking pressure and yaw deceleration effect. This mathematical relationship constitutes the core mapping relationship of the braking performance reference model, including: Extract multiple sets of braking pressure values ​​and corresponding yaw rate of change data from the standard operating condition dataset; The least squares method was used to fit the extracted data to obtain the basic functional relationship between braking pressure and yaw deceleration. Based on the rotational inertia parameters of the wind turbine, the yaw deceleration is converted into yaw braking torque, and a linear mapping between braking pressure and braking torque is established. By introducing the brake's effective radius and the number of friction pads as scale factors, the linear mapping is corrected to obtain a mathematical relationship that takes into account the specific brake structural parameters. The goodness of fit of the mathematical relationship is verified by residual analysis. When the coefficient of determination is greater than a preset threshold, the mathematical relationship is confirmed as the core mapping relationship.

[0015] Compared with the prior art, the beneficial effects of the present invention are: By using a pre-built braking performance reference model, the real-time monitored yaw rate and actual brake line pressure are used as inputs to inversely map the theoretical pressure required to complete the current braking action under standard temperature conditions. This technology establishes a dynamic performance benchmark that changes in real time with the yaw load, rather than a fixed pressure value. This allows for the clear separation and quantification of performance degradation caused by low temperatures from the overall performance of the current system, enabling compensation control to directly target the gap with ideal performance at normal temperature. This improves the pertinence and scientific nature of the compensation, avoiding the problem of being out of touch with actual load requirements when compensating based on fixed values.

[0016] By fusing ambient temperature and hydraulic oil temperature data, a comprehensive temperature characteristic value is generated. This characteristic value is then combined with real-time brake line pressure data to generate a braking system pressure loss assessment parameter. This parameter integrates external cold source input and internal system operating temperature rise, and is correlated with the system's instantaneous pressure performance. This creates a more comprehensive and accurate characteristic index that characterizes the overall state of the entire hydraulic braking circuit at low temperatures, transcending the limitations of a single temperature point. Compensation decisions based on this parameter enable the control system to move from simply "sensing temperature" to "diagnosing system state." The magnitude of the compensation directly corresponds to the currently assessed degree of system loss, thus achieving a shift from coarse empirical compensation to precise state-based compensation. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine according to the present invention. Figure 2 A flowchart for calculating pressure compensation; Figure 3 A thermogram showing the coupling influence coefficients of environmental temperature and pressure loss assessment parameters; Figure 4 The curve showing the relationship between pressure compensation and control commands; Figure 5 The curves showing the efficiency and stability of the adaptive compensation controller as a function of ambient temperature. Detailed Implementation

[0018] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. 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.

[0019] See Figure 1The low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine provided by this invention has the following overall implementation scheme: During the operation of the wind turbine, a multi-source real-time monitoring data set of the yaw braking system is acquired. This multi-source real-time monitoring data set includes ambient temperature data collected by an ambient temperature sensor, hydraulic oil temperature data collected by a hydraulic oil temperature sensor, actual pressure data of the brake pipeline collected by a brake pressure sensor, and angular velocity data of the moving body collected by a yaw state sensor. Based on the ambient temperature data and hydraulic oil temperature data, the comprehensive temperature characteristic value of the hydraulic oil is calculated, and combined with the actual pressure data of the brake pipeline, pressure loss assessment parameters of the braking system under the current low-temperature operating conditions are generated. Using the angular velocity data of the moving body and the actual pressure data of the brake pipeline, the theoretical required pressure value under standard temperature operating conditions is calculated through a pre-constructed braking performance reference model. The theoretical required pressure value is compared with the actual pressure data of the brake pipeline, and combined with the pressure loss assessment parameters of the braking system, the pressure compensation amount required to compensate for the effects of low temperature is calculated. Based on the pressure compensation amount, the corresponding hydraulic system control command is generated through a pre-trained adaptive compensation controller. The hydraulic system control command is sent to the hydraulic actuator, which drives the hydraulic actuator to dynamically adjust the braking pressure of the yaw brake, thereby achieving pressure compensation control in low-temperature environments.

[0020] In one embodiment of the present invention, the generation of comprehensive temperature characteristic values ​​and pressure loss assessment parameters is involved, using a wind turbine generator set located in a low-temperature environment as an example. The hydraulic system structural parameters of the wind turbine generator set are known: the hydraulic oil tank volume is 500 liters, the main pressure pipeline length is 15 meters and the inner diameter is 20 millimeters, and the shell material of the pipeline and the oil tank is carbon steel with a thermal conductivity of 50 W / m / Kelvin. During the operation of the wind turbine generator set, an ambient temperature sensor located outside the nacelle continuously collects ambient temperature data, a hydraulic oil temperature sensor installed inside the hydraulic oil tank continuously collects hydraulic oil temperature data, and the actual pressure data of the brake pipeline is collected by a pressure sensor installed at the brake inlet. In some embodiments, the construction of a temperature coupling model requires the above-mentioned hydraulic system structural parameters. Based on the principles of fluid mechanics and heat transfer, a differential equation describing the dynamic heat exchange relationship between the ambient temperature data and the hydraulic oil temperature data is established. The form of this differential equation is: in: This represents the hydraulic oil temperature data at time t. This represents the ambient temperature data at time t. This indicates the rate of change of hydraulic oil temperature over time. The time constant is the value of the total thermal resistance. With total heat capacity The product ( Using historical operating data recorded from a complete operating year of the wind turbine generator set, the thermal resistance in the differential equation is analyzed. With heat capacity The parameters are identified and calibrated so that the root mean square error between the hydraulic oil temperature change curve output by the temperature coupling model and the measured hydraulic oil temperature data curve is less than 0.5 Kelvin.

[0021] In practical implementation, real-time collected ambient temperature data is used as input to the temperature coupling model. This data is then substituted into the differential equations whose parameters have been identified and solved to obtain a dynamic response sequence of hydraulic oil temperature changes over a future period. The calculated dynamic response sequence is then weighted and averaged, with higher weights assigned to recent predicted temperatures, resulting in a comprehensive temperature characteristic value that represents the overall thermal state of the hydraulic oil. Comprehensive temperature characteristic value It is not an instantaneous measurement, but reflects the average thermal state of the hydraulic oil in the current and short-term future.

[0022] Optionally, the pressure loss assessment function uses a composite temperature characteristic value. Actual pressure data of brake lines As input, the pressure loss assessment function describes the quantitative relationship between the effect of hydraulic oil viscosity variation on pipeline pressure loss under a specific comprehensive temperature characteristic value. The pressure loss assessment function was obtained through prior bench testing calibration. The comprehensive temperature characteristic value is then used as input. Actual pressure data of the brake line Input the pressure loss assessment function, and the function will output a dimensionless braking system pressure loss assessment parameter after calculation. Braking system pressure loss assessment parameters The larger the value, the more significant the system pressure loss due to the increased viscosity of hydraulic oil under the current low-temperature operating conditions.

[0023] It is understandable that building a temperature coupling model is a continuous optimization process. As the operating time of the wind turbine generator accumulates, new historical operating data can be periodically used to adjust the thermal resistance. With heat capacity The parameters are re-identified and calibrated to enable the temperature coupling model to adapt to changes in the characteristics of the hydraulic system caused by slight aging or changes in oil properties. (Comprehensive temperature characteristic value) Calculation and braking system pressure loss assessment parameters The generation of this data provides the basic input for subsequent calculations of the precise pressure compensation caused by low temperature.

[0024] In one embodiment of the present invention, the calculation process of the theoretical required pressure value is described using a megawatt-class wind turbine generator operating in winter as an example. Currently, the wind turbine generator is in a yaw-facing wind state. The yaw state sensor collects the angular velocity data ω of the moving body in real time, while the brake pressure sensor collects the actual pressure data of the brake pipeline. A pre-built braking performance reference model has been stored in the control system of the wind turbine generator. The braking performance reference model reflects the relationship between the braking pressure required to achieve a specific yaw rate control and the measured pressure under standard temperature conditions (e.g., 20 degrees Celsius and hydraulic oil viscosity within the standard range).

[0025] In some embodiments, the real-time collected angular velocity data of the moving body Actual pressure data of brake lines Input the braking performance reference model along with the brake lines. The braking performance reference model is based on the input current actual pressure data of the brake lines. and angular velocity data of moving bodies Then, a reverse calculation process is performed. The purpose of the reverse calculation is to determine the braking pressure required to achieve the same braking performance as the current measured operating conditions (with pressure loss due to low temperature) under standard oil temperature and viscosity conditions.

[0026] In practical implementation, the core mapping relationship within the braking performance reference model expresses the quantitative relationship between braking pressure and yaw braking torque under standard operating conditions. This relationship can be expressed as: in: This indicates the angular velocity reached. The required braking torque This represents the braking pressure. The model is based on the currently input angular velocity data of the moving body. Determine the target braking torque The model incorporates the actual pressure data of the brake line currently being input. Considering the impact of pressure loss due to low temperature, the solution is obtained in reverse through the built-in algorithm to satisfy the condition. Braking pressure value The calculated braking pressure value This refers to the theoretical pressure value that should be provided under standard temperature operating conditions.

[0027] Optionally, the back-calculation process can rely on a parameterized function or lookup table. For example, the braking performance reference model stores a database of stable yaw decelerations corresponding to different braking pressures at standard temperatures. The model is based on the current angular velocity data of the moving body. The model calculates its trend, estimates the current yaw deceleration, matches this deceleration value against a database, and then searches for the reference pressure required to produce the same deceleration at standard temperature. Simultaneously, the model uses actual pressure data from the current brake lines. The difference between the estimated baseline pressure and the actual pressure is used to compensate for the loss of pressure transmission efficiency caused by low temperature, and finally the corrected theoretical demand pressure value is output.

[0028] It is understandable that the accuracy of the braking performance reference model directly affects the reliability of the theoretical required pressure value. Therefore, the braking performance reference model needs to be established through rigorous testing and calibration under standard experimental conditions before the wind turbine generator is put into operation. Throughout the entire life cycle of the wind turbine generator, the parameters of the braking performance reference model can be fine-tuned periodically based on actual operating data to compensate for the slow characteristic drift of the braking system caused by long-term use. After outputting the theoretical required pressure value obtained through back-calculation, this value will serve as the benchmark reference for subsequent calculations of low-temperature pressure compensation.

[0029] In one embodiment of the present invention, see [reference] Figure 2 In practical implementation, the specific calculation steps involving pressure compensation will be explained using a unit performing yaw braking as an example. At this point, the control system has already obtained the theoretical required pressure value calculated and output by the braking performance reference model. And the actual pressure data of the brake line collected in real time by sensors. Meanwhile, the system has generated braking system pressure loss assessment parameters based on ambient temperature data and hydraulic oil temperature data. .

[0030] In some embodiments, the theoretical demand pressure value is calculated. Actual pressure data of the brake line The pressure difference between them. Compare this pressure difference with the braking system pressure loss assessment parameters. Correlation analysis was performed to distinguish between the pressure loss component caused by increased viscosity at low temperatures and the pressure deviation component caused by other factors. This distinction relied on a pre-defined allocation function, which used the pressure difference and braking system pressure loss assessment parameters. As input, it outputs two independent pressure components. In specific implementation, the pressure loss component caused by increased viscosity at low temperatures is combined with a preset safety margin coefficient. Multiply them, and the product is used as the basic pressure compensation amount. Safety margin factor. The set value is greater than 1, for example, it can be set to 1.15, to cover model estimation errors and uncertainties in the system's dynamic response. The base pressure compensation is fine-tuned based on the sign and magnitude of the pressure deviation component caused by other factors. If the pressure deviation component is positive, a fine-tuning amount is added to the base pressure compensation; if the pressure deviation component is negative, a fine-tuning amount is subtracted from the base pressure compensation.

[0031] Optionally, the fine-tuning process is performed using a preset correction function, which maps the values ​​of the pressure deviation components to adjustments to the base pressure compensation. The final output pressure compensation is... Calculated using the following formula: in: This represents the final amount of stress compensation. This represents the pressure loss component caused by increased viscosity at low temperatures. This represents the preset safety margin coefficient. This represents the pressure deviation component caused by other factors. It is a sign function, and its value depends on the pressure deviation component. Positive and negative, This represents a correction factor between 0 and 1. The strength of the response used to control deviations caused by other factors. This can be understood as the safety margin factor. and correction factor The specific values ​​need to be adjusted during the system commissioning phase based on the characteristics and control requirements of the actual hydraulic system. The purpose of distinguishing between pressure loss and pressure deviation components is to ensure that the generated pressure compensation... It can precisely target the core issue of increased viscosity at low temperatures, while smoothing out other secondary disturbances, thus ensuring the accuracy and stability of compensation control. The final calculated pressure compensation amount... It will serve as the direct input command for the adaptive compensation controller.

[0032] See Figure 3This paper presents the coupling effect of ambient temperature and pressure loss assessment parameters on the system influence coefficient in the yaw braking system of a wind turbine in cold regions. The horizontal axis represents the pressure loss assessment parameter (MPa), and the vertical axis represents the ambient temperature (°C). The color depth represents the magnitude of the influence coefficient; the darker the color, the more significant the impact of low temperature and pressure loss on system performance. The graph clearly shows that in the low-temperature range of -1°C to -10°C, the influence coefficient increases sharply with the increase of the pressure loss assessment parameter, and the color transitions from light yellow to dark red. This indicates that the increased viscosity of hydraulic oil in low-temperature environments is the core cause of aggravated pipeline pressure loss. When the pressure loss assessment parameter exceeds 1.5 MPa, even in relatively mild temperature ranges (such as 0°C to 10°C), the influence coefficient increases significantly, indicating a clear synergistic effect between the pressure loss assessment parameter and ambient temperature, jointly amplifying the negative impact on the braking system performance. When the pressure loss assessment parameter is close to 2.0 MPa and the ambient temperature is approximately -1℃, the influence coefficient reaches its peak (close to 1.75). This corresponds to the most unfavorable operating point of the system under extreme low temperature and high pressure loss conditions, providing a crucial reference for the threshold setting and safety margin design of the adaptive compensation controller. The thermogram, as a visual output of the pressure loss assessment function, provides a quantitative basis for distinguishing the pressure loss component caused by increased viscosity at low temperatures from pressure deviation components caused by other factors, serving as important data support for achieving accurate low-temperature pressure compensation control.

[0033] In one embodiment of the present invention, the operation and training process of the adaptive compensation controller, as well as the execution of control commands, are involved. Taking a wind turbine generator set equipped with the aforementioned control system as an example, the pressure compensation amount is described. The pressure compensation amount has been calculated using the aforementioned steps. The input interface of the adaptive compensation controller is used. The adaptive compensation controller contains a feedforward control channel and a feedback adjustment channel.

[0034] In practical implementation, in the feedforward control channel, based on the pressure compensation amount... Based on the numerical values ​​and trends of the data, preliminary pressure control commands are generated by searching a pre-stored control quantity mapping table. The control quantity mapping table defines the pressure compensation quantity. With initial pressure control command The table establishes the correspondence between these parameters by testing the response characteristics of the hydraulic system under standard operating conditions. Refer to Table 1, which shows a simplified control variable mapping table.

[0035] Table 1: Mapping Table of Pressure Compensation Amount and Initial Pressure Control Command Initial pressure control commands The data is sent to the hydraulic system, and the actual pressure data of the brake lines after execution is collected simultaneously. This is used as a feedback signal input to the feedback control channel, which adjusts according to the target pressure. The actual pressure of feedback The deviation between them is used to calculate the adjustment control command through the built-in proportional-integral algorithm. The formula for the proportional-integral algorithm is: in: This represents the adjustment and control command at time t. This represents the pressure deviation at time t. It is the proportional gain coefficient. It is the integral gain coefficient. The integral of the pressure deviation from the initial time to time t represents the target pressure. Based on theoretical demand pressure value With pressure compensation To be determined jointly.

[0036] In some embodiments, the initial pressure control command will be... With adjustment and control commands The commands are superimposed to synthesize the final hydraulic system control commands. The composition relation is Hydraulic system control commands The variable frequency drive (VFD) transmits data to the hydraulic pump via fieldbus. The VFD adjusts the motor's operating frequency according to the pressure setpoint in the command. In some embodiments, the change in motor frequency drives the hydraulic pump to change its output flow and pressure. The adjusted hydraulic oil is delivered to the yaw brake cylinder via a control valve assembly. At the brake cylinder, a pressure sensor monitors the changes in braking pressure in real time and sends the monitored pressure data back to the control system. The control system compares the returned braking pressure data with the theoretical required pressure value. If the set target is not met, it reacquires multi-source real-time monitoring data and sequentially executes the pressure loss assessment, theoretical required pressure calculation, pressure compensation determination, and control command generation processes to generate new hydraulic system control commands until the braking pressure reaches the expected range.

[0037] Optionally, the training method for the adaptive compensation controller includes building a semi-physical simulation test platform containing a hydraulic system, sensors, and a controller. On this platform, a cold-climate environment is simulated. Under this simulated low-temperature environment, pressure compensation requirements of different amplitudes and rates of change are applied, and the hydraulic system control commands output by the adaptive compensation controller and their corresponding actual pressure regulation effects are recorded. Optionally, with the optimization goal of quickly eliminating pressure deviations and ensuring a smooth control process without overshoot, a machine learning algorithm is used to iteratively optimize the parameters internal to the adaptive compensation controller based on the recorded input-output data. The machine learning algorithm can employ reinforcement learning or neural networks to adjust parameters such as the proportional gain coefficient. Integral gain coefficient And the mapping relationships in the control quantity mapping table.

[0038] It is understandable that when the adaptive compensation controller's control performance on the test dataset meets the preset accuracy indicators, training is completed and its internal parameters are fixed. These preset accuracy indicators may include the steady-state error range of pressure regulation, the upper limit of response time, and the overshoot limit. It is also understood that the training of the adaptive compensation controller is an offline process, completed before the control system is deployed to the wind turbine generator. The trained adaptive compensation controller parameters are then loaded into the actual control unit of the wind turbine generator to achieve online pressure compensation control.

[0039] See Figure 4In the low-temperature pressure compensation control process of the yaw braking system of a cold-region wind turbine, the dynamic response relationship between the pressure compensation amount and the control command can be quantitatively analyzed through three core curves: Pressure compensation amount (solid line): This curve reflects the dynamic adjustment of the compensation amplitude required to offset the pressure loss caused by the increase in hydraulic oil viscosity at low temperatures. It exhibits significant negative fluctuations and attenuation in the 0-40s range, reflecting the system's rapid response and gradual convergence to pressure deviation during the low-temperature start-up phase; after 40s, the compensation amount turns from negative to positive and continues to rise, matching the trend of increased pressure loss caused by the continuous decrease in ambient temperature, with a compensation peak reaching 3.3 bar, verifying the adaptability of the compensation strategy to extreme operating conditions. Preliminary pressure control command (dashed line): As the output of the feedforward control channel, this command is generated based on the mapping table between the pressure compensation amount and the control amount, exhibiting a step response characteristic strongly coupled with the compensation amount. Within the 0-15s and 45-60s ranges, the command stabilizes at a saturated output of 6.0V, corresponding to the mapping relationship in the control quantity mapping table for pressure compensation ≥2.5bar. Within the 15-45s range, the command decays to 1.2V with the compensation amount, demonstrating the feedforward channel's rapid mapping capability to compensation requirements. The final hydraulic control command (dot-line): This command is formed by superimposing the preliminary control command and the PI adjustment command from the feedback adjustment channel, exhibiting smoother dynamic characteristics and higher response accuracy. It fluctuates around 8.0V within the 0-30s range and continuously climbs to 11.5V within the 30-60s range, retaining the rapidity of the feedforward channel while eliminating steady-state errors through feedback adjustment. The overshoot is controlled within 10%, meeting the accuracy and stability requirements of pressure regulation under cold conditions.

[0040] In one embodiment of the present invention, a method for constructing a braking performance reference model is described using a wind turbine generator of the same model to which the control system is to be installed. The construction work is completed in a standard temperature laboratory. The ambient temperature of the standard temperature laboratory is maintained at 20 degrees Celsius, and the humidity is controlled within the standard range to ensure that the hydraulic oil viscosity is at the nominal state defined in its specifications.

[0041] In practice, within a standard temperature laboratory, the angular velocity data of the moving body of the wind turbine generator's yaw braking system under different braking pressures are collected to form a standard operating condition dataset. During the data acquisition process, the control system applies a series of stepped braking pressure commands to the yaw brake, while a high-precision encoder records the unit's yaw angular velocity response curve in real time. Each pressure test records a stable braking pressure value and the corresponding rate of change of yaw angular velocity. Analysis of the standard operating condition dataset establishes a mathematical relationship between braking pressure and yaw deceleration effect; this mathematical relationship constitutes the core mapping relationship of the braking performance reference model.

[0042] Multiple sets of braking pressure values ​​and corresponding yaw rate of change data were extracted from a standard operating condition dataset. The least squares method was used to perform curve fitting on the extracted data to obtain the fundamental functional relationship between braking pressure and yaw deceleration. Based on the rotational inertia parameters of the wind turbine generator, yaw deceleration was converted into yaw braking torque, establishing a linear mapping relationship between braking pressure and braking torque. The brake's effective radius and the number of friction pads were introduced as scale factors to correct the linear mapping relationship, resulting in a mathematical relationship considering specific brake structural parameters. The mathematical expression for this core mapping relationship is: in: Indicates the yaw braking torque. This indicates the coefficient of friction between the brake disc and the friction pads. Indicates the number of friction plates. Indicates the effective radius of the brake. Indicates the effective working area of ​​the brake cylinder. This indicates braking pressure.

[0043] In some embodiments, the goodness of fit of the mathematical relationship is verified by residual analysis. When the coefficient of determination is greater than a preset threshold of 0.95, the mathematical relationship is confirmed as the core mapping relationship of the braking performance reference model. Other secondary factors affecting braking performance, including the range of variation of the brake disc friction coefficient and the degree of slight mechanical wear, are introduced as correction factors into the core mapping relationship. For example, the fixed friction coefficient in the core mapping relationship is used... The model is expanded to a function related to brake disc temperature and friction pad surface condition. Optionally, the model with the added correction factor is used to identify and optimize parameters through extensive experimental data covering different brake disc temperatures and different friction pad wear stages. The optimization algorithm is then used to adjust the parameters in the model to minimize the error between the model-predicted braking torque and the measured braking torque, ultimately determining the complete structure and parameters of the braking performance reference model.

[0044] Understandably, the testing conditions in a standard temperature laboratory aim to eliminate the influence of temperature variables on hydraulic oil viscosity, thereby establishing a baseline relationship between pure braking pressure and mechanical braking performance. It is also understandable that the reliability of the core mapping relationship is ensured through verification of the coefficient of determination via residual analysis, while the introduction of correction factors and parameter optimization improve the generalization ability and prediction accuracy of the braking performance reference model under actual complex operating conditions. The completed braking performance reference model will be integrated into the main controller of the wind turbine generator set for online calculation of theoretical required pressure values.

[0045] See Figure 5In the graph, the red curve represents control efficiency, and the blue curve represents control stability, both quantified as percentages. The horizontal axis represents ambient temperature (°C), and the vertical axes correspond to control efficiency and control stability (%), respectively. The curve trends show that when the ambient temperature is below 0°C, both control efficiency and stability are at low levels, decreasing linearly with decreasing temperature, falling below the baseline threshold of 60% near -5°C. This reflects the sharp increase in hydraulic oil viscosity at low temperatures, leading to increased pipeline pressure loss, delayed braking response, and consequently weakening the controller's compensation capability and system stability. When the ambient temperature is above 0°C, both curves show a significant upward trend, with the slope of the control efficiency curve being greater than that of the control stability curve. This indicates that as the temperature rises, the hydraulic oil viscosity returns to its nominal state, pressure loss decreases significantly, and the controller can execute pressure compensation commands more accurately, resulting in a rapid improvement in control efficiency. Control stability gradually improves due to the improved dynamic response of the hydraulic system, but its growth rate is relatively slow due to limitations in mechanical structure and actuator response bandwidth.

[0046] Under standard operating conditions of 20℃, the control efficiency reached 100%, and the control stability stabilized at around 95%, both meeting the design specifications and verifying the optimal performance of the adaptive compensation controller in a normal temperature environment. Further analysis of the curve intersection point (approximately 12℃) reveals that both control efficiency and control stability reach 90%. This temperature can be considered the performance inflection point of the adaptive compensation controller: when the ambient temperature is higher than this inflection point, control efficiency surpasses control stability, becoming the dominant advantage of the system performance; when the ambient temperature is lower than this inflection point, control stability becomes the key bottleneck restricting the overall system performance, requiring further improvement in low-temperature adaptability through auxiliary means such as optimizing the hydraulic oil formula or adding a preheating device.

[0047] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine, characterized in that, include: During the operation of the wind turbine, a multi-source real-time monitoring data set of the yaw braking system is acquired. The multi-source real-time monitoring data set includes ambient temperature data collected by the ambient temperature sensor, hydraulic oil temperature data collected by the hydraulic oil temperature sensor, actual pressure data of the brake pipeline collected by the brake pressure sensor, and angular velocity data of the moving body collected by the yaw state sensor. Based on the ambient temperature data and the hydraulic oil temperature data, the comprehensive temperature characteristic value of the hydraulic oil is calculated, and combined with the actual pressure data of the brake line, the pressure loss assessment parameters of the brake system under the current low temperature conditions are generated. Using the angular velocity data of the moving body and the actual pressure data of the braking line, the theoretical required pressure value under standard temperature conditions is calculated through a pre-constructed braking performance reference model. By comparing the theoretical required pressure value with the actual pressure data of the brake line, and combining the pressure loss assessment parameters of the brake system, the pressure compensation amount required to compensate for the effects of low temperature is calculated. Based on the pressure compensation amount, a corresponding hydraulic system control command is generated by a pre-trained adaptive compensation controller. The hydraulic system control command is sent to the hydraulic actuator, which drives the hydraulic actuator to dynamically adjust the braking pressure of the yaw brake, thereby achieving pressure compensation control in low-temperature environments.

2. The low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine according to claim 1, characterized in that, Based on the ambient temperature data and the hydraulic oil temperature data, the comprehensive temperature characteristic value of the hydraulic oil is calculated. Combined with the actual pressure data of the brake lines, pressure loss assessment parameters for the braking system under current low-temperature conditions are generated, including: The ambient temperature data and the hydraulic oil temperature data are input into a temperature coupling model for data fusion. The temperature coupling model is constructed based on the heat exchange law between ambient temperature and hydraulic oil temperature. The temperature coupling model outputs a comprehensive temperature characteristic value that characterizes the overall thermal state of the hydraulic oil. The comprehensive temperature characteristic value and the actual pressure data of the brake line are input into the pressure loss evaluation function, which describes the quantitative relationship between the effect of hydraulic oil viscosity change on line pressure loss under a specific comprehensive temperature characteristic value. The pressure loss assessment function is used to calculate and output the pressure loss assessment parameters of the braking system.

3. The low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine according to claim 2, characterized in that, Using the angular velocity data of the moving body and the actual pressure data of the braking line, the theoretical required pressure value under standard temperature conditions is calculated through a pre-constructed braking performance reference model, including: The angular velocity data of the moving body and the actual pressure data of the braking line are input into the braking performance reference model. The braking performance reference model reflects the relationship between the braking pressure required to achieve specific yaw rate control and the measured pressure under standard temperature conditions. The braking performance reference model is based on the input current actual pressure data of the braking line and the angular velocity data of the moving body, and reversely calculates the braking pressure required to achieve the same braking performance under standard oil temperature and viscosity conditions. The braking pressure obtained by the reverse calculation is output as the theoretical required pressure value.

4. The low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine according to claim 3, characterized in that, By comparing the theoretical required pressure value with the actual pressure data of the brake line, and combining this with the brake system pressure loss assessment parameters, the pressure compensation amount required to compensate for the effects of low temperature is calculated, including: Calculate the pressure difference between the theoretical required pressure value and the actual pressure data of the brake line; The pressure difference is correlated with the pressure loss assessment parameters of the braking system to distinguish the pressure loss component caused by increased viscosity at low temperature and the pressure deviation component caused by other factors. The pressure loss component caused by the increase in viscosity at low temperature is multiplied by a preset safety margin coefficient, and the product is used as the basic pressure compensation amount. Based on the sign and magnitude of the pressure deviation component caused by other factors, the basic pressure compensation amount is finely adjusted and corrected, and the pressure compensation amount is finally output.

5. The low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine according to claim 4, characterized in that, Based on the pressure compensation amount, a corresponding hydraulic system control command is generated through a pre-trained adaptive compensation controller, including: The hydraulic system control commands are used to adjust the output pressure and flow rate of the hydraulic pump; The pressure compensation amount is input into the input interface of the adaptive compensation controller, which internally includes a feedforward control channel and a feedback adjustment channel. In the feedforward control channel, based on the value and trend of the pressure compensation amount, a preliminary pressure control command is generated by looking up a pre-stored control quantity mapping table. The initial pressure control command is sent to the hydraulic system, and the actual pressure data of the brake line after execution is collected simultaneously and input as a feedback signal into the feedback adjustment channel. The feedback adjustment channel calculates the adjustment control command based on the deviation between the target pressure and the actual feedback pressure using a built-in proportional-integral algorithm. The initial pressure control command and the adjustment control command are superimposed to synthesize the final hydraulic system control command.

6. The low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine according to claim 5, characterized in that, The hydraulic system control command is sent to the hydraulic actuator to drive the hydraulic actuator to dynamically adjust the braking pressure of the yaw brake, including: The hydraulic system control commands are transmitted to the variable frequency drive of the hydraulic pump via a fieldbus. The variable frequency drive adjusts the operating frequency of the motor according to the pressure set value in the command. The change in motor frequency drives the hydraulic pump to change its output flow and pressure, and the adjusted hydraulic oil is delivered to the brake cylinder of the yaw brake via the control valve group. A pressure sensor installed at the brake cylinder monitors changes in brake pressure in real time and transmits the monitored pressure data back to the control system. The control system compares the returned braking pressure data with the theoretical required pressure value. If the set target is not met, it reacquires multi-source real-time monitoring data and sequentially executes the pressure loss assessment, theoretical required pressure calculation, pressure compensation determination, and control command generation process to generate new hydraulic system control commands until the braking pressure reaches the expected range.

7. The low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine according to claim 6, characterized in that, The method for constructing the pre-built braking performance reference model includes: In a standard temperature laboratory, angular velocity data of the moving body of the yaw braking system of the wind turbine under different braking pressures were collected to form a standard operating condition dataset. The standard working condition dataset is analyzed to establish a mathematical relationship between braking pressure and yaw deceleration effect. This mathematical relationship constitutes the core mapping relationship of the braking performance reference model. Other secondary factors affecting braking performance, including the range of variation in the brake disc friction coefficient and the degree of slight mechanical wear, are introduced as correction factors into the core mapping relationship. Through extensive experimental data, the parameters of the model after introducing the correction factor were identified and optimized, and the complete structure and parameters of the braking performance reference model were finally determined.

8. The low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine according to claim 7, characterized in that, The training method for the pre-trained adaptive compensation controller includes: A semi-physical simulation test platform including a hydraulic system, sensors, and controllers was built to simulate a cold-region low-temperature environment. Under simulated low-temperature conditions, pressure compensation requirements with different amplitudes and rates of change are applied, and the hydraulic system control commands output by the adaptive compensation controller and their corresponding actual pressure regulation effects are recorded. With the optimization goal of quickly eliminating pressure deviation and ensuring a smooth control process without overshoot, machine learning algorithms are used to iteratively optimize the parameters inside the adaptive compensation controller based on the recorded input and output data. When the adaptive compensation controller's control performance on the test dataset meets the preset accuracy index, training is completed and its internal parameters are solidified.

9. The low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine according to claim 8, characterized in that, The temperature coupling model is constructed based on the heat exchange law between ambient temperature and hydraulic oil temperature, including: Obtain the structural parameters of the wind turbine hydraulic system, including hydraulic tank volume, pipeline length and diameter, and material thermal conductivity; Based on the structural parameters and combined with the principles of fluid mechanics and heat transfer, a differential equation describing the dynamic heat exchange relationship between ambient temperature and hydraulic oil temperature is established. Historical operating data is used to identify and calibrate the thermal resistance and heat capacity parameters in the differential equation to ensure that the model output matches the measured oil temperature data. The real-time collected ambient temperature data is used as the model input to solve the differential equation and obtain the dynamic response of hydraulic oil temperature change. The dynamic response is weighted and averaged to output the comprehensive temperature characteristic value that characterizes the overall thermal state of the hydraulic oil.

10. The low-temperature pressure compensation control method for the yaw braking system of a cold-region wind turbine according to claim 9, characterized in that, The standard operating condition dataset is analyzed to establish a mathematical relationship between braking pressure and yaw deceleration effect. This mathematical relationship constitutes the core mapping relationship of the braking performance reference model, including: Extract multiple sets of braking pressure values ​​and corresponding yaw rate of change data from the standard operating condition dataset; The least squares method was used to fit the extracted data to obtain the basic functional relationship between braking pressure and yaw deceleration. Based on the rotational inertia parameters of the wind turbine, the yaw deceleration is converted into yaw braking torque, and a linear mapping between braking pressure and braking torque is established. By introducing the brake's effective radius and the number of friction pads as scale factors, the linear mapping is corrected to obtain a mathematical relationship that takes into account the specific brake structural parameters. The goodness of fit of the mathematical relationship is verified by residual analysis. When the coefficient of determination is greater than a preset threshold, the mathematical relationship is confirmed as the core mapping relationship.

Images