Heating method for hydraulic station of wind turbine generator

Abstract

The invention relates to the technical field of wind driven generators, in particular to a heating method for a hydraulic station of a wind turbine generator, which comprises the following steps: acquiring temperature data and control instruction data of the hydraulic station of the wind turbine generator; building a wind turbine generator hydraulic station temperature change trend prediction model, taking the current temperature data and historical temperature data of the wind turbine generator hydraulic station as input, and predicting the temperature change trend of the wind turbine generator hydraulic station through a neural network algorithm; and dynamically adjusting the control instruction data of the wind turbine generator hydraulic station based on the hybrid control strategy in combination with the wind turbine generator hydraulic station temperature change trend prediction model and the PID control algorithm, so as to realize heating control of the wind turbine generator hydraulic station. According to the method, the temperature of the hydraulic station is predicted through the neural network, and the control instruction is dynamically adjusted, so that the accuracy and response speed of temperature control are improved, the operation efficiency of a hydraulic system is optimized, the energy consumption is reduced, and the safety and reliability of equipment are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbines, and in particular to a heating method for a hydraulic station of a wind turbine set. Background Art

[0002] In modern wind turbines, the hydraulic system plays a vital role in controlling and regulating the operating performance of the generator set. Temperature changes in the hydraulic station directly affect the oil viscosity, transmission efficiency, and overall equipment performance. To ensure that the hydraulic system operates within the optimal temperature range, timely and accurate monitoring and adjustment of the hydraulic station temperature becomes particularly important. Especially in conditions of significant climatic fluctuations, hydraulic station temperature control faces even more complex challenges. Currently, traditional hydraulic system temperature monitoring typically relies on simple sensors and manual intervention, lacking real-time, accurate information feedback. This not only results in a slow response to the temperature adjustment process, but also makes it difficult to effectively respond to sudden temperature changes. Furthermore, it relies on preset control parameters and cannot adaptively adjust based on real-time data, resulting in poor temperature control under dynamic conditions. This can also cause overheating or insufficient cooling of the equipment, posing a safety and reliability risk. Summary of the Invention

[0003] The purpose of the present invention is to provide a heating method for a hydraulic station of a wind turbine generator set, which predicts the temperature of the hydraulic station through a neural network and dynamically adjusts the control instructions, thereby improving the accuracy and response speed of temperature control, optimizing the operating efficiency of the hydraulic system, reducing energy consumption, and enhancing the safety and reliability of the equipment.

[0004] The present invention is achieved through the following technical solutions:

[0005] A method for heating a hydraulic station of a wind turbine generator set, the method comprising the following steps:

[0006] Obtain temperature data and control instruction data of the wind turbine hydraulic station;

[0007] Build a wind turbine hydraulic station temperature change trend prediction model, take the current temperature data and historical temperature data of the wind turbine hydraulic station as input, and use the neural network algorithm to predict the temperature change trend of the wind turbine hydraulic station;

[0008] Based on the hybrid control strategy combined with the temperature change trend prediction model of the wind turbine hydraulic station and the PID control algorithm, the control instruction data of the wind turbine hydraulic station is dynamically adjusted to realize the heating control of the wind turbine hydraulic station.

[0009] Optionally, the temperature data of the hydraulic station of the wind turbine generator set is obtained, which is specifically:

[0010] Arrange sensors on the hydraulic station of wind turbines;

[0011] The sensor includes: a PT100 platinum resistance temperature sensor or an NTC thermistor;

[0012] The oil temperature of the hydraulic station of the wind turbine is collected based on a set period to obtain the temperature data of the hydraulic station of the wind turbine.

[0013] Optionally, the wind turbine hydraulic station temperature change trend prediction model is constructed as follows:

[0014] Obtain historical temperature data of the hydraulic station of the wind turbine generator set and preprocess the historical temperature data of the hydraulic station of the wind turbine generator set;

[0015] The pre-processed historical temperature data of the wind turbine hydraulic station is sorted according to the timestamp and divided into a training set and a test set;

[0016] A wind turbine hydraulic station temperature change trend prediction model was built based on a convolutional neural network. The model was trained using a training set and its prediction results were verified on a test set. The weights of the model were adjusted until the loss function converged to the set value.

[0017] Completed the training of the wind turbine hydraulic station temperature change trend prediction model.

[0018] Optionally, the preprocessing of the historical temperature data of the hydraulic station of the wind turbine generator set is specifically as follows:

[0019] Clean and normalize the historical temperature data of the wind turbine hydraulic station;

[0020] The historical temperature data of the hydraulic station of the wind turbine generator set is sorted in chronological order to complete the preprocessing of the historical temperature data of the hydraulic station of the wind turbine generator set.

[0021] Optionally, the loss function calculation process is:

[0022] Obtain the actual temperature value and predicted temperature value of the wind turbine hydraulic station temperature change trend prediction model;

[0023] Calculate the difference between the predicted value and the actual value of each training sample to obtain the prediction error of each training sample;

[0024] Square each prediction error;

[0025] Sum the squared errors of all training samples to get the total squared error;

[0026] The total squared error is divided by the number of training samples to obtain the average squared error, which is represented as the loss function.

[0027] Optionally, the hybrid control strategy is combined with a wind turbine hydraulic station temperature change trend prediction model and a PID control algorithm, which is specifically:

[0028] The current temperature prediction value and future temperature change trend are obtained through the wind turbine hydraulic station temperature change trend prediction model;

[0029] Calculate the temperature error based on the temperature prediction value and the set target temperature to determine the current state of the wind turbine hydraulic station;

[0030] Apply the PID control algorithm and input the temperature error into the PID controller to calculate the control output;

[0031] The control output of the PID controller is integrated with the results of the wind turbine hydraulic station temperature change trend prediction model to generate dynamically adjusted heating instructions;

[0032] The output power of the heater is adjusted according to the dynamically adjusted heating instruction to realize the heating control of the hydraulic station of the wind turbine.

[0033] Optionally, the PID control algorithm is applied to input the temperature error into the PID controller to calculate the control output, which is specifically:

[0034] Obtain the temperature prediction value and the set target temperature value to calculate the temperature error and obtain the temperature error value;

[0035] Initialize PID controller parameters, including proportional gain, integral gain, and differential gain;

[0036] Set the cumulative error variable to store the integral value of the historical temperature error and initialize it to zero;

[0037] In each control cycle, the following operations are performed:

[0038] Add the current temperature error to the accumulated error variable to update the historical temperature error;

[0039] Calculate the proportional term based on the current temperature error value and multiply the proportional gain by the current temperature error;

[0040] Calculate the integral term by multiplying the accumulated error variable by the integral gain;

[0041] Calculate the differential term by taking the difference between the current temperature error and the temperature error of the previous control cycle, dividing it by the length of the control cycle, and then multiplying it by the differential gain;

[0042] Add the proportional term, integral term, and differential term to get the current control output value.

[0043] Optionally, the temperature error is calculated as follows:

[0044] e(t)=T set -T current

[0045] Among them, e(t) is the temperature error at the current moment, T set To set the target temperature, T current is the current temperature forecast value.

[0046] Optionally, the calculation formula of the PID control algorithm is:

[0047]

[0048] Where u(t) is the output of the PID controller, K p is the proportional gain, K i is the integral gain, K d is the differential gain, is the accumulation of temperature error over time, is the rate of change of the current temperature error.

[0049] Optionally, the calculation formulas for the proportional term, integral term, and differential term are as follows:

[0050] P(t)=K p ·e(t)

[0051]

[0052] Among them, P(t) is the proportional term, I(t) is the integral term, and D(t) is the differential term.

[0053] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0054] This invention uses a deep neural network algorithm to construct a temperature trend prediction model, enabling precise prediction and dynamic control of wind turbine hydraulic station temperatures. Combined with advanced hybrid control strategies, this invention captures subtle temperature variations in real time, significantly enhancing the intelligent level of temperature control. The neural network model autonomously learns the complex mapping relationships of historical temperature data, providing more accurate prediction information for the PID control algorithm, thereby achieving smoother and more efficient temperature regulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A schematic flow chart of a heating method for a hydraulic station of a wind turbine generator set provided by the present invention;

[0056] Figure 2 This is a schematic diagram of the principle of building a temperature change trend prediction model for a wind turbine hydraulic station provided by the present invention;

[0057] Figure 3 This is a logical diagram of the hybrid control strategy provided by the present invention. DETAILED DESCRIPTION

[0058] To make the objectives, technical solutions, and advantages of the present invention more apparent, the following will provide a clear and complete description of the technical solutions of the present invention in conjunction with the accompanying drawings. It should be understood that the description is only a portion of the present invention, not all of it. The components of the present invention generally described and illustrated in the drawings herein may be arranged and designed in a variety of different configurations.

[0059] like Figure 1 As shown, the present invention provides one embodiment: a heating method for a hydraulic station of a wind turbine generator set, the method comprising the steps of:

[0060] Obtain temperature data and control instruction data of the wind turbine hydraulic station;

[0061] Build a wind turbine hydraulic station temperature change trend prediction model, take the current temperature data and historical temperature data of the wind turbine hydraulic station as input, and use the neural network algorithm to predict the temperature change trend of the wind turbine hydraulic station;

[0062] Based on the hybrid control strategy combined with the temperature change trend prediction model of the wind turbine hydraulic station and the PID control algorithm, the control instruction data of the wind turbine hydraulic station is dynamically adjusted to realize the heating control of the wind turbine hydraulic station.

[0063] Specifically, this embodiment uses a high-precision sensor to obtain the current temperature data and related control instruction data of the wind turbine hydraulic station in real time, and then builds a temperature change trend prediction model for the wind turbine hydraulic station. This embodiment uses a neural network algorithm in deep learning to conduct a comprehensive analysis of the current temperature data and historical temperature data, which can effectively capture the complex laws of temperature changes and provide accurate temperature predictions. On this basis, this embodiment combines a hybrid control strategy to combine the output of the temperature change trend prediction model with the PID control algorithm to achieve dynamic adjustment of the control instructions. In actual applications, when the model predicts that the temperature of the hydraulic station is about to exceed the set range, the system will promptly issue an adjustment instruction to optimize the heating control and ensure that the hydraulic oil remains within the optimal operating temperature range. This can improve the response speed and accuracy of the temperature control, greatly enhance the stability and operating efficiency of the wind turbine hydraulic station, reduce system energy consumption, and extend the service life of the equipment, thereby providing a strong guarantee for the efficient and safe operation of the wind power generation system.

[0064] In this embodiment, the temperature data of the hydraulic station of the wind turbine generator set is obtained as follows:

[0065] Arrange sensors on the hydraulic station of wind turbines;

[0066] The sensor includes: a PT100 platinum resistance temperature sensor or an NTC thermistor;

[0067] The oil temperature of the hydraulic station of the wind turbine is collected based on a set period to obtain the temperature data of the hydraulic station of the wind turbine.

[0068] During implementation, this embodiment uses a PT100 platinum resistance temperature sensor or an NTC thermistor, both of which have extremely high measurement accuracy, good stability and response speed, and can reflect the temperature changes of the hydraulic oil in real time. The sensors are arranged scientifically and reasonably according to the specific layout and working characteristics of the hydraulic station. Due to its linear characteristics and wide measurement range, the PT100 platinum resistance sensor is suitable for placement in the hydraulic oil flow area in the oil tank to monitor the actual temperature of the hydraulic oil in real time. The NTC thermistor can be set in the hydraulic oil return line to monitor the oil temperature after cooling and avoid temperature data distortion caused by the unreasonable layout of the temperature collection points. In order to ensure the real-time and accuracy of the temperature data, this embodiment regularly collects the oil temperature of the wind turbine hydraulic station based on a set period (such as every 5 seconds or 10 seconds). During the collection process, the temperature sensor will dynamically record the real-time temperature of the hydraulic oil and transmit the data to the central control system to form a continuous temperature data stream, which can effectively improve the temperature monitoring accuracy of the wind turbine hydraulic station and the overall operating efficiency of the system.

[0069] like Figure 2 As shown, in further implementation, the wind turbine hydraulic station temperature change trend prediction model is built, which is specifically:

[0070] Obtain historical temperature data of the hydraulic station of the wind turbine generator set and preprocess the historical temperature data of the hydraulic station of the wind turbine generator set;

[0071] The pre-processed historical temperature data of the wind turbine hydraulic station is sorted according to the timestamp and divided into a training set and a test set;

[0072] A wind turbine hydraulic station temperature change trend prediction model was built based on a convolutional neural network. The model was trained using a training set and its prediction results were verified on a test set. The weights of the model were adjusted until the loss function converged to the set value.

[0073] Completed the training of the wind turbine hydraulic station temperature change trend prediction model.

[0074] During implementation, after obtaining the historical temperature data, this embodiment performs necessary preprocessing, including: data cleaning and normalization of the historical temperature data of the wind turbine hydraulic station; arranging the historical temperature data of the wind turbine hydraulic station in chronological order, and completing the preprocessing of the historical temperature data of the wind turbine hydraulic station. Data cleaning ensures the accuracy of the temperature data, and eliminates noise and abnormal data that may appear in the measurement process through algorithms to improve the training quality of subsequent models. Interpolation methods can be used to fill missing data to maintain data continuity. Normalization processing helps to improve the efficiency of model training and make the data comparable during the training process. After completing the data preprocessing, the stored historical temperature data is sorted according to timestamps to ensure the time series of the data. Based on the overall availability of the data, the data set is divided into a training set and a test set. A hydraulic station temperature change trend prediction model is built based on a convolutional neural network (CNN). Convolutional neural networks can effectively capture the time series features in temperature data. The model is trained using the training set, and the model weights are continuously adjusted through the optimization algorithm to reduce the loss function until it converges to a preset threshold. During the training process, the model is regularly validated using the test set to assess its generalization ability on unseen data.

[0075] Specifically, the loss function calculation process is:

[0076] Obtain the actual temperature value and predicted temperature value of the wind turbine hydraulic station temperature change trend prediction model;

[0077] Calculate the difference between the predicted value and the actual value of each training sample to obtain the prediction error of each training sample;

[0078] Square each prediction error;

[0079] Sum the squared errors of all training samples to get the total squared error;

[0080] The total squared error is divided by the number of training samples to obtain the average squared error, which is represented as the loss function.

[0081] like Figure 3 As shown, in the specific implementation of this embodiment, the hybrid control strategy is combined with the wind turbine hydraulic station temperature change trend prediction model and PID control algorithm, which is specifically:

[0082] The current temperature prediction value and future temperature change trend are obtained through the wind turbine hydraulic station temperature change trend prediction model;

[0083] Calculate the temperature error based on the temperature prediction value and the set target temperature to determine the current state of the wind turbine hydraulic station;

[0084] Apply the PID control algorithm and input the temperature error into the PID controller to calculate the control output;

[0085] The control output of the PID controller is integrated with the results of the wind turbine hydraulic station temperature change trend prediction model to generate dynamically adjusted heating instructions;

[0086] The output power of the heater is adjusted according to the dynamically adjusted heating instruction to realize the heating control of the hydraulic station of the wind turbine.

[0087] The PID control algorithm is applied to input the temperature error into the PID controller to calculate the control output, which is specifically:

[0088] Obtain the temperature prediction value and the set target temperature value to calculate the temperature error and obtain the temperature error value;

[0089] Initialize PID controller parameters, including proportional gain, integral gain, and differential gain;

[0090] Set the cumulative error variable to store the integral value of the historical temperature error and initialize it to zero;

[0091] In each control cycle, the following operations are performed:

[0092] Add the current temperature error to the accumulated error variable to update the historical temperature error;

[0093] Calculate the proportional term based on the current temperature error value and multiply the proportional gain by the current temperature error;

[0094] Calculate the integral term by multiplying the accumulated error variable by the integral gain;

[0095] Calculate the differential term by taking the difference between the current temperature error and the temperature error of the previous control cycle, dividing it by the length of the control cycle, and then multiplying it by the differential gain;

[0096] Add the proportional term, integral term, and differential term to get the current control output value.

[0097] In specific implementation, this embodiment can obtain the current temperature prediction value and future temperature change trend through the trained wind turbine hydraulic station temperature change trend prediction model, and calculate the error between the obtained current temperature prediction value and the set target temperature. The calculation formula of temperature error is: e(t) = T set -T current , where e(t) is the temperature error at the current moment, T set To set the target temperature, T currentis the current temperature prediction value. After obtaining the temperature error, the PID control algorithm is applied and input into the PID controller to calculate the control output. The calculation formula of the PID control algorithm is: Where u(t) is the output of the PID controller, K p is the proportional gain, K i is the integral gain, K d is the differential gain, is the accumulation of temperature error over time, is the rate of change of the current temperature error. The PID controller effectively adjusts the output of the heater by responding to the temperature error in real time, and fuses the control output of the PID controller with the results of the temperature change trend prediction model of the wind turbine hydraulic station. Through fusion, more accurate dynamically adjusted heating instructions can be generated. Specifically, this instruction can adjust the output power of the heater according to the prediction of the temperature change trend to ensure that the hydraulic station always maintains within the set temperature range under changing environmental conditions. According to the generated dynamically adjusted heating instruction, the output power of the heater can be adjusted in real time, and the wind turbine hydraulic station can achieve efficient heating control, which not only ensures that the hydraulic oil is within the optimal operating temperature range, but also improves the operating efficiency and reliability of the entire wind turbine.

[0098] The above is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that the present invention is susceptible to various modifications and variations. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A heating method for a hydraulic station of a wind turbine generator set, characterized in that: The steps of the method include: Obtain temperature data and control instruction data of the wind turbine hydraulic station; Build a wind turbine hydraulic station temperature change trend prediction model, take the current temperature data and historical temperature data of the wind turbine hydraulic station as input, and use the neural network algorithm to predict the temperature change trend of the wind turbine hydraulic station; Based on the hybrid control strategy combined with the temperature change trend prediction model of the wind turbine hydraulic station and the PID control algorithm, the control instruction data of the wind turbine hydraulic station is dynamically adjusted to realize the heating control of the wind turbine hydraulic station.

2. The heating method for a wind turbine hydraulic station according to claim 1, characterized in that: The temperature data of the hydraulic station of the wind turbine generator set is obtained as follows: Arrange sensors on the hydraulic station of wind turbines; The sensor includes: a PT100 platinum resistance temperature sensor or an NTC thermistor; The oil temperature of the hydraulic station of the wind turbine is collected based on a set period to obtain the temperature data of the hydraulic station of the wind turbine.

3. The heating method for a wind turbine hydraulic station according to claim 1, characterized in that: The wind turbine hydraulic station temperature change trend prediction model is constructed as follows: Obtain historical temperature data of the hydraulic station of the wind turbine generator set and preprocess the historical temperature data of the hydraulic station of the wind turbine generator set; The pre-processed historical temperature data of the wind turbine hydraulic station is sorted according to the timestamp and divided into a training set and a test set; A wind turbine hydraulic station temperature change trend prediction model was built based on a convolutional neural network. The model was trained using a training set and its prediction results were verified on a test set. The weights of the model were adjusted until the loss function converged to the set value. Completed the training of the wind turbine hydraulic station temperature change trend prediction model.

4. The heating method for a wind turbine hydraulic station according to claim 3, characterized in that: The pre-processing of the historical temperature data of the hydraulic station of the wind turbine generator set is specifically as follows: Clean and normalize the historical temperature data of the wind turbine hydraulic station; The historical temperature data of the hydraulic station of the wind turbine generator set is sorted in chronological order to complete the preprocessing of the historical temperature data of the hydraulic station of the wind turbine generator set.

5. The heating method for a hydraulic station of a wind turbine generator set according to claim 3, characterized in that: The loss function calculation process is: Obtain the actual temperature value and predicted temperature value of the wind turbine hydraulic station temperature change trend prediction model; Calculate the difference between the predicted value and the actual value of each training sample to obtain the prediction error of each training sample; Square each prediction error; Sum the squared errors of all training samples to get the total squared error; The total squared error is divided by the number of training samples to obtain the average squared error, which is represented as the loss function.

6. The heating method for a wind turbine hydraulic station according to claim 3, characterized in that: The hybrid control strategy is combined with the wind turbine hydraulic station temperature change trend prediction model and PID control algorithm, which is specifically: The current temperature prediction value and future temperature change trend are obtained through the wind turbine hydraulic station temperature change trend prediction model; Calculate the temperature error based on the temperature prediction value and the set target temperature to determine the current state of the wind turbine hydraulic station; Apply the PID control algorithm and input the temperature error into the PID controller to calculate the control output; The control output of the PID controller is integrated with the results of the wind turbine hydraulic station temperature change trend prediction model to generate dynamically adjusted heating instructions; The output power of the heater is adjusted according to the dynamically adjusted heating instruction to realize the heating control of the hydraulic station of the wind turbine.

7. The heating method for a wind turbine hydraulic station according to claim 6, characterized in that: The PID control algorithm is applied to input the temperature error into the PID controller to calculate the control output, which is specifically: Obtain the temperature prediction value and the set target temperature value to calculate the temperature error and obtain the temperature error value; Initialize PID controller parameters, including proportional gain, integral gain, and differential gain; Set the cumulative error variable to store the integral value of the historical temperature error and initialize it to zero; In each control cycle, the following operations are performed: Add the current temperature error to the accumulated error variable to update the historical temperature error; Calculate the proportional term based on the current temperature error value and multiply the proportional gain by the current temperature error; Calculate the integral term by multiplying the accumulated error variable by the integral gain; Calculate the differential term by taking the difference between the current temperature error and the temperature error of the previous control cycle, dividing it by the length of the control cycle, and then multiplying it by the differential gain; Add the proportional term, integral term, and differential term to get the current control output value.

8. The method for heating a hydraulic station of a wind turbine generator set according to claim 7, characterized in that: The calculation formula of the temperature error is: e(t)=T set -T current Among them, e(t) is the temperature error at the current moment, T set To set the target temperature, T current is the current temperature forecast value.

9. The heating method for a hydraulic station of a wind turbine generator set according to claim 8, characterized in that: The calculation formula of the PID control algorithm is: Where u(t) is the output of the PID controller, K p is the proportional gain, K i is the integral gain, K d is the differential gain, is the accumulation of temperature error over time, is the rate of change of the current temperature error.

10. The heating method for a hydraulic station of a wind turbine generator set according to claim 9, characterized in that: The calculation formulas for the proportional term, integral term and differential term are as follows: P(t)=K p ·e(t) Among them, P(t) is the proportional term, I(t) is the integral term, and D(t) is the differential term.

Images