Wind power intelligent protection control system and method based on artificial intelligence

Abstract

The invention discloses a wind power intelligent protection control system and method based on artificial intelligence, and the method comprises the following steps: collecting an operation record, a rated parameter and a stress record of a wind driven generator, wherein the operation record comprises the vibration amplitude of a main shaft of the wind driven generator, the environment wind speed and the actual output power; calculating a vibration characteristic value of the wind driven generator based on the spindle vibration amplitude; obtaining a theoretical output value of the wind driven generator based on the rated parameter and the environment wind speed; calculating an output deviation value of the wind driven generator based on the actual output power and the theoretical output value; based on the stress record, obtaining damage values of the wind driven generator, summing the damage values of the wind driven generator to obtain an accumulated damage value of the wind driven generator, and generating a maintenance decision of the wind driven generator; calculating a real-time comprehensive health value of the wind driven generator based on the vibration characteristic value, the output deviation value and the accumulated damage value, and generating a control decision of the wind driven generator; and the running state is dynamically adjusted, and unnecessary shutdown is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and particularly to a wind power intelligent protection control system and method based on artificial intelligence. Background Art

[0002] A wind farm refers to a place where wind energy is used to generate electricity, usually composed of multiple wind turbines. The wind turbine utilizes the kinetic energy of the wind, drives the blades of the generator to rotate through the wind force, transfers the mechanical energy to the generator, and then converts the mechanical energy into electrical energy through the principle of electromagnetic induction. Wind energy is a natural energy source with intermittency and volatility. The changes in wind speed and direction will cause the output power of wind power generation to be unstable. With the continuous expansion of the scale of wind farms, this instability will have a significant impact on the frequency and voltage stability of the power system. Therefore, the safety and stability of the wind power generation system are crucial for the reliable operation of the overall power system.

[0003] Due to the instability of wind conditions, the rotational speed of the generator cannot be kept constant. If the wind speed is too high, the rotational speed of the generator will be too high, which will damage the generator. The existing wind power protection system adopts a simple protection logic: when the wind force is too high or mechanical problems such as bearing overheating and blade damage occur, the wind turbine is directly shut down. This simple protection logic lacks the ability of intelligent judgment and dynamic adjustment, resulting in reduced production efficiency or unnecessary shutdown of the system.

[0004] Therefore, there is an urgent need for a wind power intelligent protection control method and system based on artificial intelligence to solve the above problems. Summary of the Invention

[0005] In view of the above-mentioned prior art, the present invention aims to provide a wind power intelligent protection control method and system based on artificial intelligence, mainly to solve the technical problems existing in the above background art.

[0006] To achieve the above object, the technical solution of the embodiment of the present invention is realized as follows:

[0007] In a first aspect, the present invention provides a wind power intelligent protection control method based on artificial intelligence, including the following steps:

[0008] Collect the operation records, rated parameters of the wind turbine, and stress records corresponding to the operation records. The operation records include the main shaft vibration amplitude, ambient wind speed, and actual output power of the wind turbine.

[0009] Calculate the vibration characteristic value of the wind turbine based on the main shaft vibration amplitude.

[0010] Based on the rated parameters and the environmental wind speed, obtain the theoretical output value of the wind turbine; calculate the output deviation value of the wind turbine based on the actual output power and the theoretical output value;

[0011] Based on the stress record, obtain the damage value of the wind turbine, sum up the damage values of the wind turbine to obtain the cumulative damage value of the wind turbine, and generate a maintenance decision for the wind turbine;

[0012] Based on the vibration characteristic value, the output deviation value, and the cumulative damage value, calculate the real-time comprehensive health value of the wind turbine and generate a control decision for the wind turbine.

[0013] Optionally, the calculating the vibration characteristic value of the wind turbine based on the main shaft vibration amplitude includes:

[0014] Obtain the first data set of the main shaft vibration amplitude, calculate the average value and the standard deviation of the first data set of the main shaft vibration amplitude, set a screening range according to the average value and the standard deviation to obtain the second data set of the main shaft vibration amplitude, and calculate the average value of the second data set of the main shaft vibration amplitude to obtain the vibration characteristic value of the wind turbine;

[0015] The expression of the screening range is:

[0016] [A 1 -α×A 2 , A 1 +α×A 2

[0017] Wherein, A 1 represents the average value of the main shaft vibration amplitudes of all operation records in the first data set, A 2 represents the standard deviation of the main shaft vibration amplitudes of all operation records in the first data set, and α represents a range parameter.

[0018] Optionally, the rated parameters include the starting wind speed, the rated wind speed, the shutdown wind speed and the rated power of the wind turbine;

[0019] The obtaining the theoretical output value of the wind turbine based on the rated parameters and the environmental wind speed includes:

[0020] If the environmental wind speed is less than or equal to the starting wind speed, the theoretical output value is zero;

[0021] If the environmental wind speed is greater than the starting wind speed and the environmental wind speed is less than or equal to the rated wind speed, calculate the theoretical output value of the wind turbine, and the calculation formula is:

[0022] P theo =P r ×{(V​real -V in ) / (V r -V out )} 3

[0023] Among them, P theo represents the theoretical output value of the wind power generator, P r represents the rated power of the wind power generator, V real represents the ambient wind speed of the wind power generator, V in represents the cut-in wind speed of the wind power generator, V r represents the rated wind speed of the wind power generator, V out represents the cut-out wind speed of the wind power generator;

[0024] If the ambient wind speed is greater than the rated wind speed and the ambient wind speed is less than or equal to the cut-out wind speed, then the theoretical output value is the rated power;

[0025] If the ambient wind speed is greater than the cut-out wind speed, then the theoretical output value is zero;

[0026] The output deviation value of the wind turbine is calculated based on the actual output power and the theoretical output value, and the calculation formula is:

[0027] ΔP = P real -P theo

[0028] Among them, ΔP represents the output deviation value of the wind turbine, P real represents the actual output power, P theo represents the theoretical output value of the wind turbine.

[0029] Optionally, the stress record includes stress amplitude and number of cycles;

[0030] Based on the stress record, the damage value of the wind turbine is obtained, and the damage values of the wind turbine are summed to obtain the cumulative damage value of the wind turbine, and a maintenance decision for the wind turbine is generated, including:

[0031] The damage value of the wind turbine is calculated according to the stress amplitude and the number of cycles, and the calculation formula is:

[0032] B 1 = B 2 / {β 1 ×(B 3 ) β 2}

[0033] Among them, B 1 represents the damage value of the wind turbine, B2 represents the number of cycles, B 3 represents the stress amplitude, β 1 represents the damage constant, β 2 represents the damage index;

[0034] Obtain the stress record at the current moment and all stress records before the current moment to obtain the third data set of stress records; sum the damage values of the wind turbine at each moment in the third data set of stress records to obtain the cumulative damage value;

[0035] Obtain all cumulative damage values, arrange the all cumulative damage values in chronological order to obtain a time series, construct a neural network model, and train the neural network model to obtain a trained neural network model;

[0036] Use the trained neural network model to predict the cumulative damage value of the wind turbine, output the predicted cumulative damage value of the wind turbine, set a cumulative damage threshold, compare the predicted cumulative damage value with the cumulative damage threshold, if the predicted cumulative damage value of the wind turbine is less than or equal to the cumulative damage threshold, the maintenance decision generated for the wind turbine is not to trigger a maintenance warning;

[0037] If the predicted cumulative damage value is greater than or equal to the cumulative damage threshold, the maintenance decision generated for the wind turbine is to trigger a maintenance warning, and the wind turbine is inspected and maintained.

[0038] Optionally, generating a control decision for the wind turbine based on the vibration eigenvalue, the output deviation value, and the cumulative damage value includes:

[0039] Calculate the real-time comprehensive health value of the wind turbine at the current moment, and the calculation formula is:

[0040] C t = γ 1 ×A + γ 2 ×B + γ 3 ×ΔP

[0041] where C t represents the comprehensive health value at the current moment t, A represents the vibration eigenvalue, B represents the cumulative damage value, γ 1 represents the vibration weight coefficient, γ 2 represents the damage weight coefficient, γ 3 represents the deviation weight coefficient, and ΔP represents the output deviation value;

[0042] Set a first threshold and a second threshold, if C t is less than or equal to the first threshold, the control decision generated for the wind turbine is that the wind turbine operates normally and no control is performed on the wind turbine;

[0043] If C t is greater than the first threshold and C t is less than or equal to the second threshold, the control decision for the wind turbine is to operate at a reduced load, reducing the actual output power of the wind turbine so that the actual output power is less than or equal to the rated power. The expression is:

[0044] P real ≤P r ×(1 - δ)

[0045] where P real represents the actual output power of the wind turbine, P r represents the rated power of the wind turbine, and δ represents the load reduction coefficient;

[0046] If C t is greater than the second threshold, the control decision for the wind turbine is shutdown protection.

[0047] In a second aspect, the present invention also provides an artificial intelligence-based intelligent protection control system for wind power, which system includes: a collection module, a calculation module, a health assessment module, and a decision module;

[0048] The collection module is used to collect the operation data of the wind turbine and the stress data corresponding to the operation data period, and store the data in a database;

[0049] The calculation module is used to calculate the vibration characteristic value and output deviation value of the wind turbine according to the collected operation data;

[0050] The health assessment module is used to evaluate the health status of the wind turbine, and judge the overall health status of the wind turbine by calculating the cumulative damage value and comprehensive health value of the wind turbine;

[0051] The decision module is used to predict the cumulative damage value of the wind turbine according to the processing results of historical and real-time data, generate a maintenance decision according to the prediction result, and at the same time, generate a control decision according to the real-time comprehensive health value of the wind turbine.

[0052] Optionally, the collection module includes an operation data collection unit and a stress data collection unit;

[0053] The operation data collection unit is used to collect the operation data of the wind turbine and store it in the database; the operation data includes the main shaft vibration amplitude, environmental wind speed, and actual output power;

[0054] The stress data collection unit is used to collect the stress data of the wind turbine and store it in the database; the stress data includes the stress amplitude and the number of cycles.

[0055] Optionally, the calculation module includes a vibration feature calculation unit and an output deviation calculation unit

[0056] The vibration feature calculation unit is used to calculate the mean and standard deviation of the spindle vibration amplitude based on the selected operation records, and obtain vibration feature values;

[0057] The output deviation calculation is used to calculate the output deviation value according to the difference between the actual output power and the theoretical output power of the wind turbine.

[0058] Optionally, the health assessment module includes a damage calculation unit and a health calculation unit;

[0059] The damage calculation unit is used to calculate the damage value of each stress record, and sum the damage values corresponding to all stress records to obtain the cumulative damage value of the wind turbine;

[0060] The health calculation unit is used to calculate the comprehensive health value of the wind turbine according to the vibration feature value, output deviation value, and cumulative damage value, and evaluate the overall health status of the wind turbine.

[0061] Optionally, the decision-making module includes a maintenance decision-making unit and a control decision-making unit;

[0062] The maintenance decision-making unit is used to train a neural network model to predict the cumulative damage value according to the cumulative damage value of the wind turbine, and generate a maintenance decision;

[0063] The control decision-making unit is used to generate a control decision according to the comprehensively calculated health value in real time, and intelligently protect the wind turbine.

[0064] The beneficial effects of the present invention are as follows: An artificial intelligence-based intelligent protection control method for wind power provided by the present invention collects the operation data and stress records of the wind turbine in real time, combines a neural network model to predict the cumulative damage value of the wind turbine, monitors its operation status in real time, generates a maintenance decision in time, reduces the situation of over-maintenance or delayed maintenance, effectively extends the service life of the equipment; and by calculating the comprehensive health value of the wind turbine unit, it can intelligently judge and dynamically adjust the operation status, thus avoiding unnecessary shutdowns, improving the operation efficiency of the wind power system; it also optimizes the protection strategy of the wind turbine unit to ensure appropriate control measures are taken in high wind speeds or equipment failures, avoiding losses caused by over-shutdowns or unstable operation, enhancing the stability, safety and economy of the wind farm, and improving the comprehensive operation benefits of the wind power system. Description of the Drawings

[0065] Figure 1Schematic flowchart of a wind power intelligent protection control method based on artificial intelligence provided in an embodiment of the present invention;

[0066] Figure 2 Schematic diagram of a wind power intelligent protection control system based on artificial intelligence provided in an embodiment of the present invention. Specific embodiments

[0067] The technical solution of the present invention will be further elaborated in detail below in conjunction with the specification drawings and specific embodiments. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention. In the following description, the expression "some embodiments" is used, which describes a subset of all possible embodiments. However, it should be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments and can be combined with each other without conflict.

[0068] In the following description, a large number of specific details are given to provide a more thorough understanding of the present invention. However, it is obvious to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, some well-known technical features are not described to avoid confusion with the present invention.

[0069] It should be understood that the present invention can be implemented in different forms and should not be construed as limited to the embodiments presented herein. On the contrary, providing these embodiments will make the disclosure thorough and complete, and will fully convey the scope of the present invention to those skilled in the art. And the purpose of the terms used herein is only to describe specific embodiments and is not a limitation of the present invention. When used herein, the singular forms "a", "an" and "the" are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms "comprising" and / or "including", when used in this specification, determine the presence of the described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups. When used herein, the term "and / or" includes any and all combinations of the related listed items.

[0070] It should be further noted that when an element is referred to as being "fixed to" another element, it can be directly on the other element or there may also be an intermediate element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be an intermediate element at the same time. The terms "vertical", "horizontal", "inner", "outer", "left", "right" and similar expressions used herein are for illustrative purposes only and do not represent the only implementation.

[0071] To thoroughly understand the present invention, detailed structures will be presented in the following description to illustrate the technical solutions proposed by the present invention. The optional embodiments of the present invention are described in detail below. However, in addition to these detailed descriptions, the present invention may also have other implementations.

[0072] Embodiment 1

[0073] Please refer to the attached Figure 1 , the present invention provides an artificial intelligence-based intelligent protection control method for wind turbines, including the following steps:

[0074] Collect the operation records, rated parameters, and stress records corresponding to the operation records of the wind turbine. The operation records include the main shaft vibration amplitude, ambient wind speed, and actual output power of the wind turbine;

[0075] Calculate the vibration characteristic value of the wind turbine based on the main shaft vibration amplitude;

[0076] Based on the rated parameters and the ambient wind speed, obtain the theoretical output value of the wind turbine; calculate the output deviation value of the wind turbine based on the actual output power and the theoretical output value;

[0077] Based on the stress records, obtain the damage value of the wind turbine, sum up the damage values of the wind turbine to obtain the cumulative damage value of the wind turbine, and generate a maintenance decision for the wind turbine;

[0078] Specifically, the vibration eigenvalue is an important indicator of the operating state of a wind turbine. Changes in the vibration eigenvalue can reflect abnormal states of mechanical components, such as bearing wear, blade imbalance, or gearbox faults. By analyzing the eigenvalues of vibration signals, fault diagnosis and health monitoring of wind turbines can be achieved; the output deviation value reflects the difference between the actual output power of the wind turbine and the theoretical or designed power, and is an important indicator for evaluating the performance and control effect of the wind turbine generator set; by analyzing the output deviation value, the control strategy of the wind turbine can be adjusted, such as pitch angle control or speed regulation, to achieve maximum power capture; the damage value is an indicator for measuring the fatigue degree of wind turbine components. By calculating the cumulative damage value, the health state of the wind power generation equipment in the wind farm can be evaluated; by predicting the damage degree in advance, sudden failures can be avoided, and corresponding strategies can be formulated to reduce maintenance costs and downtime.

[0079] Therefore, based on the vibration eigenvalue, output deviation value, and cumulative damage value, calculate the real-time comprehensive health value of the wind turbine, generate the control decision of the wind turbine, so as to achieve the comprehensive monitoring and optimization of the wind power generation system and improve the overall operation efficiency of the wind farm.

[0080] As an optional implementation manner, calculating the vibration eigenvalue of the wind turbine based on the vibration amplitude of the main shaft includes:

[0081] Obtain the first dataset of the vibration amplitude of the main shaft, calculate the average value and standard deviation of the first dataset of the vibration amplitude of the main shaft, set the screening range according to the average value and standard deviation to obtain the second dataset of the vibration amplitude of the main shaft, and calculate the average value of the second dataset of the vibration amplitude of the main shaft to obtain the vibration eigenvalue of the wind turbine;

[0082] The expression of the screening range is:

[0083] [A 1 -α×A 2 ,A 1 +α×A 2

[0084] Wherein, A 1 represents the average value of the vibration amplitude of the main shaft of all operation records in the first dataset, A 2 represents the standard deviation of the vibration amplitude of the main shaft of all operation records in the first dataset, and α represents the range parameter;

[0085] Specifically, by screening the obtained first dataset of the vibration amplitude of the main shaft, outliers or noise data can be removed, which helps to improve the accuracy and reliability of the data, and realizes the basic monitoring of the operating state of the wind turbine, ensuring that subsequent analysis can be based on accurate data, avoiding misjudgment caused by abnormal data, and providing basic data for subsequent fan health assessment.​

[0086] As an optional embodiment, the rated parameters include the starting wind speed, rated wind speed, cut-out wind speed and rated power of the wind turbine;

[0087] Obtaining the theoretical output value of the wind turbine based on the rated parameters and the ambient wind speed includes:

[0088] If the ambient wind speed is less than or equal to the starting wind speed, the theoretical output value is zero;

[0089] If the ambient wind speed is greater than the starting wind speed and the ambient wind speed is less than or equal to the rated wind speed, calculate the theoretical output value of the wind turbine. The calculation formula is:

[0090] P theo = P r ×{(V real - V in ) / (V r - V out )} 3

[0091] Wherein, P theo represents the theoretical output value of the wind power generator, P r represents the rated power of the wind power generator, V real represents the ambient wind speed of the wind power generator, V in represents the starting wind speed of the wind power generator, V r represents the rated wind speed of the wind power generator, V out represents the cut-out wind speed of the wind power generator;

[0092] If the ambient wind speed is greater than the rated wind speed and the ambient wind speed is less than or equal to the cut-out wind speed, the theoretical output value is the rated power;

[0093] If the ambient wind speed is greater than the cut-out wind speed, the theoretical output value is zero;

[0094] Calculating the output deviation value of the wind turbine based on the actual output power and the theoretical output value. The calculation formula is:

[0095] ΔP = P real - P theo

[0096] Wherein, ΔP represents the output deviation value of the wind turbine, P real represents the actual output power, P theo represents the theoretical output value of the wind turbine;

[0097] Specifically, according to the relationship between the wind speed and the output power, the theoretical output value is determined, and then the output deviation value is determined. The output deviation value can reflect the difference between the actual output and the theoretical output of the wind turbine, thereby providing a basis for the subsequent health assessment of the wind turbine. At the same time, it provides an accurate benchmark for whether the wind turbine is within the normal operating range, providing a decision-making basis for the control and maintenance of the wind farm.

[0098] As an optional implementation manner, the stress record includes the stress amplitude and the number of cycles;

[0099] Based on the stress record, obtaining the damage value of the wind turbine, summing up the damage values of the wind turbine to obtain the cumulative damage value of the wind turbine, and generating the maintenance decision of the wind turbine, including:

[0100] Calculating the damage value of the wind turbine according to the stress amplitude and the number of cycles, and the calculation formula is:

[0101] B 1 =B 2 / {β 1 ×(B 3 ) β 2}

[0102] Wherein, B 1 represents the damage value of the wind turbine, B 2 represents the number of cycles, B 3 represents the stress amplitude, β 1 represents the damage constant, β 2 represents the damage index;

[0103] Specifically, every time the stress data is collected, a stress record of the wind turbine is generated and stored in the database, that is: each stress data collection corresponds to the stress data collection at each moment, and the collection of the stress data is carried out simultaneously with the collection of the operation data, so that the collected stress data is corresponding to the operation data during the corresponding period;

[0104] Exemplarily, for a certain wind turbine, the damage constant β 1 =1×10 8 , the damage index β 2 =-3, and the stress amplitude B 3 =100 of a certain stress record of the wind turbine, the number of cycles B 2 =1000. Using the formula to calculate the damage value of the wind turbine corresponding to this stress record is B 1 =1×10 -11 ;

[0105] The damage constant is directly related to the static strength and damage of the material. It reflects the service life of the material under high stress amplitudes. The longer the service life of the material, the larger the value of β 1 The value is larger; the damage index is usually a negative value, indicating that the fatigue life of the material decreases rapidly with the increase of the stress amplitude; by calculating the damage value, the overall performance and component life of the wind turbine can be understood, which is convenient for providing a basis for the subsequent control decision-making of the wind turbine;

[0106] Obtain the stress record at the current moment and all stress records before the current moment to obtain the third dataset of stress records; sum the damage values of the wind turbine at each moment in the third dataset of stress records to obtain the cumulative damage value;

[0107] Exemplarily, there are three stress records in the third dataset of stress records, and the damage values of the wind turbine corresponding to each stress record are 1×10 -11 、2.37×10 -12 and 6.25×10 -13 respectively; sum the damage values of the wind turbine corresponding to each stress record in the third dataset of stress records to get 1×10 -11 +2.37×10 -12 +6.25×10 -13 =1.61×10 -11 ; that is, 1.61×10 -11 is the cumulative damage value of the wind turbine;

[0108] Obtain all cumulative damage values, arrange the all cumulative damage values in chronological order to obtain a time series, construct a neural network model, and train the neural network model to obtain a trained neural network model;

[0109] Use the trained neural network model to predict the cumulative damage value of the wind turbine, output the predicted cumulative damage value of the wind turbine, set a cumulative damage threshold, compare the predicted cumulative damage value with the cumulative damage threshold. If the predicted cumulative damage value of the wind turbine is less than or equal to the cumulative damage threshold, the maintenance decision generated for the wind turbine is not to trigger a maintenance warning;

[0110] If the predicted cumulative damage value is greater than or equal to the cumulative damage threshold, the maintenance decision generated for the wind turbine is to trigger a maintenance warning, and the wind turbine is inspected and maintained.

[0111] Specifically, after obtaining all the cumulative damage values, arrange all the cumulative damage values in chronological order to obtain a time series. Then, construct a neural network model. Use the cumulative damage value corresponding to the last time point in the sorted time series as the output, and use the cumulative damage values corresponding to other time points in the time series as the input to train the neural network model. By arranging the cumulative damage values in chronological order and using them as input, the neural network model can better capture the time dependence in the data, and the neural network model can learn the law of how the damage value changes over time, thereby improving the accuracy of predicting future damage values. Exemplarily, the neural network model can adopt an RNN model or a Transformer model. Then, compare the predicted cumulative damage value with the set cumulative damage threshold. If the predicted cumulative damage value exceeds the cumulative damage threshold, trigger a maintenance warning to remind the operation and maintenance personnel to take corresponding measures in time to avoid accidents. Thus, the prediction of possible structural damage to the wind turbine is realized, which helps to detect potential failure risks in advance and provides data support for maintenance decision-making.

[0112] As an optional implementation manner, generating a control decision for the wind turbine based on the vibration characteristic value, the output deviation value, and the cumulative damage value includes:

[0113] Calculate the real-time comprehensive health value of the wind turbine at the current moment. The calculation formula is:

[0114] C t =γ 1 ×A + γ 2 ×B + γ 3 ×ΔP

[0115] Where C t represents the comprehensive health value at the current moment t, A represents the vibration characteristic value, B represents the cumulative damage value, γ 1 represents the vibration weight coefficient, γ 2 represents the damage weight coefficient, γ 3 represents the deviation weight coefficient, and ΔP represents the output deviation value;

[0116] Set a first threshold and a second threshold, where the first threshold is less than the second threshold. If C t is less than or equal to the first threshold, the generated control decision for the wind turbine is that the wind turbine operates normally without controlling the wind turbine;

[0117] If C t is greater than the first threshold and C tis less than or equal to the second threshold, then the generated control decision of the wind turbine is to reduce the load of the wind turbine, reduce the actual output power of the wind turbine, and make the actual output power less than or equal to the rated power. The expression is:

[0118] P real ≤P r ×(1-δ)

[0119] Among them, P real Indicates the actual output power of the wind turbine, P r represents the rated power of the wind turbine, and δ represents the load reduction factor;

[0120] Specifically, wind turbines operate in a complex natural environment and are affected by many uncontrollable factors such as random wind, wind shear, external power grid disturbances, and are bound to bear huge loads. In the control system of a wind turbine generator set, the speed of the wind turbine is closely related to the output power. Controlling the speed of the generator is one of the common ways to adjust the power. Most wind turbines use constant speed or variable speed control technology. In the case of load reduction, reducing the speed of the generator can reduce the output power. Power control can also be achieved by adjusting the windward angle of the blades, such as using the method of wind turbine pitch control. Wind turbine pitch control is to reduce the impact of wind on the impeller by changing the angle between the wind rotor blades and the wind direction, thereby reducing the power output of the wind turbine.

[0121] It should be noted that the main purpose of wind turbine load reduction is to reduce mechanical load and protect wind turbine equipment from damage; while reducing the output power of wind turbines is more to respond to grid demand or protect equipment from overload. In some cases, load reduction can be achieved by reducing output power.

[0122] If C t If the value is greater than the second threshold, the generated control decision of the wind turbine generator is shutdown protection;

[0123] Specifically, shutdown protection means that when a wind turbine encounters certain abnormal conditions or reaches a preset protection condition, the control system automatically or manually stops the operation of the generator; shutdown protection can prevent the wind turbine from continuing to operate under abnormal conditions, thereby avoiding damage to the equipment due to overload, overheating or excessive vibration; if the comprehensive health value exceeds the second threshold, shutdown protection can maintain the security of the power grid;

[0124] By comprehensively evaluating the health status of wind turbines and automatically generating control decisions, it is possible to adjust the operating mode in real time to ensure the safety and efficiency of wind turbines, avoid excessive wear and failure, and achieve intelligent protection for wind turbines.

[0125] Example 2

[0126] Please refer to the attached Figure 2 ,The present invention provides a wind power intelligent protection and control system based on artificial intelligence, including: an acquisition module, a calculation module, a health assessment module and a decision module;

[0127] The acquisition module is used to collect the operation data of the wind turbine and the stress data of the time period corresponding to the operation data, and store the data in a database;

[0128] The calculation module is used to calculate the vibration characteristic value and output deviation value of the wind turbine according to the collected operation data;

[0129] The health assessment module is used to assess the health status of the wind turbine and determine the overall health status of the wind turbine by calculating the cumulative damage value and the comprehensive health value of the wind turbine;

[0130] The decision module is used to predict the cumulative damage value of the wind turbine according to the processing results of historical and real-time data, and generate maintenance decisions according to the prediction results. At the same time, it generates control decisions according to the real-time comprehensive health value of the wind turbine.

[0131] As an optional implementation, the acquisition module includes an operation data acquisition unit and a stress data acquisition unit;

[0132] The operation data acquisition unit is used to collect the operation data of the wind turbine generator and store it in the database; the operation data includes the main shaft vibration amplitude, the ambient wind speed and the actual output power;

[0133] The stress data acquisition unit is used to acquire stress data of the wind turbine generator and store the data in a database; the stress data includes stress amplitude and cycle number.

[0134] As an optional implementation, the calculation module includes a vibration characteristic calculation unit and an output deviation calculation unit.

[0135] The vibration characteristic calculation unit is used to calculate the mean and standard deviation of the main shaft vibration amplitude according to the screened operation records to obtain the vibration characteristic value;

[0136] The output deviation calculation is used to calculate the output deviation value according to the difference between the actual output power and the theoretical output power of the wind turbine.

[0137] As an optional implementation, the health assessment module includes an injury calculation unit and a health calculation unit;

[0138] The damage calculation unit is used to calculate the damage value of each stress record, sum the damage values ​​corresponding to all stress records, and obtain the cumulative damage value of the wind turbine;

[0139] The health calculation unit is used to calculate the comprehensive health value of the wind turbine according to the vibration characteristic value, the output deviation value, and the cumulative damage value, and evaluate the overall health status of the wind turbine.

[0140] As an optional implementation, the decision module includes a maintenance decision unit and a control decision unit;

[0141] The maintenance decision unit is used to train a neural network model to predict the cumulative damage value according to the cumulative damage value of the wind turbine generator and generate a maintenance decision;

[0142] The control decision unit is used to generate a control decision according to the comprehensive health value calculated in real time, so as to provide intelligent protection for the wind turbine.

[0143] Specifically, corresponding sensors are arranged in the acquisition module, and the acquisition module includes an operation data unit and a stress data unit, which respectively acquire the operation data and stress data of the wind turbine in real time through the operation data unit and the stress data unit, and store the acquired operation data and stress data in a data set; the calculation module includes a vibration characteristic calculation unit and an output deviation calculation unit, which calculate the vibration characteristic value and the output deviation value through the vibration characteristic calculation unit and the output deviation calculation unit respectively according to the data acquired by the acquisition module; the health assessment module includes a damage calculation unit and a health calculation unit, which calculate the cumulative damage value and the comprehensive health value through the damage calculation unit and the health calculation unit respectively; the decision module includes a maintenance decision unit and a control decision unit, which generate the maintenance decision of the wind turbine and the control decision of the wind turbine respectively through the maintenance decision unit and the control decision unit, thereby realizing intelligent protection of the wind turbine.

[0144] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. The protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A wind power intelligent protection and control method based on artificial intelligence, characterized in that: The following steps are involved: Collecting the operation records, rated parameters, and stress records of the wind turbine generator corresponding to the operation records, wherein the operation records include the main shaft vibration amplitude, ambient wind speed, and actual output power of the wind turbine generator; Calculating a vibration characteristic value of the wind turbine based on the main shaft vibration amplitude; Based on the rated parameters and the ambient wind speed, obtaining a theoretical output value of the wind turbine; Calculate the output deviation value of the wind turbine based on the actual output power and the theoretical output value; Based on the stress record, a damage value of the wind turbine is obtained, the damage values ​​of the wind turbine are summed to obtain a cumulative damage value of the wind turbine, and a maintenance decision of the wind turbine is generated; Based on the vibration characteristic value, the output deviation value, and the cumulative damage value, a real-time comprehensive health value of the wind turbine is calculated, and a control decision of the wind turbine is generated.

2. The wind power intelligent protection and control method based on artificial intelligence according to claim 1 is characterized in that: The calculating the vibration characteristic value of the wind turbine generator based on the main shaft vibration amplitude comprises: Acquire a first data set of main shaft vibration amplitude, calculate the average value and standard deviation of the first data set of main shaft vibration amplitude, set a screening range according to the average value and standard deviation to obtain a second data set of main shaft vibration amplitude, calculate the average value of the second data set of main shaft vibration amplitude, and obtain the vibration characteristic value of the wind turbine; The expression of the screening range is: [A1-α×A2, A1+α×A2] Wherein, A1 represents the average value of the spindle vibration amplitude of all the running records in the first data set, A2 represents the standard deviation of the spindle vibration amplitude of all the running records in the first data set, and α represents the range parameter.

3. The wind power intelligent protection and control method based on artificial intelligence according to claim 1 is characterized in that: The rated parameters include the starting wind speed, rated wind speed, shutdown wind speed and rated power of the wind turbine; The obtaining of a theoretical output value of the wind turbine generator based on the rated parameters and the ambient wind speed includes: If the ambient wind speed is less than or equal to the starting wind speed, the theoretical output value is zero; If the ambient wind speed is greater than the starting wind speed and the ambient wind speed is less than or equal to the rated wind speed, the theoretical output value of the wind turbine is calculated using the following formula: P theo =P r ×{(V real -V in ) / (V r -V out )} 3 Among them, P theo Represents the theoretical output value of the wind turbine generator, P r Indicates the rated power of the wind turbine generator, V real Represents the ambient wind speed of the wind turbine generator, V in Indicates the starting wind speed of the wind turbine generator, V r Indicates the rated wind speed of the wind turbine generator, V out Indicates the shutdown wind speed of the wind turbine generator; If the ambient wind speed is greater than the rated wind speed and the ambient wind speed is less than or equal to the shutdown wind speed, the theoretical output value is the rated power; If the ambient wind speed is greater than the shutdown wind speed, the theoretical output value is zero; The output deviation value of the wind turbine generator is calculated based on the actual output power and the theoretical output value, and the calculation formula is: ΔP=P real -P theo Where ΔP represents the output deviation of the wind turbine, P real Indicates the actual output power, P theo Indicates the theoretical output value of the wind turbine.

4. The wind power intelligent protection and control method based on artificial intelligence according to claim 1 is characterized in that: The stress record includes stress amplitude and cycle number; The method of obtaining a damage value of the wind turbine generator based on the stress record, summing the damage values ​​of the wind turbine generator to obtain a cumulative damage value of the wind turbine generator, and generating a maintenance decision for the wind turbine generator includes: The damage value of the wind turbine is calculated according to the stress amplitude and the number of cycles, and the calculation formula is: <h2 style=";text-align:left;direction:ltr">B1 = B2 / {β1×(B3)}<h2 style=";text-align:left;direction:ltr"> β <h2 style=";text-align:left;direction:ltr"> 2} Among them, B1 represents the damage value of the wind turbine, B2 represents the number of cycles, B3 represents the stress amplitude, β1 represents the damage constant, and β2 represents the damage index; Acquire the stress record at the current moment and all stress records before the current moment to obtain a third data set of stress records; sum the damage values ​​of the wind turbine generator at each moment in the third data set of stress records to obtain the cumulative damage value; Obtaining all cumulative damage values, arranging all the cumulative damage values ​​in chronological order to obtain a time series, constructing a neural network model, and training the neural network model to obtain a trained neural network model; The trained neural network model is used to predict the cumulative damage value of the wind turbine, the predicted cumulative damage value of the wind turbine is output, a cumulative damage threshold is set, and the predicted cumulative damage value is compared with the cumulative damage threshold. If the predicted cumulative damage value of the wind turbine is less than or equal to the cumulative damage threshold, the maintenance decision of the wind turbine is generated as not triggering a maintenance warning. If the predicted cumulative damage value is greater than or equal to the cumulative damage threshold, the generated maintenance decision for the wind turbine is to trigger a maintenance warning, and inspect and maintain the wind turbine.

5. The wind power intelligent protection and control method based on artificial intelligence according to claim 1 is characterized in that: The step of generating a control decision of the wind turbine generator based on the vibration characteristic value, the output deviation value, and the cumulative damage value comprises: Calculate the real-time comprehensive health value of the wind turbine at the current moment. The calculation formula is: C t =γ1×A+γ2×B+γ3×ΔP Among them, C t represents the comprehensive health value at the current time t, A represents the vibration characteristic value, B represents the cumulative damage value, γ1 represents the vibration weight coefficient, γ2 represents the damage weight coefficient, γ3 represents the deviation weight coefficient, and ΔP represents the output deviation value; Set the first threshold and the second threshold. If C t is less than or equal to the first threshold, then the generated control decision of the wind turbine is that the wind turbine operates normally and the wind turbine is not controlled; If C t is greater than the first threshold and C t is less than or equal to the second threshold, then the generated control decision of the wind turbine is to reduce the load of the wind turbine, reduce the actual output power of the wind turbine, and make the actual output power less than or equal to the rated power. The expression is: P real ≤P r ×(1-d) Among them, P real Indicates the actual output power of the wind turbine, P r represents the rated power of the wind turbine, and δ represents the load reduction factor; If C t If the wind turbine generator is greater than the second threshold, the generated control decision is shutdown protection.

6. An artificial intelligence-based wind power intelligent protection and control system, characterized in that: include: Acquisition module, calculation module, health assessment module and decision-making module; The acquisition module is used to collect the operation data of the wind turbine and the stress data of the time period corresponding to the operation data, and store the data in a database; The calculation module is used to calculate the vibration characteristic value and output deviation value of the wind turbine according to the collected operation data; The health assessment module is used to assess the health status of the wind turbine and determine the overall health status of the wind turbine by calculating the cumulative damage value and the comprehensive health value of the wind turbine; The decision module is used to predict the cumulative damage value of the wind turbine according to the processing results of historical and real-time data, and generate maintenance decisions according to the prediction results. At the same time, it generates control decisions according to the real-time comprehensive health value of the wind turbine.

7. The wind power intelligent protection and control system based on artificial intelligence according to claim 6 is characterized in that: The acquisition module includes an operation data acquisition unit and a stress data acquisition unit; The operation data acquisition unit is used to collect the operation data of the wind turbine generator and store it in the database; the operation data includes the main shaft vibration amplitude, the ambient wind speed and the actual output power; The stress data acquisition unit is used to acquire stress data of the wind turbine generator and store the data in a database; the stress data includes stress amplitude and cycle number.

8. The wind power intelligent protection and control system based on artificial intelligence according to claim 6 is characterized in that: The calculation module includes a vibration characteristic calculation unit and an output deviation calculation unit The vibration characteristic calculation unit is used to calculate the mean and standard deviation of the main shaft vibration amplitude according to the screened operation records to obtain the vibration characteristic value; The output deviation calculation is used to calculate the output deviation value according to the difference between the actual output power and the theoretical output power of the wind turbine.

9. The wind power intelligent protection and control system based on artificial intelligence according to claim 6 is characterized in that: The health assessment module includes an injury calculation unit and a health calculation unit; The damage calculation unit is used to calculate the damage value of each stress record, sum the damage values ​​corresponding to all stress records, and obtain the cumulative damage value of the wind turbine; The health calculation unit is used to calculate the comprehensive health value of the wind turbine according to the vibration characteristic value, the output deviation value, and the cumulative damage value, and evaluate the overall health status of the wind turbine.

10. The wind power intelligent protection and control system based on artificial intelligence according to claim 6, characterized in that: The decision module includes a maintenance decision unit and a control decision unit; The maintenance decision unit is used to train a neural network model to predict the cumulative damage value according to the cumulative damage value of the wind turbine generator and generate a maintenance decision; The control decision unit is used to generate a control decision according to the comprehensive health value calculated in real time, so as to provide intelligent protection for the wind turbine.

Images