Unit yaw control method and system based on dynamic electricity price and fatigue damage coupling

Abstract

The invention discloses a unit yaw control method and system based on dynamic electricity price and fatigue damage coupling. The method comprises the following steps that a multi-generating-capacity income model is constructed according to historical operation data of a wind turbine generator based on power loss and income correlation; constructing a yaw wear cost model based on the yaw bearing fatigue loss; constructing a yaw loss function according to the correlation between the electricity price and the risk preference in the historical operation data of the wind turbine generator; acquiring a real-time yaw loss index prediction value according to the real-time electricity price prediction data and the wind speed prediction data based on a multi-generating capacity income model, a yaw wear cost model and a yaw loss function; and outputting a yaw control strategy according to the real-time yaw loss index predicted value and the yaw loss index threshold value. The method has the beneficial effect that the optimal balance between the power generation income and the service life of the equipment is realized.

Description

Technical Field

[0001] This application relates to the technical field of yaw control for wind turbines, and particularly to a yaw control method and system for wind turbines based on the coupling of dynamic electricity prices and fatigue damage. Background Art

[0002] With the continuous growth of the global demand for clean energy, wind power generation, as a sustainable energy solution, occupies an increasingly important position in the energy structure. As the core equipment of wind power generation, the operating efficiency and reliability of wind turbines directly affect the economic benefits and energy supply stability of the entire wind power generation system. In recent years, the single-unit capacity of wind turbines has been continuously increasing, the number of installed units has also been continuously increasing, and the scale of wind farms has been growing larger. However, wind turbines face many challenges during actual operation, and among them, the optimal control of the yaw system is crucial for improving the economy of wind turbines.

[0003] The core function of the current yaw control system for wind turbines is to achieve the dynamic tracking of the wind direction by the wind turbine rotor through components such as wind direction sensors and yaw drives to maximize the wind energy capture efficiency. However, with the deepening of the electricity market reform and the large-scale application of wind turbines, the following key defects have gradually emerged in traditional yaw control technology: on the one hand, frequent yaw actions will exacerbate the mechanical wear of key components such as the gearbox and yaw bearing. Excessive pursuit of wind direction tracking accuracy will lead to an increase in the annual yaw times, significantly shortening the service life cycle of components, and thus increasing the operation and maintenance costs. On the other hand, the existing yaw control logic takes the real-time wind energy capture efficiency as the single optimization goal and does not introduce the regulatory effect of dynamic electricity prices on power generation revenue. In some regions where the electricity spot market has been implemented in China, the electricity price during peak hours can reach 120% of the benchmark electricity price, while the traditional method still maintains a high-power output strategy during the low electricity price period, resulting in a significant deviation between the actual power sales revenue and the theoretical value.

[0004] Existing solutions generally lack the ability to synergistically perceive the electricity price signal and the health status of components, and are unable to actively reduce the yaw frequency during low electricity price periods to delay fatigue, or improve the tracking accuracy during high electricity price periods to obtain premium revenue, and it is difficult to dynamically adjust yaw thresholds (such as wind speed dead zone, yaw delay time) to balance power generation revenue and maintenance costs.

[0005] The above defects lead to the paradox that the existing yaw system faces "high power generation ≠ high economy" in a complex market environment, and at the same time accelerates the aging process of the wind turbine. There is an urgent need to develop a collaborative control method that integrates a dynamic electricity price response mechanism and a fatigue damage suppression strategy to achieve the optimal economy of the entire life cycle of wind turbines.

[0006] Patent "A Yaw Control Method and Related Devices", Publication Number: CN119467206A, Publication Date: February 18, 2025, specifically discloses the following: obtaining the original wind speed time series dataset, the original wind direction time series dataset, and the design parameters of the wind turbine generator set; based on the original wind speed time series dataset and the original wind direction time series dataset, obtaining the wind speed probability density function, the wind direction deviation probability density function, and the yaw control input data; based on the wind speed probability density function, the wind direction deviation probability density function, and the design parameters, determining the initial yaw control parameters; and performing simulation based on the yaw control input data and the initial yaw control parameters to determine the target yaw control parameters based on the simulation results and the preset multi-objective constraint conditions. Although this solution takes into account the differences in wind conditions of different wind farms and the differences in wind turbine generator sets, there is still a problem of being unable to balance the power generation revenue and the maintenance cost. Summary of the Invention

[0007] In view of the problem in the prior art that the power generation revenue and the maintenance cost cannot be balanced, the present application provides a yaw control method and system for a wind turbine generator set based on the coupling of dynamic electricity price and fatigue damage. By respectively establishing a multi-power generation revenue model and a yaw wear cost model to quantify the revenue brought by the increase in power generation and the loss brought by yaw wear, the balance relationship between the two is quantified based on the yaw loss function, so as to achieve the optimal balance between power generation revenue and equipment life.

[0008] To achieve the above technical objectives, a technical solution provided by the present application is a yaw control method for a wind turbine generator set based on the coupling of dynamic electricity price and fatigue damage, including the following steps: constructing a multi-power generation revenue model based on the correlation between power loss and revenue according to the historical operation data of the wind turbine generator set; constructing a yaw wear cost model based on the fatigue loss of the yaw bearing; constructing a yaw loss function according to the correlation between the electricity price and the risk preference in the historical operation data of the wind turbine generator set; obtaining a real-time yaw loss index prediction value based on the multi-power generation revenue model, the yaw wear cost model, and the yaw loss function according to the real-time electricity price prediction data and the wind speed prediction data; and outputting a yaw control strategy according to the real-time yaw loss index prediction value and the yaw loss index threshold.

[0009] Further, the constructing a multi-power generation revenue model based on the correlation between power loss and revenue according to the historical operation data of the wind turbine generator set includes: obtaining the historical yaw angle according to the wind direction and the nacelle position in the historical operation data of the wind turbine generator set; constructing a power loss function according to the wind speed and the historical yaw angle in the historical operation data of the wind turbine generator set; and constructing a multi-power generation revenue model based on the correlation between power loss and revenue according to the power loss function.

[0010] Further, the construction of the yaw wear cost model based on the fatigue loss of the yaw bearing includes: calculating the bearing fatigue loss per cycle according to the yaw maintenance cost and the designed stress cycle times of the yaw gear; constructing the yaw wear cost model according to the yaw angle and the bearing fatigue loss per cycle.

[0011] Further, the construction of the yaw wear cost model based on the fatigue loss of the yaw bearing further includes: calculating the theoretical bearing fatigue loss per cycle according to the yaw maintenance cost and the designed stress cycle times of the yaw gear; obtaining the actual yaw operation times corresponding to the environmental data according to the historical operation data of the wind turbine; constructing the yaw wear cost model according to the environmental data, the actual yaw operation times, the yaw angle, and the theoretical bearing fatigue loss per cycle.

[0012] Further, the construction of the yaw loss function according to the correlation between the electricity price and the risk preference in the historical operation data of the wind turbine includes: establishing the correlation between the electricity price and the risk preference according to the correlation between the electricity price and the wind-facing ratio in different wind speed intervals in the historical operation data of the wind turbine; constructing the yaw loss function with the correlation between the electricity price and the risk preference.

[0013] Further, the construction of the yaw loss function according to the correlation between the electricity price and the risk preference in the historical operation data of the wind turbine further includes: calculating the average yaw speed of the wind turbine according to the historical operation data of the wind turbine, and obtaining the wind-facing adjustment time according to the average yaw speed of the wind turbine; constructing the yaw loss function with the wind-facing adjustment time and the correlation between the electricity price and the risk preference.

[0014] Further, obtaining the predicted value of the real-time yaw loss index based on the multi-power generation revenue model, the yaw wear cost model, and the yaw loss function according to the real-time electricity price prediction data and the wind speed prediction data includes: obtaining the predicted value of the multi-power generation revenue according to the real-time electricity price prediction data, the wind speed prediction data, and the multi-power generation revenue model; obtaining the predicted value of the yaw wear cost according to the wind speed prediction data, the current operation parameters of the wind turbine, and the yaw wear cost model; obtaining the predicted value of the real-time yaw loss index according to the predicted value of the multi-power generation revenue, the predicted value of the yaw wear cost, and the yaw loss function.

[0015] Further, outputting the yaw control strategy according to the predicted value of the real-time yaw loss index and the yaw loss index threshold includes: obtaining the historical yaw loss index during the wind-facing process in the historical operation data of the wind turbine according to different wind speed intervals; obtaining the yaw loss index threshold corresponding to the wind speed interval according to the average value of the historical yaw loss index; retrieving the yaw loss index threshold according to the current wind speed interval, and outputting the yaw control strategy based on the comparison result between the predicted value of the real-time yaw loss index and the yaw loss index threshold.

[0016] Further, the yaw control strategy output based on the comparison result between the predicted real-time yaw loss index and the yaw loss index threshold includes: if the predicted real-time yaw loss index is greater than the yaw loss index threshold, the yaw control strategy is to execute the wind alignment mode; if the predicted real-time yaw loss index is less than or equal to the yaw loss index threshold, the yaw control strategy is to execute the error tolerance mode.

[0017] Another technical solution provided by this application is a yaw control system for wind turbines based on the coupling of dynamic electricity prices and fatigue damage, which is used to implement the above method and is connected to an electricity price prediction module and a wind speed prediction module, including: a revenue prediction unit for constructing a multi-power generation revenue model based on the correlation between power loss and revenue according to the historical operation data of the wind turbine; a wear cost prediction unit for constructing a yaw wear cost model based on the fatigue loss of the yaw bearing; a yaw loss prediction unit for obtaining a predicted real-time yaw loss index based on the multi-power generation revenue model, the yaw wear cost model, and the yaw loss function according to the real-time electricity price prediction data and the wind speed prediction data; and a yaw control unit for outputting a yaw control strategy according to the comparison between the predicted real-time yaw loss index and the yaw loss index threshold.

[0018] The beneficial effects of this application: By constructing a multi-power generation revenue model based on the correlation between power loss and revenue, the impact of yaw on power generation revenue is demonstrated. By constructing a yaw wear cost model based on the fatigue loss of the yaw bearing, the negative impact of yaw adjustment on the bearing is demonstrated. And based on the dynamic electricity price and risk preference in the historical operation data, the game relationship between power generation revenue and mechanical loss is obtained. Thus, combined with the real-time wind speed and electricity price prediction data, the predicted real-time yaw loss index within the future time window is calculated to show the loss difference between the yaw situation and the wind alignment situation, which is used as the basis for outputting the yaw control strategy to achieve the balance between power generation revenue and maintenance cost. Description of the Drawings

[0019] Figure 1 It is a schematic flow chart of the yaw control method for wind turbines based on the coupling of dynamic electricity prices and fatigue damage in this application. Detailed Embodiments

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the following further details this application with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only the best embodiments of this application, which are only used to explain this application and do not limit the protection scope of this application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of this application.

[0021] As Figure 1 shown, as Embodiment 1 of this application, the yaw control method for wind turbines based on the coupling of dynamic electricity prices and fatigue damage includes the following steps: Construct a multi-power generation revenue model based on the correlation between power loss and revenue according to the historical operation data of the wind turbine generator set; Construct a yaw wear cost model based on the fatigue loss of the yaw bearing; Construct a yaw loss function according to the correlation between electricity price and risk preference in the historical operation data of the wind turbine generator set; Based on the multi-power generation revenue model, the yaw wear cost model, and the yaw loss function, obtain the predicted value of the real-time yaw loss index according to the real-time electricity price prediction data and the wind speed prediction data; Output a yaw control strategy according to the predicted value of the real-time yaw loss index and the yaw loss index threshold.

[0022] In this embodiment, the influence of yaw on power generation revenue is demonstrated by constructing a multi-power generation revenue model according to the correlation between power loss and revenue, the negative impact of yaw adjustment on the bearing is demonstrated by constructing a yaw wear cost model based on the fatigue loss of the yaw bearing, and the game relationship between power generation revenue and mechanical loss is obtained based on the dynamic electricity price and risk preference in the historical operation data. Thus, by combining the real-time wind speed and electricity price prediction data, the predicted value of the yaw loss index within the future time window is calculated to demonstrate the loss difference between the yaw situation and the on-wind situation, which is used as the basis for outputting the yaw control strategy to achieve the balance between power generation revenue and maintenance cost.

[0023] Specifically, constructing a multi-power generation revenue model based on the correlation between power loss and revenue according to the historical operation data of the wind turbine generator set includes: Obtain the historical yaw angle according to the wind direction and nacelle position in the historical operation data of the wind turbine generator set; Construct a power loss function according to the wind speed and the historical yaw angle in the historical operation data of the wind turbine generator set; Construct a multi-power generation revenue model based on the power loss function according to the correlation between power loss and revenue.

[0024] The historical operation data of the wind turbine generator set at least includes wind speed, wind direction, temperature, active power, electricity price, yaw operation times, and the operation parameters of the wind turbine generator set. The operation parameters of the wind turbine generator set at least include the nacelle position of the wind turbine generator set, the rotor area of the wind turbine generator set, and the electromechanical conversion efficiency of the wind turbine generator set.

[0025] At this time, obtaining the historical yaw angle according to the wind direction and nacelle position in the historical operation data of the wind turbine generator set is: θ = θ W - θ N ; where θ W represents the wind direction angle, and θ N represents the nacelle position angle of the wind turbine generator set.

[0026] Construct a power loss function based on the historical operating data of a wind turbine, including the wind speed and the historical yaw angle, as follows: Among them, P loss represents the power loss rate caused by wind direction deviation, ρ represents the air density, S represents the wind turbine rotor area, C p represents the aerodynamic power coefficient, ξ represents the electromechanical conversion efficiency of the wind turbine, and v w represents the wind speed.

[0027] Based on the correlation between power loss and revenue, construct a multi-power generation revenue model according to the power loss function: R t = λ(t)·P loss ; Among them, R t represents the electricity price revenue corresponding to the power generation loss that can be recovered by the wind turbine with a yaw angle of θ when implementing the wind alignment strategy at time t; λ(t) represents the real-time electricity price at time t.

[0028] It can be understood that when initially constructing the multi-power generation revenue model using the historical operating data of the wind turbine, since the historical electricity price is known, the real-time electricity price at the target time can be directly obtained based on the historical electricity price. When obtaining the predicted value of the real-time yaw loss index, the real-time predicted electricity price at the target time is used for the prediction calculation of the multi-power generation revenue.

[0029] Construct a yaw wear cost model based on the yaw bearing fatigue loss, including: Calculate the bearing fatigue loss per cycle according to the yaw maintenance cost and the design stress cycle times of the yaw gear; Construct a yaw wear cost model according to the yaw angle and the bearing fatigue loss per cycle.

[0030] At this time, the operating parameters of the wind turbine also include the yaw maintenance cost.

[0031] Calculate the bearing fatigue loss per cycle according to the yaw maintenance cost and the design stress cycle times of the yaw gear as follows: Among them, S c represents the bearing fatigue loss per cycle, C is the yaw maintenance cost, and N is the design stress cycle times of the yaw gear.

[0032] Construct a yaw wear cost model according to the yaw angle and the bearing fatigue loss per cycle as follows: Among them, S θ represents the yaw wear cost of the yaw gear when implementing the wind alignment strategy at a yaw angle of θ.

[0033] In some other cases, constructing a yaw wear cost model based on yaw bearing fatigue loss includes: Calculating the theoretical bearing fatigue loss per cycle according to the yaw maintenance cost and the designed stress cycle times of the yaw gear; Obtaining the actual yaw action times corresponding to the environmental data based on the historical operation data of the wind turbine; Constructing a yaw wear cost model according to the environmental data, the actual yaw action times, the yaw angle, and the theoretical bearing fatigue loss per cycle.

[0034] At this time, the operation data of the wind turbine also includes environmental data. At this time, constructing a yaw wear cost model according to the environmental data, the actual yaw action times, the yaw angle, and the theoretical bearing fatigue loss per cycle includes: Calculating the actual yaw cycle times corresponding to the environmental characteristics according to the actual yaw action times and the yaw angle under different environments; Calculating the wear influence coefficient according to the actual cruise cycle times and the designed stress cycle times of the yaw gear, and constructing the correlation between the wear influence coefficient and the environmental characteristics; Constructing a yaw wear cost model according to the correlation between the wear influence coefficient and the environmental characteristics and the theoretical bearing fatigue loss per cycle.

[0035] Calculating the wear influence coefficient according to the actual cruise cycle times and the designed stress cycle times of the yaw gear as: Where κ represents the wear influence coefficient, N′ represents the actual cruise cycle times, and N′ = the total yaw angle corresponding to each actual yaw action / 360.

[0036] At this time, the actual bearing fatigue loss per cycle is the product of the theoretical bearing fatigue loss per cycle and the wear influence coefficient. When the actual cruise cycle times are less than the designed stress cycle times, the wear influence coefficient is larger, the actual bearing fatigue loss per cycle is greater than the theoretical bearing fatigue loss per cycle, and thus the actual yaw wear cost of a single yaw is greater than the theoretical yaw wear cost of a single yaw, that is, the actual yaw wear cost of a single yaw = κ * the theoretical yaw wear cost of a single yaw.

[0037] Constructing a yaw loss function according to the correlation between the electricity price and the risk preference in the historical operation data of the wind turbine includes: constructing the correlation between the electricity price and the risk preference according to the electricity price and the wind-facing ratio in different wind speed intervals in the historical operation data of the wind turbine; Constructing a yaw loss function based on the correlation between the electricity price and the risk preference.

[0038] Classify historical electricity prices according to different wind speed intervals, construct the correlation between electricity prices and risk preferences based on the proportion of wind following under different electricity price models, and then obtain the risk preference weights according to the correlation between electricity prices and risk preferences. Based on this, construct a yaw loss function to show the game relationship between power generation revenue and mechanical loss. As shown in Table 1, it is the value of the risk preference weight coefficient under a certain wind speed interval.

[0039] Table 1 Value of the risk preference weight coefficient under a certain wind speed interval

[0040] Among them, α is the percentile of the electricity price. Sort the electricity prices in the same wind speed interval, and define it as the percentile α according to the percentage position of the current electricity price in the electricity price concentration in the same wind speed interval. Set the risk preference weight coefficient according to the proportion of wind following under different percentiles. When the proportion of wind following under the percentile is high, set a smaller risk preference weight. When the proportion of wind following under the percentile is low, set a higher risk preference weight. Infer the risk preference weight through historical behavior to avoid subjective parameter setting deviation and improve the accuracy of regulation.

[0040] In some cases, constructing a yaw loss function based on the correlation between electricity prices and risk preferences in the historical operation data of wind turbines also includes: Calculate the average yaw speed of the wind turbine according to the historical operation data of the wind turbine, and obtain the wind following adjustment time according to the average yaw speed of the wind turbine; Construct a yaw loss function based on the wind following adjustment time and the correlation between electricity prices and risk preferences.

[0041] Calculate the wind following adjustment time with the average yaw speed of the wind turbine as the length of the time window T considering yaw loss. In this embodiment, according to the average yaw adjustment speed of the wind turbine in different wind speed intervals, obtain the time required to adjust the yaw angle according to the average yaw adjustment speed of the wind turbine as the wind following adjustment time, and calculate the yaw loss during the wind following adjustment time to ensure the calculation accuracy of the yaw loss.

[0042] Specifically, the yaw loss function is: Among them, Y t represents the yaw loss value at time t, R t represents the electricity price revenue corresponding to the power generation power loss that can be recovered by the wind turbine with a yaw angle of θ when implementing the wind following strategy at time t, represents the risk preference weight coefficient, S θ represents the yaw wear cost of the yaw gear when implementing the wind following strategy at a yaw angle of θ, YT Denote the yaw economic loss function considering real-time electricity price and fatigue loss within the T time window, where T represents the wind regulation time.

[0043] Based on the multi-power generation revenue model, yaw wear cost model, and yaw loss function, obtaining the predicted value of the real-time yaw loss index according to the real-time electricity price prediction data and wind speed prediction data includes: Obtaining the predicted value of multi-power generation revenue according to the real-time electricity price prediction data, wind speed prediction data, and multi-power generation revenue model; obtaining the predicted value of yaw wear cost according to the wind speed prediction data, current operating parameters of the wind turbine, and yaw wear cost model; obtaining the predicted value of the real-time yaw loss index according to the predicted value of multi-power generation revenue, the predicted value of yaw wear cost, and the yaw loss function.

[0044] Obtaining the predicted value of the yaw angle through the wind speed prediction data and the current operating parameters of the wind turbine, obtaining the predicted value of yaw wear cost according to the predicted value of the yaw angle, and retrieving the risk preference weight coefficient according to the real-time electricity price prediction data and the correlation between electricity price and risk preference. Calculating the predicted value of the real-time yaw loss index according to the predicted value of multi-power generation revenue, the predicted value of yaw wear cost, and the risk preference weight coefficient.

[0045] Outputting the yaw control strategy according to the predicted value of the real-time yaw loss index and the yaw loss index threshold includes: Obtaining the historical yaw loss index during the wind alignment process in the historical operation data of the wind turbine according to different wind speed intervals; Obtaining the yaw loss index threshold corresponding to the wind speed interval according to the average value of the historical yaw loss index; Retrieving the yaw loss index threshold according to the current wind speed interval, and outputting the yaw control strategy based on the comparison result between the predicted value of the real-time yaw loss index and the yaw loss index threshold.

[0046] Obtaining the yaw loss index for implementing the wind alignment strategy under each wind speed interval, calculating the average value, and using this as the yaw loss index threshold for the corresponding wind speed interval, so as to adapt the yaw loss index threshold to the actual working conditions and improve the accuracy of the output of the yaw control strategy.

[0047] Specifically, calculating the average value of the yaw loss index for a single yaw wind alignment process: Among them, Y i Represents the average value of the yaw loss index for a single yaw wind alignment process, and T i Represents the duration of a single yaw wind alignment process.

[0048] Furthermore, according to the average value of the average values of the yaw loss indices for all yaw wind alignment processes under the same wind speed interval: Among them, Y threshold It represents the yaw loss index threshold, and I represents all yaw-to-wind processes in the same wind speed range.

[0049] The yaw control strategy outputted based on the comparison result between the real-time yaw loss index prediction value and the yaw loss index threshold value includes: If the real-time yaw loss index prediction value is greater than the yaw loss index threshold, the yaw control strategy is to execute the windward mode; If the real-time yaw loss index prediction value is less than or equal to the yaw loss index threshold, the yaw control strategy is to execute the error tolerance mode.

[0050] If the real-time yaw loss index prediction value is greater than the yaw loss index threshold, it means that the current yaw control can exceed the average benefit in the same wind speed range in history, and the precise wind mode is started; if the real-time yaw loss index prediction value is less than or equal to the yaw loss index threshold, the error tolerance mode is enabled, which means that the current yaw control benefit is lower than or equal to the average benefit in the same wind speed range in history, and the wind is not started. In the entire error tolerance mode, the time window is continuously moved in seconds to perform the above calculation and judgment process.

[0051] It can be understood that the real-time electricity price forecast data and wind speed forecast data of the present application are obtained by utilizing the existing wind farm SCADA system monitoring data and the wind speed and electricity price forecast modules, without the need for additional hardware equipment. By changing the existing yaw system control judgment logic, it can effectively balance the wind turbine power generation revenue and maintenance costs, avoid yaw action under low electricity price conditions, reduce mechanical wear of the yaw system, extend equipment service life, and simultaneously improve the overall economic benefits and operational stability of the wind farm.

[0052] As the second embodiment of the present application, a specific embodiment is provided for illustration. Taking a 5MW wind turbine as an example, data from January 2024 to December 2024 in the SCADA system is collected, with a collection interval of 1 minute, starting from 9:00 on a certain day in January 2025. The specific process is as follows: 1) Collect historical and real-time operation data of wind turbines, including wind speed, wind direction, temperature, active power, electricity price, number of yaw actions, and wind turbine operation parameters. Screen these data for validity, remove abnormal data, and ensure the accuracy and reliability of the data. The real-time electricity price forecast value from 9:00 to 9:15 on the same day obtained by the electricity price forecast module is 0.417 yuan / kWh, and the current wind speed is 7.98m / s.

[0053] 2) Calculate the electricity price revenue of the wind turbine after performing a single wind regulation: Set the time window \(T = 180s\) according to the average yaw speed of a certain wind turbine, and obtain the predicted electricity price and wind speed data from the electricity price prediction module and wind speed prediction module of the wind farm. The electricity price revenue corresponding to the recoverable power generation loss of the wind turbine with a yaw angle of \(\theta\) when implementing the wind alignment strategy at time \(t\) is expressed as:

[0054] Calculate the yaw wear cost consumed after the wind turbine performs a single wind alignment adjustment: The single yaw maintenance cost of the wind turbine is mainly composed of the costs of multiple aspects. In this embodiment, the equipment replacement cost is mainly calculated. There is an approximately monotonically increasing relationship between the number of load cycles and the damage \(D\). Estimated according to the selling price of 256,000 yuan per unit of the yaw variable bearing applicable to 5MW wind turbines, based on the assumption that the life of the wind turbine is 20 years and the number of stress cycles is \(1.0\times10^{8}\), the cost \(S\) of a single complete stress cycle of the yaw bearing c is about 0.00256 yuan. Under the simplified assumption that other parameters change little, in order to evaluate the fatigue damage cost caused by a single yaw adjustment, based on the current yaw angle, the yaw wear cost caused by the equivalent stress cycle is

[0055] Calculate the adaptive risk preference weight coefficient: First, divide different working condition intervals according to a wind speed step of 2m / s, count the historical electricity price data in each working condition interval and calculate the probability distribution. Calculate the percentile \(\alpha\) corresponding to the predicted electricity price in the predicted wind speed corresponding working condition interval in the future time window. According to Table 1, calculate the corresponding mean value as the risk weight coefficient. As shown in Table 2, the calculated risk weight coefficient within the \(T\) time window is 0.874.

[0056] Table 2 Risk preference weight coefficient within the \(T\) time window

[0058] Calculate the yaw economic loss index \(Y\) within the current time window T : According to Calculate to obtain: Y T = 0.0948578; Calculate the threshold of the yaw economic loss index: According to the collected data, divide the working conditions according to different wind speed intervals, calculate the \(Y\) i value of all yaw wind alignment processes under each working condition, find the mean value of all \(Y\) i values, and set it as the threshold \(Y\) of the yaw economic loss index threshold .

[0057] Statistically analyze the historical operating conditions in which the wind speed is in the range of [6 - 8). Divide the data based on the change in the number of yaws. A total of I = 22,156 sets of yaw process data are obtained. Take Substitute into the calculation to obtain Yi. The calculation results are shown in Table 3, and the calculated Y threshold = 0.08407121.

[0058] Table 3 Yaw economic loss index for all yaw wind alignment processes within the historical operating conditions where the wind speed is in the range of [6 - 8)

[0061] Compare the yaw economic loss index Y within the current time window T with the yaw economic loss index threshold Y_threshold. The yaw economic loss index Y within the current time window T > Y_threshold. It can be seen that if the precise wind alignment mode is started, the yaw economic loss index within the current time window will be higher than the average value of the yaw economic loss index under the same historical wind speed conditions. It can be seen that the current execution of wind alignment is beneficial to the wind turbine, so yaw adjustment should be carried out.

[0059] As the third embodiment of the present application, a yaw control system for a wind turbine based on the coupling of dynamic electricity prices and fatigue damage is connected to an electricity price prediction module and a wind speed prediction module, and includes: A revenue prediction unit for constructing a multi - power generation revenue model based on the correlation between power loss and revenue according to the historical operation data of the wind turbine; A wear cost prediction unit for constructing a yaw wear cost model based on the fatigue loss of the yaw bearing; A yaw loss prediction unit for obtaining a real - time yaw loss index prediction value based on the multi - power generation revenue model, the yaw wear cost model, and the yaw loss function according to the real - time electricity price prediction data and the wind speed prediction data; A yaw control unit for outputting a yaw control strategy according to the real - time yaw loss index prediction value and the yaw loss index threshold.

[0060] In this embodiment, the revenue prediction unit and the wear cost prediction unit are connected to the electricity price prediction module, the wind speed prediction module, and the yaw loss prediction unit. The yaw control unit is connected to the yaw loss prediction unit and the wind turbine, and controls whether the wind turbine executes the wind alignment mode according to the yaw control strategy, so as to balance the power generation revenue and the maintenance cost.

[0061] The above-described specific embodiments are the preferred embodiments of the yaw control method and system for a unit based on the coupling of dynamic electricity price and fatigue damage of the present application, and do not limit the specific implementation scope of the present application. The scope of the present application includes but is not limited to this specific embodiment. Any equivalent changes made according to the shape and structure of the present application are within the protection scope of the present application.

Claims

1. A yaw control method for a unit based on the coupling of dynamic electricity price and fatigue damage, characterized in that: The method includes the following steps: Construct a multi-power generation revenue model based on the correlation between power loss and revenue according to the historical operation data of the wind turbine; Construct a yaw wear cost model based on the fatigue loss of the yaw bearing; Construct a yaw loss function according to the correlation between electricity price and risk preference in the historical operation data of the wind turbine; Obtain the predicted value of the real-time yaw loss index based on the multi-power generation revenue model, the yaw wear cost model, and the yaw loss function according to the real-time electricity price prediction data and the wind speed prediction data; Output a yaw control strategy according to the predicted value of the real-time yaw loss index and the yaw loss index threshold.

2. The yaw control method of the unit based on the coupling of dynamic electricity price and fatigue damage according to claim 1, wherein: The constructing a multi-power generation revenue model based on the correlation between power loss and revenue according to the historical operation data of the wind turbine includes: Obtain the historical yaw angle according to the wind direction and the nacelle position in the historical operation data of the wind turbine; Construct a power loss function according to the wind speed and the historical yaw angle in the historical operation data of the wind turbine; Construct a multi-power generation revenue model based on the correlation between power loss and revenue according to the power loss function.

3. The yaw control method of the unit based on the coupling of dynamic electricity price and fatigue damage according to claim 1, wherein: The constructing a yaw wear cost model based on the fatigue loss of the yaw bearing includes: Calculate the bearing fatigue loss in a single cycle according to the yaw maintenance cost and the designed stress cycle times of the yaw gear; Construct a yaw wear cost model according to the yaw angle and the bearing fatigue loss in a single cycle.

4. The yaw control method for a unit based on the coupling of dynamic electricity price and fatigue damage according to claim 1, characterized in that: The constructing a yaw wear cost model based on the fatigue loss of the yaw bearing further includes: Calculate the theoretical bearing fatigue loss in a single cycle according to the yaw maintenance cost and the designed stress cycle times of the yaw gear; Obtain the actual yaw operation times corresponding to the environmental data according to the historical operation data of the wind turbine; Construct a yaw wear cost model according to the environmental data, the actual yaw operation times, the yaw angle, and the theoretical bearing fatigue loss in a single cycle.

5. The yaw control method of the unit based on the coupling of dynamic electricity price and fatigue damage according to claim 1, characterized in that: The constructing a yaw loss function according to the correlation between electricity price and risk preference in the historical operation data of the wind turbine includes: Construct the correlation between electricity price and risk preference according to the correlation between electricity price and the wind-facing ratio in different wind speed intervals in the historical operation data of the wind turbine; Construct a yaw loss function based on the correlation between electricity price and risk preference.

6. The yaw control method of the unit based on the coupling of dynamic electricity price and fatigue damage according to claim 1, characterized in that: The constructing a yaw loss function according to the correlation between electricity price and risk preference in the historical operation data of the wind turbine further includes: Calculate the average yaw speed of the wind turbine according to the historical operation data of the wind turbine, and obtain the wind-facing adjustment time according to the average yaw speed of the wind turbine; Construct a yaw loss function based on the wind-facing adjustment time and the correlation between electricity price and risk preference.

7. The yaw control method of the unit based on the coupling of dynamic electricity price and fatigue damage according to claim 1, characterized in that: The obtaining the predicted value of the real-time yaw loss index based on the multi-power generation revenue model, the yaw wear cost model, and the yaw loss function according to the real-time electricity price prediction data and the wind speed prediction data includes: Obtain the predicted value of multi-power generation revenue according to the real-time electricity price prediction data, the wind speed prediction data, and the multi-power generation revenue model; Obtain the predicted value of the yaw wear cost according to the wind speed prediction data, the current operation parameters of the wind turbine, and the yaw wear cost model; obtain the predicted value of the real-time yaw loss index according to the predicted value of multi-power generation revenue, the predicted value of the yaw wear cost, and the yaw loss function.

8. The yaw control method of the unit based on the coupling of dynamic electricity price and fatigue damage according to claim 1, characterized in that: The outputting a yaw control strategy according to the predicted value of the real-time yaw loss index and the yaw loss index threshold includes: Obtain the historical yaw loss index of the windward process in the historical operation data of the wind turbine according to different wind speed intervals; Obtain the yaw loss index threshold corresponding to the wind speed interval according to the average value of the historical yaw loss index; Retrieve the yaw loss index threshold according to the current wind speed interval, and output the yaw control strategy based on the comparison result between the predicted value of the real-time yaw loss index and the yaw loss index threshold.

9. The yaw control method of the unit based on the coupling of dynamic electricity price and fatigue damage according to claim 1, characterized in that: The output of the yaw control strategy based on the comparison result between the predicted value of the real-time yaw loss index and the yaw loss index threshold includes: If the predicted value of the real-time yaw loss index is greater than the yaw loss index threshold, the yaw control strategy is to execute the windward mode; If the predicted value of the real-time yaw loss index is less than or equal to the yaw loss index threshold, the yaw control strategy is to execute the error tolerance mode.

10. A yaw control system for a unit based on the coupling of dynamic electricity prices and fatigue damage, which is used to implement the method according to any one of claims 1 to 9, and is connected to an electricity price prediction module and a wind speed prediction module, and is characterized in that: Include: A revenue prediction unit for constructing a multi-power generation revenue model based on the historical operation data of the wind turbine according to the correlation between power loss and revenue; A wear cost prediction unit for constructing a yaw wear cost model based on the fatigue loss of the yaw bearing; A yaw loss prediction unit for obtaining the predicted value of the real-time yaw loss index according to the real-time electricity price prediction data and the wind speed prediction data based on the multi-power generation revenue model, the yaw wear cost model, and the yaw loss function; A yaw control unit for outputting a yaw control strategy according to the comparison result between the predicted value of the real-time yaw loss index and the yaw loss index threshold.

Images