Wind power plant active power optimization control method and system based on health state evaluation

Abstract

The invention discloses a wind power plant active power optimization control method and system based on health state evaluation, and the method comprises the steps: training a part normal behavior model based on the historical operation data of a wind power plant SCADA system, and building a mapping relation between a key index and a part prediction temperature; on the basis of real-time operation data of a wind power plant SCADA system, a component temperature prediction value is obtained through input of a component normal behavior model; constructing a multi-dimensional state feature vector according to the actual measured value of the component temperature and the deviation between the actual measured value of the component temperature and the predicted value of the component temperature, and calculating a mahalanobis distance between the multi-dimensional state feature vector and a normal data sample space; and determining degradation state types and health degree evaluation results of the fans, and adjusting the output active power of each fan so as to dynamically optimize the active power scheduling of the wind power plant. The invention aims to optimize the active power distribution of the wind power plant, reduce the operating temperature and fatigue load of the fault fan, prolong the service life of the fan and improve the economic benefit.

Description

Technical Field

[0001] This invention relates to the field of active power optimization technology in wind farms, specifically to a method and system for optimizing and controlling active power in wind farms based on health status assessment. Background Technology

[0002] With the rapid development of wind power technology and the continuous expansion of wind farm scale, maintaining the safe, continuous, and profitable operation of wind turbines has become a major challenge. Modern wind turbines can quickly control their output active power, which has stimulated research interest in active power allocation optimization (APC) within wind farms. Many methods have been proposed for optimal scheduling within wind farms, aiming to minimize wind turbine fatigue loads and system power losses under wind power forecasting and power system load constraints, avoid frequent turbine start-ups and shutdowns, reduce operating costs, and improve the power quality of wind farm output power. However, existing power allocation methods typically do not consider the actual health state of individual wind turbines; different turbines are often treated as being under the same health state, resulting in an inability to fully reflect the specific condition of each turbine during power allocation. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for optimizing the active power control of wind farms based on health status assessment, in view of the above-mentioned problems in the prior art. The present invention aims to optimize the active power distribution of wind farms, reduce the operating temperature and fatigue load of faulty wind turbines, extend the service life of wind turbines and improve economic benefits.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A method for optimizing active power control in wind farms based on health status assessment includes the following steps: 1) Based on the historical operation data of the wind farm SCADA system, normal data samples without fault alarm records are selected. For multiple components of the wind turbine, multiple key indicators related to component temperature are selected as inputs. Component temperature is used as a label to train the normal behavior model of the component offline, so as to establish the mapping relationship between key indicators and component predicted temperature. 2) Based on the real-time operation data of the wind farm SCADA system, multiple key indicators of each component are used as inputs to the component normal behavior model to obtain the component temperature prediction value; the actual measured value of component temperature and its deviation from the component temperature prediction value are used to construct a multi-dimensional state feature vector, and the Mahalanobis distance between the multi-dimensional state feature vector and the normal data sample space is calculated. 3) The type of deterioration state of the wind turbine is determined by the difference between the Mahalanobis distance and the preset threshold, and the Mahalanobis distance at each time point in the monitoring period is mapped to the health assessment result of the wind turbine. 4) Adjust the output active power of each wind turbine according to the degradation status type and health assessment results of each wind turbine in the wind farm to achieve dynamic optimization of active power scheduling of the wind farm.

[0005] Optionally, the multiple components of the wind turbine specified in step 1) include the main bearing of the wind turbine, and part or all of the components in the nacelle.

[0006] Optionally, the functional expression of the multidimensional state feature vector in step 2) is: ; in, For the first time during the monitoring period Multidimensional state feature vectors at each time point For the first time during the monitoring period The error between the predicted component temperature and the actual measured component temperature at n time points. The monitoring period from the real-time operation data of the wind farm SCADA system Actual measured temperatures of n components at n time points. ~ These are the 1st to mth state features, where each state feature represents the error between the predicted and actual measured temperatures of a component, or the actual measured temperature of the component, where m = 2n, and n is the number of components specified for the fan. The functional expression for calculating the Mahalanobis distance between the multidimensional state feature vector and the normal data sample space in step 2) is: ; in, For the first time during the monitoring period Mahalanobis distance between the multidimensional state feature vectors at each time point and the normal data sample space For the first time during the monitoring period Multidimensional state feature vectors at each time point and Let be the mean vector and covariance matrix of the normal data sample space, respectively. for The inverse matrix, in the superscript This is a transpose operation.

[0007] Optionally, in step 3), when determining the deterioration state type of the wind turbine by the difference between the Mahalanobis distance and the preset threshold, the calculation function expression for the difference between the Mahalanobis distance and the preset threshold is: ; in, This is the difference between the Mahalanobis distance and a preset threshold. For the first time during the monitoring period Mahalanobis distance between the multidimensional state feature vectors at each time point and the normal data sample space The threshold is set; if the difference between the Mahalanobis distance of a certain wind turbine and the preset threshold is... If the value is greater than 0, the fan is determined to be in a deteriorated state; otherwise, the fan is determined to be in a healthy state.

[0008] Optionally, a preset threshold is used. The determination includes: defining events with a probability less than a preset probability as impossible events; subtracting the preset probability from 1 to obtain the value of the two-parameter Weibull probability distribution; and determining the preset threshold according to the following formula. : ; in, It is a two-parameter Weibull probability distribution. For shape parameters, This is the scale parameter.

[0009] Optionally, in step 3), when mapping the Mahalanobis distance at each time point within the monitoring period to the health assessment result of the wind turbine, the following formula is used to dynamically calculate the degree of wind turbine degradation using a sliding time window: ; in, Indicates the first Typhoon machine in Degradation rate within each monitoring period The window length represents the monitoring period. Indicates the first The difference between the Mahalanobis distance of the typhoon turbine and a preset threshold; using a fuzzy evaluation algorithm, the degradation degree is mapped to a membership degree by combining a preset membership function, and the membership degree of each typhoon turbine is calculated at the [missing value]. Comprehensive health assessment indicators within each monitoring period: ; in, For the first Typhoon machine in Comprehensive health assessment indicators within each monitoring period, The membership matrix is ​​obtained by combining the degradation degree with the membership function. Evaluation standard vector The transpose of the evaluation criterion vector It includes a set of boundary values ​​to divide the comprehensive health assessment index into multiple level ranges; the corresponding health assessment result is determined based on the level range into which each wind turbine's comprehensive health assessment index falls.

[0010] Optionally, in step 4), when adjusting the output active power of each wind turbine in the wind farm based on its degradation status type and health assessment results to achieve dynamic optimization of the wind farm's active power dispatch, this includes switching healthy wind turbines to maximum power mode for active power output, and using the comprehensive health assessment index as a load reduction factor for degraded wind turbines, multiplying their active power in the default dispatch mode by the load reduction factor to obtain their adaptive derating active power, thereby allocating wind farm power output commands according to the following formula: ; in, Output command for the power to be allocated to the wind farm within the i-th power allocation window. For a healthy number of wind turbines, For the k-th healthy wind turbine, the active power output based on the maximum power mode within the i-th power allocation window. The number of deteriorated wind turbines, For the first The comprehensive health assessment index of the typhoon turbine within the i-th power allocation window. The active power output of the kth degraded wind turbine within the i-th power allocation window based on the default scheduling mode.

[0011] Furthermore, the present invention also provides a wind farm active power optimization control system based on health status assessment, including a microprocessor and a memory interconnected thereto, wherein the microprocessor is programmed or configured to execute the wind farm active power optimization control method based on health status assessment.

[0012] Furthermore, the present invention also provides a computer-readable storage medium storing a computer program or instructions that are programmed or configured to execute the wind farm active power optimization control method based on health status assessment by a processor.

[0013] In addition, the present invention also provides a computer program product, including a computer program or instructions, which are programmed or configured to execute the wind farm active power optimization control method based on health status assessment by a processor.

[0014] Compared with existing technologies, the present invention mainly achieves the following beneficial effects: The wind farm active power optimization control method based on health status assessment includes: training a component normal behavior model by offline modeling based on historical operating data of the wind farm SCADA system, establishing a mapping relationship between key indicators and component predicted temperature; obtaining component temperature prediction values ​​by using multiple key indicators of each component as inputs to the component normal behavior model based on real-time operating data of the wind farm SCADA system; constructing a multi-dimensional state feature vector from the actual measured component temperature and its deviation from the predicted component temperature, and calculating the Mahalanobis distance between the multi-dimensional state feature vector and the normal data sample space; determining the wind turbine's degradation state type by the difference between the Mahalanobis distance and a preset threshold, and mapping the Mahalanobis distance at each time point within the monitoring period to the wind turbine's health assessment result; adjusting the output active power of each wind turbine according to its degradation state type and health assessment result to achieve dynamic optimization of the wind farm's active power scheduling. This invention not only more sensitively detects wind turbine faults but also significantly reduces wind turbine operating temperature and fatigue load, helping to extend wind turbine service life and improve economic benefits. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the basic process of the method in an embodiment of the present invention.

[0016] Figure 2 This is a schematic diagram illustrating the principle of the method in an embodiment of the present invention.

[0017] Figure 3 This refers to the membership function used in the embodiments of the present invention.

[0018] Figure 4 This is a framework diagram of the active power optimization allocation according to an embodiment of the present invention.

[0019] Figure 5 This is a simulation diagram of a healthy fan using the health assessment method in this embodiment of the invention.

[0020] Figure 6 This is a simulation diagram of a faulty fan using the health assessment method in an embodiment of the present invention.

[0021] Figure 7 This is a simulation diagram of the converter operating temperature of a wind turbine under different power control methods in an embodiment of the present invention. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be further described in detail below with reference to the accompanying drawings in the embodiments of the present invention.

[0023] like Figure 1 and Figure 2As shown, the wind farm active power optimization control method based on health status assessment in this embodiment includes the following steps: 1) Offline modeling stage: Based on the historical operation data of the wind farm SCADA system, normal data samples without fault alarm records are selected. For multiple components of the wind turbine, multiple key indicators related to component temperature are selected as inputs. The component temperature is used as a label to train the normal behavior model of the component offline, so as to establish the mapping relationship between key indicators and component predicted temperature. 2) Online monitoring phase: Based on the real-time operation data of the wind farm SCADA system, multiple key indicators of each component are used as inputs to the component normal behavior model to obtain the component temperature prediction value; the actual measured component temperature and its deviation from the component temperature prediction value are used to construct a multi-dimensional state feature vector, and the Mahalanobis distance between the multi-dimensional state feature vector and the normal data sample space is calculated. 3) Health assessment stage: The type of deterioration state of the wind turbine is determined by the difference between the Mahalanobis distance and the preset threshold, and the Mahalanobis distance at each time point within the monitoring period is mapped to the health assessment result of the wind turbine. 4) Power optimization stage: Adjust the output active power of each wind turbine according to the degradation status type and health assessment results of each wind turbine in the wind farm to achieve dynamic optimization of active power scheduling of the wind farm.

[0024] The multiple components specified in step 1) of the wind turbine can be designated according to actual needs. For example, as an optional implementation, the multiple components specified in step 1) of this embodiment include the main shaft and the nacelle. For the main shaft, multiple key indicators related to the main shaft temperature are selected from the historical operating data of the wind farm SCADA system as inputs. The main shaft temperature is used as a label to train the normal behavior model of the main shaft components offline, so as to establish a mapping relationship between the key indicators of the main shaft and the predicted temperature of the main shaft components. For the nacelle, multiple key indicators related to the nacelle temperature are selected from the historical operating data of the wind farm SCADA system as inputs. The nacelle temperature is used as a label to train the normal behavior model of the nacelle components offline, so as to establish a mapping relationship between the key indicators of the nacelle and the predicted temperature of the nacelle components.

[0025] In step 1), when selecting multiple key indicators related to component temperature as input, the correlation between component temperature and indicators can be calculated using the required method, and the correlations can be ranked to select the most strongly correlated indicators as key indicators related to component temperature. The number of key indicators can also be adjusted according to actual needs. For example, as an optional implementation, this embodiment uses the ReliefF method to calculate the correlation between component temperature and indicators, ranks the correlations, and selects the 15 most strongly correlated indicators as 15 key indicators related to component temperature. The ReliefF method is a well-known correlation calculation method. Its core principle is that key features can more clearly distinguish samples of different categories, making similar samples closer and dissimilar samples further apart.

[0026] In step 1), the normal behavior model of the component can adopt the required machine learning model as needed. For example, as an optional implementation, in this embodiment, a gated recurrent neural network (GRU) is used as the normal behavior model of the component. The gated recurrent neural network (GRU) is a multi-input single-output model. The input features are multiple key indicators in SCADA that are most relevant to the prediction target, and the output is the corresponding component prediction temperature, such as the main bearing temperature and the nacelle temperature, thereby establishing the behavioral benchmarks of the main bearing temperature and the nacelle temperature under the normal operating conditions of the wind turbine.

[0027] In step 1), the offline modeling stage establishes a normal behavior model for each component, creating a mapping relationship between key indicators and predicted component temperatures, which can then be used to predict component temperatures. In step 2), based on real-time operational data from the wind farm's SCADA system, multiple key indicators of each component are used as inputs to the normal behavior model to obtain predicted component temperatures.

[0028] The functional expression for the multidimensional state feature vector in step 2) of this embodiment is: ; in, For the first time during the monitoring period Multidimensional state feature vectors at each time point For the first time during the monitoring period The error between the predicted component temperature and the actual measured component temperature at n time points. The monitoring period from the real-time operation data of the wind farm SCADA system Actual measured temperatures of n components at n time points. ~ These are the 1st to mth state features, where each state feature represents the error between the predicted and actual measured temperatures of a component, or the actual measured temperature of the component, where m = 2n, and n is the number of components specified for the fan. The functional expression for calculating the Mahalanobis distance between the multidimensional state feature vector and the normal data sample space in step 2) is: ; in, For the first time during the monitoring period Mahalanobis distance between the multidimensional state feature vectors at each time point and the normal data sample space For the first time during the monitoring period Multidimensional state feature vectors at each time point and Let be the mean vector and covariance matrix of the normal data sample space, respectively. for The inverse matrix, in the superscript This is a transpose operation.

[0029] In step 3) of this embodiment, when determining the deterioration state type of the wind turbine by the difference between the Mahalanobis distance and the preset threshold, the calculation function expression for the difference between the Mahalanobis distance and the preset threshold is: ; in, This is the difference between the Mahalanobis distance and a preset threshold. For the first time during the monitoring period Mahalanobis distance between the multidimensional state feature vectors at each time point and the normal data sample space The threshold is set; if the difference between the Mahalanobis distance of a certain wind turbine and the preset threshold is... If the value is greater than 0, the fan's degradation state is determined to be degraded; otherwise, the fan's degradation state is determined to be healthy. The preset threshold corresponds to the Mahalanobis distance. The settings can be adjusted according to actual needs. For example, as an optional implementation, in this embodiment, a preset threshold corresponding to the Mahalanobis distance is used. A two-parameter Weibull probability distribution is used for accurate fitting. Preset threshold. The determination includes: defining events with a probability less than a preset probability as impossible events; subtracting the preset probability from 1 to obtain the value of the two-parameter Weibull probability distribution; and determining the preset threshold according to the following formula. : ; in, It is a two-parameter Weibull probability distribution. For shape parameters, This is the scaling parameter. For example, events with a probability less than 0.05 are generally considered impossible. Therefore, the value of the two-parameter Weibull probability distribution can be defined as 0.95. Substituting this value into the function expression of the two-parameter Weibull probability distribution yields the preset threshold. The value of overcomes the shortcoming of the traditional threshold method where the error value characteristics do not change significantly.

[0030] In step 3) of this embodiment, when mapping the Mahalanobis distance at each time point within the monitoring period to the health assessment result of the wind turbine, the following formula is used to dynamically calculate the degree of deterioration of the wind turbine using a sliding time window: ; in, Indicates the first Typhoon machine in Degradation rate within each monitoring period The window length represents the monitoring period. Indicates the first The difference between the Mahalanobis distance of the typhoon generator and the preset threshold; using a fuzzy evaluation algorithm, the degradation degree is mapped to the [0,1] interval by combining the preset membership function, and the membership degree of each typhoon generator is calculated in the [0,1] interval. Comprehensive health assessment indicators within each monitoring period: ; in, For the first Typhoon machine in Comprehensive health assessment indicators within each monitoring period, The membership matrix is ​​obtained by combining the degradation degree with the membership function. Evaluation standard vector The transpose of the evaluation criterion vector It includes a set of boundary values ​​to divide the comprehensive health assessment index into multiple level ranges; the corresponding health assessment result is determined based on the level range into which each wind turbine's comprehensive health assessment index falls.

[0031] Figure 3 The membership functions used in this embodiment are g1 to g4, which represent the degree of degradation. They include the membership functions of the four health assessment results: "Good", "Attention", "Warning" and "Danger". The membership functions used for "Attention" and "Warning" are trigonometric functions, while the membership functions used for "Good" and "Danger" are semi-trapezoidal functions.

[0032] Introducing evaluation criterion vectors The aim is to quantify health into a single indicator while simultaneously conducting a fuzzy comprehensive evaluation of health, maximizing the utilization of information from the four membership degrees, and facilitating the optimal allocation of active power using this comprehensive evaluation indicator. This is achieved through the evaluation standard vector. Integrating four membership degrees into a single comprehensive health evaluation index simplifies the assessment of health status, quantifies fuzzy evaluation results, and enables intuitive grading and rapid decision-making regarding health status. As an optional implementation method, this embodiment uses an evaluation standard vector. The value can be: ; When the membership matrix corresponding to the degradation degree at a certain moment is When the value is "attention", the membership degree of the corresponding evaluation is the largest. That is, the health assessment result of the fuzzy evaluation algorithm for the wind turbine at this moment is "attention", and the corresponding comprehensive health evaluation index is 0.5×0.8+0.6×0.6+0×0.4+0×0.2=0.76.

[0033] Evaluation Criterion Vector The system includes a set of boundary values ​​to divide the comprehensive health assessment index into multiple levels: greater than or equal to 0.8, between [0.6, 0.8), between [0.4, 0.6), and less than 0.4. Ultimately, if the comprehensive health assessment index is greater than or equal to 0.8, the health assessment result is "Good"; if it is between [0.6, 0.8), the result is "Caution"; if it is between [0.4, 0.6), the result is "Warning"; and if it is less than 0.4, the result is "Danger". Since 0.76 falls within [0.6, 0.8), the final health assessment result for the wind turbine is also "Caution".

[0034] In step 4) of this embodiment, when adjusting the output active power of each wind turbine based on its degradation status type and health assessment results to achieve dynamic optimization of the wind farm's active power scheduling, the dynamic optimization wind farm active power scheduling strategy employs adaptive de-capacity operation for degraded turbines while increasing the power quota of healthy turbines. This includes switching healthy turbines to maximum power mode for active power output, and using the comprehensive health assessment index as a load reduction factor for degraded turbines, multiplying their active power under the default scheduling mode by the load reduction factor to obtain their adaptive de-capacity active power. The wind farm power output command is then allocated according to the following formula: ; in, Output command for the power to be allocated to the wind farm within the i-th power allocation window. For a healthy number of wind turbines, For the k-th healthy wind turbine, the active power output based on the maximum power mode within the i-th power allocation window. The number of deteriorated wind turbines, For the first The comprehensive health assessment index of the typhoon turbine within the i-th power allocation window. Let the active power output of the k-th degraded wind turbine be based on the default scheduling mode within the i-th power allocation window. Let the time period of the power allocation window be denoted as . The time period of the monitoring cycle is as follows: This allows for the integration of the power allocation window's time period. The time period of the monitoring cycle Optimize the rolling time of the wind turbine. Referring to the formula for allocating power output commands to the wind farm mentioned above, it can be seen that for... For a healthy wind turbine, the active power output is based on the maximum power mode. For example, MPPT technology is used to extract maximum power from the wind until the rated power is reached, thereby improving the power quota of the healthy wind turbine; for For a deteriorated wind turbine, its comprehensive health assessment index is used as a load reduction factor, thereby achieving the goal of adaptive capacity reduction operation for the deteriorated wind turbine.

[0035] Figure 4 This is a framework diagram of the active power optimization allocation in this embodiment. After receiving the wind farm power output command to be allocated, the power control center uses the comprehensive health assessment index of the deteriorated wind turbine as the load reduction factor, and multiplies the active power of the turbine in the default scheduling mode by the load reduction factor to obtain its adaptive derating active power. Based on the above formula, the power output command of the wind farm is allocated to N wind turbines WT1 to WT2. N , where P1~P N Then, the power allocated to N wind turbines (i.e., or Meanwhile, the wind farm's SCADA system monitors N wind turbines WT1 to WT2. N Status monitoring is performed to obtain real-time operational data. The health assessment method described in this embodiment is used to obtain the WT1~WT values ​​for N wind turbines. N The comprehensive health assessment indicators are sent to the power control center for power allocation in the next cycle.

[0036] The active power optimization control method for wind farms based on health status assessment in this embodiment was simulated and verified. The results are as follows: Figures 5-7 As shown. Figure 5The simulation diagram of a healthy wind turbine using the method of this embodiment shows that although the Mahalanobis distance of the healthy wind turbine momentarily exceeds the threshold, this is due to the abnormality of instantaneous SCADA data acquisition or the transient operation process of the wind turbine. Since the sliding time window comprehensively considers the Mahalanobis distance error within three days of the monitoring period, the deterioration of the unit remains at a very low level. It can be seen that this method can correctly assess the state of a healthy wind turbine and has a certain anti-interference capability. Figure 6 The simulation diagram of the faulty wind turbine using the method of this embodiment shows that the deterioration degree of the unit with the faulty wind turbine will increase significantly before the fault is cleared. Compared with the anomaly monitoring method that directly uses error value and simple threshold, the method of this embodiment detects the state anomaly by using Mahalanobis distance statistical outlier detection technology, which overcomes the disadvantage of the insignificant change of error value characteristics when using the traditional threshold method. Figure 7 The figures show simulations of the converter operating temperature of wind turbines under different power control methods according to this embodiment. For wind turbines with fault records, the proposed power optimization method automatically controls the turbine to operate in derating mode during the fault period. Compared to traditional proportional power allocation, this embodiment's method can maintain the converter temperature at a lower level, preventing further damage to components due to excessively high temperatures during the fault period. Simultaneously, the lower power output significantly reduces the turbine's torque, thereby reducing fatigue load and better protecting faulty turbines.

[0037] In summary, to address the technical problems existing in current wind farm active power allocation technologies, this embodiment establishes a normal behavior model of each key component of the wind turbine in the offline stage using a neural network. By constructing a multi-dimensional state feature vector through the model's output predicted values ​​and actual measured values, anomalies are monitored using Mahalanobis distance, and a fuzzy evaluation algorithm is employed to quantify the overall health status of the wind turbine. This optimizes the active power allocation of the wind farm, reduces the operating temperature and fatigue load of faulty wind turbines, and helps extend the service life of wind turbines and improve economic benefits.

[0038] Furthermore, this embodiment also provides a wind farm active power optimization control system based on health status assessment, including a microprocessor and a memory interconnected, wherein the microprocessor is programmed or configured to execute the wind farm active power optimization control method based on health status assessment. This embodiment also provides a computer-readable storage medium storing a computer program or instructions programmed or configured to execute the wind farm active power optimization control method based on health status assessment via a processor. This embodiment also provides a computer program product including a computer program or instructions programmed or configured to execute the wind farm active power optimization control method based on health status assessment via a processor.

[0039] The present invention can implement all or part of the processes in the methods of the above embodiments, or it can be implemented by hardware related to computer program instructions. The computer program can be stored in a computer-readable storage medium or provided for online installation as a product in a software marketplace. When the computer program is executed by a processor, it can implement the steps of the above method embodiments. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable storage media include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. The memory is used to store computer programs and / or modules. The processor implements various functions by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may include high-speed random access memory, as well as non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital (SD) cards, flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0040] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A method for optimizing active power control in wind farms based on health status assessment, characterized in that, Includes the following steps: 1) Based on the historical operation data of the wind farm SCADA system, normal data samples without fault alarm records are selected. For multiple components of the wind turbine, multiple key indicators related to component temperature are selected as inputs. Component temperature is used as a label to train the normal behavior model of the component offline, so as to establish the mapping relationship between key indicators and component predicted temperature. 2) Based on the real-time operation data of the wind farm SCADA system, multiple key indicators of each component are used as inputs to the component normal behavior model to obtain the component temperature prediction value; the actual measured value of component temperature and its deviation from the component temperature prediction value are used to construct a multi-dimensional state feature vector, and the Mahalanobis distance between the multi-dimensional state feature vector and the normal data sample space is calculated. 3) The type of deterioration state of the wind turbine is determined by the difference between the Mahalanobis distance and the preset threshold, and the Mahalanobis distance at each time point in the monitoring period is mapped to the health assessment result of the wind turbine. 4) Adjust the output active power of each wind turbine according to the degradation status type and health assessment results of each wind turbine in the wind farm to achieve dynamic optimization of active power scheduling of the wind farm.

2. The wind farm active power optimization control method based on health status assessment according to claim 1, characterized in that, The multiple components specified in step 1) of the wind turbine include the main bearing of the wind turbine and part or all of the components in the nacelle.

3. The wind farm active power optimization control method based on health status assessment according to claim 1, characterized in that, The functional expression for the multidimensional state feature vector in step 2) is: ； in, For the first time during the monitoring period Multidimensional state feature vectors at each time point For the first time during the monitoring period The error between the predicted component temperature and the actual measured component temperature at n time points. The monitoring period from the real-time operation data of the wind farm SCADA system Actual measured temperatures of n components at n time points. ~ These are the 1st to mth state features, where each state feature represents the error between the predicted and actual measured temperatures of a component, or the actual measured temperature of the component, where m = 2n, and n is the number of components specified for the fan. The functional expression for calculating the Mahalanobis distance between the multidimensional state feature vector and the normal data sample space in step 2) is: ； in, For the first time during the monitoring period Mahalanobis distance between the multidimensional state feature vectors at each time point and the normal data sample space For the first time during the monitoring period Multidimensional state feature vectors at each time point and Let be the mean vector and covariance matrix of the normal data sample space, respectively. for The inverse matrix, in the superscript This is a transpose operation.

4. The wind farm active power optimization control method based on health status assessment according to claim 1, characterized in that, In step 3), when determining the deterioration state type of the wind turbine by the difference between the Mahalanobis distance and the preset threshold, the calculation function expression for the difference between the Mahalanobis distance and the preset threshold is: ； in, This is the difference between the Mahalanobis distance and a preset threshold. For the first time during the monitoring period Mahalanobis distance between the multidimensional state feature vectors at each time point and the normal data sample space The threshold is set; if the difference between the Mahalanobis distance of a certain wind turbine and the preset threshold is... If the value is greater than 0, the fan is determined to be in a deteriorated state; otherwise, the fan is determined to be in a healthy state.

5. The wind farm active power optimization control method based on health status assessment according to claim 4, characterized in that, Preset threshold The determination includes: defining events with a probability less than a preset probability as impossible events; subtracting the preset probability from 1 to obtain the value of the two-parameter Weibull probability distribution; and determining the preset threshold according to the following formula. : ； in, It is a two-parameter Weibull probability distribution. For shape parameters, This is the scale parameter.

6. The wind farm active power optimization control method based on health status assessment according to claim 1, characterized in that, In step 3), when mapping the Mahalanobis distance at each time point within the monitoring period to the health assessment results of the wind turbine, the following formula is used to dynamically calculate the degree of wind turbine degradation using a sliding time window: ； in, Indicates the first Typhoon machine in Degradation rate within each monitoring period The window length represents the monitoring period. Indicates the first The difference between the Mahalanobis distance of the typhoon turbine and a preset threshold; using a fuzzy evaluation algorithm, the degradation degree is mapped to a membership degree by combining a preset membership function, and the membership degree of each typhoon turbine is calculated at the [missing value]. Comprehensive health assessment indicators within each monitoring period: ； in, For the first Typhoon machine in Comprehensive health assessment indicators within each monitoring period, The membership matrix is ​​obtained by combining the degradation degree with the membership function. Evaluation standard vector The transpose of the evaluation criterion vector It includes a set of boundary values ​​to divide the comprehensive health assessment index into multiple level ranges; the corresponding health assessment result is determined based on the level range into which each wind turbine's comprehensive health assessment index falls.

7. The wind farm active power optimization control method based on health status assessment according to claim 6, characterized in that, In step 4), adjusting the output active power of each wind turbine in the wind farm based on its degradation status and health assessment results to achieve dynamic optimization of the wind farm's active power dispatch includes switching healthy turbines to maximum power mode for active power output, and using the comprehensive health assessment index as a load reduction factor for degraded turbines, multiplying their active power in the default dispatch mode by the load reduction factor to obtain their adaptive derating active power. The wind farm's power output command is then allocated according to the following formula: ； in, Output command for the power to be allocated to the wind farm within the i-th power allocation window. For a healthy number of wind turbines, For the k-th healthy wind turbine, the active power output based on the maximum power mode within the i-th power allocation window. The number of deteriorated wind turbines, For the first The comprehensive health assessment index of the typhoon turbine within the i-th power allocation window. The active power output of the kth degraded wind turbine within the i-th power allocation window based on the default scheduling mode.

8. A wind farm active power optimization control system based on health status assessment, comprising a microprocessor and a memory interconnected, characterized in that, The microprocessor is programmed or configured to execute the wind farm active power optimization control method based on health status assessment as described in any one of claims 1 to 7.

9. A computer-readable storage medium storing a computer program or instructions, characterized in that, The computer program or instructions are programmed or configured to execute, via a processor, the wind farm active power optimization control method based on health status assessment as described in any one of claims 1 to 7.

10. A computer program product, comprising a computer program or instructions, characterized in that, The computer program or instructions are programmed or configured to execute, via a processor, the wind farm active power optimization control method based on health status assessment as described in any one of claims 1 to 7.

Images