Standardized analysis method and device for operation difficulty and operation level of wind turbine generator and station

Abstract

The invention provides a standardized analysis method and device for operation difficulty and operation level of a wind turbine generator and a station, and the method comprises the steps: obtaining basic attribute data, operation condition data and operation index data of a plurality of wind power devices and stations to which the wind power devices belong, so as to construct a training data set; a standard value calculation model is constructed based on machine learning, the input of the standard value calculation model is basic attribute data and operation condition data, the output of the standard value calculation model is operation index data, and the training data set is utilized to train the standard value calculation model to obtain a trained standard value calculation model; inputting the basic attribute data and the operation condition data of the to-be-analyzed wind power equipment in multiple periods into a trained standard value calculation model to obtain operation index standard values of the to-be-analyzed wind power equipment in corresponding periods; and on the basis of the operation index standard values, the same-period theoretical optimal values and the current-period operation actual values of the to-be-analyzed wind power equipment, operation difficulty coefficients and operation level coefficients corresponding to the to-be-analyzed wind power equipment and the to-be-analyzed station are obtained through calculation.

Description

Technical Field

[0001] This invention relates to the field of wind power generation technology, and in particular to a standardized analysis method and apparatus for the operational difficulty and operational level of wind turbine generators and wind farms. Background Technology

[0002] Under my country's "3060" target, new energy industries such as wind power are developing rapidly. Compared with traditional power plants, wind power equipment and stations are affected by multiple factors such as operation and maintenance level, regional differences, resource and climate conditions, equipment selection, and power plant technology type. The actual operating level is greatly affected by objective factors beyond subjective operation capabilities, making it difficult to distinguish the operational difficulty under given conditions. The analysis of operating level and capability is more difficult and cannot effectively guide the operation and maintenance of power plants.

[0003] Several comprehensive analysis methods for wind farms exist in the prior art. For example, Chinese invention patent application CN113610443A discloses a method for assessing the health and service quality of wind turbine generators. This invention, by comprehensively considering the severity, frequency, and duration of various faults in wind turbine generators, uses Failure Mode and Effects Analysis (FMEA) to define and calculate the health of the wind turbine generators, achieving a comprehensive quantitative assessment of their health status. Furthermore, by comprehensively considering performance and reliability indicators such as the health of the wind turbine generators, equivalent available hours, availability, mean time between failures (MTBF), mean maintenance hours, and power loss, a multi-attribute decision-making method is used to define and calculate the service quality index of the wind turbine generators, achieving a comprehensive quantitative assessment of the overall service status of the wind turbine generators. The advantage of this invention lies in its ability to automatically calculate the comprehensive impact of faults on the reliability of wind turbine generators and to comprehensively quantitatively assess the overall service status of wind turbine generators. However, this method only supports the evaluation of the absolute value of relevant indicators, cannot eliminate the influence of uncontrollable factors, and is difficult to objectively evaluate the evaluation subjects affected by different objective factors; it only supports the evaluation of the health status of wind turbine equipment and cannot be extended to the site; it only uses the fault status of the unit to evaluate the health level, and does not include operational indicators that reflect equipment performance, control strategies, etc.

[0004] For example, Chinese invention patent application CN118462500A discloses a comprehensive performance evaluation method for smart wind farms. This method includes: constructing a dynamic adaptive evaluation index system, real-time monitoring and updating technical parameters, and introducing machine learning to assist index optimization; an objective and data-driven evaluation process, standardized power curve evaluation, intelligent fault diagnosis and repair time prediction; comprehensive evaluation integrating multi-source data and intelligent algorithms, multi-source data integration, deep learning performance evaluation model, model interpretation and transparent evaluation; intelligent decision support and continuous optimization, automated generation of performance evaluation reports, analysis of key influencing factors, improvement suggestions, and visualization. This invention effectively solves the shortcomings of existing technologies and fully reflects the intelligent characteristics of smart wind farms. However, this method has drawbacks: it only uses site monitoring data for machine learning model construction, without considering the influence of external objective factors and internal management factors; the required monitoring data has high requirements for the level of intelligence of the site, which is not conducive to low-cost implementation.

[0005] Therefore, existing technologies need a method for calculating and comparing standard values ​​of operating levels that comprehensively consider the subjective, objective, variable, and immutable conditions of wind power equipment and stations, in order to achieve standardized analysis that eliminates uncontrollable differences. Summary of the Invention

[0006] The present invention aims to at least partially solve one of the technical problems in the related art.

[0007] Therefore, the first objective of this invention is to propose a standardized analysis method for the operational difficulty and operational level of wind turbine generators and wind farms, so as to achieve accurate standardized analysis of the operational difficulty and operational level of wind turbine generators and wind farms.

[0008] The second objective of this invention is to provide a standardized analysis device for the operational difficulty and operational level of wind turbine generators and wind farms.

[0009] The third objective of this invention is to provide an electronic device.

[0010] The fourth objective of this invention is to provide a computer-readable storage medium.

[0011] To achieve the above objectives, the first aspect of this invention proposes a standardized analysis method for the operational difficulty and operational level of wind turbine generators and wind farms, comprising:

[0012] Acquire basic attribute data, operating condition data, and operating indicator data of multiple wind power devices and their respective wind farms to construct a training dataset;

[0013] A standard value calculation model is constructed based on machine learning. The input of the standard value calculation model is basic attribute data and operating condition data, and the output is operating indicator data. The standard value calculation model is trained using the training dataset to obtain a trained standard value calculation model.

[0014] The basic attribute data and operating condition data of the wind power equipment to be analyzed in multiple periods are input into the trained standard value calculation model to obtain the corresponding period's standard values ​​of the operating indicators of the wind power equipment to be analyzed.

[0015] Based on the standard values ​​of the operating indicators of the wind power equipment to be analyzed in each period, the theoretical optimal values ​​of the same period, and the actual operating values ​​of the current period, the operating difficulty coefficient and operating level coefficient of the wind power equipment and the wind farm to be analyzed are calculated.

[0016] In the method of the first aspect of the present invention, the operating difficulty coefficient and operating level coefficient corresponding to the wind power equipment and the wind farm to be analyzed are calculated based on the standard values ​​of the operating indicators of the wind power equipment to be analyzed in each period, the theoretical optimal values ​​of the same period, and the actual operating values ​​of the current period. The calculation includes: calculating the operating difficulty coefficient of the wind power equipment to be analyzed based on the standard values ​​of the operating indicators of the wind power equipment to be analyzed and the theoretical optimal values ​​of the same period; calculating the operating difficulty coefficient of the wind farm to be analyzed based on the standard values ​​of the operating indicators of the wind farm to be analyzed and the theoretical optimal values ​​of the same period; calculating the operating level coefficient of the wind power equipment to be analyzed based on the standard values ​​of the operating indicators of the wind power equipment to be analyzed in the current period, the theoretical optimal values ​​of the current period, and the actual operating values ​​of the current period; and calculating the operating level coefficient of the wind farm to be analyzed based on the standard values ​​of the operating indicators of the wind farm to be analyzed in the current period, the theoretical optimal values ​​of the current period, and the actual operating values ​​of the current period.

[0017] In the method of the first aspect of the present invention, the standard value calculation model is trained using the training dataset by employing a k-fold training and cross-validation method.

[0018] In the method of the first aspect of the present invention, the basic attribute data and the operating condition data each include multiple parameters. The parameters included in the basic attribute data and the operating condition data are filtered and dimensionality reduced. The filtered and dimensionality-reduced basic attribute data and operating condition data are used as inputs to the standard value calculation model to train the standard value calculation model.

[0019] In the method of the first aspect of the present invention, correlation analysis and association rule discovery methods are used to filter and reduce the dimensionality of the parameters included in the basic attribute data and the operating condition data.

[0020] In the method of the first aspect of the present invention, the basic attribute data includes a variety of predetermined attribute parameters such as province and city, topography, equipment model, station type, years of operation, and spatial distribution of equipment; the operating conditions include objective uncontrollable condition parameters and controllable condition parameters. The objective uncontrollable condition parameters include a variety of parameters such as wind speed, temperature, weather, and power grid obstruction, while the controllable condition parameters include a variety of parameters such as the number of maintenance personnel, cost input, and planned downtime.

[0021] In the method of the first aspect of the present invention, the operational indicator data includes at least one key indicator parameter among power generation, multiple power loss, time availability, and energy utilization.

[0022] To achieve the above objectives, a second aspect of the present invention provides a standardized analysis device for the operational difficulty and operational level of wind turbine generators and wind farms, comprising:

[0023] The acquisition module is used to acquire basic attribute data, operating condition data, and operating indicator data of multiple wind power devices and their respective wind farms in order to construct a training dataset.

[0024] The modeling module is used to build a standard value calculation model based on machine learning. The input of the standard value calculation model is basic attribute data and running condition data, and the output is running indicator data. The standard value calculation model is trained using the training dataset to obtain a trained standard value calculation model.

[0025] The calculation module is used to input the basic attribute data and operating condition data of the wind power equipment to be analyzed from multiple periods into the trained standard value calculation model to obtain the corresponding period's standard values ​​of the operating indicators of the wind power equipment to be analyzed.

[0026] The analysis module is used to calculate the operation difficulty coefficient and operation level coefficient of the wind power equipment and the site under analysis based on the standard values ​​of the operation indicators of the wind power equipment under analysis in each period, the theoretical optimal values ​​of the same period, and the actual operation values ​​of the current period.

[0027] To achieve the above objectives, a third aspect of the present invention provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer execution instructions; the processor executes the computer execution instructions stored in the memory to implement the method proposed in the first aspect of the present invention.

[0028] To achieve the above objectives, a fourth aspect of the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method proposed in the first aspect of the present invention.

[0029] The present invention provides a standardized analysis method, device, electronic equipment, and storage medium for wind turbine and wind farm operation difficulty and operation level. It acquires basic attribute data, operating condition data, and operation index data of multiple wind turbines and their associated wind farms to construct a training dataset. A standard value calculation model is built based on machine learning. The input of the standard value calculation model is the basic attribute data and operating condition data, and the output is the operation index data. The standard value calculation model is trained using the training dataset to obtain a trained standard value calculation model. The basic attribute data and operating condition data of the wind turbines to be analyzed from multiple periods are input into the trained standard value calculation model to obtain the corresponding period's standard values ​​of the operation indexes for the wind turbines to be analyzed. Based on the standard values ​​of the operation indexes for each period of the wind turbines to be analyzed, the theoretical optimal values ​​for the same period, and the actual operating values ​​for the current period, the operation difficulty coefficient and operation level coefficient corresponding to the wind turbines and wind farms to be analyzed are calculated. In this context, a standardized value calculation model was constructed by comprehensively considering various data sources, including basic attribute data, operating condition data, and operating index data of wind turbines and their associated power plants. This model, after training, yielded the corresponding period's standardized operating index values ​​for the wind turbines under analysis. By combining the theoretically optimal values ​​for the same period with the actual operating values ​​for the current period, the operational difficulty and level of the wind turbines and power plants under analysis were determined. This allows for a more accurate and standardized analysis of the operational difficulty and level of wind turbines and power plants.

[0030] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0031] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0032] Figure 1 This is a flowchart illustrating a standardized analysis method for the operational difficulty and operational level of wind turbine generators and wind farms provided in an embodiment of the present invention.

[0033] Figure 2 This is a schematic diagram illustrating the specific process of the screening and dimensionality reduction processing provided in the embodiments of the present invention.

[0034] Figure 3 This is a schematic diagram illustrating the specific process for obtaining the operational difficulty coefficient and operational level coefficient provided in an embodiment of the present invention.

[0035] Figure 4 This is a block diagram of a standardized analysis device for the operation difficulty and operation level of wind turbine generators and wind farms, provided in an embodiment of the present invention. Detailed Implementation

[0036] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0037] The following describes, with reference to the accompanying drawings, a standardized analysis method and apparatus for the operational difficulty and operational level of wind turbine generators and wind farms according to embodiments of the present invention.

[0038] This invention provides a standardized analysis method for the operational difficulty and operational level of wind turbines and wind farms, so as to achieve accurate standardized analysis of the operational difficulty and operational level of wind turbines and wind farms.

[0039] Figure 1 This is a flowchart illustrating a standardized analysis method for the operational difficulty and operational level of wind turbine generators and wind farms, provided in an embodiment of the present invention.

[0040] like Figure 1 As shown, the standardized analysis method for the operational difficulty and operational level of this wind turbine and its power station includes the following steps:

[0041] Step S101: Obtain basic attribute data, operating condition data, and operating indicator data of multiple wind power devices and their respective wind farms to construct a training dataset.

[0042] In step S101, multiple wind turbines and their associated wind farms include the objects to be analyzed. The objects to be analyzed include the wind turbines and the wind farms to be analyzed. It should be noted that a certain number of wind turbines and their associated wind farms can be selected as the objects to be analyzed from all the acquired data, or all wind turbines and their associated wind farms can be analyzed separately. For example, data from 18,253 wind turbines and their associated 463 wind farms can be used as the overall sample, and all wind turbines and wind farms can be analyzed; this is not limited here.

[0043] In step S101, the basic attribute data includes, but is not limited to, various predetermined attribute parameters such as province / city, topography, equipment model, station type, years of operation, and spatial distribution of equipment.

[0044] In step S101, the operating conditions include objective uncontrollable condition parameters and controllable condition parameters. Objective uncontrollable condition parameters include, but are not limited to, various parameters such as wind speed, temperature, weather, and power grid obstruction. Controllable condition parameters include, but are not limited to, various parameters such as the number of maintenance personnel, cost input, and planned shutdown.

[0045] In step S101, the operational indicator data includes, but is not limited to, various key indicator parameters such as power generation, multiple power losses, time availability, and energy utilization. Among them, multiple power losses include, but are not limited to, power losses due to fault outages, power losses due to self-degradation, power losses due to performance issues, power losses due to planned outages, and other power losses.

[0046] In step S101, the acquired operating condition data includes current operating condition data and historical operating condition data for multiple periods.

[0047] In step S101, the acquired operational indicator data includes historical operational indicator data from multiple periods.

[0048] In step S101, the data obtained from multiple wind power devices and their respective sites also include the theoretical optimal values ​​of operating indicators for the current period and multiple historical periods under given operating conditions.

[0049] In step S101, the acquired basic attribute data, operating condition data, and operating indicator data can be processed by extracting features to obtain corresponding feature values. The feature values ​​of each type of data are then used to represent each type of data for subsequent training and calculation.

[0050] In step S101, all parameters included in the acquired basic attribute data, operating condition data, and operating indicator data can be divided into qualitative data and quantitative data from a qualitative and quantitative perspective.

[0051] In step S101, a training dataset is constructed based on the basic attribute data, operating condition data, and operating index data of multiple wind turbines and their respective wind farms. Specifically, the wind turbines and their respective wind farms included in the training dataset may or may not include the objects to be analyzed. Historical basic attribute data and operating condition data from multiple periods in the training dataset are used as model input variables, and the corresponding operating index data are used as the corresponding labels. For parameters in the training dataset that are qualitative data, these parameters (i.e., their feature values) can be quantized and encoded; for parameters in the training dataset that are quantitative data, these parameters (i.e., their feature values) can be normalized, thereby constructing a training dataset for subsequent steps.

[0052] Step S102: Construct a standard value calculation model based on machine learning. The input of the standard value calculation model is basic attribute data and operating condition data, and the output is operating indicator data. The standard value calculation model is trained using the training dataset to obtain the trained standard value calculation model.

[0053] In step S102, the machine learning model can be a model that supports regression of multiple continuous variables, such as the random forest regression model used in this embodiment of the invention. In some embodiments, a neural network-based model can also be used, which is not limited here.

[0054] In step S102, the input to the standard value calculation model is basic attribute data and operating condition data, and the output is operating index data. The output operating index data includes key operating indicators such as power generation, power loss due to fault outage, power loss due to self-degradation, power loss due to performance, power loss due to planned outage, other power loss, time availability, and energy utilization rate.

[0055] In step S102, when training the standard value calculation model using the training dataset, a k-fold training and cross-validation method can be employed. For example, basic attribute data and historical operating condition data can be used as input variables, and key operating indicators as output variables to construct a random forest regression model. The overall sample, including the object to be analyzed, is divided into k sets. One set is used as the test sample set, and the remaining sets are used as the training sample sets to train k models. The training parameters are determined with the goal of achieving the optimal average accuracy of the k models. The standard values ​​of the operating indicators of the object to be analyzed are calculated using models whose training sample sets do not include the object to be analyzed, supporting subsequent analysis. This achieves the calculation of the standard values ​​of key operating indicators for each wind turbine under given objective and immutable conditions. The test sample set and the training sample set constitute the training dataset.

[0056] In step S102, after obtaining the trained standard value calculation model, the input variables of the object to be analyzed during the analysis period can be input into the model, and the output of the model is the standard value of the index of the object to be analyzed during the analysis period.

[0057] In step S102, considering that the basic attribute data and the operating condition data each include multiple parameters, in order to improve the efficiency and accuracy of model training, the parameters included in the basic attribute data and the operating condition data are filtered and dimensionality reduced. The filtered and dimensionality-reduced basic attribute data and operating condition data are then used as input to the standard value calculation model for training. Specifically, correlation analysis and association rule discovery methods are used to filter and reduce the dimensionality of the parameters included in the basic attribute data and the operating condition data.

[0058] Figure 2 This is a schematic diagram illustrating the specific process of the screening and dimensionality reduction process provided in an embodiment of the present invention.

[0059] Specifically, such as Figure 2As shown, the screening and dimensionality reduction process includes: performing linear and nonlinear correlation analysis on the quantitative input variables (i.e., parameters belonging to quantitative data in basic attribute data and operating condition data) in the input variables of the standard value calculation model to obtain the linear and nonlinear correlation coefficients between each input variable and with the output variable, i.e., the key operating indicators (step S201); performing association rule discovery analysis on the qualitative input variables (i.e., parameters belonging to qualitative data in basic attribute data and operating condition data) in the input variables of the standard value calculation model to obtain the frequent itemsets between each input variable and with the output variable, i.e., the key operating indicator intervals (step S202); and screening input variables based on the linear and nonlinear correlation coefficients and frequent itemset analysis results between input variables and with the output variable (step S203). During screening, parameters with correlation coefficients greater than a set threshold can be retained (one type) and the remaining types deleted; itemsets in the frequent itemsets can be retained (one type) and the remaining types deleted.

[0060] In step S102, for each type of parameter in the basic attribute data and operating condition data after dimensionality reduction and filtering, corresponding parameters are obtained to obtain a new training dataset. Then, the standard value calculation model is trained. It is easy to understand that after training, when using this model to calculate standard values ​​subsequently, the types of input parameters are consistent with the types of parameters in the basic attribute data and operating condition data after dimensionality reduction and filtering.

[0061] In step S102, optionally, in the construction of the standard value calculation model, based on the analysis objectives based on management needs, some basic attributes and operating conditions are selectively selected for input or non-input for training, so as to realize the calculation of standard values ​​under different premises.

[0062] Step S103: Input the basic attribute data and operating condition data of the wind power equipment to be analyzed from multiple periods into the trained standard value calculation model to obtain the corresponding period's standard values ​​of the operating indicators of the wind power equipment to be analyzed.

[0063] In step S103, for each period, the basic attribute data and operating condition data of the wind power equipment to be analyzed in that period are input into the trained standard value calculation model to obtain the standard values ​​of the operating indicators of the wind power equipment to be analyzed in that period. Since the operating indicator data includes various key parameters such as power generation, power loss due to fault outage, power loss due to self-degradation, power loss due to performance degradation, power loss due to planned outage, other power losses, time availability, and energy utilization, the standard values ​​of the operating indicators include standard values ​​for power generation, power loss due to fault outage, power loss due to self-degradation, power loss due to performance degradation, power loss due to planned outage, other power losses, time availability, and energy utilization.

[0064] Step S104: Based on the standard values ​​of the operating indicators of the wind power equipment to be analyzed in each period, the theoretical optimal values ​​of the same period, and the actual operating values ​​of the current period, calculate the operating difficulty coefficient and operating level coefficient of the wind power equipment and the wind farm to be analyzed.

[0065] Figure 3 This is a schematic diagram illustrating the specific process for obtaining the operational difficulty coefficient and operational level coefficient provided in an embodiment of the present invention.

[0066] In step S104, the operational difficulty coefficient and operational level coefficient of the wind power equipment to be analyzed and the wind farm to be analyzed are calculated based on the standard values ​​of the operational indicators of the wind power equipment to be analyzed in each period, the theoretical optimal values ​​of the same period, and the actual operating values ​​of the current period. This includes: calculating the operational difficulty coefficient of the wind power equipment to be analyzed based on the standard values ​​of the operational indicators of the wind power equipment to be analyzed and the theoretical optimal values ​​of the same period (step S301); calculating the operational difficulty coefficient of the wind farm to be analyzed based on the standard values ​​of the operational indicators of the wind farm to be analyzed and the theoretical optimal values ​​of the same period (step S302); calculating the operational level coefficient of the wind power equipment to be analyzed based on the standard values ​​of the operational indicators of the wind power equipment to be analyzed in the current period, the theoretical optimal values ​​of the current period, and the actual operating values ​​of the current period (step S303); and calculating the operational level coefficient of the wind farm to be analyzed based on the standard values ​​of the operational indicators of the wind farm to be analyzed in the current period, the theoretical optimal values ​​of the current period, and the actual operating values ​​of the current period (step S304).

[0067] Specifically, in step S301, the standard values ​​of the operating indicators of each wind turbine based on historical data are compared with the theoretical optimal values ​​for the same period to obtain the results of the turbine operation difficulty analysis (i.e., the operation difficulty coefficient). The operation difficulty coefficient includes the power generation difficulty coefficient, the difficulty coefficient for controlling the power loss due to fault outage, the difficulty coefficient for controlling the power loss due to self-degradation, the difficulty coefficient for controlling the power loss due to performance loss, the difficulty coefficient for controlling the power loss due to planned outage, the difficulty coefficient for controlling other power losses, the difficulty coefficient for time availability, and the difficulty coefficient for energy utilization.

[0068] Taking the power generation difficulty coefficient of the i-th wind turbine in any phase of the power plant j as an example, the power generation difficulty coefficient of the i-th wind turbine in the power plant j satisfies:

[0069]

[0070] In the formula, D 0，j，i This represents the power generation difficulty coefficient of the i-th wind turbine in station j, where the subscript 0 corresponds to the power generation index; This represents the standard value of power generation obtained by the i-th wind turbine in station j using a trained standard value calculation model. This represents the theoretically optimal value of the concurrent power generation of the i-th wind turbine in wind farm j. When D 0，j，iA value >1 indicates that the difficulty of operating this wind turbine simultaneously is greater than the overall level of the training sample, and vice versa. It should be noted that the time availability difficulty coefficient and energy utilization difficulty coefficient can be obtained by referring to the power generation difficulty coefficient.

[0071] For any given period, the control difficulty coefficients corresponding to the following power loss indicators, with serial numbers m ranging from 1 to 5: power loss due to fault outage, power loss due to self-derating, power loss due to performance issues, power loss due to planned outage, and other power loss indicators, can be determined in the following ways:

[0072]

[0073] Among them, D m，j，i Let m be the control difficulty coefficient for the following power loss indicators in the i-th wind turbine in the station j, including power loss due to fault shutdown, power loss due to self-derating, power loss due to performance issues, power loss due to planned shutdown, and other power loss indicators, when m is taken from 1 to 5. This represents the standard values ​​of power loss due to fault shutdown, power loss due to self-derating, power loss due to performance issues, power loss due to planned shutdown, and other power losses obtained by the i-th wind turbine in station j using a trained standard value calculation model. Let D represent the theoretical optimal values ​​for the synchronous fault outage power loss, synchronous self-derating power loss, synchronous performance loss power loss, synchronous planned outage power loss, and synchronous other power loss power for the i-th wind turbine in station j. m，j，i A value >1 indicates that the difficulty of controlling this loss of the wind turbine is greater than the overall level of the training samples, and vice versa.

[0074] In step S301, the operational difficulty coefficients corresponding to the above-mentioned operational indicator data can be compared and ranked horizontally among multiple units within the same time period, or the changes in the operational difficulty coefficients of the same unit in different time periods can be compared.

[0075] In step S302, the operational difficulty coefficient of the analyzed wind turbine 0 is calculated based on the standard values ​​of its operational indicators and the theoretical optimal values ​​for the same period. Specifically, according to the analysis indicator calculation logic and the affiliation between the turbine and the wind turbine, the standard values ​​and theoretical optimal values ​​of the wind turbine's indicators are calculated by accumulating the theoretical optimal values ​​and standard values ​​of the wind turbine. By comparing the standard values ​​of the operational indicators of each wind turbine based on historical data with the theoretical optimal values ​​for the same period, the operational difficulty analysis results of the wind turbines are obtained.

[0076] Taking the control difficulty coefficients corresponding to the unit's operational difficulty coefficients, including those for power loss due to fault outage, power loss due to self-degradation, power loss due to performance issues, power loss due to planned outage, and other power loss indicators, as an example, the standard and theoretical optimal values ​​of the operational indicators for station j can be obtained by aggregating the corresponding indicator results of all units within the station, based on the calculation of the indicators themselves. For any given period, the following method is used in this embodiment:

[0077]

[0078] in, Let m be the theoretical optimal and standard values ​​of the following power loss indicators for station j when m ranges from 1 to 5: power loss due to fault outage, power loss due to self-degradation, power loss due to performance issues, power loss due to planned outage, and other power losses. Let I be the total number of wind turbines included in station j. Then, referring to the above formula for calculating the control difficulty coefficient, the control difficulty coefficients corresponding to the power loss indicators for fault outage, power loss due to self-degradation, power loss due to performance issues, power loss due to planned outage, and other power losses for station j are obtained. The power generation difficulty coefficient of station j can be calculated by summing the theoretical optimal and standard values ​​of power generation and then classifying the power generation difficulty coefficients of the aforementioned wind turbines.

[0079] In step S303, the current standard values ​​and actual values ​​of the operating indicators of each piece of equipment are compared to obtain the standardized analysis results of the equipment operating level (i.e., the operating level coefficient). The operating level coefficient includes the power generation level coefficient, the control level coefficient for power loss due to fault outage, the control level coefficient for power loss due to self-degradation, the control level coefficient for power loss due to performance loss, the control level coefficient for power loss due to planned outage, the control level coefficient for other power loss, the time availability level coefficient, and the energy utilization level coefficient.

[0080] Taking the power generation level coefficient of the i-th wind turbine in any phase of the power plant j as an example, the power generation level coefficient of the i-th wind turbine in the power plant j satisfies:

[0081]

[0082] In the formula, S 0，j，i This represents the power generation level coefficient of the i-th wind turbine in station j, where the subscript 0 corresponds to the power generation index; A represents the standard value of power generation obtained by the i-th wind turbine in station j using a trained standard value calculation model; 0，j，i This represents the actual power generation of the i-th wind turbine in wind farm j during the same period. When S 0，j，iA value >1 indicates that the wind turbine's standardized power generation level is higher than the overall level of the training sample, and vice versa. It should be noted that the time availability level coefficient and energy utilization level coefficient can be obtained by referring to the power generation level coefficient. Referring to this formula, the required current power generation level coefficient for the i-th wind turbine in station j can be obtained by using the current standard value and the current actual value of the operating indicators.

[0083] For any given period, the control level coefficients corresponding to the following power loss indicators, with serial numbers m ranging from 1 to 5: power loss due to fault outage, power loss due to self-derating, power loss due to performance issues, power loss due to planned outage, and other power loss indicators, can be determined respectively using the following methods:

[0084]

[0085] Among them, S m，j，i Let m be the control level coefficient for the power loss due to fault shutdown, power loss due to self-derating, power loss due to performance, power loss due to planned shutdown, and other power loss indicators corresponding to the i-th wind turbine in station j when m is taken from 1 to 5. This represents the standard values ​​of power loss due to fault shutdown, power loss due to self-derating, power loss due to performance issues, power loss due to planned shutdown, and other power losses obtained by the i-th wind turbine in station j using a trained standard value calculation model. Let A represent the theoretical optimal values ​​for the synchronous fault outage power loss, synchronous self-derating power loss, synchronous performance loss power loss, synchronous planned outage power loss, and synchronous other power loss power for the i-th wind turbine in station j. m，j，i This represents the actual values ​​of the concurrent fault outage power loss, concurrent self-derating power loss, concurrent performance loss, concurrent planned outage power loss, and concurrent other power loss for the i-th wind turbine in station j. When S m，j，i A value >1 indicates that the wind turbine's loss control level is higher than the overall level of the training sample, and vice versa. Referring to this formula, the operating level coefficients for the current period's various power losses of the i-th wind turbine in station j can be obtained using the current period's standard value, current period's theoretical optimal value, and current period's actual value of the operating indicators.

[0086] In step S303, the operating level coefficients corresponding to the above-mentioned operating indicator data can be compared and ranked horizontally among multiple units within the same time period, or the changes in the operating level coefficients of the same unit in different time periods can be compared.

[0087] In step S304, the operating level coefficient of the analyzed wind turbine is calculated based on the current standard value, current theoretical optimal value, and current actual operating value of the operating indicators. Specifically, according to the calculation logic of the analysis indicators and the affiliation between the turbine and the wind turbine, the standard value, theoretical optimal value, and actual value of the same indicators for the wind turbine are calculated based on the standard value, theoretical optimal value, and actual value of the wind turbine. By comparing the current standard value, theoretical optimal value, and actual value of the operating indicators for each wind turbine, the standardized analysis results of the operating level of the wind turbine are obtained.

[0088] Taking the unit's operating level coefficient, which includes the control level coefficient corresponding to indicators such as power loss due to fault outage, power loss due to self-degradation, power loss due to performance issues, power loss due to planned outage, and other power loss indicators, as an example, the theoretical optimal value and actual value of station j can be obtained by aggregating the corresponding indicator results of all units within the station, based on the calculation of the indicators themselves. For any given period, the following method is adopted in this embodiment:

[0089]

[0090] in, A m，j Let m be the theoretical optimal and actual values ​​of the following power loss indicators for station j when m ranges from 1 to 5: power loss due to fault outage, power loss due to self-degradation, power loss due to performance issues, power loss due to planned outage, and other power losses. Let I be the total number of wind turbines included in station j. Then, referring to the above formula for calculating the control level coefficient, the control level coefficients corresponding to the power loss indicators for fault outage, power loss due to self-degradation, power loss due to performance issues, power loss due to planned outage, and other power losses for station j are obtained. The power generation level coefficient of station j can be calculated by summing the theoretical optimal and standard values ​​of power generation and then classifying the power generation level coefficients of the wind turbines mentioned above. Therefore, by using the current standard value, current theoretical optimal value, and current actual value of the operating indicators, the required current power generation level coefficient and the current operating level coefficients for various power losses of station j can be obtained.

[0091] To achieve the above embodiments, the present invention also proposes a standardized analysis device for the operational difficulty and operational level of wind turbine generators and wind farms. This device can be implemented through software, hardware, or a combination of both.

[0092] Figure 4 This is a block diagram of a standardized analysis device for the operation difficulty and operation level of wind turbine generators and wind farms, provided in an embodiment of the present invention.

[0093] like Figure 4 As shown, the standardized analysis device for the operational difficulty and operational level of wind turbine units and power stations includes an acquisition module 10, a modeling module 20, a calculation module 30, and an analysis module 40, wherein:

[0094] The acquisition module 10 is used to acquire basic attribute data, operating condition data, and operating indicator data of multiple wind power equipment and their respective wind farms in order to construct a training dataset.

[0095] Modeling module 20 is used to build a standard value calculation model based on machine learning. The input of the standard value calculation model is basic attribute data and running condition data, and the output is running indicator data. The standard value calculation model is trained using the training dataset to obtain the trained standard value calculation model.

[0096] The calculation module 30 is used to input the basic attribute data and operating condition data of the wind power equipment to be analyzed from multiple periods into the trained standard value calculation model to obtain the corresponding period's standard values ​​of the operating indicators of the wind power equipment to be analyzed.

[0097] Analysis module 40 is used to calculate the operation difficulty coefficient and operation level coefficient of the wind power equipment and the site to be analyzed based on the standard values ​​of the operation indicators of the wind power equipment to be analyzed in each period, the theoretical optimal values ​​of the same period, and the actual operation values ​​of the current period.

[0098] Furthermore, in one possible implementation of this invention, the basic attribute data in the acquisition module 10 includes various predetermined attribute parameters such as province / city, topography, equipment model, station type, years of operation, and spatial distribution of equipment; the operating conditions include objective uncontrollable condition parameters and controllable condition parameters. The objective uncontrollable condition parameters include various parameters such as wind speed, temperature, weather, and power grid obstruction, while the controllable condition parameters include various parameters such as the number of maintenance personnel, cost input, and planned downtime.

[0099] Furthermore, in one possible implementation of this invention, the operational indicator data in the acquisition module 10 includes at least one key indicator parameter among power generation, multiple power loss, time availability, and energy utilization.

[0100] Furthermore, in one possible implementation of this invention, the modeling module 20 employs k-fold training and cross-validation when training the standard value calculation model using the training dataset.

[0101] Furthermore, in one possible implementation of this invention, the basic attribute data and operating condition data in the modeling module 20 each include multiple parameters. The parameters included in the basic attribute data and operating condition data are filtered and dimensionality reduced. The filtered and dimensionality-reduced basic attribute data and operating condition data are used as inputs to the standard value calculation model to train the standard value calculation model.

[0102] Furthermore, in one possible implementation of this invention, the modeling module 20 employs correlation analysis and association rule discovery methods to filter and reduce the dimensions of the parameters included in the basic attribute data and operating condition data.

[0103] Furthermore, in one possible implementation of this invention, the analysis module 40 calculates the operation difficulty coefficient and operation level coefficient corresponding to the wind power equipment and the wind farm to be analyzed based on the standard values ​​of the operation indicators of the wind power equipment to be analyzed in each period, the theoretical optimal value of the same period, and the actual operation value of the current period. This includes: calculating the operation difficulty coefficient of the wind power equipment to be analyzed based on the standard values ​​of the operation indicators of the wind power equipment to be analyzed and the theoretical optimal value of the same period; calculating the operation difficulty coefficient of the wind farm to be analyzed based on the standard values ​​of the operation indicators of the wind farm to be analyzed and the theoretical optimal value of the same period; calculating the operation level coefficient of the wind power equipment to be analyzed based on the standard values ​​of the operation indicators of the wind power equipment to be analyzed in the current period, the theoretical optimal value of the current period, and the actual operation value of the current period; and calculating the operation level coefficient of the wind farm to be analyzed based on the standard values ​​of the operation indicators of the wind farm to be analyzed in the current period, the theoretical optimal value of the current period, and the actual operation value of the current period.

[0104] It should be noted that the explanation of the aforementioned embodiment of the standardized analysis method for the operational difficulty and operational level of wind turbines and wind farms also applies to the standardized analysis device for the operational difficulty and operational level of wind turbines and wind farms in this embodiment, and will not be repeated here. It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on a standardized analysis device for wind turbines and wind farms. In other embodiments of the present invention, a standardized analysis device for wind turbines and wind farms may include more or fewer components than illustrated, or combine certain components, or split certain components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0105] In this embodiment of the invention, a training dataset is constructed by acquiring basic attribute data, operating condition data, and operating index data of multiple wind turbines and their associated wind farms. A standard value calculation model is built based on machine learning. The input of the standard value calculation model is the basic attribute data and operating condition data, and the output is the operating index data. The standard value calculation model is trained using the training dataset to obtain a trained standard value calculation model. The basic attribute data and operating condition data of the wind turbines to be analyzed from multiple periods are input into the trained standard value calculation model to obtain the corresponding period's standard values ​​of the operating indexes for the wind turbines to be analyzed. Based on the standard values ​​of the operating indexes for each period of the wind turbines to be analyzed, the theoretical optimal values ​​for the same period, and the actual operating values ​​for the current period, the operating difficulty coefficient and operating level coefficient corresponding to the wind turbines and wind farms to be analyzed are calculated. In this case, a standard value calculation model is constructed by comprehensively considering various data such as basic attribute data, operating condition data, and operating index data of the wind turbines and their associated wind farms. The trained standard value calculation model is used to obtain the corresponding period's standard values ​​of the operating indexes for the wind turbines to be analyzed. The analysis of the operating difficulty and operating level corresponding to the wind turbines and wind farms to be analyzed is obtained by combining the theoretical optimal values ​​for the same period and the actual operating values ​​for the current period. This enabled a standardized analysis of the operational difficulty and operational level of wind turbine units and power stations.

[0106] The method and apparatus of this invention are applied to the standardized analysis of the operational difficulty and level of wind turbines and wind farms in the field of wind power generation. The aim is to achieve standardized analysis of the operational difficulty and level of wind turbines and wind farms under the combined influence of complex subjective and objective factors, so as to support industry operation management decisions. It improves the objectivity and guidance of the analysis results and provides decision support for the optimized operation of wind power systems. The main technical problems to be solved are the feature extraction of multiple influencing factors, the establishment of a standardized model (i.e., a standard value calculation model), and the analysis method of operational difficulty and level.

[0107] The beneficial effects of the method and apparatus of the present invention are as follows:

[0108] 1) This invention integrates the basic attributes and operating conditions of wind power equipment and wind farms, considering both subjective and objective factors, to calculate the difficulty coefficient of key operating indicators under given conditions. It can analyze the inherent conditions for early-stage decision-making at wind farms and simulate key indicators for new projects, supporting project decision-making. 2) Based on realistic conditions and management needs, this invention can eliminate objectively uncontrollable or unconsidered constraints and calculate standard values ​​for key indicators. By comparing actual values ​​with expected values, it achieves standardized analysis of subjective operating capabilities under given conditions. 3) This invention uses correlation analysis and association rule analysis to filter and reduce the dimensionality of training data, improving training efficiency and accuracy. 4) The wind farm-level operating indicators analyzed by this invention can be drilled down at the equipment level to identify problematic equipment and their causes. It also allows for horizontal and vertical comparisons, as well as comparisons with historical cycles, facilitating understanding development trends, identifying problems, and assigning responsibility. 5) This invention is highly scalable and customizable, applicable to the standardized analysis of various common operating indicators. It can achieve targeted analysis for different management needs within the framework of this solution and can be implemented at low cost.

[0109] To implement the above embodiments, the present invention also proposes an electronic device, including: a processor and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided in the foregoing embodiments.

[0110] To implement the above embodiments, the present invention also proposes a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the methods provided in the foregoing embodiments.

[0111] To implement the above embodiments, the present invention also proposes a computer program product, including a computer program that, when executed by a processor, implements the methods provided in the foregoing embodiments.

[0112] In the foregoing descriptions of the embodiments, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0113] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0114] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of the invention pertain.

[0115] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0116] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any of the following techniques known in the art, or a combination thereof: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0117] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.

[0118] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0119] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A standardized analysis method for the operational difficulty and operational level of wind turbine generators and wind farms, characterized in that, include: Acquire basic attribute data, operating condition data, and operating indicator data of multiple wind power devices and their respective wind farms to construct a training dataset; A standard value calculation model is constructed based on machine learning. The input of the standard value calculation model is basic attribute data and operating condition data, and the output is operating indicator data. The standard value calculation model is trained using the training dataset to obtain a trained standard value calculation model. The basic attribute data and operating condition data of the wind power equipment to be analyzed in multiple periods are input into the trained standard value calculation model to obtain the corresponding period's standard values ​​of the operating indicators of the wind power equipment to be analyzed. Based on the standard values ​​of the operating indicators of the wind power equipment to be analyzed in each period, the theoretical optimal values ​​of the same period, and the actual operating values ​​of the current period, the operating difficulty coefficient and operating level coefficient of the wind power equipment and the wind farm to be analyzed are calculated.

2. The standardized analysis method for the operational difficulty and operational level of wind turbine generators and power stations according to claim 1, characterized in that, Based on the standard values ​​of operating indicators for each period of the wind power equipment to be analyzed, the theoretical optimal values ​​for the same period, and the actual operating values ​​for the current period, the operating difficulty coefficient and operating level coefficient corresponding to the wind power equipment and the wind farm to be analyzed are calculated, including: The operational difficulty coefficient of the wind power equipment to be analyzed is calculated based on the standard values ​​of the operating indicators of the wind power equipment to be analyzed and the theoretical optimal values ​​of the same period. The operational difficulty coefficient of the site under analysis is calculated based on the standard values ​​of the operational indicators and the theoretical optimal values ​​for the same period. The operating level coefficient of the wind power equipment to be analyzed is calculated based on the current standard value, current theoretical optimal value and current actual operating value of the operating indicators of the wind power equipment to be analyzed. The operational level coefficient of the site under analysis is calculated based on the current standard value, current theoretical optimal value, and current actual value of the operating indicators.

3. The standardized analysis method for the operational difficulty and operational level of wind turbine units and power stations according to claim 1, characterized in that, The standard value calculation model is trained using the training dataset with k-fold training and cross-validation.

4. The standardized analysis method for the operational difficulty and operational level of wind turbine generators and power stations according to claim 1, characterized in that, The basic attribute data and the operating condition data each include multiple parameters. The parameters included in the basic attribute data and the operating condition data are filtered and dimensionality reduced. The filtered and dimensionality-reduced basic attribute data and operating condition data are used as inputs to the standard value calculation model to train the standard value calculation model.

5. The standardized analysis method for the operational difficulty and operational level of wind turbine generators and wind farms according to claim 4, characterized in that, The parameters included in the basic attribute data and the operating condition data are filtered and reduced in dimensionality using correlation analysis and association rule discovery methods.

6. The standardized analysis method for the operational difficulty and operational level of wind turbine generators and power stations according to claim 1, characterized in that, Basic attribute data includes various predetermined attribute parameters such as province and city, topography, equipment model, station type, years of operation, and spatial distribution of equipment; operating conditions include objective uncontrollable conditions and controllable conditions. Objective uncontrollable conditions include various parameters such as wind speed, temperature, weather, and power grid obstruction, while controllable conditions include various parameters such as the number of maintenance personnel, cost input, and planned downtime.

7. The standardized analysis method for the operational difficulty and operational level of wind turbine units and power stations according to claim 1, characterized in that, Operational performance data includes at least one key indicator parameter among power generation, multiple power loss parameters, time availability, and energy utilization.

8. A standardized analysis device for the operational difficulty and operational level of wind turbine generators and power stations, characterized in that, include: The acquisition module is used to acquire basic attribute data, operating condition data, and operating indicator data of multiple wind power devices and their respective wind farms in order to construct a training dataset. The modeling module is used to build a standard value calculation model based on machine learning. The input of the standard value calculation model is basic attribute data and running condition data, and the output is running indicator data. The standard value calculation model is trained using the training dataset to obtain a trained standard value calculation model. The calculation module is used to input the basic attribute data and operating condition data of the wind power equipment to be analyzed from multiple periods into the trained standard value calculation model to obtain the corresponding period's standard values ​​of the operating indicators of the wind power equipment to be analyzed. The analysis module is used to calculate the operation difficulty coefficient and operation level coefficient of the wind power equipment and the site under analysis based on the standard values ​​of the operation indicators of the wind power equipment under analysis in each period, the theoretical optimal values ​​of the same period, and the actual operation values ​​of the current period.

9. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

Images