Fan operation and maintenance evaluation method and device

Abstract

The invention discloses a fan operation and maintenance evaluation method and device, and relates to the technical field of fans, and the method comprises the steps: obtaining the operation data of each fan in a target unit; identifying the fan state of each fan at each data point according to the operation data of each fan, and determining the power band of each fan according to the fan state of each data point; determining the dispersion of each fan according to the power band of each fan; determining the standard degree of each fan according to the power band of each fan; for each fan model, taking the power curve of the fan with the highest standard degree as a theoretical power curve of the corresponding fan model; calculating each electric quantity index of each fan according to the theoretical power curve of each fan model; and screening out the to-be-operated and to-be-maintained fan according to the dispersion degree, the standard reaching degree and each electric quantity index of each fan. The effectiveness of fan operation and maintenance evaluation is improved.

Description

Technical Field

[0001] The present application relates to the field of wind turbine technology, and in particular to a method and device for evaluating wind turbine operation and maintenance. Background Art

[0002] With the continuous growth of global energy demand and increasing awareness of environmental protection, improving energy efficiency has become a key strategy for alleviating resource constraints, reducing environmental pollution, and promoting sustainable development. As the core link in modern energy transmission and conversion, the energy efficiency of electric fields directly affects the economic and environmental friendliness of the entire energy system.

[0003] Related wind turbine evaluation methods primarily assess wind turbine power generation based on the manufacturer's theoretical power curve and the turbine system's built-in power calculation method. However, this theoretical power curve is measured in a standard test wind farm under relatively harsh testing conditions. In actual operation, the unit's power output is affected by a variety of factors, including the unit's operating status and environment. These factors can cause the wind turbine's actual power curve to deviate from the theoretical power curve, resulting in a significant discrepancy between actual power generation and expected values. Therefore, effective wind turbine intelligent operation and maintenance evaluation technology is needed. Summary of the Invention

[0004] The purpose of this application is to provide a wind turbine operation and maintenance evaluation method and device, which can improve the effectiveness of wind turbine operation and maintenance evaluation.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a wind turbine operation and maintenance evaluation method, the wind turbine operation and maintenance evaluation method comprising:

[0007] Obtain the operating data of each fan in the target unit;

[0008] Identifying the fan status of each fan at each data point based on the operating data of each fan, and determining the power band of each fan based on the fan status at each data point;

[0009] Determine the discreteness of each fan according to its power band;

[0010] Determine the compliance of each fan according to its power band;

[0011] For each fan model, the power curve of the fan with the highest degree of compliance is used as the theoretical power curve of the corresponding fan model;

[0012] Calculate the power indicators of each fan according to the theoretical power curve of each fan model;

[0013] The wind turbines to be operated and maintained are selected based on their discreteness, compliance and power indicators.

[0014] Optionally, after obtaining the operating data of each wind turbine in the target unit, the wind turbine operation and maintenance evaluation method further includes:

[0015] The operating data of each wind turbine is stored in a time series database; the operating data includes wind speed, rotation speed and power; the power is the generated power of the wind turbine.

[0016] Optionally, identifying a fan state of each fan at each data point based on the operating data of each fan, and determining a power band of each fan based on the fan state at each data point, specifically includes:

[0017] Identify the operating data of each fan into normal state, abnormal state and shutdown state according to the set rules;

[0018] For the operating data of any wind turbine in normal state, a data scatter plot is obtained in the Cartesian coordinate system with wind speed as the horizontal axis and power as the vertical axis;

[0019] Dividing the data scatter plot into consecutive equally spaced intervals according to the vertical axis to obtain a plurality of power intervals;

[0020] Clustering the wind speeds within each power interval, and classifying the wind speeds according to the clustering results; the classification includes a first abnormal area, a normal area, and a second abnormal area distributed in sequence from the origin to the direction of increasing wind speed on the horizontal axis;

[0021] Identifying a power-limited state on the operating data of the second abnormal area to obtain a corrected normal state;

[0022] The power band of each fan is determined according to the corrected normal state.

[0023] Optionally, identifying the power-limiting state of the operating data of the second abnormal area to obtain a corrected normal state specifically includes:

[0024] Calculate the range value of each power interval;

[0025] Calculate the normalized weight of each power interval;

[0026] Determine the power-limiting state identification tag value for each power interval based on the extreme value and normalized weight of each power interval;

[0027] The power-limiting state identification tag value obtained for each power interval is processed with the 3σ maximum value abnormal value. If the power-limiting state identification tag value of the i-th power interval is greater than the 3σ maximum value, the normal state of the data in the second abnormal area of ​​the i-th power interval is corrected to the power-limiting state. Otherwise, no correction is performed.

[0028] Optionally, the discreteness of each wind turbine is determined according to the power band of each wind turbine, specifically including:

[0029] For any wind turbine power band, within the preset effective wind speed range, the wind speed on the horizontal axis is divided into equal intervals to obtain multiple wind speed intervals;

[0030] Calculate the coefficient of variation of power in each wind speed interval;

[0031] The average value of the coefficient of variation of power in each wind speed range is taken as the dispersion of the wind turbine.

[0032] Optionally, determining the compliance of each wind turbine according to the power band of each wind turbine includes:

[0033] For any wind turbine power band, within the preset effective wind speed range, the wind speed on the horizontal axis is divided into equal intervals to obtain multiple wind speed intervals;

[0034] Calculate the degree of compliance for each wind speed range.

[0035] Optionally, the power indicators of each wind turbine are calculated based on the theoretical power curve of each wind turbine model, specifically including:

[0036] Calculate the theoretical power generation of each fan within a set time period based on the theoretical power curve of each fan model;

[0037] Calculate the actual power generation of any wind turbine according to the operating data of the wind turbine within a set time period;

[0038] The power generation loss due to shutdown is obtained according to the difference between the theoretical power generation and the actual power generation of the wind turbine in the shutdown state within the set time period;

[0039] Obtaining power loss due to power restriction based on the difference between the theoretical power generation and the actual power generation of the wind turbine in the power restriction state within the set time period;

[0040] Obtaining abnormal power loss according to the difference between the theoretical power generation and the actual power generation of the wind turbine in the abnormal state within the set time period;

[0041] The actual power generation of the wind turbine is obtained according to the actual power generation of the wind turbine, the power generation loss due to shutdown, the power generation loss due to power restriction and the power generation loss due to abnormality within the set time period;

[0042] Calculate the performance loss of the wind turbine based on the theoretical power generation and the actual power generation of the wind turbine within the set time period;

[0043] The energy utilization rate of the wind turbine is calculated based on the actual power generation and the actual power generation expected during the set time period of the wind turbine.

[0044] Optionally, after selecting wind turbines to be operated and maintained based on the dispersion, compliance and power indicators of each wind turbine, the wind turbine operation and maintenance evaluation method further includes:

[0045] Generate a wind turbine operation and maintenance report, which includes a status identification scatter plot of each wind turbine, a wind turbine dispersion ranking table, a wind turbine compliance ranking table, a wind turbine power index ranking table, a wind turbine power index bar chart, and the number and model of the wind turbine to be operated and maintained.

[0046] Optionally, the wind turbine operation and maintenance report is displayed through a visual interface, and the visual interface is also used to delete, query, preview and download the wind turbine operation and maintenance report.

[0047] In a second aspect, the present application provides a wind turbine operation and maintenance evaluation device, which applies the wind turbine operation and maintenance evaluation method, and includes:

[0048] An operating data acquisition module is used to obtain the operating data of each fan in the target unit;

[0049] A power band determination module, configured to identify the fan status of each fan at each data point based on the operating data of each fan, and determine the power band of each fan based on the fan status at each data point;

[0050] A dispersion determination module, configured to determine the dispersion of each fan according to the power band of each fan;

[0051] A compliance determination module is used to determine the compliance of each wind turbine according to the power band of each wind turbine;

[0052] A theoretical power curve determination module is used to determine, for each wind turbine model, the power curve of the wind turbine with the highest degree of compliance as the theoretical power curve of the corresponding wind turbine model;

[0053] The power index determination module is used to calculate the power index of each fan according to the theoretical power curve of each fan model;

[0054] The wind turbine screening module for operation and maintenance is used to screen out wind turbines for operation and maintenance based on the discreteness, compliance and power indicators of each wind turbine.

[0055] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0056] The present application provides a method and device for evaluating wind turbine operation and maintenance, which identifies the fan status of each fan at each data point based on the operating data of each fan, determines the power band of each fan based on the fan status of each data point, and for each fan model, uses the power curve of the fan with the highest degree of compliance as the theoretical power curve of the corresponding fan model, that is, determines the theoretical power curve of each fan based on the actual operating data, thereby improving the effectiveness of the theoretical power curve of each fan, and thus improving the effectiveness of the fan operation and maintenance evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 A flow chart of a wind turbine operation and maintenance evaluation method provided in one embodiment of the present application;

[0059] Figure 2 A complete flow chart of a wind turbine operation and maintenance evaluation method provided in one embodiment of the present application;

[0060] Figure 3 A schematic diagram of a "factory"-shaped power band before status recognition provided by an embodiment of the present application;

[0061] Figure 4 This is a schematic diagram of the "factory"-shaped power band after status recognition provided by an embodiment of the present application. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0063] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0064] This application provides a wind turbine operation and maintenance evaluation method, such as Figure 1 As shown, the wind turbine operation and maintenance evaluation method includes:

[0065] Step 101: Obtain operating data of each wind turbine in the target unit.

[0066] Step 102: Identify the fan status of each fan at each data point based on the operating data of each fan, and determine the power band of each fan based on the fan status of each data point.

[0067] Step 103: Determine the dispersion of each wind turbine according to the power band of each wind turbine.

[0068] Step 104: Determine the compliance level of each wind turbine according to the power band of each wind turbine.

[0069] Step 105: For each wind turbine model, the power curve of the wind turbine with the highest degree of compliance is used as the theoretical power curve of the corresponding wind turbine model.

[0070] Step 106: Calculate the power indicators of each wind turbine based on the theoretical power curve of each wind turbine model.

[0071] Step 107: Filter out wind turbines to be operated and maintained based on the discreteness, compliance degree, and power indicators of each wind turbine.

[0072] like Figure 2 As shown, the wind turbine operation and maintenance evaluation method includes data modeling, data acquisition, data storage, data analysis, automatic report generation, visual display and system services.

[0073] In an exemplary embodiment, after step 101, the wind turbine operation and maintenance evaluation method further includes: storing the operating data of each wind turbine in a time series database; the operating data includes wind speed, rotation speed and power; and the power is the generated power of the wind turbine.

[0074] Based on business characteristics, the operation data of the wind turbine is obtained, where the operation data mainly includes the real-time speed of the wind turbine, the real-time power of the wind turbine, the real-time wind speed corresponding to the real-time speed of the wind turbine and the power generation, and the real-time collection time.

[0075] The wind turbine speed refers to the number of times the wind turbine blades rotate per minute, typically expressed in revolutions per minute (RPM). The wind turbine's real-time power refers to the actual electrical power generated by the wind turbine, typically measured in kilowatts (kW). Wind speed refers to the speed of air relative to the ground, typically measured in meters per second (m / s). The fields required for data modeling are as follows: wind turbine model, wind turbine number, real-time data collection time, wind turbine's real-time speed, wind turbine's real-time power, and real-time wind speed.

[0076] Based on the communication protocol of the wind farm operation and maintenance Supervisory Control and Data Acquisition System (SCADA), through the relevant protocol class library of related technologies, data is collected from the wind farm operation and maintenance SCADA according to the fields required for data modeling to obtain real-time collected operation data.

[0077] The time series database provides a data foundation for adding, deleting, modifying and querying data.

[0078] In one exemplary embodiment, operating data is used as input to identify the status of a single wind turbine based on wind turbine dimensions (wind speed, rotational speed, and power). Specifically, the three-dimensional data of wind speed, rotational speed, and power are classified to determine the status of the wind turbine. The status of the wind turbine can be classified into normal, abnormal, shutdown, and power-limited states.

[0079] Wherein, step 102 specifically includes:

[0080] Step 1021: Identify the operating data of each wind turbine as a normal state, an abnormal state, and a shutdown state according to a set rule.

[0081] The setting rules include shutdown state identification rules and abnormal state identification rules.

[0082] Based on the distribution of fan operation data, determine whether the fan is in a shutdown state. During the operation of the fan, its wind speed, rotation speed, and power are all greater than 0. The shutdown state identification rule is that if the wind speed, rotation speed, and power are all less than 0 or equal to 0, the corresponding fan is marked as a stopped state.

[0083] The values ​​less than 0 indicate abnormal values ​​due to abnormalities in the sensor or acquisition system. This part of the data is also marked as downtime data.

[0084] Based on the wind turbine operating parameters provided by the manufacturer, the abnormal state of the wind turbine is determined. Specifically, the wind turbine designed by the manufacturer can operate normally to generate electricity when the wind speed is between the cut-in wind speed and the cut-out wind speed. In other words, when the wind speed is not between the cut-in wind speed and the cut-out wind speed, the wind turbine cannot operate normally to generate electricity. Its real-time power should be 0. In other words, if the wind speed is not between the cut-in wind speed and the cut-out wind speed, and the real-time power of the wind turbine is greater than 0, the wind turbine will be determined to be in an abnormal state. In other words, the abnormal state identification rule is that if the wind speed is not between the cut-in wind speed and the cut-out wind speed, and the power is greater than 0, the wind turbine will be determined to be in an abnormal state.

[0085] The fan in the operating data other than the abnormal state and the shutdown state is determined to be in a normal state.

[0086] Step 1022: For the normal state of any wind turbine operating data, use wind speed as the horizontal axis and power as the vertical axis to obtain a data scatter plot in the Cartesian coordinate system. At this time, the data points are distributed in a "factory"-shaped band, which is called the power band. The normal state of the wind turbine is as follows: Figure 3 As shown by the green dots in the figure, the abnormal state is as follows Figure 3 As shown in the middle blue point, the shutdown status is as follows Figure 3 As shown by the black dot in the middle, the power limit status is as follows Figure 3 As shown in red dots.

[0087] Step 1023: Divide the data scatter plot into consecutive equal intervals according to the vertical axis to obtain a plurality of power intervals. In each power interval, the power values ​​of the data scatter points satisfy a normal distribution within a certain interval.

[0088] Specifically, the power axis is divided into intervals of 20 kW to obtain multiple power intervals.

[0089] Step 1024: Cluster the wind speeds within each power interval and classify the wind speeds based on the clustering results; the classification includes a first abnormal region, a normal region, and a second abnormal region, distributed sequentially along the horizontal axis from the origin to increasing wind speed. This application uses clustering to classify scattered data points into normal values ​​and abnormal values, that is, to classify wind turbines into normal and abnormal states.

[0090] Step 1025: Identify the power-limited state of the operating data of the second abnormal area to obtain a corrected normal state.

[0091] Step 1026: Determine the power band of each wind turbine according to the corrected normal state.

[0092] The clustering of wind speeds in each power interval specifically includes: clustering the wind speeds in each power interval using K-Means++ (K-Means Clustering Algorithm, k-means clustering algorithm) to divide the data into normal values ​​and abnormal values.

[0093] The K-Means++ algorithm is an improved version of the K-Means algorithm, which aims to solve the sensitivity problem of the initial center point selection in the K-Means algorithm. In addition, regarding the hyperparameters set by the K-Means++ algorithm, the number of clusters used in the K-Means++ algorithm of this application is 3, and the maximum number of iterations is set to 1000 for the status identification of the fan. At this time, the data of each power interval will be divided into 3 categories. The fan scatter points on the left side of the "factory"-shaped power band are in an abnormal state (the first abnormal area), the fan scatter points on the right side of the "factory"-shaped power band are also in an abnormal state (the second abnormal area), and the remaining fan scatter points are in the middle of the "factory"-shaped position and are in a normal state (normal area).

[0094] The step of identifying the power-limited state of the operating data of the second abnormal area to obtain a corrected normal state specifically includes:

[0095] Calculate the extreme difference of wind speed in each power interval, that is, the maximum wind speed minus the minimum wind speed in each power interval. The specific calculation formula can be referred to as follows:

[0096]

[0097] Among them, r i Indicates the extreme difference in wind speed in the i-th power interval, v ik represents the wind speed corresponding to the kth data point in the i-th power interval, and m represents the number of data points in the i-th power interval.

[0098] Calculate the normalized weight of each power interval. The specific calculation method can be referred to the following formula:

[0099]

[0100] Among them, w i represents the normalized weight of the i-th power interval, len i represents the number of data points in the i-th power interval, and n represents the total number of power intervals involved in the power curtailment state identification step.

[0101] The power limit state identification tag value for each power interval is determined based on the range value and normalized weight of each power interval. The specific calculation method can be referred to the following formula:

[0102] l i =r i *w i

[0103] where l i Indicates the power limit state identification tag value of the i-th power interval.

[0104] For the power-limiting state identification tag values ​​obtained for each power interval, perform 3σ maximum value abnormal value processing. If the power-limiting state identification tag value of the i-th power interval is greater than the 3σ maximum value, the normal state of the data in the second abnormal area of ​​the i-th power interval is corrected to the power-limiting state. Otherwise, no correction is performed. The specific correction method can refer to the following conditions:

[0105]

[0106] Among them, l i_revise represents the corrected power-limiting state identification tag value of the i-th power interval, l i Indicates the power-limiting state identification tag value of the i-th power interval before correction. 1 indicates that the data point corresponding to the power interval needs to be corrected to the power-limiting state, and 0 indicates that no correction is required. mean Indicates the average value of the power limit state identification tag value of each power interval involved in this step, l std Indicates the standard deviation of the power curtailment identification tag values ​​for each power interval involved in this step.

[0107] According to the fan unit model dimension, traverse each fan of the model and identify the power limit status of each fan.

[0108] The "factory" shaped power band after the fan status is identified is as follows Figure 4 shown.

[0109] In an exemplary embodiment, step 103 specifically includes:

[0110] Step 1031: For any wind turbine power band, within a preset effective wind speed range, the wind speed on the horizontal axis is divided into equal intervals to obtain a plurality of wind speed intervals.

[0111] Step 1032: Calculate the coefficient of variation of the power in each wind speed interval. The calculation formula of the coefficient of variation is as follows:

[0112]

[0113] in, is the coefficient of variation of power in the i'th wind speed interval, P i'j Indicates the power of the jth data point in the i'th wind speed interval, N i' represents the number of data points in the i'th wind speed interval, represents the power standard deviation in the i'th wind speed interval, Represents the average power in the i'th wind speed interval.

[0114] The specific division interval of the wind speed interval can be determined according to the data distribution range and the actual recognition effect. After a large amount of data verification, this application uses the wind speed interval of 0.5m / s to divide the wind speed axis.

[0115] Step 1033: taking the average value of the coefficient of variation of the power in each wind speed interval as the dispersion of the wind turbine.

[0116] According to the fan unit model dimension, traverse each fan of the model and repeat the above steps to obtain the discreteness of each fan. According to the discreteness, rank the fans of the traversed model in order. The smaller the discreteness value, the better.

[0117] In an exemplary embodiment, step 104 specifically includes:

[0118] Step 1041: For any wind turbine power band, the wind speed on the horizontal axis is divided into equal intervals within a preset effective wind speed range to obtain multiple wind speed intervals. The preset effective wind speed range is the data range between the cut-in wind speed and the cut-out wind speed.

[0119] Step 1042: Calculate the degree of compliance for each wind speed interval.

[0120]

[0121] Among them, the power curve represents the maximum power of the wind turbine in each wind speed range, and the average power curve represents the average value of the power curves of each wind turbine of this model in the same wind speed range. i is the power corresponding to the i-th wind speed interval of the power curve, M i is the power corresponding to the i-th wind speed interval of the average power curve. The specific division interval of the wind speed interval is the same as the discreteness calculation.

[0122] The Weibull distribution of the wind speed over the entire field is calculated, and the calibration degree is corrected using the Weibull distribution as the weight coefficient to obtain the final calibration degree.

[0123] Based on the Weibull distribution, the Weibull weight coefficient of the power curve in each wind speed range is calculated. The specific formula is as follows:

[0124]

[0125] Among them, P i is the power corresponding to the i-th wind speed interval of the power curve, λ is the scale parameter, and k' is the shape parameter. The values ​​of λ and k' determine the shape and scale of the distribution.

[0126] The final degree of compliance for each wind speed interval is obtained by multiplying the degree of compliance for each wind speed interval by the Weibull weight coefficient for each wind speed interval.

[0127] The average value of the power compliance in each wind speed range is taken as the compliance of the wind turbine.

[0128] The Weibull distribution parameter settings can be determined based on the data distribution range and actual results. This application sets the scale parameter to 3 and the shape parameter to 4.

[0129] According to the fan unit model dimension, traverse each fan of the model and repeat the above steps to obtain the compliance value of each fan. Rank them in descending order according to the compliance value. The larger the compliance value, the better.

[0130] In an exemplary embodiment, step 106 specifically includes:

[0131] Step 1061: Calculate the theoretical power generation of each wind turbine within a set time period based on the theoretical power curve of each wind turbine model.

[0132] According to the wind turbine unit model dimension, traverse each wind turbine of the model and calculate the theoretical power generation. The theoretical power generation is the power generation generated by all the electrical energy that can be converted by the wind turbine, estimated by the theoretical power curve of the model and the wind turbine wind speed data. The specific calculation formula is as follows:

[0133]

[0134] Among them, E t is the calculated theoretical power generation, n is the number of data points within the cut-in wind speed and cut-out wind speed range, v k is the wind speed of the kth data point, P t (v k ) is the theoretical power curve at wind speed v k Theoretical power at T k is the time step corresponding to the data point.

[0135] Step 1062: Calculate the actual power generation of any wind turbine according to the operating data of the wind turbine within the set time period.

[0136] According to the wind turbine unit model dimension, traverse each wind turbine of the model and calculate the actual power generation. The actual power generation is the power generated during the actual operation of the wind turbine, which is calculated using the wind turbine power data. The specific calculation formula is as follows:

[0137]

[0138] Among them E r is the actual power generation obtained by calculation, n is the number of data points within the range of cut-in wind speed and cut-out wind speed, v k is the wind speed of the kth data point, P(v k) is the actual power curve at wind speed v k The actual power at T k is the time step corresponding to the data point.

[0139] Step 1063: Obtain the power generation loss due to shutdown according to the difference between the theoretical power generation and the actual power generation of the wind turbine in the shutdown state within the set time period.

[0140] According to the wind turbine unit model dimension, each wind turbine of the model is traversed to calculate the power generation loss due to shutdown. The power generation loss due to shutdown is the power generation loss caused by wind turbine shutdown. It is obtained by subtracting the power generation obtained by substituting the shutdown state point in the state identification into the wind turbine theoretical power curve and the actual power generation at the shutdown state point. The specific calculation formula is as follows:

[0141]

[0142] Among them, E p_loss is the calculated power generation loss due to shutdown, and n1 is the number of shutdown state data points.

[0143] According to the above method, the shutdown loss power generation data of each wind turbine of each model is obtained, and then the wind turbines of the same model are ranked in order according to the shutdown loss power generation, and the smaller the shutdown loss power generation value, the better.

[0144] Step 1064: Obtain power generation loss due to power restriction based on the difference between the theoretical power generation and the actual power generation of the wind turbine in the power restriction state within the set time period.

[0145] According to the wind turbine unit model dimension, each wind turbine of the model is traversed to calculate the power generation loss due to power curtailment. The power generation loss due to power curtailment is the power generation loss caused by wind turbine power curtailment. That is, the power generation loss obtained by substituting the power curtailment state point in the state identification into the wind turbine theoretical power curve and the actual power generation at the power curtailment state point is calculated. The specific calculation formula can be referred to as follows:

[0146]

[0147] Among them E l_loss is the calculated power generation loss due to power curtailment, and n2 is the number of power curtailment state data points.

[0148] Step 1065: Obtain abnormal power loss according to the difference between the theoretical power generation and the actual power generation of the wind turbine in the abnormal state within the set time period.

[0149] According to the wind turbine unit model dimension, each wind turbine of the model is traversed to calculate the abnormal power loss. The abnormal power loss is the power loss caused by the wind turbine abnormality. That is, the power loss obtained by substituting the abnormal state point in the state identification into the wind turbine theoretical power curve and the actual power generation at the abnormal state point is obtained by subtracting it. The specific calculation formula is as follows:

[0150]

[0151] Among them, E a_loss is the calculated abnormal power loss, and n3 is the number of abnormal state data points.

[0152] According to the above method, the abnormal power loss data of each wind turbine of each model is obtained, and then the wind turbines of the same model are ranked in order according to the abnormal power loss data. The smaller the abnormal power loss data, the better.

[0153] Step 1066: Obtain the actual power generation required of the wind turbine according to the actual power generation, power generation loss due to shutdown, power generation loss due to power restriction, and power generation loss due to abnormality within the set time period.

[0154] According to the wind turbine unit model dimension, each wind turbine of the model is traversed to calculate the actual power generation. The actual power generation is the power generation of the wind turbine under completely normal operation under the current wind speed conditions. It is the power generation that is summarized after taking into account the power loss due to shutdown state, power restriction state, and abnormal state based on the actual power generation. The specific calculation formula is as follows:

[0155] E r_should =E r +E p_loss +E l_loss +E a_loss

[0156] Among them, E r_should is the actual power generation amount calculated, E r is the actual power generation, E p_loss is the power generation loss due to shutdown, E l_loss is the power generation loss due to power restriction, E a_loss Abnormal power loss.

[0157] Step 1067: Calculate the performance loss of the wind turbine according to the theoretical power generation and the actual power generation of the wind turbine within the set time period.

[0158] Based on the wind turbine model dimension, traverse each wind turbine of the model and calculate the performance loss. The performance loss is the percentage of power generation loss caused by unit performance deviation. The specific calculation formula is as follows:

[0159]

[0160] Among them, loss is the calculated performance loss, E t is the theoretical power generation, E r_should is the actual power generation.

[0161] According to the above method, the performance data of each fan of each model is obtained, and then the fans of the same model are ranked in order of performance loss, and the smaller the performance loss data, the better.

[0162] Step 1068: Calculate the energy utilization rate (EBA) of the wind turbine according to the actual power generation and the actual power generation expected by the wind turbine within the set time period.

[0163] According to the wind turbine unit model dimension, traverse each wind turbine of the model and calculate EBA. EBA is the ratio of the wind turbine's actual power generation to its expected power generation. The specific calculation formula is as follows:

[0164]

[0165] Among them E r is the actual power generation, E r_should is the actual power generation.

[0166] According to the above method, the EBA data of each fan of each model is obtained, and then the fans of the same model are ranked in reverse order according to EBA, and the larger the EBA data, the better.

[0167] In an exemplary embodiment, step 107 specifically includes:

[0168] (1) Discreteness assessment.

[0169] Step 103 obtains the dispersion ranking of each wind turbine for each model. A higher wind turbine's dispersion ranking, i.e., a smaller dispersion value, indicates a higher sensitivity in the wind turbine's power response to wind speed changes and a higher power generation efficiency. Conversely, the wind turbines ranked last in the dispersion ranking have lower sensitivity and are considered inefficient for that model. A preset number of wind turbines at the end of the dispersion ranking are designated as those awaiting maintenance.

[0170] (2) Assessment of degree of compliance.

[0171] Step 104 determines the compliance ranking of each wind turbine for each model. A positive compliance value indicates that the wind turbine's power curve is superior to the average power curve for that model. A higher compliance value indicates a closer approximation to the theoretical power curve, and thus a higher power generation efficiency. Similarly, the wind turbines ranked last in the compliance ranking have power curves that deviate significantly from the average power curve for that model, making them inefficient within that model. A preset number of wind turbines ranked last in the compliance ranking are designated as those awaiting maintenance.

[0172] (3) Electricity decomposition and assessment.

[0173] a. Assessment of power generation loss due to shutdown

[0174] Step 106 obtains a ranking of power generation lost due to downtime for each wind turbine of each model. A wind turbine with a lower ranking for power generation lost due to downtime, for example, has a relatively high power generation loss due to objective factors such as downtime. This indicates that the wind turbine is inefficient for this model and its power generation can be improved through maintenance and overhaul. A preset number of wind turbines ranked last in the power generation lost due to downtime are designated as wind turbines to be maintained.

[0175] b. Assessment of power generation loss due to power restriction

[0176] Through step 106, the ranking of power generation loss due to power curtailment for each wind turbine of each model can be obtained. The lower the ranking of power generation loss due to power curtailment, the greater the power generation loss of the wind turbine due to objective factors such as power curtailment. This indicates that the wind turbine is inefficient for that model and its power generation can be improved through operation and maintenance. The wind turbines with the lowest power generation loss ranking at the end of the preset number are designated as wind turbines to be operated and maintained.

[0177] c. Abnormal power loss assessment

[0178] Step 106 obtains a ranking of abnormal power loss for each wind turbine of each model. A wind turbine with a lower abnormal power loss ranking indicates a greater power loss due to objective reasons such as failure or abnormal operation. This indicates that the wind turbine is considered inefficient for that model and its power generation can be improved through maintenance. A preset number of wind turbines at the end of the abnormal power loss ranking are designated as wind turbines to be maintained.

[0179] d. Performance loss evaluation

[0180] Step 106 obtains a performance loss ranking for each wind turbine of each model. A lower wind turbine performance loss ranking indicates a greater loss of power generation due to its own performance deviation, making it an inefficient wind turbine for that model. Control parameter optimization can be used to improve power generation. A preset number of wind turbines at the bottom of the wind turbine performance loss ranking are designated as wind turbines awaiting maintenance.

[0181] e.EBA assessment

[0182] Through step 106, the EBA ranking of each wind turbine of each model can be obtained. The higher the wind turbine EBA ranking, the higher the wind turbine's ability to utilize wind energy. Conversely, the lower the wind turbine EBA ranking, the more power loss the wind turbine has caused due to objective reasons such as failures and abnormal operation. This indicates that the wind turbine is an inefficient wind turbine of that model. From the perspective of operation and maintenance, it can be explored whether the wind turbine has the potential to increase power generation. The wind turbines with the last preset number of wind turbine EBA rankings are designated as wind turbines to be operated and maintained.

[0183] In an exemplary embodiment, after step 107, the wind turbine operation and maintenance evaluation method further includes:

[0184] Generate a wind turbine operation and maintenance report, which includes a status identification scatter plot of each wind turbine, a wind turbine dispersion ranking table, a wind turbine compliance ranking table, a wind turbine power index ranking table, a wind turbine power index bar chart, and the number and model of the wind turbine to be operated and maintained.

[0185] Based on data classification, this application obtains a scatter plot of the wind turbine's operating status, a fan's dispersion ranking, a fan's compliance ranking, a fan's power decomposition ranking, and information on inefficient wind turbines. Combined with the pypdf library in Python, the above information is integrated and processed, and finally an energy efficiency analysis report for the wind farm to be analyzed is automatically generated.

[0186] The framework of the wind farm energy efficiency analysis report is as follows:

[0187] (I) Overview: This mainly includes the wind turbine model to be analyzed, the turbine ID for the specific model, the analysis data time, and basic wind turbine information. Basic information includes: turbine hub height, impeller diameter, rated power, rated wind speed, cut-in wind speed, cut-out wind speed, and the longitude and latitude of the wind turbine location.

[0188] (2) Technical route: mainly includes the technical framework of the analysis report, specifically: data modeling, data collection, data storage, data analysis, automatic report generation, visual display, system services and the specific technologies corresponding to each module.

[0189] (3) State identification: mainly includes the state identification scatter diagram of the fan obtained in step 102.

[0190] (IV) Discreteness ranking: mainly includes the fan discreteness ranking table obtained in step 103.

[0191] (5) Compliance ranking: mainly includes the wind turbine compliance ranking table obtained in step 104.

[0192] (6) Electricity decomposition ranking: mainly includes the ranking table and bar chart of each electricity decomposition index of the wind turbine obtained in step 106.

[0193] (VII) Conclusion: This mainly includes the numbers of the wind turbines to be operated and maintained obtained in step 107 and their corresponding models.

[0194] The wind turbine operation and maintenance report is displayed through a visual interface, and the visual interface is also used to delete, query, preview and download the wind turbine operation and maintenance report.

[0195] This application will obtain multi-dimensional analysis information of wind turbine operation data, combine it with the front-end framework to display the analysis data, and finally form a front-end UI interface of the wind turbine intelligent operation and maintenance evaluation system that can be displayed on the web and is operational.

[0196] The UI system functions are as follows:

[0197] State Identification: Mainly displays the state identification scatter plot of the wind turbine obtained in step 102, and the state identification scatter plot of the wind turbine analyzed historically can be deleted and queried on this page.

[0198] Discreteness analysis: mainly displays the fan discreteness ranking form obtained in step 103, and can delete and query the fan discreteness ranking form of historical analysis on this page.

[0199] Compliance analysis: mainly displays the wind turbine compliance ranking form obtained in step 104, and can delete and query the wind turbine compliance ranking form of historical analysis on this page.

[0200] Power decomposition analysis: mainly displays the ranking table and bar chart of each power decomposition index of the wind turbine obtained in step 106, and can delete and query the ranking table and bar chart of each power decomposition index of the wind turbine in historical analysis on this page.

[0201] Report Query: Mainly displays the obtained wind farm energy efficiency analysis report, and can delete, query, preview and download the historical wind farm energy efficiency analysis report on this page.

[0202] Trend analysis: Mainly uses bar charts, line charts, pie charts and other charts to display the statistical analysis results of a single wind turbine in different time dimensions obtained from data analysis. It can more intuitively show the data trend changes of the wind turbine over a period of time, and then quickly locate the abnormalities that may occur in the wind turbine at a specific time point; at the same time, the statistical analysis results of the historical analysis of wind turbines can also be deleted and queried on this page.

[0203] Model Analysis: Mainly uses bar charts, line charts, pie charts and other charts to display the statistical analysis results of multiple fans of the same model obtained through data analysis. It can more intuitively show the data trend changes of multiple fans of the same model, and then quickly locate inefficient fans of the model. This page also allows deletion and query of the statistical analysis results of multiple fans of the same model under historical analysis.

[0204] This application also includes back-end system services for building a wind turbine intelligent operation and maintenance evaluation system using a back-end framework.

[0205] The main functions of the system services are as follows:

[0206] (1) Processing client requests: mainly processing query, deletion, download and other requests made by web clients on the front end.

[0207] (2) Executing business logic: Mainly executing the functional logic corresponding to the visual display, and configuring each function in the form of a scheduled task, that is, the execution time of the functional task corresponding to the visual display can be flexibly configured according to needs.

[0208] (3) Database interaction: mainly stores the data results generated in the system service in the relational database, and performs add, delete, modify and query operations on the database according to client requests.

[0209] (IV) Module communication: mainly uses RESTful API to communicate between the functional module services mentioned in steps 101 to 107 and the visualization step.

[0210] This application can accurately locate inefficient wind turbines, display relevant analysis data and automatically generate energy efficiency analysis reports, thereby optimizing resource allocation and identifying inefficient wind turbines. Compared with traditional operation and maintenance methods, it greatly improves the analysis efficiency of wind farm operation and maintenance personnel.

[0211] This application implements a method for identifying inefficient fans based on the actual theoretical power curve of the fan and then performing statistical analysis. The data results obtained can be more realistic and more reliable.

[0212] This application is based on many statistical analysis methods, such as using the K-Means++ algorithm to identify the status of the wind turbine, using discrete statistical indicators to evaluate the sensitivity of the wind turbine to changes in wind speed, using scale-reaching statistical indicators to evaluate the performance differences between wind turbine units, and using the wind turbine status obtained based on state identification to evaluate the power generation of the wind turbine. These methods can evaluate the status of the wind turbine from multiple dimensions, overcoming the problems of low accuracy of analysis results and inaccurate positioning of inefficient fans caused by using a single dimension to identify inefficient fans in traditional methods.

[0213] Based on the same inventive concept, embodiments of the present application also provide a wind turbine operation and maintenance evaluation device for implementing the aforementioned wind turbine operation and maintenance evaluation method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the wind turbine operation and maintenance evaluation device provided below can be found in the above-described limitations of the wind turbine operation and maintenance evaluation method and will not be further elaborated here.

[0214] In an exemplary embodiment, the present application provides a wind turbine operation and maintenance evaluation device, wherein the wind turbine operation and maintenance evaluation device applies the wind turbine operation and maintenance evaluation method, and the wind turbine operation and maintenance evaluation device includes:

[0215] The operating data acquisition module is used to obtain the operating data of each fan in the target unit.

[0216] The power band determination module is used to identify the fan status of each fan at each data point based on the operating data of each fan, and determine the power band of each fan based on the fan status of each data point.

[0217] The dispersion determination module is used to determine the dispersion of each fan according to the power band of each fan.

[0218] The compliance determination module is used to determine the compliance of each wind turbine according to the power band of each wind turbine.

[0219] The theoretical power curve determination module is used to use the power curve of the fan with the highest degree of compliance as the theoretical power curve of the corresponding fan model for each fan model.

[0220] The power index determination module is used to calculate the power index of each fan according to the theoretical power curve of each fan model.

[0221] The wind turbine screening module for operation and maintenance is used to screen out wind turbines for operation and maintenance based on the discreteness, compliance and power indicators of each wind turbine.

[0222] In an exemplary embodiment, the present application provides a wind turbine operation and maintenance evaluation device including: a data acquisition module, a data storage module, a data analysis module, an automatic report generation module, a visualization display module and a system service module. The data analysis module performs data modeling based on the real-time operation data of the wind turbine, and then identifies the operation status of the wind turbine in combination with business rules and KMeans++ clustering algorithm, and analyzes the sensitivity of the power response of each wind turbine of the same model to the change of wind speed in combination with the discreteness statistical index, and then quantifies the difference in power curves between different wind turbines of the same model in combination with the standardization statistical index, and then obtains the actual theoretical power curve of the model based on the standardization calculation result, and calculates the power generation status of the wind turbine in different states in combination with the operation status of the wind turbine, and finally locates the inefficient wind turbine and gives corresponding operation and maintenance suggestions.

[0223] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0224] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A wind turbine operation and maintenance evaluation method, characterized in that: The wind turbine operation and maintenance evaluation method includes: Obtain the operating data of each fan in the target unit; Identifying the fan status of each fan at each data point based on the operating data of each fan, and determining the power band of each fan based on the fan status at each data point; Determine the discreteness of each fan according to its power band; Determine the compliance of each fan according to its power band; For each fan model, the power curve of the fan with the highest degree of compliance is used as the theoretical power curve of the corresponding fan model; Calculate the power indicators of each fan according to the theoretical power curve of each fan model; The wind turbines to be operated and maintained are selected based on their discreteness, compliance and power indicators.

2. The wind turbine operation and maintenance evaluation method according to claim 1, characterized in that: After obtaining the operating data of each wind turbine in the target unit, the wind turbine operation and maintenance evaluation method further includes: The operating data of each wind turbine is stored in a time series database; the operating data includes wind speed, rotation speed and power; the power is the generated power of the wind turbine.

3. The wind turbine operation and maintenance evaluation method according to claim 2, characterized in that: The fan status of each fan at each data point is identified based on the operating data of each fan, and the power band of each fan is determined based on the fan status at each data point, specifically including: Identify the operating data of each fan into normal state, abnormal state and shutdown state according to the set rules; For the operating data of any wind turbine in normal state, a data scatter plot is obtained in the Cartesian coordinate system with wind speed as the horizontal axis and power as the vertical axis; Dividing the data scatter plot into consecutive equally spaced intervals according to the vertical axis to obtain a plurality of power intervals; Clustering the wind speeds within each power interval, and classifying the wind speeds according to the clustering results; the classification includes a first abnormal area, a normal area, and a second abnormal area distributed in sequence from the origin to the direction of increasing wind speed on the horizontal axis; Identifying a power-limited state on the operating data of the second abnormal area to obtain a corrected normal state; The power band of each fan is determined according to the corrected normal state.

4. The wind turbine operation and maintenance evaluation method according to claim 3, characterized in that: Identifying the power-limited state of the operating data of the second abnormal area to obtain a corrected normal state specifically includes: Calculate the range value of each power interval; Calculate the normalized weight of each power interval; Determine the power-limiting state identification tag value for each power interval based on the extreme value and normalized weight of each power interval; The power-limiting state identification tag value obtained for each power interval is processed with the 3σ maximum value abnormal value. If the power-limiting state identification tag value of the i-th power interval is greater than the 3σ maximum value, the normal state of the data in the second abnormal area of ​​the i-th power interval is corrected to the power-limiting state. Otherwise, no correction is performed.

5. The wind turbine operation and maintenance evaluation method according to claim 3, characterized in that: The discreteness of each fan is determined according to its power band, including: For any wind turbine power band, within the preset effective wind speed range, the wind speed on the horizontal axis is divided into equal intervals to obtain multiple wind speed intervals; Calculate the coefficient of variation of power in each wind speed interval; The average value of the coefficient of variation of power in each wind speed range is taken as the dispersion of the wind turbine.

6. The wind turbine operation and maintenance evaluation method according to claim 3, characterized in that: The compliance of each fan is determined based on its power band, including: For any wind turbine power band, within the preset effective wind speed range, the wind speed on the horizontal axis is divided into equal intervals to obtain multiple wind speed intervals; Calculate the degree of compliance for each wind speed range.

7. The wind turbine operation and maintenance evaluation method according to claim 2, characterized in that: Calculate the power indicators of each fan based on the theoretical power curve of each fan model, including: Calculate the theoretical power generation of each fan within a set time period based on the theoretical power curve of each fan model; Calculate the actual power generation of any wind turbine according to the operating data of the wind turbine within a set time period; The power generation loss due to shutdown is obtained according to the difference between the theoretical power generation and the actual power generation of the wind turbine in the shutdown state within the set time period; Obtaining power loss due to power restriction based on the difference between the theoretical power generation and the actual power generation of the wind turbine in the power restriction state within the set time period; Obtaining abnormal power loss according to the difference between the theoretical power generation and the actual power generation of the wind turbine in the abnormal state within the set time period; The actual power generation of the wind turbine is obtained according to the actual power generation of the wind turbine, the power generation loss due to shutdown, the power generation loss due to power restriction and the power generation loss due to abnormality within the set time period; Calculate the performance loss of the wind turbine based on the theoretical power generation and the actual power generation of the wind turbine within the set time period; The energy utilization rate of the wind turbine is calculated based on the actual power generation and the actual power generation expected during the set time period of the wind turbine.

8. The wind turbine operation and maintenance evaluation method according to claim 1, characterized in that: After selecting the wind turbines to be operated and maintained based on the dispersion, compliance and power indicators of each wind turbine, the wind turbine operation and maintenance evaluation method further includes: Generate a wind turbine operation and maintenance report, which includes a status identification scatter plot of each wind turbine, a wind turbine dispersion ranking table, a wind turbine compliance ranking table, a wind turbine power index ranking table, a wind turbine power index bar chart, and the number and model of the wind turbine to be operated and maintained.

9. The wind turbine operation and maintenance evaluation method according to claim 8, characterized in that: The wind turbine operation and maintenance report is displayed through a visual interface, and the visual interface is also used to delete, query, preview and download the wind turbine operation and maintenance report.

10. A wind turbine operation and maintenance evaluation device, characterized in that: The wind turbine operation and maintenance evaluation device applies the wind turbine operation and maintenance evaluation method according to any one of claims 1 to 9, and the wind turbine operation and maintenance evaluation device includes: An operating data acquisition module is used to obtain the operating data of each fan in the target unit; A power band determination module, configured to identify the fan status of each fan at each data point based on the operating data of each fan, and determine the power band of each fan based on the fan status at each data point; A dispersion determination module, configured to determine the dispersion of each fan according to the power band of each fan; A compliance determination module is used to determine the compliance of each wind turbine according to the power band of each wind turbine; A theoretical power curve determination module is used to determine, for each wind turbine model, the power curve of the wind turbine with the highest degree of compliance as the theoretical power curve of the corresponding wind turbine model; The power index determination module is used to calculate the power index of each fan according to the theoretical power curve of each fan model; The wind turbine screening module for operation and maintenance is used to screen out wind turbines for operation and maintenance based on the discreteness, compliance and power indicators of each wind turbine.

Images