Method and device for calculating power loss of wind turbine generator set

Abstract

Disclosed are a method and device for calculating the power loss of a wind turbine generator set. The method comprises: acquiring the power, influencing factor data, and device status data of the wind turbine generator set in real time; determining whether the wind turbine generator set is in a normal or abnormal state based on the device status data; and, in response to the wind turbine generator set being in an abnormal state, calculating the power loss of the wind turbine generator set based on the power and influencing factor data. The present invention classifies the influencing factor data based on the degree of its impact on power, and calculates the power loss of the wind turbine generator set based on the power and the classified influencing factor data, thereby increasing the accuracy of the calculated power loss.

Description

Technical Field

[0001] The present invention relates to a method and device for calculating the lost electricity of a wind turbine generator set, and more particularly, to a method and device for calculating the lost electricity of a wind turbine generator set capable of more accurately calculating the lost electricity. Background Art

[0002] With the rapid development of renewable energy generation, power generation companies and electricity users are paying increasing attention to various power generation indicators. Power loss is one of the most important indicators, directly impacting the vital interests of power generation companies and electricity users, and significantly affecting the reliability and economic efficiency of the entire system. Therefore, accurately calculating power loss is of great practical significance.

[0003] Current methods for calculating power loss include theoretical device power generation and benchmark device power generation. However, the theoretical device power generation method, provided by the manufacturer, represents the maximum power generation capacity of the device under ideal conditions and does not consider the impact of actual on-site conditions on power generation. The benchmark device power generation method designates one or more devices as benchmark devices, failing to account for differences between the benchmark device and other devices. Furthermore, calculations become impossible if the benchmark device experiences a malfunction. Summary of the Invention

[0004] The object of the present invention is to provide a method and device for calculating the lost power of a wind turbine generator set, which can more accurately calculate the lost power.

[0005] According to one embodiment of the present invention, there is provided a method for calculating the power loss of a wind turbine generator set, comprising: acquiring power, influencing factor data, and device status data of the wind turbine generator set in real time; judging whether the wind turbine generator set is in a normal state or an abnormal state based on the device status data; and calculating the power loss of the wind turbine generator set based on the power and influencing factor data in response to the wind turbine generator set being in an abnormal state.

[0006] Optionally, the influencing factor data includes data of a plurality of variables, wherein the plurality of variables are initially classified into basic influencing factor variables, new influencing factor variables, low-influencing factor variables, and uncategorized variables.

[0007] Optionally, the multiple variables include at least wind speed, wind direction, temperature, air density, humidity and light intensity.

[0008] Optionally, the calculation method also includes: in response to the wind turbine being in a normal state, classifying the influencing factor data based on the degree of influence of the influencing factor data on power, wherein, in response to the wind turbine being in an abnormal state, the step of calculating the lost electricity of the wind turbine based on the power and the influencing factor data includes: in response to the wind turbine being in an abnormal state, calculating the lost electricity of the wind turbine based on the power and the classified influencing factor data.

[0009] Optionally, in response to the wind turbine being in a normal state, the step of classifying the influencing factor data based on the degree of influence of the influencing factor data on power includes: in response to the wind turbine being in a normal state, classifying low-influence factor variables and uncategorized variables whose values change with power changes as pending new influencing factor variables; verifying each pending new influencing factor variable; classifying the pending new influencing factor variables that pass the verification as new influencing factor variables, and classifying the pending new influencing factor variables that fail the verification as low-influence factor variables, thereby updating the classification of the influencing factor data.

[0010] Optionally, in response to the wind turbine being in a normal state, the step of classifying low-impact factor variables and uncategorized variables whose values change with power changes as pending new influencing factor variables includes: obtaining from a database multiple sets of low-impact factor variables and uncategorized variables whose values change with power changes while the values of basic influencing factor variables remain unchanged; and classifying the variables in the intersection of the multiple sets as pending new influencing factor variables.

[0011] Optionally, the step of verifying each pending new influencing factor variable includes: for any pending new influencing factor variable, obtaining a first fitting function for calculating power by performing multivariate fitting on the power obtained from the database, the value of the basic influencing factor variable, the value of the new influencing factor variable and the value of any pending new influencing factor variable; obtaining multiple calculated powers by substituting the value of the basic influencing factor variable, the value of the new influencing factor variable and the value of any pending new influencing factor variable obtained in real time multiple times when the wind turbine generator set is in a normal state into the first fitting function; if the difference between the multiple calculated powers and the corresponding multiple real-time obtained powers is less than the first threshold and the number of times is greater than the second threshold, the verification is passed; otherwise, the verification is failed.

[0012] Optionally, in response to the wind turbine being in a normal state, the step of classifying the influencing factor data based on the degree of influence of the influencing factor data on power includes: in response to the wind turbine being in a normal state, classifying new influencing factor variables whose values remain unchanged with changes in power as pending low-influence factor variables; verifying each pending low-influence factor variable; classifying the pending low-influence factor variables that have passed the verification as low-influence factor variables, and classifying the pending low-influence factor variables that have not passed the verification as new influencing factor variables, thereby updating the classification of the influencing factor data.

[0013] Optionally, in response to the wind turbine generator set being in a normal state, the step of classifying a new influencing factor variable whose value remains unchanged with power changes as a pending low-influence factor variable includes: obtaining from a database multiple sets of new influencing factor variables whose values remain unchanged with power changes when the values of basic influencing factor variables remain unchanged; and classifying the variables in the intersection of the multiple sets as pending low-influence factor variables.

[0014] Optionally, the step of verifying each pending low-impact factor variable includes: obtaining a second fitting function for calculating power by performing multivariate fitting on the power obtained from the database, the values of the basic influencing factor variables and the values of the new influencing factor variables; obtaining multiple calculated powers by substituting the values of the basic influencing factor variables and the values of the new influencing factor variables obtained in real time multiple times when the wind turbine generator set is in a normal state into the second fitting function; if the difference between the multiple calculated powers and the corresponding multiple real-time obtained powers is less than the third threshold and the number of times is greater than the fourth threshold, the verification is passed; otherwise, the verification fails.

[0015] Optionally, in response to the wind turbine being in an abnormal state, the step of calculating the lost electricity of the wind turbine based on the power and classified influencing factor data includes: obtaining a third fitting function for calculating the power by performing multivariate fitting on the power, the values of the basic influencing factor variables and the values of the new influencing factor variables obtained from the database; obtaining the calculated power by substituting the values of the basic influencing factor variables and the values of the new influencing factor variables obtained in real time when the wind turbine is in an abnormal state into the third fitting function; calculating the lost electricity of the wind turbine based on the calculated power, the power obtained in real time when the wind turbine is in an abnormal state and the time information when the wind turbine is in the abnormal state.

[0016] Optionally, based on the calculated power, the power obtained in real time when the wind turbine is in an abnormal state, and the time information when the wind turbine is in the abnormal state, the step of calculating the lost electricity of the wind turbine includes: calculating the lost electricity of the wind turbine by integrating the difference between the calculated power and the power obtained in real time when the wind turbine is in the abnormal state according to the time information.

[0017] Optionally, the method further includes: in response to the wind turbine being in a normal state, storing the real-time acquired power and the results of binning and valuing the real-time acquired influencing factor data as historical influencing factor data and historical power data in a database.

[0018] According to one embodiment of the present invention, there is provided a device for calculating the lost electricity of a wind turbine generator set, comprising: an acquirer configured to acquire power, influencing factor data, and device status data of the wind turbine generator set in real time; a judger configured to judge, based on the device status data, whether the wind turbine generator set is in a normal state or an abnormal state; and a calculator configured to calculate the lost electricity of the wind turbine generator set based on the power and influencing factor data in response to the wind turbine generator set being in an abnormal state.

[0019] Optionally, the influencing factor data includes data of a plurality of variables, wherein the plurality of variables are initially classified into basic influencing factor variables, new influencing factor variables, low-influencing factor variables, and uncategorized variables.

[0020] Optionally, the multiple variables include at least wind speed, wind direction, temperature, air density, humidity and light intensity.

[0021] Optionally, the computing device further includes: a classifier configured to classify the influencing factor data based on the degree of influence of the influencing factor data on the power in response to the wind turbine being in a normal state, wherein the calculator is further configured to calculate the lost electricity of the wind turbine based on the power and the classified influencing factor data in response to the wind turbine being in an abnormal state.

[0022] Optionally, the classifier is also configured to: in response to the wind turbine being in a normal state, classify low-impact factor variables and uncategorized variables whose values change with power changes as pending new influencing factor variables; verify each pending new influencing factor variable; classify the pending new influencing factor variables that pass the verification as new influencing factor variables, and classify the pending new influencing factor variables that fail the verification as low-impact factor variables, thereby updating the classification of influencing factor data.

[0023] Optionally, the classifier is also configured to: obtain from the database multiple sets of low-influence factor variables and uncategorized variables whose values change with power while the values of basic influencing factor variables remain unchanged; and classify the variables in the intersection of the multiple sets as new influencing factor variables to be determined.

[0024] Optionally, the classifier is further configured to: for any one of the pending new influencing factor variables, obtain a first fitting function for calculating the power by performing multivariate fitting on the power obtained from the database, the value of the basic influencing factor variable, the value of the new influencing factor variable and the value of any one of the pending new influencing factor variables; obtain multiple calculated powers by substituting the values of the basic influencing factor variable, the value of the new influencing factor variable and the value of any one of the pending new influencing factor variables obtained in real time multiple times when the wind turbine generator set is in a normal state into the first fitting function; if the difference between the multiple calculated powers and the corresponding multiple real-time obtained powers is less than the first threshold and the number of times is greater than the second threshold, the verification is passed; otherwise, the verification is failed.

[0025] Optionally, the classifier is also configured to: in response to the wind turbine being in a normal state, classify new influencing factor variables whose values remain unchanged with power changes as pending low-influence factor variables; verify each pending low-influence factor variable; classify the pending low-influence factor variables that pass the verification as low-influence factor variables, and classify the pending low-influence factor variables that fail the verification as new influencing factor variables.

[0026] Optionally, the classifier is further configured to: obtain from the database multiple sets of new influencing factor variables whose values remain unchanged as power changes while the values of basic influencing factor variables remain unchanged; and classify variables in the intersection of the multiple sets as pending low-influence factor variables.

[0027] Optionally, the classifier is also configured to: obtain a second fitting function for calculating power by performing multivariate fitting on the power, the values of the basic influencing factor variables and the values of the new influencing factor variables obtained from the database; obtain multiple calculated powers by substituting the values of the basic influencing factor variables and the values of the new influencing factor variables obtained in real time multiple times when the wind turbine generator set is in a normal state into the second fitting function; if the difference between the multiple calculated powers and the corresponding multiple real-time obtained powers is less than the third threshold and the number of times is greater than the fourth threshold, the verification is passed; otherwise, the verification is failed.

[0028] Optionally, the calculator is configured to: obtain a third fitting function for calculating power by performing multivariate fitting on the power obtained from the database, the values of the basic influencing factor variables, and the values of the new influencing factor variables; obtain the calculated power by substituting the values of the basic influencing factor variables and the values of the new influencing factor variables obtained in real time when the wind turbine is in an abnormal state into the third fitting function; calculate the lost electricity of the wind turbine based on the calculated power, the power obtained in real time when the wind turbine is in an abnormal state, and the time information when the wind turbine is in the abnormal state.

[0029] Optionally, the calculator is further configured to calculate the lost power of the wind turbine generator set by integrating the difference between the calculated power and the power acquired in real time when the wind turbine generator set is in an abnormal state according to the time information.

[0030] Optionally, the device also includes: a memory, configured to: in response to the wind turbine generator set being in a normal state, store the real-time acquired power and the results of binning and valuing the real-time acquired influencing factor data as historical influencing factor data and historical power data in a database.

[0031] According to one embodiment of the present inventive concept, a computer-readable storage medium storing a computer program is provided. When the computer program is executed by a processor, the method for calculating the power loss of a wind turbine generator set as described above is implemented.

[0032] According to an embodiment of the present invention, a computing device is provided, including: a processor; and a memory storing a computer program. When the computer program is executed by the processor, the method for calculating the power loss of a wind turbine generator set as described above is implemented.

[0033] The present invention classifies influencing factor data based on their degree of influence on power, and calculates the wind turbine's power loss based on the power and the classified influencing factor data. This results in a more accurate calculation of power loss. Furthermore, because the calculation of wind turbine power loss does not require the use of benchmark equipment, even if individual equipment fails, the calculation of power loss is not affected, resulting in greater adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The above and / or other aspects of the present disclosure will become clearer and more easily understood through the following detailed description taken in conjunction with the accompanying drawings, in which:

[0035] Figure 1 is a flow chart illustrating a method for calculating power loss of a wind turbine generator system according to an embodiment of the present inventive concept;

[0036] Figure 2is a flow chart illustrating a method for determining new influencing factor variables according to an embodiment of the present inventive concept;

[0037] Figure 3 is a flow chart illustrating a method for determining a new influencing factor variable to be determined according to an embodiment of the present inventive concept;

[0038] Figure 4 is a flow chart illustrating a method for verifying a pending new influencing factor variable according to an embodiment of the present inventive concept;

[0039] Figure 5 is a flowchart illustrating a method for determining low-impact factor variables according to an embodiment of the present inventive concept;

[0040] Figure 6 is a flow chart illustrating a method for determining a pending low-impact factor variable according to an embodiment of the present inventive concept;

[0041] Figure 7 is a flow chart illustrating a method for verifying a pending low-impact factor variable according to an embodiment of the present inventive concept;

[0042] Figure 8 is a flowchart illustrating a method for calculating power loss according to an embodiment of the present inventive concept;

[0043] Figure 9 is a block diagram illustrating a device for calculating a lost amount of power of a wind turbine generator system according to an embodiment of the present inventive concept. DETAILED DESCRIPTION

[0044] Hereinafter, embodiments of the inventive concept will be described in detail with reference to the accompanying drawings.

[0045] Figure 1 is a flowchart illustrating a method for calculating lost power of a wind turbine generator system according to an embodiment of the present inventive concept.

[0046] Reference Figure 1 In step S1, the power, influencing factor data, and device status data of the wind turbine generator set are acquired in real time. In one example, the power, influencing factor data, and device status data of the wind turbine generator set may be acquired at predetermined time intervals. For example, the predetermined time interval may be 6 seconds.

[0047] The power of a wind turbine generator set may represent the actual output power or actual power generation of the wind turbine generator set. When calculating the power loss of a single wind turbine generator set, the power of a wind turbine generator set may represent the actual output power or actual power generation of the single wind turbine generator set. Accordingly, when calculating the total power loss of all wind turbine generator sets in an entire wind farm, the power of a wind turbine generator set may represent the actual total output power or actual total power generation of all wind turbine generator sets in the entire wind farm.

[0048] The influencing factor data may include data on multiple variables. Specifically, the influencing factor data may include data on multiple variables that can affect the power of the wind turbine generator set. In one example, the influencing factor data may include or primarily include data on environmental variables. For example, the multiple variables may include at least wind speed, wind direction, temperature, air density, humidity, light intensity, etc. However, the example is not limited to this, and the multiple variables may also include other variables (or other environmental variables) in addition to wind speed, wind direction, temperature, air density, humidity, light intensity, etc.

[0049] Multiple variables may be initially or pre-classified into basic influencing factor variables, new influencing factor variables, and low-influence factor variables. Basic influencing factor variables may represent variables that have a significant and / or inevitable impact on the power of a wind turbine generator set. In one example, basic influencing factor variables may include wind speed, wind direction, etc. However, the example is not limited to this, and basic influencing factor variables may also include other variables (or other environmental variables) in addition to wind speed, wind direction, etc. New influencing factor variables may represent variables that have a certain degree of impact on the power of a wind turbine generator set. In one example, new influencing factor variables may include temperature, air density, etc. However, the example is not limited to this, and new influencing factor variables may also include other variables (or other environmental variables) in addition to temperature, air density, etc. Low-influence factor variables may represent variables that have an insignificant and / or negligible impact on the power of a wind turbine generator set. In one example, low-influence factor variables may include light intensity, etc. However, the example is not limited to this, and low-influence factor variables may also include other variables (or other environmental variables) in addition to light intensity, etc. In one example, the basic influencing factor variables, new influencing factor variables, and low-influencing factor variables may be classified according to engineering experience, or may be determined according to results of previous classification of influencing factor data.

[0050] In addition, multiple variables may also include uncategorized variables. Uncategorized variables may represent variables whose degree of influence on the power of a wind turbine is still unclear. For example, uncategorized variables may be variables that have not been classified after the influencing factors were previously classified or newly introduced variables. In one example, a variable whose degree of influence on the power of a wind turbine is still unclear (i.e., an uncategorized variable) may be initially classified as any one of a new influencing factor variable and a low-influence factor variable, and then the classification of the new influencing factor variable and the low-influence factor variable may be updated by a subsequent process of classifying the influencing factor data. That is, multiple variables may or may not include uncategorized variables according to actual needs, and the variable whose degree of influence on the power of a wind turbine is still unclear may exist in multiple variables in other classification forms besides uncategorized variables.

[0051] Furthermore, the classification of basic influencing factor variables, new influencing factor variables, low-influence factor variables, and uncategorized variables can be achieved by adding corresponding data labels to the data of multiple variables in the database. Therefore, reclassifying influencing factor data can be achieved by modifying the data labels of multiple variables. In one example, low-influence factor variables and new influencing factor variables can be converted to each other under specific conditions. For example, specific conditions may include specific seasons, specific uncategorized variables, etc.

[0052] Furthermore, in one example, in addition to obtaining real-time wind turbine power, influencing factor data, and device status data, other wind turbine data can also be obtained in real time. This other data can be data on variables that have no impact on the functioning of the wind turbine. For example, this other data can include data on the wind turbine's voltage, current, capacitance, resistance, and the like.

[0053] Device status data can indicate whether the device is operating normally. In one example, the device status data can be used to categorize the wind turbine's status into nine states based on several factors, including the wind turbine's communication status, fault code, warning mode word, operating status word, shutdown mode word, and power limit mode word. Table 1 below details the names and meanings of these nine states.

[0054]

Table 1

[0055]

[0056] In step S2, based on the device status data, it is determined whether the wind turbine generator set is in a normal state or an abnormal state. For example, when the device status data indicates a normal power generation state, the wind turbine generator set may be determined to be in a normal state; when the device status data indicates at least one of a power restriction state, a wind turbine maintenance state, a technical standby state, a remote shutdown state, a grid fault state, a fault shutdown state, a local shutdown state, and other power restriction states, the wind turbine generator set may be determined to be in an abnormal state.

[0057] In step S3, in response to the wind turbine generator set being in an abnormal state, the power loss of the wind turbine generator set is calculated based on the power and influencing factor data. Figure 8 Step S3 is described in detail.

[0058] In step S4, in response to the wind turbine being in a normal state, the influencing factor data is classified based on the degree of influence of the influencing factor data on the power. It should be understood that step S4 is an optional step. That is, when the wind turbine is in a normal state, step S4 can be selectively performed. For example, when the wind turbine is in a normal state, step S4 can be performed in response to a user instruction or in response to the introduction of a new variable or in response to meeting a certain time interval condition. In addition, after executing step S4, when the wind turbine is in an abnormal state again, step S3 may further include: in response to the wind turbine being in an abnormal state, calculating the lost power of the wind turbine based on the power and the classified influencing factor data. Step S4 may include step S6 of determining new influencing factor variables and step S7 of determining low influencing factor variables. In the following, reference will be made to Figure 2 The step S6 of determining the new influencing factor variables will be described in detail and will be referred to Figure 5 The step S7 of determining the low-influence factor is described in detail.

[0059] In one example, in response to the wind turbine being in a normal state, a step S5 of storing data may be further included. Step S5 may include: in response to the wind turbine being in a normal state, storing the power obtained in real time and the result of binning the influencing factor data obtained in real time as historical influencing data and historical power data in the database. For example, the first power obtained at the first time and the first result of binning the first influencing factor data obtained in real time may be stored in the database as the first data, and the second power obtained at the second time after the first time and the second result of binning the second influencing factor data obtained in real time may be stored in the database as the second data.

[0060] In another example, in addition to obtaining the power, influencing factor data and equipment status data of the wind turbine generator set in real time, other data of the wind turbine generator set can also be obtained in real time, step S5 may include: in response to the wind turbine generator set being in a normal state, storing the real-time obtained power, the results of binning and averaging the real-time obtained influencing factor data, and the results of averaging the real-time obtained other data in the database as historical influencing factor data, historical power data and historical other data.

[0061] In addition, the above-mentioned step S5 of storing data, step S6 of determining new influencing factor variables, and step S7 of determining low-influencing factor variables can be executed in parallel, in any order, or at different frequencies, and this application does not impose any specific restrictions on this.

[0062] Figure 2 1 is a flow chart illustrating a method for determining new influencing factor variables according to an embodiment of the present invention. Specifically, Figure 2 This is a detailed description of the above step S6 of determining new influencing factor variables.

[0063] Reference Figure 2 , in step S61, in response to the wind turbine being in a normal state, low-impact factor variables and uncategorized variables whose values change with power changes are classified as pending new influencing factor variables. The pending new influencing factor variables may represent factors that may have a certain impact on the power of the wind turbine. In other words, the pending new influencing factor variables may represent variables that may be ultimately classified as new factor influencing variables. In one example, low-impact factor variables and uncategorized variables whose values change with power changes within a predetermined time period stored in the database may be classified as pending new influencing factor variables. Here, the predetermined time period may be one or more days, one or more weeks, one or more months, one quarter or more quarters, or one or more years. In addition, when there are no uncategorized variables, step S61 may be performed only on the low-impact factor variables. Optionally, in one example, step S61 may also include classifying uncategorized variables whose values do not change with power changes as low-impact factor variables. In the following, reference will be made to Figure 3 Step S61 is described in detail.

[0064] Figure 3 1 is a flow chart illustrating a method for determining a new influencing factor variable to be determined according to an embodiment of the present invention. Specifically, Figure 3 This is a detailed description of step S61 of classifying low-impact factor variables and unclassified variables whose values change with power changes as new influencing factor variables to be determined in response to the wind turbine generator being in a normal state.

[0065] Reference Figure 3In step S611, multiple sets of low-impact factor variables and uncategorized variables whose values vary with power when the value of the basic influencing factor variable remains unchanged are obtained from the database. In step S612, the variables in the intersection of these multiple sets are classified as pending new influencing factor variables. For example, when the basic influencing factor variable is wind speed, temperature and air density, whose values vary with power when the wind speed is 1 meter per second (m / s), can be obtained from the database as a first set. Temperature, air density, and light intensity, whose values vary with power when the wind speed is 1.5 m / s, can be obtained from the database as a second set. Variables in the intersection of the first and second sets (i.e., temperature and air density) can then be classified as pending new influencing factor variables. Furthermore, if no uncategorized variables exist, steps S611 and S612 can be performed only on the low-impact factor variables. Optionally, in one example, one of step S611 and step S612 may further include classifying a non-categorical variable whose value does not change with power as a low-influence factor variable.

[0066] Return to reference Figure 2 In step S62, each new influencing factor variable to be determined is verified. In one example, each new influencing factor variable to be determined can be verified separately. Figure 4 Step S62 is described in detail.

[0067] Figure 4 1 is a flow chart illustrating a method for verifying a pending new influencing factor according to an embodiment of the present invention. Specifically, Figure 4 This is a detailed description of the above step S62 of verifying each of the new influencing factor variables to be determined.

[0068] Reference Figure 4In step S621, for any undetermined new influencing factor variable, a first fitting function for calculating power is obtained by performing a multivariate fitting on the power, the values of the basic influencing factor variables, the values of the new influencing factor variables, and the values of the any undetermined new influencing factor variables obtained from the database. In one example, a multivariate fitting can be performed on the historical power, historical values of the basic influencing factor variables, historical values of the new influencing factor variables, and historical values of the any undetermined new influencing factor variables stored in the database over a predetermined time period to obtain a first fitting function representing the mapping relationship between the basic influencing factor variables, the new influencing factor variables, the any undetermined new influencing factor variables, and power. For example, when verifying light intensity as a undetermined new influencing factor variable, a multivariate fitting can be performed on the historical power, historical values of wind speed and wind direction as basic influencing factor variables, historical values of temperature as a new influencing factor variable, and historical values of light intensity as the any undetermined new influencing factor variable stored in the database over a year to obtain a first fitting function representing the mapping relationship between wind speed, wind direction, temperature, light intensity, and power.

[0069] In step S622, multiple calculated powers are obtained by substituting the values of the basic influencing factor variables, the values of the new influencing factor variables, and the value of any one of the pending new influencing factor variables, which are obtained in real time multiple times when the wind turbine generator set is in a normal state, into the first fitting function. In one example, the values of the basic influencing factor variables, the values of the new influencing factor variables, and the value of any one of the pending new influencing factor variables, which are obtained in real time a predetermined number of times, can be substituted into the first fitting function fitted in step S621 to obtain the predetermined number of calculated powers. Here, the predetermined number of times can be a predefined value. For example, the values of wind speed and wind direction as basic influencing factor variables, the values of the new influencing factor variables, and the value of light intensity as any one of the pending new influencing factor variables, which are obtained in real time 100,000 times, can be substituted into the first fitting function fitted in step S621 to obtain 100,000 calculated powers.

[0070] In step S623, if the number of times the difference between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the first threshold is greater than the second threshold, the verification is passed, otherwise the verification is not passed. If the number of times the difference between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the first threshold is greater than the second threshold, it indicates that any one of the pending new influencing factor variables affects the power of the wind turbine generator set to a certain extent. Therefore, in one example, if the probability that the error between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the first threshold is greater than the second threshold, the verification is passed, otherwise the verification is not passed. Here, the first threshold and the second threshold can be percentages. For example, if the ratio of the number of times the error between one hundred thousand calculated powers and the corresponding one hundred thousand real-time acquired powers is less than 5% to the total number of times (i.e., one hundred thousand times) is greater than 95%, the verification is passed, otherwise the verification is not passed.

[0071] Return to reference Figure 2 In step S63, the new influencing factor variables that have passed the verification are classified as new influencing factor variables, and the new influencing factor variables that have not passed the verification are classified as low-influencing factor variables, thereby updating the classification of the influencing factor data. In one example, the label of the data of any one of the new influencing factor variables that have passed the verification in step S623 stored in the database can be classified (i.e., reclassified) as a new influencing factor variable, and any one of the new influencing factor variables that have not passed the verification in step S623 stored in the database can be classified (i.e., reclassified) as a low-influencing factor variable. For example, when the illumination intensity as any one of the new influencing factor variables that have passed the verification in step S623 passes the verification, the illumination intensity can be classified (i.e., reclassified) from the new influencing factor variable to the new influencing factor variable; when the illumination intensity as any one of the new influencing factor variables that have not passed the verification in step S623 fails to pass the verification, the illumination intensity can be classified (i.e., reclassified) from the new influencing factor variable to the low-influencing factor variable.

[0072] Furthermore, after completing verification of the illumination intensity as any one of the pending new influencing factor variables, verification may be continued for other of the pending new influencing factor variables in sequence until all of the pending new influencing factor variables are verified. For example, verification may be continued for air density as any one of the pending new influencing factor variables.

[0073] Figure 5 1 is a flow chart illustrating a method for determining low-impact factor variables according to an embodiment of the present invention. Specifically, Figure 5 This is a detailed description of the above step S7 of determining low-impact factor variables.

[0074] Reference Figure 5 In step S71, in response to the wind turbine generator set being in a normal state, a new influencing factor variable whose value does not change with power changes is classified as a pending low-influence factor variable. The pending low-influence factor variable may represent a variable that may not have a significant impact on the power of the wind turbine generator set and / or whose impact on the power of the wind turbine generator set is negligible. In other words, the pending low-influence factor variable may represent a variable that may be ultimately classified as a low-factor influencing variable. In one example, a new influencing factor variable whose value does not change with power changes within a predetermined time period stored in the database may be classified as a pending low-influence factor variable. In the following, reference will be made to Figure 6 Step S71 is described in detail.

[0075] Figure 6 1 is a flow chart illustrating a method for determining a variable of a low-impact factor to be determined according to an embodiment of the present invention. Specifically, Figure 6 This is a detailed description of the step S71 of classifying the new influencing factor variable whose value remains unchanged with power change as a pending low-influencing factor variable in response to the wind turbine generator being in a normal state.

[0076] Reference Figure 6 In step S711, multiple sets of new influencing factor variables whose values remain unchanged as power changes when the values of the basic influencing factor variables remain unchanged are obtained from the database; in step S712, the variables in the intersection of the multiple sets are classified as pending low-influencing factor variables. For example, when the basic influencing factor variable is wind speed, humidity and light intensity, whose values remain unchanged as power changes when the wind speed value is 1 meter per second (m / s), can be obtained from the database as a first set as new influencing factor variables, and humidity, light intensity, and temperature, whose values remain unchanged as power changes when the wind speed value is 1.5m / s, can be obtained from the database as a second set; then, the variables in the intersection of the first set and the second set (i.e., humidity and light intensity) can be classified as pending low-influencing factor variables.

[0077] Return to reference Figure 5 In step S72, each low-impact factor variable to be determined is verified. Figure 7 Step S72 is described in detail.

[0078] Figure 7 1 is a flow chart illustrating a method for verifying a variable of a pending low-impact factor according to an embodiment of the present invention. Specifically, Figure 7 This is a detailed description of the above step S72 of verifying each of the pending low-impact factor variables.

[0079] Reference Figure 7In step S721, a second fitting function for calculating power is obtained by performing a multivariate fit on the power, the values of the basic influencing factor variables, and the values of the new influencing factor variables obtained from the database. In one example, a multivariate fit can be performed on the historical power, the historical values of the basic influencing factor variables, and the historical values of the new influencing factor variables stored in the database over a predetermined time period to obtain a second fitting function representing the mapping relationship between the basic influencing factor variables, the new influencing factor variables, and power. It should be understood that since the new influencing factor variables, whose values remain unchanged with power, have been classified as pending low-impact factor variables in sub-step S712 of step S71, the new influencing factor variables in sub-step S721 no longer include all variables corresponding to the pending low-impact factor variables. For example, a multivariate fit can be performed on the historical power over a year, the historical values of wind speed and wind direction as basic influencing factor variables, and the historical values of air density as a new influencing factor variable stored in the database to obtain a second fitting function representing the mapping relationship between wind speed, wind direction, air density, and power.

[0080] In step S722, multiple calculated powers are obtained by substituting the values of the basic influencing factor variables and the values of the new influencing factor variables, acquired multiple times in real time while the wind turbine is in a normal state, into the second fitting function. In one example, the values of the basic influencing factor variables and the values of the new influencing factor variables acquired a predetermined number of times in real time can be substituted into the second fitting function fitted in step S721 to obtain the predetermined number of calculated powers. For example, the values of wind speed and wind direction as basic influencing factor variables and the value of air density as a new influencing factor variable acquired 100,000 times in real time can be substituted into the second fitting function fitted in step S721 to obtain 100,000 calculated powers.

[0081] In step S723, if the number of times the difference between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the third threshold is greater than the fourth threshold, then the verification is passed, otherwise the verification is not passed. If the number of times the difference between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the third threshold is greater than the fourth threshold, it indicates that the variable classified as a pending low-impact factor has no effect on the power of the wind turbine generator set. Therefore, in one example, if the probability that the error between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the third threshold is greater than the fourth threshold, then the verification is passed, otherwise the verification is not passed. Here, the third threshold may be the same as or different from the first threshold, and the fourth threshold may be the same as or different from the second threshold. For example, if the ratio of the number of times the error between one hundred thousand calculated powers and the corresponding one hundred thousand real-time acquired powers is less than 5% to the total number of times (i.e., one hundred thousand times) is greater than 95%, then the verification is passed, otherwise the verification is not passed.

[0082] Return to reference Figure 5 , in step S73, the low-impact factor variables to be determined that have passed verification are classified as low-impact factor variables, and the low-impact factor variables to be determined that have not passed verification are classified as new influencing factor variables, thereby realizing the updating of the classification of influencing factor data. In one example, when verified, all low-impact factor variables to be determined can be classified (i.e., reclassified) as low-impact factor variables. When not verified, any one of the low-impact factor variables to be determined can be added to the fitting process of the second fitting function in step S721, and any one of the low-impact factor variables to be determined that are obtained in real time can be substituted into the second fitting function for verification similar to step S723 in step S722; when verified, any one of the low-impact factor variables to be determined can be classified (i.e., reclassified) as low-impact factor variables, and when not verified, any one of the low-impact factor variables to be determined can be classified (i.e., reclassified or restored) as new influencing factor variables. For example, when the illumination intensity and temperature classified as pending low-impact factor variables in step S723 pass the verification, the illumination intensity and temperature can be classified as low-impact factor variables. When the illumination intensity and temperature classified as pending low-impact factor variables in step S723 fail the verification, a second fitting function for calculating power can be obtained by performing multivariate fitting on the power obtained from the database, the value of the basic influencing factor variable, the value of the new influencing factor variable, and the value of the illumination intensity; multiple calculated powers can be obtained by substituting the values of the basic influencing factor variable, the value of the new influencing factor variable, and the value of the illumination intensity obtained in real time multiple times when the wind turbine generator set is in a normal state into the second fitting function; if the difference between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the third threshold and the number of times is greater than the fourth threshold, the verification is passed; otherwise, the verification is failed; when the verification is passed, the illumination intensity is classified (i.e., reclassified) as a low-impact factor variable; when the verification is failed, the illumination intensity is classified (i.e., reclassified or restored) as a new influencing factor variable; and then, a similar separate verification process is performed on the temperature.

[0083] Figure 8 1 is a flow chart illustrating a method for calculating power loss according to an embodiment of the present invention. Specifically, Figure 8 This is a detailed description of the above-mentioned step S3 of calculating the power loss of the wind turbine generator set based on the power and influencing factor data in response to the wind turbine generator set being in an abnormal state.

[0084] Reference Figure 8In step S31, a third fitting function for calculating power is obtained by performing a multivariate fitting on the power, the values of the basic influencing factor variables, and the values of the new influencing factor variables obtained from the database. In one example, a multivariate fitting can be performed on the historical power, the historical values of the basic influencing factor variables, and the historical values of the new influencing factor variables stored in the database over a predetermined time period to obtain a third fitting function representing the mapping relationship between the basic influencing factor variables, the new influencing factor variables, and power. For example, a multivariate fitting can be performed on the historical power over a year, the historical values of wind speed and wind direction as the basic influencing factor variables, and the historical values of air density as the new influencing factor variable stored in the database to obtain a third fitting function representing the mapping relationship between wind speed, wind direction, air density, and power.

[0085] In step S32, the calculated power is obtained by substituting the values of the basic influencing factor variables and the values of the new influencing factor variables, acquired in real time when the wind turbine is in an abnormal state, into the third fitting function. In one example, the values of the basic influencing factor variables and the values of the new influencing factor variables acquired in real time a predetermined number of times can be substituted into the third fitting function fitted in step S31 to obtain the predetermined number of calculated powers. For example, the values of wind speed and wind direction as basic influencing factor variables and the value of air density as a new influencing factor variable acquired in real time 100,000 times can be substituted into the third fitting function fitted in step S31 to obtain 100,000 calculated powers.

[0086] In step S33, the power loss of the wind turbine is calculated based on the calculated power, the power obtained in real time when the wind turbine is in the abnormal state, and the time information when the wind turbine is in the abnormal state. In one example, the power loss of the wind turbine can be calculated by integrating the difference between the calculated power and the power obtained in real time when the wind turbine is in the abnormal state according to the time information.

[0087] In the above reference Figures 2 to 8 In the steps of obtaining the first fitting function, the second fitting function, and the third fitting function, a multivariate fitting can be performed on the historical power within a predetermined time period stored in the database, the historical values of the basic influencing factor variables, the historical values of the new influencing factor variables, and the historical values of other data to obtain a third fitting function representing the mapping relationship between the basic influencing factor variables, the new influencing factor variables, the other data, and the power. For example, a multivariate fitting can be performed on the historical power within one year stored in the database, the historical values of wind speed and wind direction as the basic influencing factor variables, the historical values of air density as the new influencing factor variable, and the historical values of voltage, capacitance, and resistance as other parameters to obtain a third fitting function representing the mapping relationship between wind speed, wind direction, temperature, air density, voltage, capacitance, resistance, and power.

[0088] Figure 9 is a block diagram illustrating an apparatus for calculating a lost amount of power of a wind turbine generator system according to an embodiment of the present inventive concept.

[0089] Reference Figure 9 The device 100 for calculating the power loss of a wind turbine generator set may include an acquirer 110 , a determiner 120 and a calculator 130 .

[0090] The acquirer 110 may be configured to acquire the power, influencing factor data, and device status data of the wind turbine in real time. In one example, the power, influencing factor data, and device status data of the wind turbine may be acquired at predetermined time intervals. For example, the predetermined time interval may be 6 seconds.

[0091] The power of a wind turbine generator set may represent the actual output power or actual power generation of the wind turbine generator set. The influencing factor data may include data on multiple variables. In one example, the multiple variables may include at least wind speed, wind direction, temperature, air density, humidity, and light intensity. However, this example is not limited to this, and the multiple variables may also include other influencing factor variables in addition to wind speed, wind direction, temperature, air density, humidity, and light intensity.

[0092] The influencing factor data may include data on multiple variables. Specifically, the influencing factor data may include data on multiple variables that can affect the power of the wind turbine generator set. In one example, the influencing factor data may include or primarily include data on environmental variables. For example, the multiple variables may include at least wind speed, wind direction, temperature, air density, humidity, light intensity, etc. However, the example is not limited to this, and the multiple variables may also include other variables (or other environmental variables) in addition to wind speed, wind direction, temperature, air density, humidity, light intensity, etc.

[0093] Multiple variables may be initially or pre-classified into basic influencing factor variables, new influencing factor variables, and low-influence factor variables. Basic influencing factor variables may represent variables that have a significant and / or inevitable impact on the power of a wind turbine generator set. In one example, basic influencing factor variables may include wind speed, wind direction, etc. However, the example is not limited to this, and basic influencing factor variables may also include other variables (or other environmental variables) in addition to wind speed, wind direction, etc. New influencing factor variables may represent variables that have a certain degree of impact on the power of a wind turbine generator set. In one example, new influencing factor variables may include temperature, air density, etc. However, the example is not limited to this, and new influencing factor variables may also include other variables (or other environmental variables) in addition to temperature, air density, etc. Low-influence factor variables may represent variables that have an insignificant and / or negligible impact on the power of a wind turbine generator set. In one example, low-influence factor variables may include light intensity, etc. However, the example is not limited to this, and low-influence factor variables may also include other variables (or other environmental variables) in addition to light intensity, etc. In one example, the basic influencing factor variables, new influencing factor variables, and low-influencing factor variables may be classified according to engineering experience, or may be determined according to results of previous classification of influencing factor data.

[0094] In addition, multiple variables may also include uncategorized variables. Uncategorized variables may represent variables whose degree of influence on the power of a wind turbine is still unclear. For example, uncategorized variables may be variables that have not been classified after the influencing factors were previously classified or newly introduced variables. In one example, a variable whose degree of influence on the power of a wind turbine is still unclear (i.e., an uncategorized variable) may be initially classified as any one of a new influencing factor variable and a low-influence factor variable, and then the classification of the new influencing factor variable and the low-influence factor variable may be updated by a subsequent process of classifying the influencing factor data. That is, multiple variables may or may not include uncategorized variables according to actual needs, and the variable whose degree of influence on the power of a wind turbine is still unclear may exist in multiple variables in other classification forms besides the uncategorized variables.

[0095] Furthermore, the classification of basic influencing factor variables, new influencing factor variables, low-influence factor variables, and uncategorized variables can be achieved by adding corresponding data labels to the data of multiple variables in the database. Therefore, reclassifying influencing factor data can be achieved by modifying the data labels of multiple variables. In one example, low-influence factor variables and new influencing factor variables can be converted to each other under specific conditions. For example, specific conditions may include specific seasons, specific uncategorized variables, etc.

[0096] In addition, in one example, the acquirer 110 may also be configured to acquire other data of the wind turbine in real time. The other data may be data of variables that have no impact on the function of the wind turbine. For example, the other data may include data of the voltage, current, capacitance, resistance, etc. of the wind turbine.

[0097] Device status data can indicate whether the device is operating normally. In one example, the device status data can be used to categorize the wind turbine's status into nine states based on several factors, including the wind turbine's communication status, fault code, warning mode word, operating status word, shutdown mode word, and power limit mode word. Table 1 above details the names and meanings of these nine states.

[0098] The determiner 120 may be configured to determine whether the wind turbine generator set is in a normal state or an abnormal state based on the device status data. For example, the determiner 120 may be configured to determine that the wind turbine generator set is in a normal state when the device status data indicates a normal power generation state; and to determine that the wind turbine generator set is in an abnormal state when the device status data indicates at least one of a power curtailment state, a wind turbine maintenance state, a technical standby state, a remote shutdown state, a grid fault state, a fault shutdown state, a local shutdown state, and other power curtailment states.

[0099] The calculator 130 may be configured to calculate the power loss of the wind turbine generator set based on the power and influencing factor data in response to the wind turbine generator set being in an abnormal state. Specifically, in one example, the calculator 130 may be configured to perform the following processes 1, 2, and 3.

[0100] In process 1, the calculator 130 may be configured to perform a multivariate fit on the power, the values of the basic influencing factor variables, and the values of the new influencing factor variables obtained from the database to obtain a third fitting function for calculating the power. In one example, the calculator 130 may be configured to perform a multivariate fit on the historical power, the historical values of the basic influencing factor variables, and the historical values of the new influencing factor variables stored in the database within a predetermined time period to obtain a third fitting function representing the mapping relationship between the basic influencing factor variables, the new influencing factor variables, and the power. For example, the calculator 130 may be configured to perform a multivariate fit on the historical power within a year, the historical values of wind speed and wind direction as the basic influencing factor variables, and the historical values of air density as the new influencing factor variable stored in the database to obtain a third fitting function representing the mapping relationship between wind speed, wind direction, air density, and power.

[0101] In process 2, calculator 130 may be configured to obtain calculated power by substituting the values of the basic influencing factor variables and the values of the new influencing factor variables, acquired in real time when the wind turbine is in an abnormal state, into a third fitting function. In one example, calculator 130 may be configured to substitute the values of the basic influencing factor variables and the values of the new influencing factor variables acquired in real time a predetermined number of times into the third fitting function fitted in process 1, to obtain the predetermined number of calculated powers. For example, calculator 130 may be configured to substitute the values of wind speed and wind direction, which are basic influencing factor variables, and the value of air density, which is a new influencing factor variable, acquired in real time 100,000 times into the third fitting function fitted in process 1, to obtain 100,000 calculated powers.

[0102] In process 3, the calculator 130 may be configured to calculate the lost power of the wind turbine generator set based on the calculated power, the power obtained in real time when the wind turbine generator set is in the abnormal state, and the time information when the wind turbine generator set is in the abnormal state. In one example, the calculator 130 may be configured to calculate the lost power of the wind turbine generator set by integrating the difference between the calculated power and the power obtained in real time when the wind turbine generator set is in the abnormal state according to the time information.

[0103] In addition, the device 100 may further include a classifier 130. Classifier 130 may be configured to classify the influencing factor data based on the degree of impact of the influencing factor data on power, in response to the wind turbine being in a normal state. It should be understood that classifier 130 is an optional configuration of the device 100. That is, when the wind turbine is in a normal state, classifier 130 may be selectively enabled to perform the above processing. For example, when the wind turbine is in a normal state, classifier 130 may be enabled to perform the above processing in response to a user instruction, the introduction of a new variable, or the satisfaction of a certain time interval condition. Furthermore, after classifier 130 is enabled, if the wind turbine is in an abnormal state again, calculator 130 may be further configured to calculate the power loss of the wind turbine based on the power and the classified influencing factor data, in response to the wind turbine being in an abnormal state. Classifier 130 may be configured to perform the following processing of determining new influencing factor variables and determining low-influence factor variables.

[0104] In one example, the process of determining new influencing factor variables may include the following process 4, process 5, and process 6.

[0105] In process 4, classifier 130 may be configured to, in response to the wind turbine being in a normal state, classify low-impact factor variables and uncategorized variables whose values change with power as pending new influencing factor variables. Uncategorized new influencing factor variables may represent factors that may have a certain impact on the power of the wind turbine. In other words, uncategorized new influencing factor variables may represent variables that may ultimately be classified as new factor influencing variables. In one example, classifier 130 may be configured to classify low-impact factor variables and uncategorized variables whose values change with power within a predetermined time period stored in the database as pending new influencing factor variables. Here, the predetermined time period may be one or more days, one or more weeks, one or more months, one or more quarters, or one or more years. Furthermore, classifier 130 may be configured to perform process 4 only on low-impact factor variables if no uncategorized variables exist. Optionally, in one example, process 4 may also include classifying uncategorized variables whose values do not change with power as low-impact factor variables. In one example, process 4 may include the following processes 41 and 42.

[0106] In process 41, the classifier 130 may be configured to obtain from the database multiple sets of low-impact factor variables and uncategorized variables whose values change with power when the value of the basic influencing factor variable remains unchanged; in process 42, the classifier 130 may be configured to classify the variables in the intersection of the multiple sets as pending new influencing factor variables. For example, the classifier 130 may be configured to: when the basic influencing factor variable is wind speed, the temperature and air density whose values change with power when the wind speed value is 1 meter per second (m / s) may be obtained from the database as a first set as low-impact factor variables and uncategorized variables; the temperature, air density, and light intensity whose values change with power when the wind speed value is 1.5 m / s may be obtained from the database as a second set; then, the variables in the intersection of the first set and the second set (i.e., temperature and air density) may be classified as pending new influencing factor variables. In addition, the classifier 130 may be configured to perform processes 41 and 42 only on low-influence factor variables when no uncategorized variables exist. Optionally, in one example, one of processes 41 and 42 may further include classifying uncategorized variables whose values do not change with power as low-influence factor variables.

[0107] In process 5, the classifier 130 may be configured to verify each pending new influencing factor variable. In one example, the classifier 130 may be configured to verify each pending new influencing factor variable separately. In one example, process 5 may include the following processes 51, 52, and 53.

[0108] In process 51, the classifier 130 may be configured to perform a multivariate fit on the power, the values of the basic influencing factor variables, the values of the new influencing factor variables, and the values of the new influencing factor variables obtained from the database for any pending new influencing factor variable, thereby obtaining a first fitting function for calculating power. In one example, the classifier 130 may be configured to perform a multivariate fit on the historical power, historical values of the basic influencing factor variables, historical values of the new influencing factor variables, and historical values of the new influencing factor variables stored in the database over a predetermined time period to obtain a first fitting function representing a mapping relationship between the basic influencing factor variables, the new influencing factor variables, the new influencing factor variables, and power. For example, the classifier 130 may be configured to perform a multivariate fit on the historical power, historical values of wind speed and wind direction as basic influencing factor variables, historical values of temperature as new influencing factor variables, and historical values of light intensity as the new influencing factor variables stored in the database over a period of one year, while verifying light intensity as the new influencing factor variable, to obtain a first fitting function representing a mapping relationship between wind speed, wind direction, temperature, light intensity, and power.

[0109] In process 52, the classifier 130 may be configured to obtain multiple calculated powers by substituting the values of the basic influencing factor variables, the values of the new influencing factor variables, and the value of any one of the pending new influencing factor variables, acquired multiple times in real time when the wind turbine generator set is in a normal state, into the first fitting function. In one example, the classifier 130 may be configured to substitute the values of the basic influencing factor variables, the values of the new influencing factor variables, and the value of any one of the pending new influencing factor variables acquired a predetermined number of times in real time into the first fitting function fitted in process 51, to obtain the predetermined number of calculated powers. Here, the predetermined number of times may be a predefined value. For example, the classifier 130 may be configured to substitute the values of wind speed and wind direction as basic influencing factor variables, the values of the new influencing factor variables, and the value of light intensity as any one of the pending new influencing factor variables acquired 100,000 times in real time into the first fitting function fitted in process 51, to obtain 100,000 calculated powers.

[0110] In process 53, the classifier 130 may be configured to pass the verification if the number of times the difference between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the first threshold is greater than the second threshold, otherwise the verification is failed. If the number of times the difference between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the first threshold is greater than the second threshold, it indicates that any of the pending new influencing factor variables affects the power of the wind turbine generator set to a certain extent. Therefore, in one example, the classifier 130 may be configured to pass the verification if the probability that the error between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the first threshold is greater than the second threshold, otherwise the verification is failed. Here, the first threshold and the second threshold can be percentages. For example, the classifier 130 may be configured to pass the verification if the ratio of the number of times the error between 100,000 calculated powers and the corresponding 100,000 real-time acquired powers is less than 5% to the total number of times (i.e., 100,000 times) is greater than 95%, otherwise the verification is failed.

[0111] In process 6, the classifier 130 may be configured to classify the pending new influencing factor variables that have passed verification as new influencing factor variables, and classify the pending new influencing factor variables that have not passed verification as low-influence factor variables, thereby updating the classification of the influencing factor data. In one example, the classifier 130 may be configured to classify (i.e., reclassify) the label of the data of any one of the pending new influencing factor variables that have passed verification in process 53 stored in the database as a new influencing factor variable, and classify (i.e., reclassify) the label of the data of any one of the pending new influencing factor variables that have passed verification in process 53 stored in the database as a low-influence factor variable. For example, the classifier 130 can be configured to: when the illumination intensity as any one of the pending new influencing factor variables passes verification in processing 53, classify (i.e., reclassify) the illumination intensity from the pending new influencing factor variable to the new influencing factor variable; when the illumination intensity as any one of the pending new influencing factor variables fails to pass verification in processing 53, classify (i.e., reclassify) the illumination intensity from the pending new influencing factor variable to a low influencing factor variable.

[0112] Furthermore, the first classifier 150 may be configured to, after completing verification of the illumination intensity as any one of the pending new influencing factor variables, continue to sequentially verify the other pending new influencing factor variables until all of the pending new influencing factor variables are verified. For example, the first classifier 150 may be configured to continue to verify the air density as any one of the pending new influencing factor variables.

[0113] In one example, the classifier 130 may be further configured to perform the following process 7, process 8, and process 9.

[0114] In process 7, the classifier 130 may be configured to, in response to the wind turbine being in a normal state, classify a new influencing factor variable whose value remains unchanged as the power changes as a pending low-influence factor variable. The pending low-influence factor variable may represent a variable that may not have a significant impact on the power of the wind turbine and / or whose impact on the power of the wind turbine is negligible. In other words, the pending low-influence factor variable may represent a variable that may be ultimately classified as a low-factor influencing variable. In one example, the classifier 130 may be configured to classify a new influencing factor variable whose value remains unchanged as the power changes within a predetermined time period stored in the database as a pending low-influence factor variable. In one example, process 7 may include the following processes 71 and 72.

[0115] In process 71, the classifier 130 may be configured to obtain from the database multiple sets of new influencing factor variables whose values remain unchanged as power changes when the values of the basic influencing factor variables remain unchanged; in process 72, the classifier 130 may be configured to classify the variables in the intersection of the multiple sets as pending low-influencing factor variables. For example, the classifier 130 may be configured to: when the basic influencing factor variable is wind speed, obtain from the database humidity and light intensity as the first set as new influencing factor variables whose values remain unchanged as power changes when the wind speed value is 1 meter per second (m / s), and obtain from the database humidity, light intensity, and temperature as the second set as new influencing factor variables whose values remain unchanged as power changes when the wind speed value is 1.5m / s; then, classify the variables in the intersection of the first set and the second set (i.e., humidity and light intensity) as pending low-influencing factor variables.

[0116] In process 8, the classifier 130 may be configured to verify each of the undetermined low-impact factor variables. In one example, process 8 may include the following processes 81, 82, and 83.

[0117] In process 81, the classifier 130 can be configured to perform a multivariate fit on the power, the values of the basic influencing factor variables, and the values of the new influencing factor variables obtained from the database to obtain a second fitting function for calculating power. In one example, the classifier 130 can be configured to perform a multivariate fit on the historical power, the historical values of the basic influencing factor variables, and the historical values of the new influencing factor variables stored in the database over a predetermined time period to obtain a second fitting function representing the mapping relationship between the basic influencing factor variables, the new influencing factor variables, and power. It should be understood that since the new influencing factor variables, whose values remain unchanged with power changes, have been classified as pending low-influence factor variables in subprocess 72 of process 7, the new influencing factor variables in subprocess 81 no longer include all variables corresponding to the pending low-influence factor variables. For example, the classifier 130 can be configured to perform a multivariate fit on the historical power over a year, the historical values of wind speed and wind direction as the basic influencing factor variables, and the historical values of air density as the new influencing factor variable stored in the database to obtain a second fitting function representing the mapping relationship between wind speed, wind direction, air density, and power.

[0118] In process 82, the classifier 130 may be configured to obtain multiple calculated powers by substituting the values of the basic influencing factor variables and the values of the new influencing factor variables, which are obtained multiple times in real time when the wind turbine generator system is in a normal state, into the second fitting function. In one example, the classifier 130 may be configured to substitute the values of the basic influencing factor variables and the values of the new influencing factor variables, which are obtained a predetermined number of times in real time, into the second fitting function fitted in process 81, to obtain the predetermined number of calculated powers. For example, the classifier 130 may be configured to substitute the values of wind speed and wind direction as basic influencing factor variables, which are obtained 100,000 times in real time, and the value of air density as a new influencing factor variable, into the second fitting function fitted in step S721, to obtain 100,000 calculated powers.

[0119] In process 83, the classifier 130 may be configured to pass the verification if the number of times the difference between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the third threshold is greater than the fourth threshold, otherwise the verification is failed. If the number of times the difference between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the third threshold is greater than the fourth threshold, it indicates that the variable classified as a pending low-impact factor has no effect on the power of the wind turbine generator set. Therefore, in one example, the classifier 130 may be configured to pass the verification if the probability that the error between the multiple calculated powers and the corresponding multiple real-time acquired powers is less than the third threshold is greater than the fourth threshold, otherwise the verification is failed. Here, the third threshold may be the same as or different from the first threshold, and the fourth threshold may be the same as or different from the second threshold. For example, the classifier 130 may be configured to pass the verification if the ratio of the number of times the error between 100,000 calculated powers and the corresponding 100,000 real-time acquired powers is less than 5% to the total number of times (i.e., 100,000 times) is greater than 95%, otherwise the verification is failed.

[0120] In processing 9, classifier 130 can be configured to classify the low-influence factor variables to be determined by verification as low-influence factor variables, and classify the low-influence factor variables to be determined that have not been verified as new influencing factor variables, thereby realizing the updating of the classification of influencing factor data. In one example, classifier 130 can be configured to classify (i.e., reclassify) all low-influence factor variables to be determined as low-influence factor variables when verified. In addition, classifier 130 can be configured to, when not verified, in processing 81, any one low-influence factor variable to be determined is added to the fitting process of the second fitting function, and in processing 82, any one low-influence factor variable to be determined obtained in real time is substituted into the second fitting function for verification similar to processing 83; when verified, any one low-influence factor variable to be determined is classified (i.e., reclassified) as low-influence factor variable, and when not verified, any one low-influence factor variable to be determined is classified (i.e., reclassified or restored) as new influencing factor variable. For example, the classifier 130 may be configured to classify the light intensity and temperature as low-impact factor variables when the light intensity and temperature classified as pending low-impact factor variables in the process 83 pass verification. In addition, the classifier 130 can be configured to obtain a second fitting function for calculating power by performing multivariate fitting on the power, the value of the basic influencing factor variable, the value of the new influencing factor variable and the value of the light intensity obtained from the database when the light intensity and temperature classified as pending low-impact factor variables in processing 83 fail to pass verification; obtain multiple calculated powers by substituting the values of the basic influencing factor variable, the value of the new influencing factor variable and the value of the light intensity obtained in real time multiple times when the wind turbine generator set is in a normal state into the second fitting function; if the difference between the multiple calculated powers and the corresponding multiple real-time obtained powers is less than the third threshold and the number of times is greater than the fourth threshold, then the verification is passed, otherwise the verification is failed; when the verification is passed, the light intensity is classified (i.e., reclassified) as a low-impact factor variable, and when the verification is failed, the light intensity is classified (i.e., reclassified or restored) as a new influencing factor variable; and then, a similar separate verification process is performed on the temperature.

[0121] In addition, the computing device 100 may further include a memory 150. In one example, the memory 150 may be configured to: in response to the wind turbine being in a normal state, store the power obtained in real time and the result of binning the influencing factor data obtained in real time as historical influencing factor data and historical power data in the database. For example, the memory 150 may be configured to store the first power obtained at the first time and the first result of binning the first influencing factor data obtained in real time as the first data in the database, and store the second power obtained at the second time after the first time and the second result of binning the second influencing factor data obtained in real time as the second data in the database.

[0122] In another example, when other data of the wind turbine can be obtained in real time in addition to the power, influencing factor data and equipment status data of the wind turbine, the memory 150 can also be configured to: in response to the wind turbine being in a normal state, store the power obtained in real time, the results of binning and averaging the influencing factor data obtained in real time, and the results of averaging the other data obtained in real time as historical influencing factor data, historical power data and historical other data in the database.

[0123] In addition, the memory 150 and the classifier 130 may perform their respective operations in parallel, in any order, or at different frequencies, and this application does not impose any specific limitations on this.

[0124] Furthermore, it should be understood that the examples of basic influencing factor variables, new influencing factor variables, low influencing factor variables, pending new influencing factor variables, and pending low influencing factor variables listed in the above examples are merely examples and are not intended to be limiting. Specifically, which variables belong to basic influencing factor variables, new influencing factor variables, low influencing factor variables, pending new influencing factor variables, or pending low influencing factor variables will be determined by specific calculations or by pre-setting.

[0125] In addition, it should be understood that the various units in the device according to the exemplary embodiments of the present invention can be implemented as hardware components and / or software components. Those skilled in the art can, for example, use a field programmable gate array (FPGA) or an application specific integrated circuit (ASIC) to implement the various units according to the processing performed by the defined various units.

[0126] Furthermore, the method for calculating power loss of a wind turbine generator set according to an embodiment of the present invention can be implemented as program instructions in a computer-readable storage medium. Those skilled in the art can implement the program instructions based on the description of the above method. When the program instructions are executed on a computer, the above method of the present invention is implemented.

[0127] The present invention classifies influencing factor data based on their degree of influence on power, and calculates the wind turbine's power loss based on the power and the classified influencing factor data. This results in a more accurate calculation of power loss. Furthermore, because the calculation of wind turbine power loss does not require the use of benchmark equipment, even if individual equipment fails, the calculation of power loss is not affected, resulting in greater adaptability.

[0128] While the invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims

1. A method for calculating the power loss of a wind turbine generator set, wherein: include: Acquiring power, influencing factor data, and device status data of a wind turbine generator set in real time, wherein the influencing factor data includes data of a plurality of variables, wherein the plurality of variables are initially classified into basic influencing factor variables, new influencing factor variables, and low-influencing factor variables; Based on the equipment status data, determine whether the wind turbine generator set is in a normal state or an abnormal state; In response to the wind turbine generator set being in an abnormal state, the power loss of the wind turbine generator set is calculated based on the power and influencing factor data, The calculation method further includes: in response to the wind turbine generator set being in a normal state, classifying the influencing factor data based on the degree of influence of the influencing factor data on power, In response to the wind turbine generator set being in an abnormal state, the step of calculating the power loss of the wind turbine generator set based on the power and influencing factor data includes: In response to the wind turbine generator set being in an abnormal state, the power loss of the wind turbine generator set is calculated based on the power and the classified influencing factor data. wherein the uncategorized variables among the plurality of variables are initially classified into the new influencing factor variables or the low-influence factor variables, and the classification of the new influencing factor variables and the low-influence factor variables is updated by the step of classifying the influencing factor data based on the degree of influence of the influencing factor data on power; In response to the wind turbine generator set being in an abnormal state, the step of calculating the power loss of the wind turbine generator set based on the power and the classified influencing factor data includes: Obtaining a third fitting function for calculating power by performing multivariate fitting on the power obtained from the database, the value of the basic influencing factor variable, and the value of the new influencing factor variable; The calculated power is obtained by substituting the values of the basic influencing factor variables and the values of the new influencing factor variables obtained in real time when the wind turbine generator set is in an abnormal state into the third fitting function; Based on the calculated power, the power obtained in real time when the wind turbine is in an abnormal state, and the time information of the wind turbine in the abnormal state, the power loss of the wind turbine is calculated. In response to the wind turbine generator set being in a normal state, the step of classifying the influencing factor data based on the degree of influence of the influencing factor data on power includes: In response to the wind turbine being in a normal state, low-influence factor variables and uncategorized variables whose values change with power changes are classified as new influencing factor variables to be determined; Verify each new influencing factor variable to be determined; The pending new influencing factor variables that have passed verification are classified as new influencing factor variables, and the pending new influencing factor variables that have not passed verification are classified as low-influencing factor variables, thereby achieving an update of the classification of the influencing factor data.

2. The calculation method according to claim 1, wherein: The multiple variables include at least wind speed, wind direction, temperature, air density, humidity and light intensity.

3. The calculation method according to claim 2, wherein: In response to the wind turbine being in a normal state, the steps of classifying low-influence factor variables and uncategorized variables whose values change with power changes as new influencing factor variables to be determined include: Acquire multiple sets of low-influence factor variables and unclassified variables whose values change with power changes when the value of the basic influencing factor variable remains unchanged from the database; Variables in the intersection of the multiple sets are classified as pending new influencing factor variables.

4. The calculation method according to claim 2, wherein: The steps for verifying each new influencing factor variable to be determined include: For any undetermined new influencing factor variable, a first fitting function for calculating power is obtained by performing multivariate fitting on the power obtained from the database, the value of the basic influencing factor variable, the value of the new influencing factor variable, and the value of the any undetermined new influencing factor variable; Obtaining multiple calculated powers by substituting the values of the basic influencing factor variables, the values of the new influencing factor variables, and the value of any one of the to-be-determined new influencing factor variables, which are obtained multiple times in real time when the wind turbine generator set is in a normal state, into a first fitting function; If the number of times that the differences between the plurality of calculated powers and the corresponding plurality of real-time acquired powers are less than the first threshold is greater than the second threshold, the verification is passed; otherwise, the verification is failed.

5. The calculation method according to claim 1, wherein: In response to the wind turbine generator being in a normal state, the step of classifying the influencing factor data based on the degree of influence of the influencing factor data on power includes: In response to the wind turbine being in a normal state, the new influencing factor variable whose value remains unchanged as the power changes is classified as a pending low-influence factor variable; Verify each pending low-impact factor variable; The pending low-impact factor variables that have passed verification are classified as low-impact factor variables, and the pending low-impact factor variables that have not passed verification are classified as new impact factor variables, thereby updating the classification of impact factor data.

6. The calculation method according to claim 5, wherein: In response to the wind turbine generator being in a normal state, the steps of classifying a new influencing factor variable whose value remains unchanged as the power changes as a pending low-influence factor variable include: Acquire from a database multiple sets of new influencing factor variables whose values remain unchanged as power changes while the values of basic influencing factor variables remain unchanged; Variables in the intersection of the multiple sets are classified as undetermined low-impact factor variables.

7. The calculation method according to claim 6, wherein: The steps to verify each low-impact factor variable include: Obtaining a second fitting function for calculating power by performing multivariate fitting on the power obtained from the database, the value of the basic influencing factor variable, and the value of the new influencing factor variable; Obtaining multiple calculated powers by substituting the values of the basic influencing factor variables and the values of the new influencing factor variables obtained multiple times in real time when the wind turbine generator set is in a normal state into the second fitting function; If the number of times that the differences between the plurality of calculated powers and the corresponding plurality of real-time acquired powers are less than the third threshold is greater than the fourth threshold, the verification is passed; otherwise, the verification is not passed.

8. The calculation method according to claim 1, wherein: The steps of calculating the power loss of the wind turbine generator set based on the calculated power, the power obtained in real time when the wind turbine generator set is in an abnormal state, and the time information when the wind turbine generator set is in the abnormal state include: The power loss of the wind turbine generator set is calculated by integrating the difference between the calculated power and the power acquired in real time when the wind turbine generator set is in an abnormal state according to the time information.

9. The calculation method according to claim 1, further comprising: In response to the wind turbine generator set being in a normal state, the real-time acquired power and the results of bin-based value extraction of the real-time acquired influencing factor data are stored in the database as historical influencing factor data and historical power data.

10. A device for calculating power loss of a wind turbine generator set, comprising: an acquirer configured to acquire in real time the power, influencing factor data and device status data of the wind turbine generator set, wherein the influencing factor data includes data of multiple variables; a determiner configured to determine whether the wind turbine generator set is in a normal state or an abnormal state based on the device status data; a calculator configured to calculate the power loss of the wind turbine generator set based on the power and influencing factor data in response to the wind turbine generator set being in an abnormal state; The classifier is configured to classify the influencing factor data based on the degree of influence of the influencing factor data on power in response to the wind turbine being in a normal state, wherein the multiple variables are initially classified into basic influencing factor variables, new influencing factor variables, and low-influence factor variables, and the uncategorized variables among the multiple variables are initially classified into the new influencing factor variables or the low-influence factor variables, and the classification of the new influencing factor variables and the low-influence factor variables is updated by the step of classifying the influencing factor data based on the degree of influence of the influencing factor data on power. The calculator is further configured to calculate the power loss of the wind turbine generator set based on the power and the classified influencing factor data in response to the wind turbine generator set being in an abnormal state. The calculator is configured as follows: Obtaining a third fitting function for calculating power by performing multivariate fitting on the power obtained from the database, the value of the basic influencing factor variable, and the value of the new influencing factor variable; The calculated power is obtained by substituting the values of the basic influencing factor variables and the values of the new influencing factor variables obtained in real time when the wind turbine generator set is in an abnormal state into the third fitting function; Based on the calculated power, the power obtained in real time when the wind turbine is in an abnormal state, and the time information of the wind turbine in the abnormal state, the power loss of the wind turbine is calculated. Among them, the classifier is also configured as: In response to the wind turbine being in a normal state, low-influence factor variables and uncategorized variables whose values change with power changes are classified as new influencing factor variables to be determined; Verify each new influencing factor variable to be determined; The pending new influencing factor variables that have passed verification are classified as new influencing factor variables, and the pending new influencing factor variables that have not passed verification are classified as low-influencing factor variables, thereby achieving an update of the classification of the influencing factor data.

11. The computing device of claim 10, wherein: The multiple variables include at least wind speed, wind direction, temperature, air density, humidity and light intensity.

12. The computing device of claim 10, wherein: The classifier is also configured to: Acquire multiple sets of low-influence factor variables and unclassified variables whose values change with power changes when the value of the basic influencing factor variable remains unchanged from the database; Variables in the intersection of the multiple sets are classified as pending new influencing factor variables.

13. The computing device of claim 10, wherein: The classifier is also configured to: For any undetermined new influencing factor variable, a first fitting function for calculating power is obtained by performing multivariate fitting on the power obtained from the database, the value of the basic influencing factor variable, the value of the new influencing factor variable, and the value of the any undetermined new influencing factor variable; Obtaining multiple calculated powers by substituting the values of the basic influencing factor variables, the values of the new influencing factor variables, and the value of any one of the to-be-determined new influencing factor variables, which are obtained multiple times in real time when the wind turbine generator set is in a normal state, into a first fitting function; If the number of times that the differences between the plurality of calculated powers and the corresponding plurality of real-time acquired powers are less than the first threshold is greater than the second threshold, the verification is passed; otherwise, the verification is failed.

14. The computing device of claim 10 , wherein the classifier is further configured to: In response to the wind turbine being in a normal state, the new influencing factor variable whose value remains unchanged as the power changes is classified as a pending low-influence factor variable; Verify each pending low-impact factor variable; The pending low-impact factor variables that passed the verification were classified as low-impact factor variables, and the pending low-impact factor variables that failed the verification were classified as new impact factor variables.

15. The computing device of claim 14, wherein: The classifier is also configured to: Acquire from a database multiple sets of new influencing factor variables whose values remain unchanged as power changes while the values of basic influencing factor variables remain unchanged; Variables in the intersection of the multiple sets are classified as undetermined low-impact factor variables.

16. The computing device of claim 15, wherein: The classifier is also configured to: Obtaining a second fitting function for calculating power by performing multivariate fitting on the power obtained from the database, the value of the basic influencing factor variable, and the value of the new influencing factor variable; Obtaining multiple calculated powers by substituting the values of the basic influencing factor variables and the values of the new influencing factor variables obtained multiple times in real time when the wind turbine generator set is in a normal state into the second fitting function; If the number of times that the differences between the plurality of calculated powers and the corresponding plurality of real-time acquired powers are less than the third threshold is greater than the fourth threshold, the verification is passed; otherwise, the verification is not passed.

17. The computing device of claim 10, wherein: The calculator is also configured to: The power loss of the wind turbine generator set is calculated by integrating the difference between the calculated power and the power acquired in real time when the wind turbine generator set is in an abnormal state according to the time information.

18. The computing device of claim 10, further comprising: The memory is configured to: in response to the wind turbine generator set being in a normal state, store the real-time acquired power and the results of binning and valuing the real-time acquired influencing factor data as historical influencing factor data and historical power data in a database.

19. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for calculating the power loss of a wind turbine generator set according to any one of claims 1 to 9 is implemented.

20. A computing device comprising: processor; A memory storing a computer program, which, when executed by a processor, implements the method for calculating the power loss of a wind turbine generator set according to any one of claims 1 to 9.

Images