An energy-saving control method for axial flow fans in wind turbine generator sets

Abstract

This invention relates to the field of wind turbine generator technology, specifically disclosing an energy-saving control method for axial flow wind turbines in wind turbine generator sets. The method includes the following steps: acquiring the operating parameters of the wind turbine generator set from historical data and performing time-period analysis to determine the operating status of the wind turbine generator set, wherein the operating status includes high-efficiency operating periods and low-efficiency operating periods; based on the determined operating status of the wind turbine generator set, performing regularity analysis and judgment to generate regular and irregular signals; and based on the regularity analysis and judgment results, predicting the operating status of the wind turbine generator set for future periods. This invention identifies low-efficiency periods of the wind turbine generator set by analyzing historical data, thereby reducing the mechanical energy input of the axial flow wind turbine during low-efficiency periods and reducing energy waste during inefficient power generation. By predicting high-efficiency and low-efficiency periods, the high-energy-consumption phase during startup is reduced, lowering additional energy losses in the equipment, thus achieving energy-saving goals.

Description

Technical Field

[0001] This invention relates to the field of wind turbine generator technology, and more specifically to an energy-saving control method for axial flow fans in wind turbine generators. Background Technology

[0002] Wind turbine generators are key equipment for converting wind energy into electrical energy. They mainly consist of a wind turbine, a generator, and a tower. The wind turbine rotates under the action of wind, converting wind energy into mechanical energy, which is then transmitted to the generator through a transmission system and finally converted into electrical energy output.

[0003] However, when monitoring wind turbine operating data, the method of judging the operating status of the unit based on a simple power threshold cannot effectively distinguish between high and low operating efficiency. As a result, during some periods that appear to be operating normally but are actually inefficient, axial wind turbines are still consuming a lot of electrical energy, resulting in low energy utilization efficiency. In the process of analyzing historical operating data, researchers have noticed that existing technologies rarely conduct in-depth research on the patterns of high-efficiency and low-efficiency periods. This means that when predicting future operating status, only a relatively simple prediction method can be used, and the prediction results deviate significantly from the actual situation, failing to provide a reliable basis for the operation of axial wind turbines. Summary of the Invention

[0004] The purpose of this invention is to provide an energy-saving control method for axial flow fans in wind turbine generator sets, so as to solve the technical problems mentioned above.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] This invention provides an energy-saving control method for axial flow fans in wind turbine generator sets, specifically including the following steps:

[0007] Obtain the operating parameters of the wind turbine generators from historical data and perform time period analysis to determine the operating status of the wind turbine generators. The operating status includes high-efficiency operating periods and low-efficiency operating periods.

[0008] Based on the determined operating status of the wind turbine generator set, regularity analysis and judgment are performed to generate regular and irregular signals;

[0009] Based on the results of regularity analysis, the operating status of wind turbine generators is predicted for future periods, and the axial flow wind turbines are controlled based on the predicted status for future periods.

[0010] As a further aspect of the present invention: the process of determining the operating status of the wind turbine generator set is as follows:

[0011] The historical data contains historical output power data, which is divided into several time intervals, with each time interval constituting a period.

[0012] By analyzing the historical output power sequence {P1, P2, ..., P} for each time period, n The calculations are performed to obtain the power deviation ratio and dispersion value for each time period.

[0013] By processing the power deviation ratio and the dispersion value, an operating efficiency judgment coefficient is obtained.

[0014] Periods in which the operational efficiency judgment coefficient is less than the operational efficiency judgment coefficient threshold are marked as high-efficiency periods, and periods in which the operational efficiency judgment coefficient is greater than or equal to the operational efficiency judgment coefficient threshold are marked as low-efficiency periods.

[0015] As a further aspect of the present invention: the process for obtaining the power deviation ratio is as follows:

[0016] The historical output power data for each time period are summed and averaged to obtain the power mean. The power mean is then compared with the rated power of the wind turbine generator set. The absolute value of the difference is taken and then compared with the rated power of the wind turbine generator set to obtain the power deviation ratio.

[0017] As a further aspect of the present invention: the process of obtaining the dispersion value is as follows:

[0018] Extract the maximum and minimum output power values ​​from the historical output power data for each time period, calculate the difference, and obtain the output power range.

[0019] Through the formula: The discrete representation value LS is calculated, where P max-min This indicates a significant difference in output power. The power mean, P i The value represents the i-th historical output power value in the historical output power value sequence, where n represents the number of historical output power values, and PJ represents the average of all adjacent output power values ​​of the i-th historical output power value.

[0020] As a further aspect of the present invention: the process of performing regularity analysis and judgment is as follows:

[0021] By processing and calculating the high-efficiency and low-efficiency periods, the regularity characterization values ​​of the high-efficiency period group and the low-efficiency period group are obtained respectively.

[0022] The pattern judgment coefficient is obtained by weighted summation of the pattern representation values ​​of the high-efficiency time period group and the low-efficiency time period group.

[0023] If the pattern recognition coefficient is less than the pattern recognition coefficient threshold, a pattern signal is generated;

[0024] If the regularity judgment coefficient is greater than or equal to the regularity judgment coefficient threshold, an irregular signal is generated.

[0025] As a further aspect of the present invention: the process for obtaining the regularity characterization value of the efficient time period group is as follows:

[0026] Consecutive high-efficiency operating periods are combined to obtain high-efficiency operating period groups;

[0027] The number of high-efficiency periods in each high-efficiency period group is counted to obtain a sequence of high-efficiency period count values ​​for all high-efficiency period groups. The average value of the high-efficiency period count is then calculated.

[0028] The difference between each value in the high-efficiency period quantity value sequence and the average quantity of the high-efficiency period is calculated, and the absolute value of the difference is taken to obtain the high-efficiency quantity deviation value.

[0029] The high-efficiency quantity deviation value is compared with the high-efficiency quantity deviation limit. High-efficiency time period groups with high-efficiency quantity deviation values ​​less than or equal to the high-efficiency quantity deviation limit are extracted and marked as concentrated high-efficiency time period groups. The number of concentrated high-efficiency time period groups is counted and then the ratio is calculated with the total number of high-efficiency time period groups to obtain the high-efficiency concentrated quantity ratio.

[0030] Extract the maximum and minimum values ​​from the sequence of high-efficiency period values, calculate the difference, and then calculate the ratio of the difference to the mean of the high-efficiency period values ​​to obtain the change ratio of high-efficiency values.

[0031] The efficient concentration ratio and the efficient numerical change ratio are weighted and summed to obtain the regularity characterization value of the efficient time period group.

[0032] As a further aspect of the present invention: the process for obtaining the regularity characterization value of the inefficient time period group is as follows:

[0033] Consecutive periods of inefficient operation are integrated to obtain groups of inefficient operation periods, and the number of inefficient periods in each group of inefficient operation periods is counted to obtain a sequence of inefficient period count values ​​for all groups of inefficient operation periods.

[0034] Then, the values ​​in the inefficient period quantity value sequence are summed and averaged to obtain the average quantity of inefficient periods;

[0035] The difference between each value in the inefficient period quantity value sequence and the average quantity of the inefficient period is calculated, and the absolute value of the difference is taken to obtain the inefficient quantity deviation value.

[0036] The inefficient quantity deviation value is compared with the inefficient quantity deviation limit. The efficient time period group with the inefficient quantity deviation value less than or equal to the inefficient quantity deviation limit is extracted and marked as the concentrated inefficient time period group. The number of concentrated inefficient time period groups is counted and then the ratio is calculated with the total number of inefficient time period groups to obtain the inefficient concentrated quantity ratio.

[0037] Extract the maximum and minimum values ​​from the sequence of inefficient time periods, calculate the difference, and then calculate the ratio of the difference to the mean of the inefficient time periods to obtain the change ratio of inefficient values.

[0038] The weighted summation of the ratio of inefficient concentrations and the ratio of changes in inefficient values ​​yields the characteristic value of the inefficient time period group.

[0039] As a further aspect of the present invention: the process of predicting the operating status of the wind turbine generator set for future periods is as follows:

[0040] Based on regular signals, a fixed-interval pattern prediction method is used to predict the operational status of future time periods, specifically:

[0041] Obtain the average number of efficient and inefficient time periods, and use these average numbers as intervals. Based on the current time period's operating status and the intervals, obtain the operating status of future time periods.

[0042] Based on irregular signals, the moving average method is used to predict the operational status of future periods.

[0043] As a further aspect of the present invention: the prediction method based on irregular signals is as follows:

[0044] Set the window size k for the moving average method, where the window size k is adjusted based on an adjustment factor;

[0045] Obtain historical operating status data of the wind turbine generator set, i.e., {s1, s2, ..., s...} m}, where s j This represents the running status data for the j-th time period;

[0046] Moving average M j The calculation formula is: Where j = 1, 2, ..., m-k+1;

[0047] Where p represents the summation index variable, used to iterate through the data points in the interval from j+k-1, and then for the corresponding s p The summation is performed, where j represents the starting index of the moving average calculation, m is the total number of historical running state data, and S... p This represents the p-th running status data;

[0048] Set a prediction threshold. If the moving average is greater than the prediction threshold, the prediction period is marked as an efficient period. If the moving average is less than or equal to the prediction threshold, the prediction period is marked as an inefficient period.

[0049] As a further aspect of the present invention: the process of obtaining the adjustment coefficient is as follows:

[0050] Volatility Influence Factor F B The calculation formula is: Where bcg and jzg represent the standard deviation and mean of the number of efficient time periods, respectively, and bcd and jzd represent the standard deviation and mean of the number of inefficient time periods, respectively.

[0051] Comprehensive Influence Factor F of Concentrated Change JB The calculation formula is: F JB =a1*rjg+a2*rjd-b1*rbg-b2*rbd, where a1, a2, b1, and b2 are weighting coefficients, rjg and rbg represent the ratio of efficient concentrations and the ratio of efficient numerical changes, respectively, and rjd and rbd represent the ratio of inefficient concentrations and the ratio of inefficient numerical changes, respectively.

[0052] The adjustment coefficient is obtained by calculating the ratio of the comprehensive impact factor of concentrated changes to the impact factor of fluctuations.

[0053] The beneficial effects of this invention are:

[0054] (1) This invention determines the inefficient period of wind power generation by analyzing historical data, thereby reducing the mechanical energy input of axial wind turbines during the inefficient period and reducing the waste of electrical energy in the inefficient power generation process. By predicting the high-efficiency and low-efficiency periods, the high-energy consumption stage during the start-up process is reduced, and the additional energy loss of the equipment is reduced, thereby achieving the purpose of energy saving.

[0055] (2) This invention analyzes whether the occurrence time of high-efficiency and low-efficiency periods is regular, which facilitates the subsequent prediction of high-efficiency and low-efficiency periods. At the same time, the prediction method can be selected according to the regularity results, and the operation of the axial flow fan can be arranged more accurately. During the low-efficiency period, the speed of the axial flow fan can be reduced or the blade angle can be adjusted to reduce unnecessary mechanical energy input, effectively avoid the waste of electrical energy in the operation process that cannot be efficiently converted into electrical energy, and reduce the ineffective energy consumption of the fan during the low-output period. Attached Figure Description

[0056] The invention will now be further described with reference to the accompanying drawings.

[0057] Figure 1 This is a flowchart of an energy-saving control method for an axial flow fan in a wind turbine generator set according to the present invention;

[0058] Figure 2 This is an architecture diagram of an axial flow fan energy-saving control system for a wind turbine generator set according to the present invention. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] Example 1:

[0061] Please see Figure 1 , Figure 2 As shown in the embodiment of the present invention, an energy-saving control method for an axial flow wind turbine of a wind power generator set includes the following steps:

[0062] Step 1: Obtain the operating parameters of the wind turbine generator from historical data and perform time period analysis to determine the high-efficiency and low-efficiency operating periods of the wind turbine generator;

[0063] In some embodiments, historical operating parameters are collected from data sources such as monitoring systems and sensor networks of wind turbine generator sets.

[0064] The historical operating parameters include, but are not limited to: wind speed data obtained by the anemometer, output power data obtained by the power measurement device, and ambient temperature data obtained by the temperature sensor;

[0065] In one possible embodiment, the efficient and inefficient operating periods of the wind turbine are determined based on output power data. The specific process is as follows:

[0066] Historical data is divided into several time intervals, with each time interval being a period of time. The time intervals are set by those skilled in the art based on experience and the operating characteristics of wind turbine generators.

[0067] For any given time period, obtain the historical output power value sequence for each time period: {P1, P2, ..., P...} n};

[0068] The historical output power data for each time period are summed and averaged to obtain the power average. The difference between the power average and the rated power of the wind turbine generator set is calculated. The absolute value of the difference is taken and compared with the rated power of the wind turbine generator set to obtain the power deviation ratio.

[0069] Extract the maximum and minimum output power values ​​from the historical output power data for that period and calculate the difference to obtain the output power range.

[0070] Through the formula: The discrete representation value LS is calculated, where P max-minThis indicates a significant difference in output power. The power mean, P i The output power value represents the i-th output power value in the historical output power value sequence, n represents the number of historical output power values, and PJ represents the average of all adjacent output power values ​​of the i-th historical output power value.

[0071] By calculating the power deviation ratio and discrete characterization values, we can see that: First, the operating efficiency judgment coefficient is obtained by multiplying the power deviation ratio and discrete characterization values, which comprehensively reflects the operating efficiency and stability of the wind turbine generator set during that period. Second, by comprehensively considering the maximum, minimum, average, and adjacent average values ​​of the output power through discrete characterization values, we can accurately capture subtle fluctuations in power output and more accurately reflect the stability of output power changes. Third, it can serve as an early warning indicator for potential equipment faults. Compared to existing technologies that usually only detect faults when they occur and power output is significantly abnormal, if a wind turbine generator set experiences an early, minor fault, it may be apparent when it causes subtle fluctuations in power output.

[0072] In the operation of wind turbine generator sets, extreme value differences can quickly reflect the maximum fluctuation range of the operating status. For example, the drastic changes in power when encountering strong gusts can be more clearly reflected by this discrete characterization value. When analyzing output power fluctuations, the changes in adjacent power values ​​are very important for judging the operational stability. This method can better reflect the continuity and trend of output power changes.

[0073] It should be noted that if i = 1, then PJ = P2; if i = n, then PJ = P2. n-1 ;

[0074] The operating efficiency judgment coefficient is obtained by multiplying the power deviation ratio and the dispersion value.

[0075] Set a threshold for the operational efficiency judgment coefficient, wherein the threshold for the operational efficiency judgment coefficient is set by implementers in this field based on extensive data analysis and experience.

[0076] Periods with an operational efficiency judgment coefficient less than the operational efficiency judgment coefficient threshold are marked as high-efficiency periods, and periods with an operational efficiency judgment coefficient greater than or equal to the operational efficiency judgment coefficient threshold are marked as low-efficiency periods.

[0077] Step 2: Based on the determined high-efficiency and low-efficiency operating periods of the wind turbine generator sets, conduct regularity analysis and make regularity judgments;

[0078] In some implementation schemes, the periods of high-efficiency operation and low-efficiency operation of wind turbine generators are identified;

[0079] Consecutive high-efficiency periods are integrated to obtain high-efficiency period groups, and the number of high-efficiency periods in each high-efficiency period group is counted to obtain a sequence of high-efficiency period count values ​​for all high-efficiency period groups.

[0080] For example, if there is a sequence of efficient time periods [1-3 o'clock, 3-5 o'clock, 7-9 o'clock, 9-11 o'clock], it can be integrated into two efficient time period groups: [1-5 o'clock] and [7-11 o'clock].

[0081] Then, the values ​​in the high-efficiency period quantity value sequence are summed and averaged to obtain the average quantity of high-efficiency periods;

[0082] The difference between each value in the high-efficiency period quantity value sequence and the average quantity of the high-efficiency period is calculated, and the absolute value of the difference is taken to obtain the high-efficiency quantity deviation value.

[0083] The high-efficiency quantity deviation value is compared with the high-efficiency quantity deviation limit. High-efficiency time period groups with high-efficiency quantity deviation values ​​less than or equal to the high-efficiency quantity deviation limit are extracted and marked as concentrated high-efficiency time period groups. The number of concentrated high-efficiency time period groups is counted and then the ratio is calculated with the total number of high-efficiency time period groups to obtain the high-efficiency concentrated quantity ratio.

[0084] Extract the maximum and minimum values ​​from the sequence of high-efficiency period values, calculate the difference, and then calculate the ratio of the difference to the mean of the high-efficiency period values ​​to obtain the change ratio of high-efficiency values.

[0085] The efficient concentration ratio and the efficient numerical change ratio are weighted and summed to obtain the regularity characterization value of the efficient time period group.

[0086] Consecutive periods of inefficient operation are integrated to obtain groups of inefficient operation periods, and the number of inefficient periods in each group of inefficient operation periods is counted to obtain a sequence of inefficient period count values ​​for all groups of inefficient operation periods.

[0087] Then, the values ​​in the inefficient period quantity value sequence are summed and averaged to obtain the average quantity of inefficient periods;

[0088] The difference between each value in the inefficient period quantity value sequence and the average quantity of the inefficient period is calculated, and the absolute value of the difference is taken to obtain the inefficient quantity deviation value.

[0089] The inefficient quantity deviation value is compared with the inefficient quantity deviation limit. The efficient time period group with the inefficient quantity deviation value less than or equal to the inefficient quantity deviation limit is extracted and marked as the concentrated inefficient time period group. The number of concentrated inefficient time period groups is counted and then the ratio is calculated with the total number of inefficient time period groups to obtain the inefficient concentrated quantity ratio.

[0090] Extract the maximum and minimum values ​​from the sequence of inefficient time periods, calculate the difference, and then calculate the ratio of the difference to the mean of the inefficient time periods to obtain the change ratio of inefficient values.

[0091] The weighted summation of the ratio of inefficient concentrations and the ratio of changes in inefficient values ​​yields the characteristic value of the inefficient time period group.

[0092] The pattern judgment coefficient is obtained by weighted summation of the pattern representation values ​​of the high-efficiency time period group and the low-efficiency time period group.

[0093] Set a threshold for the pattern judgment coefficient. If the pattern judgment coefficient is less than the threshold, it indicates that the high-efficiency period and the low-efficiency period appear regularly in the historical data, generating a pattern signal.

[0094] If the regularity judgment coefficient is greater than or equal to the regularity judgment coefficient threshold, it indicates that the efficient and inefficient periods do not appear regularly in the historical data, generating irregular signals.

[0095] Determining whether the occurrence of high-efficiency and low-efficiency periods is regular is to facilitate subsequent prediction of high-efficiency and low-efficiency periods. At the same time, based on the regularity results, the prediction method can be selected to more accurately arrange the operation of axial flow fans. During low-efficiency periods, the speed of axial flow fans can be reduced or the blade angle can be adjusted to reduce unnecessary mechanical energy input, effectively avoid the waste of electrical energy in the process of operation that cannot be efficiently converted into electrical energy, and reduce the ineffective energy consumption of fans during low-output periods.

[0096] In addition to predicting future high-efficiency and low-efficiency periods, practitioners in this field can also compare and judge other wind turbine generators based on the regularity of high-efficiency and low-efficiency periods. This allows them to discover the advantages and disadvantages of the wind turbine generator in operation and management, thereby optimizing scheduling, achieving energy saving and efficiency improvement of the entire wind power generation system, and improving the synergistic energy saving level of axial flow wind turbines among different units.

[0097] Step 3: Based on the results of regularity analysis, predict the operating status of the wind turbine generator set for future periods;

[0098] In some embodiments, based on the generated regular signal, the average number of efficient time periods and the average number of inefficient time periods are obtained, and the average number of efficient time periods and the average number of inefficient time periods are used as the interval number.

[0099] Based on the number of intervals, predict the operational status for future periods;

[0100] For example, if the average number of high-efficiency periods is 3 and the average number of low-efficiency periods is 1, then the predicted sequence of the operating status of the wind turbine generator is: {high efficiency, high efficiency, high efficiency, low efficiency, high efficiency, high efficiency, low efficiency, ...};

[0101] Based on the generated irregular signals, the moving average method is used to predict the state of wind turbine generators for future periods. The specific process is as follows:

[0102] Set the window size k for the moving average method. The choice of window size k needs to be adjusted according to the characteristics of the data and experience. Generally, different values ​​can be used to observe the prediction effect and choose a suitable value, such as k=5 or k=7.

[0103] Obtain historical operating status data of the wind turbine generator set, i.e., {s1, s2, ..., s...} m}, where s j This represents the operational status data for the j-th time period, where 1 indicates a high-efficiency period and 0 indicates a low-efficiency period.

[0104] Moving average M j The calculation formula is: Where j = 1, 2, ..., m-k+1;

[0105] Where p represents the summation index variable, used to iterate through the data points in the interval from j+k-1, and to calculate the corresponding s. p The summation is performed, where j represents the starting index of the moving average calculation, m is the total number of historical running state data, and S... p This represents the p-th running status data;

[0106] Set a prediction threshold, where 1 represents a high-efficiency period and 0 represents a low-efficiency period, and the prediction threshold can be set to 0.5.

[0107] If the moving average is greater than the prediction threshold, the prediction period is marked as an efficient period; if the moving average is less than or equal to the prediction threshold, the prediction period is marked as an inefficient period.

[0108] The window size k of the moving average method can be adjusted according to data characteristics and experience. In different wind power generation scenarios, the data change patterns and fluctuations are different. By flexibly adjusting the window size, it is possible to better adapt to different data characteristics. For example, it can be adjusted based on the number of high-efficiency time groups and low-efficiency time groups.

[0109] The specific process is as follows: by calculating the fluctuation impact factor and the comprehensive impact factor of concentrated changes, the adjustment coefficient is obtained;

[0110] Furthermore, the volatility impact factor F B The calculation formula is: Where bcg and jzg represent the standard deviation and mean of the number of efficient time periods, respectively, and bcd and jzd represent the standard deviation and mean of the number of inefficient time periods, respectively.

[0111] Comprehensive Influence Factor F of Concentrated Change JB The calculation formula is: F JB =a1*rjg+a2*rjd-b1*rbg-b2*rbd, where a1, a2, b1, and b2 are weighting coefficients, rjg and rbg represent the ratio of efficient concentrations and the ratio of efficient numerical changes, respectively, and rjd and rbd represent the ratio of inefficient concentrations and the ratio of inefficient numerical changes, respectively.

[0112] The adjustment coefficient C is obtained by calculating the ratio of the comprehensive impact factor of concentrated changes to the impact factor of fluctuations.

[0113] It should be noted that since this calculation process is based on irregular signals, the fluctuation influence factor is not zero;

[0114] If the initial window is K0, then the window after adjustment based on the number of efficient time period groups and inefficient time period groups is KT. The formula for calculating window KT is: KT=K0*(1+C);

[0115] If the adjustment factor is greater than zero, the window size increases; if the adjustment factor is less than zero, the window size decreases.

[0116] It should be noted that in practical applications, a large amount of historical data is used to repeatedly test and optimize the weight coefficients a1, a2, b1, and b2 in the above formula to ensure that the adjustment coefficients can accurately adjust the window size of the moving average method according to different wind power generation scenarios, thereby improving the accuracy of predicting the future operating status of wind turbine generators.

[0117] Under irregular signals, the operating status of wind turbine generators is not obvious. Adjusting the window size can flexibly cope with this situation. Based on data characteristics such as the number of high-efficiency time groups and low-efficiency time groups, the window is dynamically adjusted. When the number of high-efficiency and low-efficiency time groups fluctuates greatly, the window is reduced, and when the number is relatively stable, the window is increased, thereby improving the adaptability of the forecast to irregular data and making the forecast results more reliable.

[0118] Frequent start-up and shutdown of axial flow fans can increase energy consumption and shorten equipment life due to the high current surge and mechanical stress changes at the moment of startup. By accurately predicting the high-efficiency and low-efficiency periods, the start-up and shutdown times of the fans can be rationally planned to avoid unnecessary starts, reduce the high-energy-consumption phase during startup, and reduce the additional energy loss caused by frequent start-up and shutdown.

[0119] Based on predictions of future operating conditions, the operation of axial flow fans can be rationally planned and controlled, such as reducing fan speed, adjusting blade angle, and reducing the number of fans in operation during periods of low efficiency.

[0120] The technical solution of this invention is mainly as follows: First, by dividing historical data into time periods, the power deviation ratio and discrete characterization value of each time period are calculated to obtain the operating efficiency judgment coefficient. Based on this coefficient, the high-efficiency and low-efficiency time periods are determined. Then, for the integrated continuous high-efficiency and low-efficiency time periods, the regular characterization values ​​of the high-efficiency time period group and the low-efficiency time period group are calculated respectively. The two are weighted and summed to obtain the regularity judgment coefficient. Based on the comparison result with the regularity judgment coefficient threshold, a regular signal or an irregular signal is generated. Finally, based on the regular signal, the future operating status is predicted by the interval number of the average number of high-efficiency time periods and the average number of low-efficiency time periods. Based on the irregular signal, the future operating status is predicted by the moving average method.

[0121] This allows for precise identification of high-efficiency and low-efficiency periods, enabling rational planning of axial fan operation. During low-efficiency periods, fan speed can be reduced, blade angles adjusted, and the number of fans reduced to decrease mechanical energy input, avoid energy waste, and reduce ineffective energy consumption during low-output periods. At the same time, accurate prediction of high-efficiency and low-efficiency periods helps to rationally plan fan start-up and shutdown times, reduce unnecessary starts, avoid high current surges and mechanical stress changes during startup, and reduce additional energy losses caused by frequent start-ups and shutdowns.

[0122] Example 2:

[0123] Based on Example 1, please refer to Figure 2 As shown in the embodiment of the present invention, an energy-saving control system for an axial flow wind turbine of a wind turbine generator set specifically includes:

[0124] Operation status determination module: acquires the operating parameters of the wind turbine generator from historical data, performs time period analysis, and determines the high-efficiency and low-efficiency operating periods of the wind turbine generator;

[0125] Pattern recognition module: Based on the judgment results of the high-efficiency and low-efficiency operation periods of wind turbine generators, it performs pattern analysis and makes pattern judgments;

[0126] Operation status prediction module: Based on the regularity judgment results, predict the operation status of wind turbine generators for future periods.

[0127] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. An energy-saving control method for axial flow fans in a wind turbine generator set, characterized in that, Specifically, the following steps are included: Obtain the operating parameters of the wind turbine generators from historical data and perform time period analysis to determine the operating status of the wind turbine generators. The operating status includes high-efficiency operating periods and low-efficiency operating periods. Based on the determined operating status of the wind turbine generator set, regularity analysis and judgment are performed to generate regular and irregular signals; Based on the results of regularity analysis and judgment, the operating status of wind turbine generators is predicted for future periods, and the axial flow wind turbines are controlled based on the predicted status for future periods. The process of performing regularity analysis and judgment is as follows: By processing and calculating the high-efficiency and low-efficiency periods, the regularity characterization values ​​of the high-efficiency period group and the low-efficiency period group are obtained respectively. The regularity characteristics of the high-efficiency time period group and the low-efficiency time period group are weighted and calculated to obtain the regularity judgment coefficient. If the pattern judgment coefficient is less than the pattern judgment coefficient threshold, a pattern signal is generated; if the pattern judgment coefficient is greater than or equal to the pattern judgment coefficient threshold, an irregular signal is generated.

2. The energy-saving control method for axial flow fans of a wind turbine generator set according to claim 1, characterized in that, The process of determining the operating status of the wind turbine generator set is as follows: The historical data contains historical output power data, which is divided into several time intervals, with each time interval being a period. By analyzing the historical output power sequence for each time period The calculations yielded the power deviation ratio and dispersion value for each time period. By processing the power deviation ratio and the dispersion value, an operating efficiency judgment coefficient is obtained. Periods in which the operational efficiency judgment coefficient is less than the operational efficiency judgment coefficient threshold are marked as high-efficiency periods, and periods in which the operational efficiency judgment coefficient is greater than or equal to the operational efficiency judgment coefficient threshold are marked as low-efficiency periods.

3. The energy-saving control method for axial flow fans of a wind turbine generator set according to claim 2, characterized in that, The process of obtaining the power deviation ratio is as follows: The historical output power data for each time period are summed and averaged to obtain the power mean. The power mean is then compared with the rated power of the wind turbine generator set. The absolute value of the difference is taken and then compared with the rated power of the wind turbine generator set to obtain the power deviation ratio.

4. The energy-saving control method for the axial flow fan of a wind turbine generator set according to claim 3, characterized in that, The process for obtaining the dispersion value is as follows: Extract the maximum and minimum output power values ​​from the historical output power data for each time period, calculate the difference, and obtain the output power range. Through the formula: The discrete representation value LS is calculated, where, This indicates a significant difference in output power. The power mean is represented. The value represents the i-th historical output power value in the historical output power value sequence, where n represents the number of historical output power values, and PJ represents the average of all adjacent output power values ​​of the i-th historical output power value.

5. The energy-saving control method for axial flow fans of a wind turbine generator set according to claim 1, characterized in that, The process for obtaining the regularity representation values ​​of the high-efficiency time period group is as follows: Consecutive high-efficiency operating periods are combined to obtain high-efficiency operating period groups; The number of high-efficiency periods in each high-efficiency period group is counted to obtain a sequence of high-efficiency period count values ​​for all high-efficiency period groups. The average value of the high-efficiency period count is then calculated. The difference between each value in the high-efficiency period quantity value sequence and the average quantity of the high-efficiency period is calculated, and the absolute value of the difference is taken to obtain the high-efficiency quantity deviation value. Extract high-efficiency time periods whose high-efficiency quantity deviation values ​​are less than or equal to the high-efficiency quantity deviation limit, mark them as concentrated high-efficiency time periods, count their number, and then calculate the ratio of the high-efficiency concentrated quantity ratio to the total number of high-efficiency time periods. Extract the maximum and minimum values ​​from the sequence of high-efficiency period values, calculate the difference, and then calculate the ratio of the difference to the mean of the high-efficiency period values ​​to obtain the change ratio of high-efficiency values. The efficient concentration ratio and the efficient numerical change ratio are weighted and summed to obtain the regularity characterization value of the efficient time period group.

6. The energy-saving control method for axial flow fans of a wind turbine generator set according to claim 5, characterized in that, The process for obtaining the regularity representation value of the inefficient time period group is as follows: Consecutive periods of inefficient operation are integrated to obtain groups of inefficient operation periods, and the number of inefficient periods in each group of inefficient operation periods is counted to obtain a sequence of inefficient period count values ​​for all groups of inefficient operation periods. Then, the values ​​in the inefficient period quantity value sequence are summed and averaged to obtain the average quantity of inefficient periods; The difference between each value in the inefficient period quantity value sequence and the average quantity of the inefficient period is calculated, and the absolute value of the difference is taken to obtain the inefficient quantity deviation value. The inefficient quantity deviation value is compared with the inefficient quantity deviation limit. The efficient time period group with the inefficient quantity deviation value less than or equal to the inefficient quantity deviation limit is extracted, marked as the concentrated inefficient time period group, and its quantity is counted. Then, the ratio is calculated with the total quantity of the inefficient time period group to obtain the inefficient concentrated quantity ratio. Extract the maximum and minimum values ​​from the sequence of inefficient time periods, calculate the difference, and then calculate the ratio of the difference to the mean of the inefficient time periods to obtain the change ratio of inefficient values. The weighted summation of the ratio of inefficient concentrations and the ratio of changes in inefficient values ​​yields the characteristic value of the inefficient time period group.

7. The energy-saving control method for axial flow fans of a wind turbine generator set according to claim 6, characterized in that, The process of predicting the operating status of wind turbine generators for future periods is as follows: Based on regular signals, the average number of efficient time periods and the average number of inefficient time periods are obtained. The average number of efficient time periods and the average number of inefficient time periods are used as the interval number. Based on the current operating status of the time period and the interval number, the operating status of the future time period is obtained. Based on irregular signals, the moving average method is used to predict the operational status of future periods.

8. The energy-saving control method for axial flow fans of a wind turbine generator set according to claim 7, characterized in that, The prediction method based on irregular signals is as follows: Set the window size k for the moving average method, where the window size k is adjusted based on an adjustment factor; Obtain historical operating status data of wind turbine generators, i.e. ,in, This represents the running status data for the j-th time period; Moving average The calculation formula is: Where j = 1, 2, ..., m-k+1; Where p represents the summation index variable, used to iterate through the data points in the interval from j+k-1, and thus calculate the corresponding... The summation is performed, where j represents the starting index of the moving average calculation, and m is the total number of historical running state data. This represents the p-th running status data; If the moving average is greater than the prediction threshold, the prediction period is marked as an efficient period; if the moving average is less than or equal to the prediction threshold, the prediction period is marked as an inefficient period.

9. The energy-saving control method for the axial flow fan of a wind turbine generator set according to claim 8, characterized in that, The process of obtaining the adjustment coefficient is as follows: Volatility Influence Factors The calculation formula is: Where bcg and jzg represent the standard deviation and mean of the number of efficient time periods, respectively, and bcd and jzd represent the standard deviation and mean of the number of inefficient time periods, respectively. Comprehensive Influence Factors of Concentrated Changes The calculation formula is: Where a1, a2, b1, and b2 are weighting coefficients, rjg and rbg represent the ratio of efficient concentrations and the ratio of efficient numerical changes, respectively, and rjd and rbd represent the ratio of inefficient concentrations and the ratio of inefficient numerical changes, respectively. The adjustment coefficient is obtained by calculating the ratio of the comprehensive impact factor of concentrated changes to the impact factor of fluctuations.

Images