Main transformer cooling system fault alarm threshold calculation method and system, and main transformer cooling system fault alarm method and system

Abstract

The invention discloses a main transformer cooling system fault alarm threshold calculation and alarm method and system, and the method comprises the steps: obtaining the characteristic variable data and working condition characteristic data of a plurality of wind generating sets, carrying out the statistical analysis of the characteristic variable data, and determining a characteristic variable mean a1 corresponding to each characteristic variable of each wind generating set; classifying the wind generating sets through the working condition characteristic data, determining the wind generating sets belonging to the same working condition, performing statistical analysis on the characteristic variable data of the wind generating sets belonging to the same working condition, and determining a characteristic variable mean a2 corresponding to each characteristic variable of the wind generating sets belonging to the same working condition. And determining a fault alarm threshold value corresponding to each characteristic variable of the wind generating set according to the corresponding characteristic variable mean value a1 and characteristic variable mean value a2. And comparing the mean value difference of the characteristic variables of the wind generating set with a fault alarm threshold value, and if the mean value difference does not meet an alarm condition corresponding to the fault alarm threshold value, outputting an alarm signal.

Description

Technical Field

[0001] The invention relates to the technical field of main transformer fault monitoring of a wind turbine generator set, and in particular to a main transformer cooling system fault alarm threshold calculation, alarm method and system. Background Art

[0002] The main transformer cooling device is an important component of the main transformer. Its function is to dissipate the heat inside the main transformer through the cooling device. It is a key component to ensure the stable operation of the main transformer. The abnormality of the main transformer cooling system will affect the heat dissipation inside the main transformer. Once a failure occurs, the wind turbine will be shut down, resulting in power generation loss and economic loss. However, the existing SCADA monitoring system cannot identify potential abnormalities in the main transformer cooling system. It can only wait for the unit to report a fault or discover abnormalities in the cooling system of the main transformer during inspection. It is urgent to study a method for realizing early potential fault alarm of the main transformer cooling system by using existing sensors and their data without adding hardware monitoring equipment, so as to help operation and maintenance personnel to achieve preventive maintenance and reduce power generation loss of wind turbines. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present invention proposes a main transformer cooling system fault alarm threshold calculation, alarm method and system, which can realize the fault alarm of the main transformer cooling system through the existing characteristic data of the wind turbine generator set. The specific technical solution is as follows:

[0004] In a first aspect, a method for calculating a fault alarm threshold of a main transformer cooling system is provided. In a first implementable manner of the first aspect, the method includes:

[0005] Acquire characteristic variable data and operating condition characteristic data of multiple wind turbine generator sets within a monitoring period, wherein the characteristic variable data reflects the performance of the main transformer cooling system, and the operating condition characteristic data reflects the operating condition of the wind turbine generator set;

[0006] Perform statistical analysis on the characteristic variable data of each wind turbine generator set respectively to determine the characteristic variable mean a1 corresponding to each wind turbine generator set;

[0007] Classifying all wind turbine generator sets according to the operating condition characteristic data to determine wind turbine generator sets belonging to the same operating condition;

[0008] Perform statistical analysis on the characteristic variable data of wind turbine generator sets under the same working conditions respectively, and determine the characteristic variable mean a2 corresponding to the wind turbine generator sets under the same working conditions;

[0009] According to the corresponding characteristic variable mean a1 and characteristic variable mean a2, the fault alarm threshold corresponding to the wind turbine generator set is determined.

[0010] In combination with the first implementable method of the first aspect, in the second implementable method of the first aspect, the characteristic variable data includes: transformer coolant inlet temperature, transformer U-phase winding temperature, transformer V-phase winding temperature, transformer W-phase winding temperature and / or transformer coolant inlet pressure.

[0011] In combination with the first implementable manner of the first aspect, in a third implementable manner of the first aspect, obtaining characteristic variable data and operating condition characteristic data of a plurality of wind turbine generator sets includes:

[0012] The characteristic variable data and the operating condition characteristic data are preprocessed to remove abnormal data in the characteristic variable data and the operating condition characteristic data.

[0013] In combination with the third implementable manner of the first aspect, in a fourth implementable manner of the first aspect, preprocessing the characteristic variable data and the operating condition characteristic data includes:

[0014] Delete the repeated characteristic variable data and operating condition characteristic data at all time points, as well as the characteristic variable data and operating condition characteristic data with missing values;

[0015] Eliminate the characteristic variable data and operating condition characteristic data corresponding to the time point when the wind turbine generator set is not generating electricity;

[0016] The characteristic variable data and the operating condition characteristic data corresponding to the time point when the characteristic variable data exceeds the preset range are deleted.

[0017] In combination with the first implementable manner of the first aspect, in a fifth implementable manner of the first aspect, classifying all wind turbine generator sets according to the operating condition characteristic data includes:

[0018] According to the operating condition characteristic data, the DBSCAN algorithm is used to classify the operating condition of each wind turbine generator set, and wind turbine generator sets belonging to the same operating condition are determined.

[0019] In combination with the first implementable manner of the first aspect, in a sixth implementable manner of the first aspect, determining a fault alarm threshold corresponding to the wind turbine generator set includes:

[0020] According to the corresponding characteristic variable mean a1 and characteristic variable mean a2, the average value of the mean difference corresponding to the wind turbine generator set is calculated, and combined with the preset fixed value, the fault alarm threshold corresponding to the wind turbine generator set is determined.

[0021] In a second aspect, in a first implementable manner of the second aspect, a main transformer cooling system fault alarm method is provided, including:

[0022] The fault alarm threshold calculation method for the main transformer cooling system as described in any one of the first to sixth possible implementations of the first aspect is used to determine the fault alarm threshold corresponding to the wind turbine generator set.

[0023] In combination with the first implementable method of the second aspect, the second implementable method of the second aspect also includes: determining whether the operating condition characteristic data and characteristic variable data exceed the set threshold value, and outputting an alarm signal in response to the operating condition characteristic data and characteristic variable data exceeding the set threshold value.

[0024] In a third aspect, a main transformer cooling system fault alarm threshold calculation system is provided, comprising:

[0025] A characteristic data acquisition module is configured to acquire characteristic variable data and operating condition characteristic data of multiple wind turbine generator sets within a monitoring period, wherein the characteristic variable data reflects the performance of the main transformer cooling system, and the operating condition characteristic data reflects the operating condition of the wind turbine generator set;

[0026] The unit mean value calculation module is configured to perform statistical analysis on the characteristic variable data of each wind turbine generator set respectively to determine the characteristic variable mean value a1 corresponding to each wind turbine generator set;

[0027] A unit operating condition classification module is configured to classify all wind turbine generator sets according to the operating condition characteristic data, and determine wind turbine generator sets belonging to the same operating condition;

[0028] The working condition mean value calculation module is configured to perform statistical analysis on the characteristic variable data of the wind turbine generator sets under the same working condition, and determine the characteristic variable mean value a2 corresponding to the wind turbine generator sets under the same working condition;

[0029] The alarm threshold calculation module is configured to determine the fault alarm threshold corresponding to the wind turbine generator set according to the corresponding characteristic variable mean a1 and characteristic variable mean a2.

[0030] In a fourth aspect, a main transformer cooling system fault alarm system is provided, comprising:

[0031] The threshold calculation module is configured to determine the fault alarm threshold corresponding to the wind turbine generator set by adopting the main transformer cooling system fault alarm threshold calculation method as described in any one of the first to sixth possible implementation methods of the first aspect.

[0032] Beneficial effect: By adopting the main transformer cooling system fault alarm threshold calculation, alarm method and system of the present invention, the existing sensors and data of the wind turbine can be used to perform abnormal alarm of the main transformer cooling system. No new sensors and monitoring devices are needed to accurately determine the abnormal problems of the wind turbine main transformer cooling system, thereby reducing the equipment investment cost. At the same time, through early alarm of faults, it helps the wind farm to achieve predictive maintenance, reduces the difficulty of fault inspection and troubleshooting, and reduces the power generation loss of the wind turbine. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the specific implementation of the present invention, the following will briefly introduce the drawings required for use in the specific implementation. In all the drawings, each element or part is not necessarily drawn according to the actual scale.

[0034] Figure 1 A flow chart of a method for calculating a fault alarm threshold value of a main transformer cooling system provided by an embodiment of the present invention;

[0035] Figure 2 A flow chart of a main transformer cooling system fault alarm method provided by an embodiment of the present invention;

[0036] Figure 3 A system block diagram of a main transformer cooling system fault alarm threshold calculation system provided by an embodiment of the present invention;

[0037] Figure 4 A system block diagram of a main transformer cooling system fault alarm system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0039] like Figure 1 The flowchart of the main transformer cooling system fault alarm threshold calculation method shown in FIG. 1 includes:

[0040] Step 1, obtaining characteristic variable data and operating condition characteristic data of multiple wind turbine generator sets within a monitoring period, wherein the characteristic variable data reflects the performance of the main transformer cooling system, and the operating condition characteristic data reflects the operating condition of the wind turbine generator set;

[0041] Step 2: Statistically analyze the characteristic variable data of each wind turbine generator set to determine the characteristic variable mean a1 corresponding to each wind turbine generator set;

[0042] Step 3: classify all wind turbine generator sets according to the operating condition characteristic data to determine wind turbine generator sets belonging to the same operating condition;

[0043] Step 4: Statistically analyze the characteristic variable data of the wind turbine generator sets under the same working conditions to determine the characteristic variable mean a2 corresponding to the wind turbine generator sets under the same working conditions;

[0044] Step 5: Determine the fault alarm threshold corresponding to the wind turbine generator set according to the corresponding characteristic variable mean a1 and characteristic variable mean a2.

[0045] Specifically, first, the characteristic variable data of the main transformer cooling system during the monitoring period and the operating characteristic data of the wind turbine can be monitored by using the existing sensors of each wind turbine in the wind farm. Then, the characteristic variable data of each wind turbine can be statistically analyzed to determine the characteristic variable mean a1 corresponding to each characteristic variable of the main transformer cooling system of each wind turbine. The specific calculation formula is as follows:

[0046]

[0047] Among them, x n is the characteristic variable data corresponding to the characteristic variable, and N is the number of characteristic variable data corresponding to the characteristic variable.

[0048] Afterwards, based on the operating condition characteristic data, an existing classification algorithm, such as a clustering algorithm, can be used to classify the operating conditions of all wind turbines to determine wind turbines with the same or similar operating conditions. Then, the characteristic variable data of wind turbines with the same or similar operating conditions are statistically analyzed to determine the characteristic variable mean a2 corresponding to the wind turbines with the same operating conditions. The specific calculation formula is as follows:

[0049]

[0050] Where M is the number of wind turbines with the same or similar operating conditions, N m is the number of characteristic variable data corresponding to the mth wind turbine generator set in the wind turbine generator sets with the same or similar working conditions, It is the characteristic variable data corresponding to the mth wind turbine among the wind turbines with the same or similar operating conditions.

[0051] Finally, the fault alarm threshold corresponding to the wind turbine can be determined by combining the characteristic variable mean a2 corresponding to the wind turbines under the same working conditions and the characteristic variable mean a1 corresponding to the main transformer cooling system of each wind turbine. The specific calculation formula is as follows:

[0052]

[0053] in, are the characteristic variable mean a1 and characteristic variable mean a2 corresponding to wind turbine l, and L is the number of wind turbines in the wind farm.

[0054] After determining the fault alarm threshold corresponding to the main transformer cooling system, the mean difference d corresponding to each wind turbine generator set can be l Compared with the fault alarm threshold α, if the mean difference d l If the alarm condition corresponding to the fault alarm threshold α is met, an alarm signal is output. Otherwise, no alarm signal is issued. In this way, the abnormality of the cooling system of the main transformer of the wind turbine can be accurately judged without adding new sensors and monitoring devices, which reduces the equipment investment cost. At the same time, through the early alarm of faults, it helps the wind farm to achieve predictive maintenance, reduces the difficulty of fault inspection and troubleshooting, and reduces the loss of wind turbine power generation.

[0055] In this embodiment, optionally, the characteristic variable data includes: transformer coolant inlet temperature, transformer U-phase winding temperature, transformer V-phase winding temperature, transformer W-phase winding temperature and / or transformer coolant inlet pressure.

[0056] Specifically, during alarm monitoring, the existing sensors of each wind turbine in the wind farm can be used to simultaneously monitor the rotor speed of the wind turbine, the active power of the power grid, and the characteristic variable data corresponding to characteristic variables such as the transformer coolant inlet temperature of the main transformer cooling system, the transformer U-phase winding temperature, the transformer V-phase winding temperature, the transformer W-phase winding temperature and / or the transformer coolant inlet pressure.

[0057] Among them, the rotor speed directly reflects the efficiency of the rotor in capturing wind energy and the mechanical operating status, and the grid active power directly reflects the power generation capacity and operating status of the wind turbine. The rotor speed and grid active power reflect the working conditions of the wind turbine from the perspective of mechanical energy and electrical energy output respectively. By monitoring the changes in these two parameters, we can fully understand the operating status and performance characteristics of the wind turbine. The working conditions of wind turbines can be classified according to these working condition indicators to ensure the accuracy of the working condition classification of wind turbines.

[0058] The inlet temperature of the transformer coolant directly reflects the efficiency of the cooling system in absorbing heat from the main transformer. The inlet pressure of the transformer coolant is often directly related to the performance of the cooling water pump in the cooling system. The temperature of the transformer U-phase winding, the temperature of the transformer V-phase winding, and the temperature of the transformer W-phase winding and their uniformity directly reflect the cooling system's effect on the heat dissipation inside the transformer. By making alarm judgments based on the fault alarm thresholds corresponding to each characteristic variable, the operating status of the main transformer cooling system can be comprehensively evaluated, thereby effectively judging whether there is a fault in the main transformer cooling system and avoiding underreporting.

[0059] In this embodiment, optionally, characteristic variable data and operating condition characteristic data of multiple wind turbine generator sets are obtained, including:

[0060] The characteristic variable data and the operating condition characteristic data are preprocessed to remove abnormal data in the characteristic variable data and the operating condition characteristic data.

[0061] Specifically, after obtaining the characteristic variable data and operating condition characteristic data of the wind turbine generator set within the monitoring period, in order to improve the alarm accuracy, the obtained characteristic variable data and operating condition characteristic data can be preprocessed to eliminate data with obviously low quality and data with obvious faults.

[0062] In this embodiment, optionally, preprocessing the characteristic variable data and the operating condition characteristic data includes:

[0063] Delete the repeated characteristic variable data and operating condition characteristic data at all time points, as well as the characteristic variable data and operating condition characteristic data with missing values;

[0064] Eliminate the characteristic variable data and operating condition characteristic data corresponding to the time point when the wind turbine generator set is not generating electricity;

[0065] The characteristic variable data and the operating condition characteristic data corresponding to the time point when the characteristic variable data exceeds the preset range are deleted.

[0066] Specifically, when preprocessing the characteristic variable data and the operating condition characteristic data. First, the characteristic variable data and the operating condition characteristic data corresponding to all time points in the monitoring time period can be traversed, and the characteristic variable data and the operating condition characteristic data corresponding to the time points with duplicates and missing values ​​can be deleted. Then, the active power of the power grid corresponding to the remaining time points is traversed. If the active power of the power grid is less than 0, it means that the wind turbine is not generating electricity at this time, and the corresponding characteristic variable data and the operating condition characteristic data at this time are deleted. Finally, the characteristic variable data corresponding to the remaining time points are traversed. If the characteristic variable data obviously exceeds the preset range, for example, at a certain time point, the transformer coolant inlet temperature, the transformer U-phase winding temperature, the transformer V-phase winding temperature, and the transformer W-phase winding temperature exceed 250 degrees Celsius or are lower than negative 100 degrees Celsius, then the characteristic variable data and the operating condition characteristic data corresponding to the time point are deleted.

[0067] In this embodiment, optionally, all wind turbine generator sets are classified according to the operating condition characteristic data, including:

[0068] According to the operating condition characteristic data, the DBSCAN algorithm is used to classify the operating condition of each wind turbine generator set, and wind turbine generator sets belonging to the same operating condition are determined.

[0069] Specifically, the DBSCAN algorithm can be used to classify the working conditions of each wind turbine in the wind farm. Specifically, the rotor speed of the wind turbine and the active power of the power grid are used as samples, and a sample set is constructed based on the samples corresponding to all wind turbines in the wind farm.

[0070] First set the radius to r and the minimum cluster number to R. Randomly select a sample y in the sample set i , calculate the number of samples n that fall within the radius r i , if n i >R, then it is considered that sample y i It is the core point i Otherwise, it is considered that sample y i is a non-core point, and randomly select samples again to find the core point. Repeat this process until a core point is found. If no core point is found after traversing all samples, then all samples y i They are all classified into the noise set o and cannot be clustered.

[0071] If the core point s is found i , at the core point s i Create a cluster c a , add all samples within radius r to cluster c a Then for cluster c a Except for the core point s i All points except for cluster c are examined in turn. a The sample points s in j is also a core point, then the sample point s j All samples within radius r are added to cluster c a Repeat the steps until all samples are classified into clusters, and the classification of the working conditions of each wind turbine generator set can be completed. In each cluster, the distance between each sample and the surrounding samples is calculated, and the 5 closest samples can be selected as the units with similar working conditions to the sample. For samples that are not classified into any cluster, they are classified into the noise set and do not participate in the threshold calculation.

[0072] In this embodiment, optionally, determining a fault alarm threshold corresponding to the wind turbine generator set includes:

[0073] According to the corresponding characteristic variable mean a1 and characteristic variable mean a2, the average value of the mean difference corresponding to the wind turbine generator set is calculated, and combined with the preset fixed value, the fault alarm threshold corresponding to the wind turbine generator set is determined.

[0074] Specifically, considering the volatility of different characteristic variables and the sensitivity of adjusting the threshold, a fixed value is added to the mean difference to adjust the strictness of the threshold and reduce false alarms. The fault alarm thresholds of different characteristic variables are adjusted in different ways.

[0075] Among them, the specific calculation formulas for the fault alarm thresholds corresponding to the transformer coolant inlet temperature, transformer U-phase winding temperature, transformer V-phase winding temperature, and transformer W-phase winding temperature are as follows:

[0076]

[0077] Among them, β is a fixed value corresponding to the transformer coolant inlet temperature, the transformer U-phase winding temperature, the transformer V-phase winding temperature, and the transformer W-phase winding temperature, which can be set to 6.

[0078] The specific calculation formula for the low-pressure fault alarm threshold corresponding to the transformer coolant inlet pressure is as follows:

[0079]

[0080] The specific calculation formula for the high-voltage fault alarm threshold corresponding to the transformer coolant inlet pressure is as follows:

[0081]

[0082] Among them, χ is the fixed value corresponding to the transformer coolant inlet pressure, which can be set to 2.

[0083] like Figure 2 The flowchart of the main transformer cooling system fault alarm method shown in the figure includes: using the above-mentioned main transformer cooling system fault alarm threshold calculation method to determine the fault alarm threshold corresponding to the wind turbine generator set.

[0084] Specifically, the above fault alarm threshold calculation method can be used to calculate the fault alarm threshold corresponding to each wind turbine in the wind farm. Then, it can be determined whether each wind turbine meets the alarm condition corresponding to the fault alarm threshold according to the corresponding characteristic variable data. If it meets the alarm condition, an alarm signal is issued, otherwise, no alarm signal is issued.

[0085] For example, the mean difference corresponding to the transformer coolant inlet temperature, the transformer U-phase winding temperature, the transformer V-phase winding temperature or the transformer W-phase winding temperature is compared with the corresponding fault alarm threshold α′. If the mean difference d l If the fault alarm threshold is exceeded, an alarm signal is issued.

[0086] Alternatively, the mean difference corresponding to the transformer coolant inlet pressure is compared with the corresponding fault alarm thresholds α″ and α″′, respectively. If it is less than the fault alarm threshold α″ or exceeds the fault alarm threshold α″′, an alarm signal is issued.

[0087] In this embodiment, optionally, it also includes: determining whether the operating condition characteristic data and characteristic variable data exceed a set threshold value, and outputting an alarm signal in response to the operating condition characteristic data and characteristic variable data exceeding the set threshold value.

[0088] Specifically, when the operating condition characteristic data and characteristic variable data are preprocessed, the grid active power and transformer coolant inlet temperature corresponding to all time points in the monitoring time period can be traversed. If the grid active power exceeds the set threshold, and the transformer coolant inlet temperature exceeds the set threshold, and the temperature of the transformer three-phase winding is lower than the set threshold, an alarm signal is directly issued without further calculating the early warning threshold, and an alarm signal is issued in a timely manner.

[0089] like Figure 3 The system block diagram of the main transformer cooling system fault alarm threshold calculation system shown in the figure, the calculation system includes:

[0090] A characteristic data acquisition module is configured to acquire characteristic variable data and operating condition characteristic data of multiple wind turbine generator sets within a monitoring period, wherein the characteristic variable data reflects the performance of the main transformer cooling system, and the operating condition characteristic data reflects the operating condition of the wind turbine generator set;

[0091] The unit mean value calculation module is configured to perform statistical analysis on the characteristic variable data of each wind turbine generator set respectively to determine the characteristic variable mean value a1 corresponding to each wind turbine generator set;

[0092] A unit operating condition classification module is configured to classify all wind turbine generator sets according to the operating condition characteristic data, and determine wind turbine generator sets belonging to the same operating condition;

[0093] The working condition mean value calculation module is configured to perform statistical analysis on the characteristic variable data of the wind turbine generator sets under the same working condition, and determine the characteristic variable mean value a2 corresponding to the wind turbine generator sets under the same working condition;

[0094] The alarm threshold calculation module is configured to determine the fault alarm threshold corresponding to the wind turbine generator set according to the corresponding characteristic variable mean a1 and characteristic variable mean a2.

[0095] Specifically, the calculation system includes a characteristic data acquisition module, a unit mean value calculation module, a unit operating condition classification module, an operating condition mean value calculation module and an alarm threshold value calculation module. Among them, the characteristic data acquisition module can use the existing sensors of each wind turbine in the wind farm to monitor the characteristic variable data of the main transformer cooling system during the monitoring period, as well as the operating condition characteristic data of the wind turbine. Then, the unit mean value calculation module can perform statistical analysis on the characteristic variable data of each wind turbine, and determine the characteristic variable mean value corresponding to the main transformer cooling system of each wind turbine.

[0096] The unit operating condition classification module can classify the operating conditions of all wind turbines based on the operating condition characteristic data using existing classification algorithms, such as clustering algorithms, to determine wind turbines with the same or similar operating conditions. The operating condition mean value calculation module can statistically analyze the characteristic variable data of wind turbines with the same or similar operating conditions to determine the mean of the characteristic variables corresponding to the wind turbines with the same operating conditions. The alarm threshold calculation module can combine the mean of the characteristic variables corresponding to the wind turbines with the same operating conditions and the mean of the characteristic variables corresponding to the main transformer cooling system of each wind turbine to determine the fault alarm threshold corresponding to the wind turbine.

[0097] like Figure 4 The system block diagram of the main transformer cooling system fault alarm system shown in FIG. 1 includes:

[0098] A threshold calculation module is configured to use the above-mentioned main transformer cooling system fault alarm threshold calculation method to determine the fault alarm threshold corresponding to the wind turbine generator set;

[0099] Specifically, the alarm system includes a threshold calculation module and a fault alarm module, wherein the threshold calculation module uses the above-mentioned fault alarm threshold calculation method to calculate the fault alarm threshold corresponding to each wind turbine in the wind farm. The fault alarm module can determine whether each wind turbine meets the alarm condition corresponding to the fault alarm threshold according to the corresponding characteristic variable data, and if so, an alarm signal is issued, otherwise, no alarm signal is issued.

[0100] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for calculating the fault alarm threshold of a main transformer cooling system, characterized in that: include: Acquire characteristic variable data and operating condition characteristic data of multiple wind turbine generator sets within a monitoring period, wherein the characteristic variable data reflects the performance of the main transformer cooling system, and the operating condition characteristic data reflects the operating condition of the wind turbine generator set; Perform statistical analysis on the characteristic variable data of each wind turbine generator set to determine the characteristic variable mean a1 corresponding to each wind turbine generator set; Classifying all wind turbine generator sets according to the operating condition characteristic data to determine wind turbine generator sets belonging to the same operating condition; Perform statistical analysis on the characteristic variable data of wind turbine generator sets under the same working conditions respectively, and determine the characteristic variable mean a2 corresponding to the wind turbine generator sets under the same working conditions; According to the corresponding characteristic variable mean a1 and characteristic variable mean a2, the fault alarm threshold corresponding to the wind turbine generator set is determined.

2. The method for calculating the fault alarm threshold of the main transformer cooling system according to claim 1 is characterized in that: The characteristic variable data include: transformer coolant inlet temperature, transformer U-phase winding temperature, transformer V-phase winding temperature, transformer W-phase winding temperature and / or transformer coolant inlet pressure.

3. The method for calculating the fault alarm threshold of the main transformer cooling system according to claim 1 is characterized in that: Obtain characteristic variable data and operating condition characteristic data of multiple wind turbines, including: The characteristic variable data and the operating condition characteristic data are preprocessed to remove abnormal data in the characteristic variable data and the operating condition characteristic data.

4. The method for calculating the fault alarm threshold of the main transformer cooling system according to claim 3 is characterized in that: Preprocessing the characteristic variable data and the operating condition characteristic data includes: Delete the repeated characteristic variable data and operating condition characteristic data at all time points, as well as the characteristic variable data and operating condition characteristic data with missing values; Eliminate the characteristic variable data and operating condition characteristic data corresponding to the time point when the wind turbine generator set is not generating electricity; The characteristic variable data and the operating condition characteristic data corresponding to the time point when the characteristic variable data exceeds the preset range are deleted.

5. The method for calculating the fault alarm threshold of the main transformer cooling system according to claim 1 is characterized in that: All wind turbine generator sets are classified according to the operating condition characteristic data, including: According to the operating condition characteristic data, the DBSCAN algorithm is used to classify the operating condition of each wind turbine generator set, and wind turbine generator sets belonging to the same operating condition are determined.

6. The method for calculating the fault alarm threshold of the main transformer cooling system according to claim 1, characterized in that: Determining a fault alarm threshold corresponding to the wind turbine generator set includes: According to the corresponding characteristic variable mean a1 and characteristic variable mean a2, the average value of the mean difference corresponding to the wind turbine generator set is calculated, and combined with the preset fixed value, the fault alarm threshold corresponding to the wind turbine generator set is determined.

7. A main transformer cooling system fault alarm method, characterized in that: include: The fault alarm threshold calculation method for the main transformer cooling system as described in any one of claims 1 to 6 is used to determine the fault alarm threshold corresponding to the wind turbine generator set.

8. The main transformer cooling system fault alarm method according to claim 7, characterized in that: Also includes: It is determined whether the operating condition characteristic data and the characteristic variable data exceed a set threshold value, and in response to the operating condition characteristic data and the characteristic variable data exceeding the set threshold value, an alarm signal is output.

9. A main transformer cooling system fault alarm threshold calculation system, characterized in that: include: A characteristic data acquisition module is configured to acquire characteristic variable data and operating condition characteristic data of multiple wind turbine generator sets within a monitoring period, wherein the characteristic variable data reflects the performance of the main transformer cooling system, and the operating condition characteristic data reflects the operating condition of the wind turbine generator set; The unit mean value calculation module is configured to perform statistical analysis on the characteristic variable data of each wind turbine generator set respectively to determine the characteristic variable mean value a1 corresponding to each wind turbine generator set; A unit operating condition classification module is configured to classify all wind turbine generator sets according to the operating condition characteristic data, and determine wind turbine generator sets belonging to the same operating condition; The working condition mean value calculation module is configured to perform statistical analysis on the characteristic variable data of the wind turbine generator sets under the same working condition, and determine the characteristic variable mean value a2 corresponding to the wind turbine generator sets under the same working condition; The alarm threshold calculation module is configured to determine the fault alarm threshold corresponding to the wind turbine generator set according to the corresponding characteristic variable mean a1 and characteristic variable mean a2.

10. A main transformer cooling system fault alarm system, characterized in that: include: The threshold calculation module is configured to use the main transformer cooling system fault alarm threshold calculation method as described in any one of claims 1-6 to determine the fault alarm threshold corresponding to the wind turbine generator set.

Images