Wind power station fan health state assessment method and computer equipment

Abstract

The invention belongs to the technical field of wind power generation, and particularly relates to a wind power station fan health state assessment method and computer equipment. The method comprises the following steps: acquiring the generated power of a to-be-detected fan under the current working condition, if the generated power is within the longest power interval under the corresponding working condition, judging that the health state of the fan is good, otherwise, judging that the health state of the fan is not good; wherein the longest power interval is determined according to historical generated power data of the draught fan, the interval serving as the longest power interval meets the longest power interval judgment condition, and the interval with the interval width smaller than that of the interval does not meet the longest power interval judgment condition; the longest power interval judgment condition comprises that the ratio of the number of the historical generated power data in the interval to the total number of the historical generated power data is larger than a longest power interval proportion threshold value. On the premise of ensuring the fan health state evaluation precision, the evaluation steps are simplified, and the evaluation efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wind power generation, and in particular relates to a method for evaluating the health status of a wind turbine in a wind power station and a computer device. Background Art

[0002] Wind turbine generator sets usually operate continuously all year round in an open-air environment at an altitude of 80 meters. Affected by the external environment, compared with general ground equipment, wind turbines not only have a high failure rate, but are also difficult to detect and deal with in the first time.

[0003] In order to timely evaluate and detect the health status of wind turbines in wind power stations, the Chinese invention patent application with application publication number CN115788797A proposed a technical solution that inputs the status data of wind turbine operation into a prediction model to predict the normal power of the wind turbine, and then compares the current power with the predicted normal power to calculate the health of the wind turbine. However, the construction of the prediction model in this solution requires the use of a large number of samples, and the use of samples to construct a fitting function or train a neural network to obtain a prediction model is cumbersome and complicated, resulting in low prediction efficiency. Summary of the invention

[0004] The purpose of the present invention is to provide a method and computer equipment for evaluating the health status of a wind turbine in a wind power station, so as to solve the technical problem that when a prediction model is used to predict the health status of a wind turbine, the construction steps of the prediction model are relatively complicated, which affects the efficiency of the prediction of the health status of the wind turbine.

[0005] In order to solve the above technical problems, the present invention provides a method for evaluating the health status of a wind turbine in a wind power station, comprising: Obtain the power generation of the wind turbine under the current working condition. If the power generation is within the longest power interval under the corresponding working condition, the health state of the wind turbine is determined to be good; otherwise, the health state of the wind turbine is determined to be not good. Among them, the longest power interval is determined based on the historical power generation data of the wind turbine. The interval serving as the longest power interval meets the longest power interval judgment condition, and the interval with a smaller interval width than this interval does not meet the longest power interval judgment condition; the longest power interval judgment condition includes: the ratio of the number of historical power generation data falling in this interval to the total number of historical power generation data is greater than the longest power interval proportion threshold.

[0006] Furthermore, the longest power interval determination condition also includes: the ratio is the largest under the interval width and there is no other interval with the same ratio as that of the interval.

[0007] Furthermore, if the current power generation of the wind turbine to be tested is greater than the upper limit of the longest power interval under the corresponding working condition and less than the maximum power generation, or the current power generation is less than the lower limit of the longest power interval and greater than the minimum power generation under the corresponding working condition, the health status of the wind turbine is determined to be medium; If the current power generation of the wind turbine to be tested is greater than the maximum power generation under the corresponding working condition, or the current power generation is less than the minimum power generation under the corresponding working condition, the health status of the wind turbine is judged to be poor; Among them, the upper limit of the longest power interval is less than the maximum power generation power, and the lower limit of the longest power interval is greater than the minimum power generation power.

[0008] Furthermore, the current operating conditions include the current wind speed level and wind turbine model.

[0009] Furthermore, if the health status of the fan to be tested is medium and the duration is greater than the set time threshold, an alarm will be issued to prompt immediate maintenance; if the health status of the fan to be tested is medium and the duration is less than or equal to the set time threshold, a prompt will be given that maintenance is required.

[0010] Furthermore, when it is determined that the health status of the fan to be tested is poor, an alarm is issued to prompt immediate maintenance.

[0011] The present invention is a pioneering invention, and its beneficial effects are as follows: after studying the historical power generation data of the wind turbine, the present invention found that: under any working condition, the power generation data distribution of the wind turbine is approximately normal distribution, the data is relatively scattered near the maximum and minimum values, and the data is relatively concentrated near the median. Based on this, the wind power station wind turbine health status assessment method of the present invention finds an interval with a smaller width and the most concentrated data in the interval from the historical power generation data of the wind turbine under the corresponding working condition, which is the longest power interval, so that the longest power interval determined is located near the median; using the longest power interval to judge the health status of the wind turbine, it is only necessary to judge whether the power generation of the wind turbine to be tested falls within the longest power interval. Therefore, the wind power station wind turbine health status assessment method of the present invention does not need to establish a prediction model, and eliminates the steps of construction, training and verification when establishing the prediction model. Therefore, the prediction method of the present invention simplifies the assessment steps and improves the assessment efficiency while ensuring the accuracy of the wind turbine health status assessment.

[0012] To solve the above technical problems, the present invention further provides a computer device, including a processor, which implements the steps of the wind power station wind turbine health status assessment method of the present invention when executing a computer program.

[0013] The present invention is a pioneering invention, and its beneficial effects are as follows: after studying the historical power generation data of the wind turbine, the present invention found that: under any working condition, the distribution of the power generation data of the wind turbine is approximately normal distribution, the data is relatively scattered near the maximum and minimum values, and the data is relatively concentrated near the median. Based on this, the computer device of the present invention searches for an interval with a smaller width and the most concentrated data in the interval from the historical power generation data of the wind turbine under the corresponding working condition, which is the longest power interval, so that the longest power interval determined is located near the median; using the longest power interval to judge the health status of the wind turbine, it is only necessary to judge whether the power generation of the wind turbine to be tested falls within the longest power interval. Therefore, the computer device of the present invention, the processor does not need to establish a prediction model when executing a computer program, eliminating the steps of construction, training and verification when establishing the prediction model, simplifying the evaluation steps and improving the evaluation efficiency while ensuring the accuracy of the wind turbine health status evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a flow chart of a method for evaluating the health status of wind turbines in a wind power station according to a method embodiment of the present invention. DETAILED DESCRIPTION

[0015] After studying the historical power generation data of the wind turbine, the present invention found that: under any working condition, the power generation data distribution of the wind turbine is approximately normal distribution, the data is relatively scattered near the maximum and minimum values, and the data is relatively concentrated near the median. Based on this, the wind power station wind turbine health status assessment method and computer equipment of the present invention find an interval with a smaller width and the most concentrated data in the interval from the historical power generation data of the wind turbine under the corresponding working condition, which is the longest power interval, so that the longest power interval determined in this way is located near the median; using this longest power interval to judge the health status of the wind turbine, it is only necessary to judge whether the power generation of the wind turbine to be tested falls within the longest power interval. There is no need to establish a prediction model, which saves the steps of construction, training and verification when establishing the prediction model. Under the premise of ensuring the accuracy of the wind turbine health status assessment, the assessment steps are simplified and the assessment efficiency is improved.

[0016] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0017] Method Example: By analyzing the historical power generation data of the wind turbine, it can be found that the distribution of the historical power generation data of the wind turbine is approximately normal distribution. The power generation of the wind turbine in a good health state is relatively dense and concentrated near the median, while the power generation of the wind turbine in an unhealthy state is scattered near the maximum and minimum values. Based on this, the wind power station wind turbine health status assessment method of the present invention determines the longest power interval based on the historical power generation data of the wind turbine, and uses the longest power interval to assess the health status of the wind turbine to be tested. Figure 1 As shown, the following steps are included: Step 1: Determine the longest power range and the maximum and minimum values ​​of the wind turbine power generation under the same operating conditions based on historical data.

[0018] According to historical data, the power generation sample values ​​of the normal operating wind turbines under different working conditions are sorted out, and the longest power range, maximum value and minimum value of the power generation under different working conditions in the sample values ​​are determined. The upper limit of the longest power range is less than the maximum power generation, and the lower limit of the longest power range is greater than the minimum power generation.

[0019] The different working conditions in this embodiment include the model of the fan and the wind speed, wherein the wind speed is divided into different levels according to the set step size. In this embodiment, the step size is set to 0.2 m / s, and the wind speed is divided into V1, V2, ... Vn, a total of n levels.

[0020] Assume that the wind turbine model under a certain working condition is M, the wind speed is Vi, and the historical data of the wind turbine power generation under this working condition are P1, P2, P3...Pm, a total of m data. Determine the maximum value Pmax and the minimum value Pmin, as well as the longest power interval [Plmin, Plmax] from the m data, where Plmin is the lower limit of the longest power interval and Plmax is the upper limit of the longest power interval.

[0021] The longest power interval of the present invention should satisfy the longest power interval determination condition, and there is no interval satisfying the longest power interval determination condition under a smaller interval width. The longest power interval determination condition includes: the ratio of the number of historical power generation data falling in the interval to the total number of historical power generation data is greater than the longest power interval proportion threshold.

[0022] The longest power interval that meets the above conditions must be an interval with the smallest width and the highest density of historical power generation data. Only by using such a longest power interval to evaluate the health status of the wind turbine can the accuracy of the evaluation be guaranteed.

[0023] There may be more than one interval that meets the above conditions. In order to determine the longest power interval with higher evaluation accuracy, the longest power interval determination condition in this embodiment also includes: the ratio is the largest under the interval width and there is no other interval with the same ratio as the interval. By adding the longest power interval determination condition, it is ensured that the optimal longest power interval is found. As other implementation methods, the longest power interval determination condition may not include the conditions newly added in this embodiment, and any one of the multiple intervals that meet the longest power interval determination condition can be selected as the longest power interval.

[0024] The reference value range of the longest power interval percentage threshold is set to 70%-80%. In this embodiment, the longest power interval percentage threshold is set to 75%. The smaller the longest power interval percentage threshold is set, the higher the prediction sensitivity is, but the possibility of false alarms will also increase. According to this principle, the specific value of the longest power interval percentage threshold can be set according to actual needs.

[0025] In this embodiment, the longest power interval can be determined by the following process: 1. According to the initial interval width, traverse the historical data under the corresponding working conditions to find out whether there is a longest power interval with a width equal to the initial interval width.

[0026] In order to improve the accuracy of the prediction of the health status of the wind turbine, the width of the longest power interval should be as small as possible. Therefore, the initial interval width is a small value, and the longest power interval with the smallest width is preferentially searched. If a certain interval can be found according to the initial width, and the number of data in the interval accounts for more than 75% of the total number of historical data (the longest power interval proportion threshold), then this interval is the longest power interval.

[0027] If the longest power interval that meets the requirements cannot be found according to the initial interval width, then enter process 2 to continue searching.

[0028] 2. Increase the original interval width by the set step size to obtain a new interval width, and find the longest power interval under this interval width.

[0029] If the longest power interval that meets the requirements is still not found under the interval width, process 2 is executed again to continue searching until the longest power interval that meets the requirements is found, and the iteration is stopped.

[0030] 3. After finding the interval that meets the requirements through the above process. If there are at least two intervals that meet the requirements of the longest power interval under a certain interval width, the interval with the largest ratio of the number of data in the interval to the total number of historical data is taken as the longest power interval. If there are at least two intervals that meet the requirements of the longest power interval under a certain interval width, and the ratio of the number of data in the interval to the total number of historical data is also equal, then further increase the interval width, and find the interval that meets the requirements and has the largest ratio of the number of data in the interval to the total number of historical data as the longest power interval.

[0031] Step 2: Obtain the real-time operating conditions and power generation of the wind turbine.

[0032] The real-time operating conditions of the wind turbine to be tested and the power generation under the corresponding conditions are obtained through the actual data collected. In this embodiment, the real-time operating conditions include the motor model and wind speed, where the wind speed is expressed in the form of wind speed level.

[0033] Step 3: Evaluate the health status of the wind turbine based on the real-time operating conditions, power generation and parameters determined in step 1 of the wind turbine to be tested.

[0034] 1. Determine whether the real-time power generation of the wind turbine to be tested is greater than the maximum power generation value under the working condition or less than the minimum power generation value under the working condition.

[0035] If Pi>Pmax, or Pi<Pmin, it indicates that the health status of the fan to be tested is poor, and an alarm prompts that immediate maintenance is required, ending the evaluation process; otherwise, the subsequent evaluation process continues to evaluate the health status of the fan.

[0036] Among them, Pi is the real-time power generation of the wind turbine under the current operating conditions, Pmax is the maximum power generation under the corresponding conditions, and Pmin is the minimum power generation under the corresponding conditions.

[0037] 2. Determine whether the real-time power generation of the wind turbine to be tested is outside the longest power range.

[0038] If Pi>Plmax, or Pi<Plmin, the real-time power generation of the wind turbine to be tested is outside the longest power range, which means that the health status of the wind turbine to be tested is medium at this moment, and the evaluation process is terminated. At this time, the health status of the fan is medium and the duration is less than or equal to the set time threshold t0, indicating that maintenance is required; if the health status of the wind turbine to be tested is continuously evaluated as medium, when the duration is greater than the set time threshold t0, an alarm prompts that immediate maintenance is required.

[0039] If Plmin≤Pi≤Plmax, and the power generation of the wind turbine to be tested is within the longest power range, it means that the wind turbine to be tested is in good health, and the evaluation process ends.

[0040] Pi is the real-time power generation of the wind turbine under the current operating conditions, Plmax is the upper limit of the longest power interval under the corresponding conditions, and Plmin is the upper limit of the longest power interval under the corresponding conditions.

[0041] Computer equipment example: A computer device of the present invention includes a processor, and the processor implements the steps of the method described in the method embodiment when executing a computer program. The specific process, principle and beneficial effects of the method have been described in detail in the method embodiment, and will not be repeated in this embodiment.

[0042] The processor may be a microprocessor MCU, a programmable logic device FPGA or other processing device.

[0043] The wind turbine health status assessment method and computer device of the wind power station of the present invention use the longest power interval to assess the health status of the wind turbine, and the determined longest power interval is the interval with the smallest width that meets the longest power interval judgment condition. The power generation of the wind turbine in good health mostly falls within this interval. Therefore, by comparing this interval with the current power generation of the wind turbine to be tested, the health status of the wind turbine to be tested can be assessed. The assessment method is simple, the assessment efficiency is high, and the accuracy of the assessment can be guaranteed. In this embodiment, in order to find the longest power interval with the highest assessment accuracy, "the ratio under the width of the interval is the largest and there is no other interval with the same ratio as the interval" is added as the longest power interval judgment condition, and the best interval that meets the condition can be found as the longest power interval, thereby improving the assessment accuracy. Furthermore, in this embodiment, the maximum and minimum power generation values ​​in the historical power generation data are used to further subdivide the health status of the wind turbine to be tested, and different enhancements and alarms are made after the subdivision, so that the wind power station operation and maintenance personnel can understand the health status and maintenance of the wind turbine in time.

Claims

1. A method for evaluating the health status of a wind turbine in a wind power station, characterized in that: include: Obtain the power generation of the wind turbine under the current working condition. If the power generation is within the longest power interval under the corresponding working condition, the health state of the wind turbine is determined to be good; otherwise, the health state of the wind turbine is determined to be not good. Among them, the longest power interval is determined based on the historical power generation data of the wind turbine. The interval serving as the longest power interval meets the longest power interval judgment condition, and the interval with a smaller interval width than this interval does not meet the longest power interval judgment condition; the longest power interval judgment condition includes: the ratio of the number of historical power generation data falling in this interval to the total number of historical power generation data is greater than the longest power interval proportion threshold.

2. The wind turbine health status assessment method of a wind power station according to claim 1, characterized in that: The longest power interval determination condition also includes: the ratio is the largest under the interval width and there is no other interval with the same ratio as that of the interval.

3. The wind turbine health status assessment method of a wind power station according to claim 1, characterized in that: If the current power generation of the wind turbine to be tested is greater than the upper limit of the longest power interval under the corresponding working condition and less than the maximum power generation, or the current power generation is less than the lower limit of the longest power interval and greater than the minimum power generation under the corresponding working condition, the health status of the wind turbine is determined to be medium; If the current power generation of the wind turbine to be tested is greater than the maximum power generation under the corresponding working condition, or the current power generation is less than the minimum power generation under the corresponding working condition, the health status of the wind turbine is judged to be poor; Among them, the upper limit of the longest power interval is less than the maximum power generation power, and the lower limit of the longest power interval is greater than the minimum power generation power.

4. The method for evaluating the health status of a wind turbine in a wind power station according to any one of claims 1 to 3, characterized in that: The current operating conditions include the current wind speed level and fan model.

5. The wind turbine health status assessment method of a wind power station according to claim 3, characterized in that: If the health status of the fan to be tested is medium and the duration is greater than the set time threshold, an alarm will be issued to prompt immediate maintenance; if the health status of the fan to be tested is medium and the duration is less than or equal to the set time threshold, a prompt will be issued that maintenance is required.

6. The wind turbine health status assessment method of a wind power station according to claim 3, characterized in that: When it is determined that the health status of the fan to be tested is poor, an alarm will be issued to prompt immediate maintenance.

7. A computer device comprising a processor, characterized in that: The processor implements the steps of the method for evaluating the health status of a wind turbine in a wind power station according to any one of claims 1 to 6 when executing the computer program.

Images