Method and device for evaluating running state of lubricating grease of main bearing of wind turbine generator

Abstract

The invention relates to the technical field of wind power generation, in particular to a wind turbine generator main bearing lubricating grease operation state evaluation method and device, and the method comprises the steps: 1, obtaining the impact factor content of to-be-evaluated main bearing lubricating grease; step 2, obtaining an influence factor quantitative data set in a plurality of historical samples of the main bearing lubricating grease, verifying whether the influence factor quantitative data set accords with normal distribution, and if yes, executing step 3; if not, executing the step 4; 3, comparing the influence factor content of the to-be-evaluated main bearing lubricating grease with a first evaluation threshold value to obtain an evaluation result; and 4, comparing the influence factor content of the to-be-evaluated main bearing lubricating grease with a second evaluation threshold to obtain an evaluation result. The accuracy and reliability of main bearing lubricating grease operation detection state evaluation can be improved.

Description

Technical Field

[0001] This invention relates to the field of wind power generation technology, specifically to a method and apparatus for evaluating the operating status of lubricating grease in the main bearing of a wind turbine. Background Technology

[0002] Currently, the operation and testing of grease in the main bearings of wind turbine generators mainly refer to the National Energy Administration standard NB / T 10111, which is primarily used to assess the performance of the grease inside the bearing. However, with the iterative optimization of bearing structures, the indicators for assessing the operating condition of the main bearing grease proposed in the original National Energy Administration standard NB / T 10111 deviate significantly from the actual bearing operation monitoring results. Furthermore, an increasing number of wind farm owners are using this standard to assess the internal wear condition of the main bearings. Because the standard values ​​are severely inconsistent with the actual wear conditions of the main bearings, the effectiveness and accuracy of wind farm main bearing operation and maintenance are low, leading to over-maintenance and waste of maintenance resources, which is detrimental to the normal operation and maintenance work of wind farm maintenance personnel. Summary of the Invention

[0003] The purpose of this invention is to provide a method and apparatus for evaluating the operating status of grease in the main bearing of a wind turbine, which can improve the accuracy and reliability of evaluating the operating status of the grease in the main bearing.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present invention provides a method for evaluating the operating condition of the lubricating grease in the main bearing of a wind turbine, comprising:

[0006] Step 1: Obtain quantitative data on the influencing factors of the main bearing grease to be evaluated;

[0007] Step 2: Obtain the quantitative dataset of influencing factors from multiple historical samples of main bearing grease, and verify whether the quantitative dataset of influencing factors conforms to a normal distribution. If yes, proceed to Step 3; otherwise, proceed to Step 4.

[0008] Step 3: Compare the quantitative data of the influencing factors of the main bearing grease to be evaluated with the first evaluation threshold to obtain the evaluation result; the first evaluation threshold a i =μ+i×σ, i=1,2,3,4,μ is the mean of the impact factor quantification dataset, and σ is the standard deviation of the impact factor quantification dataset;

[0009] Step four: Compare the quantitative data of the influencing factors of the main bearing grease to be evaluated with the second evaluation threshold to obtain the evaluation result; the second evaluation threshold is the quantitative data of the influencing factors when the cumulative distribution probability is a preset value, and the second evaluation threshold is greater than the mean of the quantitative dataset of the influencing factors.

[0010] Furthermore, the influencing factors include iron (Fe), copper (Cu), silicon (Si), and the quantitative index of ferromagnetic particles (PQ).

[0011] Furthermore, it also includes: Step 5: Take the lowest level among the evaluation results obtained for each influencing factor as the evaluation result of the operating status of the grease in the main bearing of the wind turbine. The lower the level, the worse the operating status of the grease.

[0012] Furthermore, the second evaluation threshold in step four includes b1 and b2, where b1 is the quantitative data of the impact factor corresponding to the cumulative distribution probability being the first preset value, and b2 is the quantitative data of the impact factor corresponding to the cumulative distribution probability being the second preset value, wherein the second preset value is greater than the first preset value.

[0013] Furthermore, the first preset value is 90%, and the second preset value is 98%.

[0014] Furthermore, the quantitative data x of the influencing factors of the main bearing grease to be evaluated are compared with the first evaluation threshold a. i The comparison yields the following evaluation results:

[0015] When x > a4, the evaluation result is Grade 1;

[0016] When a3 < x ≤ a4, the evaluation result is grade two;

[0017] When a2 < x ≤ a3, the evaluation result is level three;

[0018] When a1 < x ≤ a2, the evaluation result is level four;

[0019] When x≤a1, the evaluation result is level five;

[0020] The quantitative data x of the influencing factors of the main bearing grease to be evaluated is compared with the second evaluation threshold to obtain the evaluation results, which specifically include:

[0021] When x > b2, the evaluation result is grade two;

[0022] When b1 < x ≤ b2, the evaluation result is level three;

[0023] When x≤b1, the evaluation result is level four;

[0024] The operating status of the lubricating grease is ranked from worst to best as follows: Level 1, Level 2, Level 3, Level 4, and Level 5.

[0025] Furthermore, it also includes: after obtaining the evaluation results, updating the quantitative data of the influencing factors of the main bearing grease to be evaluated to the historical samples, and recalculating the first evaluation threshold or the second evaluation threshold.

[0026] Secondly, the present invention provides an evaluation device for the operating condition of grease in the main bearing of a wind turbine, comprising:

[0027] The first acquisition module is used to acquire quantitative data on the influencing factors of the main bearing grease to be evaluated.

[0028] The second acquisition module is used to acquire a quantitative dataset of influence factors from multiple historical samples of main bearing grease.

[0029] The validation module is used to verify whether the impact factor quantification dataset conforms to a normal distribution.

[0030] The first comparison module is used to compare the quantitative data of the influencing factors of the main bearing grease to be evaluated with a first evaluation threshold to obtain the evaluation result; the first evaluation threshold a i =μ+i×σ, i=1,2,3,4,μ is the mean of the impact factor quantification dataset, and σ is the standard deviation of the impact factor quantification dataset;

[0031] The second comparison module is used to compare the quantitative data of the influencing factors of the main bearing grease to be evaluated with the second evaluation threshold to obtain the evaluation result; the second evaluation threshold is the quantitative data of the influencing factors when the cumulative distribution probability is a preset value, and the second evaluation threshold is greater than the mean of the quantitative dataset of influencing factors.

[0032] Furthermore, it also includes a determination module, which uses the lowest level among the evaluation results obtained from each influencing factor as the evaluation result of the operating condition of the grease in the main bearing of the wind turbine. The lower the level, the worse the operating condition of the grease.

[0033] Furthermore, it also includes an update module, which, after obtaining the evaluation results, updates the quantitative data of the influencing factors of the main bearing grease to be evaluated to the historical samples and recalculates the first evaluation threshold or the second evaluation threshold.

[0034] The present invention has the following unexpected beneficial effects:

[0035] This invention performs statistical analysis on a quantitative dataset of influencing factors from multiple historical samples of main bearing grease, and uses probability density distribution and cumulative probability distribution to delineate the boundary lines of various indicators, thereby obtaining the evaluation results of the operating status of wind turbine main bearing grease, which improves the accuracy and reliability of the evaluation of the operating status of main bearing grease. Attached Figure Description

[0036] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention.

[0037] Figure 1 This is a flowchart illustrating the method for evaluating the operating status of the lubricating grease in the main bearing of a wind turbine according to an embodiment of the present invention.

[0038] Figure 2 This is a schematic diagram of the probability density distribution of Fe element content provided as an example of the present invention. Detailed Implementation

[0039] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0040] In one embodiment, the present invention discloses a method for evaluating the operating status of lubricating grease in the main bearing of a wind turbine, see [link to relevant documentation]. Figure 1 As shown, the method includes:

[0041] Step 1: Obtain quantitative data on the influencing factors of the main bearing grease to be evaluated;

[0042] Step 2: Obtain the quantitative dataset of influencing factors from multiple historical samples of main bearing grease, and verify whether the quantitative dataset of influencing factors conforms to a normal distribution. If yes, proceed to Step 3; otherwise, proceed to Step 4.

[0043] Step 3: Compare the quantitative data of the influencing factors of the main bearing grease to be evaluated with the first evaluation threshold to obtain the evaluation result; the first evaluation threshold a i =μ+i×σ, i=1,2,3,4,μ is the mean of the impact factor quantification dataset, and σ is the standard deviation of the impact factor quantification dataset;

[0044] Step four: Compare the quantitative data of the influencing factors of the main bearing grease to be evaluated with the second evaluation threshold to obtain the evaluation result; the second evaluation threshold is the quantitative data of the influencing factors when the cumulative distribution probability is a preset value, and the second evaluation threshold is greater than the mean of the quantitative dataset of the influencing factors.

[0045] This invention performs statistical analysis on a quantitative dataset of influencing factors from multiple historical samples of main bearing grease, and uses probability density distribution and cumulative probability distribution to delineate the boundary lines of various indicators, thereby obtaining the evaluation results of the operating status of wind turbine main bearing grease, which improves the accuracy and reliability of the evaluation of the operating status of main bearing grease.

[0046] The existing NB / T 10111 standard uses a fixed threshold for operational status assessment, which cannot cope with changes in operating conditions after bearing structural optimization (such as new sealing designs and material upgrades). This invention analyzes the normal distribution characteristics of historical sample data and sets differentiated assessment thresholds: normally distributed data uses a dynamic interval constructed based on the mean and standard deviation, while non-normal data uses the quantified data value corresponding to a preset distribution probability as the standard. This allows the threshold to be adjusted synchronously with bearing technology iterations, effectively avoiding assessment bias caused by standard lag.

[0047] Furthermore, by accurately assessing the actual operating status of the lubricating grease, this invention can reasonably extend the maintenance cycle, reduce unnecessary lubricating grease consumption, labor input, and equipment downtime, thereby lowering operation and maintenance costs. Simultaneously, by combining real-time assessment results, it can dynamically adjust maintenance operations such as grease replenishment and maintenance timing: reducing grease supply when no replenishment is needed, and strengthening monitoring when key attention is required, avoiding resource idleness or waste caused by a one-size-fits-all approach to operation and maintenance, and improving the utilization efficiency of operation and maintenance resources.

[0048] It should be noted that the influencing factors described in this invention are characteristic parameters directly related to wear, such as the proportion of metal particle types and abrasive particle morphology correlation factors, forming a multi-dimensional diagnostic system to avoid one-sided judgments caused by a single indicator and improve the accuracy of wear condition identification.

[0049] In a preferred embodiment of the present invention, the influencing factors include iron (Fe), copper (Cu), silicon (Si), and the quantitative index of ferromagnetic particles (PQ).

[0050] Iron (Fe) is a major component of core load-bearing components such as rolling elements (balls and raceways) and spindles in bearings. Changes in its content directly reflect the degree of fatigue wear and abrasive wear in the rolling contact area. Compared to traditional standards that only focus on grease performance, the introduction of Fe can directly relate to the health status of the core moving parts of the bearing, avoiding the problem of overlooking internal metal wear due to normal grease indicators.

[0051] Copper (Cu) primarily originates from non-rolling components such as bearing cages (made of copper alloy) and bushings. An abnormally high Cu content can precisely indicate problems such as cage wear (e.g., friction caused by cage breakage or deformation) and abnormal bushing clearance. These types of failures are often not directly reflected in the Fe content, making them easy to miss by traditional standards. Coordinated monitoring of Cu and Fe can differentiate between rolling element wear and non-rolling element wear, avoiding misdiagnosing cage failure as rolling element wear and providing clear component location information for subsequent maintenance.

[0052] Silicon (Si) is not a metallic component of the bearing itself. Its main sources are external dust, impurities that seep in due to seal failure, and residual particles from the grease production process. When Si content is abnormal, the cause can be directly traced to external factors such as seal system failure and environmental dust intrusion, rather than bearing fatigue wear itself. If Si is high but Fe and Cu are normal, maintenance should focus on replacing seals and cleaning the lubrication system, without disassembling the bearing. If Si and Fe / Cu levels rise simultaneously, it indicates that contamination has caused severe wear, and the source of contamination should be addressed before repairing worn components. Traditional standards do not include contaminants like Si, which can easily lead to misjudging contamination damage as bearing failure, resulting in excessive disassembly or incorrect maintenance direction.

[0053] Compared to simple Fe element content, the ferromagnetic particle quantitative index (PQ) offers a more comprehensive ability to quantify wear. Fe content only reflects the total amount of iron and cannot distinguish particle size. The PQ index, through magnetic field sensing, can simultaneously quantify the total amount and volume distribution of ferromagnetic particles. When the proportion of large particles is high, the PQ value spikes significantly, providing earlier warnings of fatal failures such as severe spalling and rolling element breakage. Furthermore, the PQ index is more sensitive to low concentrations of ferromagnetic particles than conventional elemental analysis, enabling early detection of minor wear and preventing traditional standards from missing warning windows due to insufficient Fe content.

[0054] As a preferred embodiment of the present invention, it further includes: Step 5: taking the lowest level among the evaluation results obtained for each influencing factor as the evaluation result of the operating status of the grease in the main bearing of the wind turbine, the lower the level, the worse the operating status of the grease.

[0055] The operating condition of the main bearing grease is affected by multiple factors, and a severe abnormality in any one factor can directly lead to fatal failures such as bearing seizure and spalling. The risk of such failures does not decrease even if other factors are normal. This step, by adopting the strict principle of taking the lowest level, can directly identify the weakest risk point. Even if most factors are at the normal level, as long as one factor has the lowest assessment level, the final conclusion will be based on that low level. This logic completely avoids the problem of missing individual serious hidden dangers due to a majority of normal conditions. It is especially suitable for the high-risk characteristic of wind turbine main bearing failure leading to high downtime losses, providing a rigid guarantee for the safe operation of equipment.

[0056] In a preferred embodiment of the present invention, the second evaluation threshold in step four includes b1 and b2, where b1 is the quantitative data of the impact factor corresponding to the cumulative distribution probability being a first preset value, and b2 is the quantitative data of the impact factor corresponding to the cumulative distribution probability being a second preset value, wherein the second preset value is greater than the first preset value.

[0057] Step four addresses the quantification of influencing factors that do not conform to a normal distribution. This type of data often exhibits narrow concentration intervals and dispersed extreme values. If a single threshold is used, either the threshold is too strict, causing normal fluctuations to be misjudged as abnormal, or the threshold is too wide, missing key risks. The dual threshold design of b1 and b2 anchors the slightly deviated and severely deviated intervals of the data distribution through two different cumulative distribution probabilities, making the threshold more closely match the actual distribution pattern of non-normal data.

[0058] The settings of b1 and b2 essentially establish a three-level state classification: normal, warning, and fault. Below b1 is normal, between b1 and b2 is a warning, and above b2 is a fault, breaking the limitations of the traditional binary judgment of either good or bad based on a single threshold. When the quantitative data of the impact factor of the sample to be evaluated is between b1 and b2 (warning), immediate shutdown and maintenance are not required. Preventive measures such as increasing monitoring frequency and replenishing clean grease can be taken to avoid excessive maintenance. When the content is above b2 (fault), an emergency warning is triggered, requiring immediate shutdown and inspection, such as disassembling bearings to check for wear and replacing seals to block contamination and prevent the fault from worsening.

[0059] It should be noted that the non-normal distribution characteristics of different influencing factors are different. The dual threshold design of b1 and b2 can adapt to the abnormal patterns of different factors by adjusting the preset cumulative distribution probability. There is no need to change the threshold setting logic. Only the probability parameters need to be fine-tuned to meet the evaluation requirements of different influencing factors. This further enhances the universality of the present invention for different wind farm operating conditions and different bearing structures, and reduces the adaptation cost of the method implementation.

[0060] In a preferred embodiment of the present invention, the first preset value is 90%, and the second preset value is 98%.

[0061] The 90% cumulative probability (b1) corresponds to the upper limit of most normal data. In non-normal datasets, about 90% of normal operating samples will be below this value, and only 10% of samples may slightly exceed b1 due to slight fluctuations (such as short-term environmental dust or occasional uneven mixing of lubricating grease). This threshold can accurately distinguish between normal and slight deviations, avoiding misjudging normal fluctuations as abnormal.

[0062] The 98% cumulative probability (b2) corresponds to the critical line of extreme anomalies. Only 2% of the samples will be higher than this value. Such samples are almost always accompanied by substantial problems. This can completely eliminate the interference of random fluctuations and avoid misjudging serious anomalies as minor deviations.

[0063] In a preferred embodiment of the present invention, the quantitative data x of the influencing factor of the main bearing grease to be evaluated is compared with the first evaluation threshold a. i The comparison yields the following evaluation results:

[0064] When x > a4, the evaluation result is Grade 1;

[0065] When a3 < x ≤ a4, the evaluation result is grade two;

[0066] When a2 < x ≤ a3, the evaluation result is level three;

[0067] When a1 < x ≤ a2, the evaluation result is level four;

[0068] When x≤a1, the evaluation result is level five; the operating status of the lubricating grease from poor to good corresponds to the following levels: level one, level two, level three, level four and level five.

[0069] Normally distributed data exhibits more regular fluctuations and a more continuous data distribution. The five-level classification can finely distinguish the gradient differences between extremely poor (Level 1), fault (Level 2), warning (Level 3), normal (Level 4), and excellent (Level 5). This fine-grained classification can accurately identify the gradual deterioration of the grease's condition. For example, a drop from Level 4 to Level 3, although not reaching a fault state, indicates a slight decline in condition, allowing for early initiation of preventative maintenance and avoiding emergency treatment only when it reaches Level 1, thus extending the effective life of the grease.

[0070] The quantitative data x of the influencing factor of the main bearing grease to be evaluated is compared with the second evaluation threshold to obtain the evaluation results, which specifically include: when x > b2, the evaluation result is level two; when b1 < x ≤ b2, the evaluation result is level three; when x ≤ b1, the evaluation result is level four.

[0071] Non-normal data often exhibits a pattern of majority concentration and minority extremes. The three-level classification eliminates redundant subdivisions and directly focuses on faults (Level 2), warnings (Level 3), and normal conditions (Level 4). This risk-oriented classification avoids misjudgments due to minor fluctuations caused by excessive subdivision.

[0072] Meanwhile, this preferred implementation clearly defines the correspondence between lower levels and worse conditions, and provides clear threshold ranges for different levels, completely solving the pain points of traditional assessments where indicator values ​​need to be converted and it is difficult to judge the condition. Maintenance personnel do not need to analyze complex statistical principles; they only need to compare the data to be assessed (x) with the first or second assessment threshold to directly determine the level. For example, seeing level one indicates that the lubricating grease condition is extremely poor, requiring an emergency shutdown and grease replacement; seeing level five indicates the optimal condition, requiring no additional maintenance.

[0073] The lower the level number, the worse it is, which aligns with the intuitive understanding of industrial scenarios. For example, a level one warning indicates the highest risk, avoiding operational errors caused by the level number being inversely related to the status. This intuitiveness significantly reduces the judgment cost for maintenance personnel and is especially suitable for rapid on-site decision-making.

[0074] As a preferred embodiment of the present invention, it further includes: after obtaining the evaluation results, updating the quantitative data of the influence factors of the main bearing grease to be evaluated to the historical samples, and recalculating the first evaluation threshold or the second evaluation threshold.

[0075] The accuracy of the threshold depends on the representativeness and data volume of historical samples. Traditional standards, when established, have limited sample sizes and often consist of laboratory or specific wind field data, making it difficult to cover all industry conditions. Similarly, when this invention is initially applied to a single wind field, the historical sample size is also small, potentially leading to biases in the initial threshold calculation. This implementation achieves two major optimizations by continuously updating the sample: First, as the number of evaluations increases, the historical sample size expands (e.g., from an initial 100 groups to 1000 groups), making the statistical results (mean, standard deviation, cumulative probability) closer to the true distribution, avoiding threshold bias due to insufficient sample size. Second, if a batch of data to be evaluated is an occasional anomaly, after updating the sample, the recalculated threshold will smooth out the occasional error through statistical regularity, preventing significant threshold fluctuations due to a single abnormal sample and ensuring threshold stability.

[0076] The following analysis and explanation will be based on specific examples.

[0077] 1. The following is an example of calculating the probability density distribution of Fe element based on the latest oil and fat testing data sample library. The process for calculating the probability density distribution of Cu, PQ, and Si elements is similar:

[0078] (1) Assume that the latest oil testing data sample library contains a total of 1,000 samples, that is, there are 1,000 oil testing data samples;

[0079] (2) Extract the Fe element from these 1000 oil test data samples and statistically analyze them in ascending order to obtain the statistical results. Assume that among them, 50 samples have a Fe element test result of 100 mg / kg, 50 samples have a Fe element test result of 200 mg / kg, 100 samples have a Fe element test result of 500 mg / kg, 400 samples have a Fe element test result of 1000 mg / kg, 200 samples have a Fe element test result of 1200 mg / kg, 100 samples have a Fe element test result of 2000 mg / kg, 50 samples have a Fe element test result of 4000 mg / kg, and 50 samples have a Fe element test result of 6000 mg / kg.

[0080] (3) The probability of the number of samples containing each Fe content relative to the 1000 samples of oil and fat test data was calculated, as shown in Table 1:

[0081] Table 1. Distribution probability of Fe content

[0082] Fe content probability 100 0.05 200 0.05 500 0.10 1000 0.40 1200 0.20 2000 0.10 4000 0.05 6000 0.05

[0083] (4) Draw a probability density distribution map of Fe element based on the probability of the number of samples containing each Fe content relative to the number of samples in 1000 oil and fat test data samples, such as Figure 1 As shown, the probability density distribution of Fe element can be obtained.

[0084] 2. Based on the probability density distributions of the four elements Fe, Cu, PQ and Si, determine whether the probability density distributions of the four elements Fe, Cu, PQ and Si follow a normal distribution.

[0085] Continuing with the previous example, such as Figure 2 As shown, based on the probability density distribution curve, it can be seen that the probability density distribution of Fe element does not follow a normal distribution. At this time, it is necessary to calculate the content value of Fe element at the cumulative distribution probability of 90% and 98%.

[0086] 3. Based on the judgment results of whether the probability density distributions of the four elements Fe, Cu, PQ and Si follow a normal distribution, the content values ​​of the four elements Fe, Cu, PQ and Si under different cumulative distribution probabilities are calculated respectively.

[0087] (1) Preferably, if the probability density distributions of the four elements Fe, Cu, PQ and Si respectively follow a normal distribution, then calculate the content values ​​of the four elements Fe, Cu, PQ and Si respectively at the cumulative distribution probabilities of 68.27.% (μ+1×σ), 94.45% (μ+2×σ), 99.73% (μ+3×σ) and 99.99% (μ+4×σ).

[0088] Specifically, ① if the probability density distribution of Fe follows a normal distribution, then calculate the content values ​​of Fe at cumulative distribution probabilities of 68.27%, 94.45%, 99.73%, and 99.99%.

[0089] ②If the probability density distribution of Cu follows a normal distribution, calculate the content of Cu at cumulative distribution probabilities of 68.27%, 94.45%, 99.73%, and 99.99%.

[0090] ③ If the probability density distribution of Si follows a normal distribution, calculate the content values ​​of Si at cumulative distribution probabilities of 68.27%, 94.45%, 99.73%, and 99.99%.

[0091] ④ If the probability density distribution of the PQ index follows a normal distribution, calculate the content values ​​of the PQ index at cumulative distribution probabilities of 68.27%, 94.45%, 99.73%, and 99.99%.

[0092] (2) If the probability density distributions of the four elements Fe, Cu, PQ and Si do not follow a normal distribution, calculate the content values ​​of the four elements Fe, Cu, PQ and Si at cumulative distribution probabilities of 90% and 98% respectively.

[0093] Specifically, ① if the probability density distribution of Fe does not follow a normal distribution, then calculate the content of Fe at cumulative distribution probabilities of 90% and 98%.

[0094] ②If the probability density distribution of Cu does not follow a normal distribution, calculate the content of Cu at cumulative distribution probabilities of 90% and 98%.

[0095] ③ If the probability density distribution of Si element does not follow a normal distribution, calculate the content of Si element at cumulative distribution probabilities of 90% and 98%.

[0096] ④ If the probability density distribution of the PQ index does not follow a normal distribution, then calculate the content values ​​of the PQ index at cumulative distribution probabilities of 90% and 98%.

[0097] Continuing with the previous example, the probability density distribution of Fe does not follow a normal distribution. The content values ​​of Fe at a cumulative probability distribution of 90% and at a cumulative probability distribution of 98% are calculated respectively and used as the dividing threshold.

[0098] As shown in Table 1, the Fe content value is 2000 when the cumulative probability distribution is 90%, and the Fe content value is 6000 when the cumulative probability distribution is 98%.

[0099] 4. Based on the content values ​​of Fe, Cu, PQ and Si under different cumulative distribution probabilities, determine the respective boundary standards for Fe, Cu, PQ and Si.

[0100] Continuing with the previous example, the boundary standards for Fe element are determined based on the content values ​​corresponding to the cumulative probability of Fe at 90% and 98%, namely b1 = 2000 and b2 = 6000. Based on 2000 and 6000 as the boundary lines, the operating status of the main bearing grease is divided into normal, warning, and fault.

[0101] When x > 6000, the evaluation result is Level 2, i.e., fault; when 2000 < x ≤ 6000, the evaluation result is Level 3, i.e., warning; when x ≤ 2000, the evaluation result is Level 4, i.e. normal.

[0102] By comparing these parameters, the state information of Fe in the grease of the main bearing to be evaluated can be obtained. Similarly, the state information of Cu, Si, and PQ in the grease of the main bearing to be evaluated can be obtained.

[0103] 5. Based on the state information of the four elements Fe, Cu, PQ and Si in the grease of the wind turbine main bearing, determine the evaluation result of the operating state of the wind turbine main bearing.

[0104] Specifically, the most abnormal state among the state information of Fe element, Cu element, Si element, and PQ index in the grease of the wind turbine main bearing is used as the evaluation result of the operating state of the grease of the wind turbine main bearing.

[0105] For example, assuming that the Fe element in the grease of the wind turbine main bearing follows a normal distribution, and the state information of the iron element is excellent; the copper element in the grease does not follow a normal distribution, and the state information of the copper element is normal; the silicon element in the grease follows a normal distribution, and the state information of the silicon element is good; the PQ index in the grease does not follow a normal distribution, and the state information of the PQ index is normal. Then, the evaluation result of the operating state of the wind turbine main bearing grease is the state with the highest abnormality level among the four elements Fe, Cu, PQ, and Si in the grease, i.e., normal.

[0106] In one embodiment, the present invention discloses an evaluation device for the operating status of grease in the main bearing of a wind turbine, comprising:

[0107] The first acquisition module is used to acquire quantitative data on the influencing factors of the main bearing grease to be evaluated.

[0108] The second acquisition module is used to acquire a quantitative dataset of influence factors from multiple historical samples of main bearing grease.

[0109] The validation module is used to verify whether the impact factor quantification dataset conforms to a normal distribution.

[0110] The first comparison module is used to compare the quantitative data of the influencing factors of the main bearing grease to be evaluated with a first evaluation threshold to obtain the evaluation result; the first evaluation threshold a i =μ+i×σ, i=1,2,3,4,μ is the mean of the impact factor quantification dataset, and σ is the standard deviation of the impact factor quantification dataset;

[0111] The second comparison module is used to compare the quantitative data of the influencing factors of the main bearing grease to be evaluated with the second evaluation threshold to obtain the evaluation result; the second evaluation threshold is the quantitative data of the influencing factors when the cumulative distribution probability is a preset value, and the second evaluation threshold is greater than the mean of the quantitative dataset of influencing factors.

[0112] In a preferred embodiment of the present invention, the device further includes a determining module, which is used to take the lowest level among the evaluation results obtained for each influencing factor as the evaluation result of the operating status of the grease in the main bearing of the wind turbine. The lower the level, the worse the operating status of the grease.

[0113] In a preferred embodiment of the present invention, the device further includes an update module, which, after obtaining the evaluation results, updates the quantitative data of the influence factors of the main bearing grease to be evaluated to the historical samples and recalculates the first evaluation threshold or the second evaluation threshold.

[0114] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention. Furthermore, the embodiments of the present invention and the features thereof can be combined with each other unless otherwise specified.

Claims

1. A method of evaluating the operating condition of a grease of a main bearing of a wind turbine, characterized in that The method comprises the following steps: Step 1: obtaining the content of influencing factors of the main bearing grease to be evaluated; Step 2: obtaining the influencing factor quantitative data set in the plurality of historical samples of the main bearing grease, and verifying whether the influencing factor quantitative data set conforms to the normal distribution, if yes, executing Step 3; if not, executing Step 4; In step three, the impact factor content of the main bearing grease to be evaluated is compared with the first evaluation threshold to obtain an evaluation result; the first evaluation threshold a i = μ + i x σ, i = 1, 2, 3, 4, μ is the mean of the impact factor quantization data set, and σ is the standard deviation of the impact factor quantization data set. Step 4: comparing the content of the influencing factors of the main bearing grease to be evaluated with the second evaluation threshold to obtain an evaluation result; the second evaluation threshold is the influencing factor quantitative data corresponding to a preset value of the cumulative distribution probability, and the second evaluation threshold is greater than the mean value of the influencing factor quantitative data set.

2. A method of evaluating the operating condition of a grease of a main bearing of a wind turbine generator according to claim 1, characterized in that: The influencing factors include iron Fe, copper Cu, silicon Si and ferromagnetic particle quantitative index PQ.

3. A method of evaluating the operating condition of a grease of a main bearing of a wind turbine generator according to claim 2, characterized in that, Further comprising Step 5: taking the lowest grade in the evaluation result corresponding to each influencing factor as the evaluation result of the running state of the main bearing grease of the wind turbine generator, and the lower the grade, the worse the running state of the grease.

4. The method of assessing the operating condition of the grease of the main bearing of a wind turbine generator according to claim 1, characterized in that: The second evaluation threshold in Step 4 comprises b1 and b2, b1 is the influencing factor quantitative data corresponding to a first preset value of the cumulative distribution probability, and b2 is the influencing factor quantitative data corresponding to a second preset value of the cumulative distribution probability, the second preset value being greater than the first preset value.

5. A method of assessing the operating condition of a grease of a main bearing of a wind turbine generator according to claim 4, characterized in that: The preset value is 90%, and the second preset value is 98%.

6. A method of assessing the operating condition of a grease of a main bearing of a wind turbine generator according to claim 4, characterized in that: quantifying data x of an influencing factor of the main bearing grease to be evaluated and a first evaluation threshold a i performing a comparison, obtaining an evaluation result, specifically comprising: When x>a4, the evaluation result is grade one; When a3 When a2 When a1 When x≤a1, the evaluation result is grade five. Comparing the influencing factor quantitative data x of the main bearing grease to be evaluated with the second evaluation threshold to obtain an evaluation result specifically comprises: When x>b2, the evaluation result is grade two; When b1 When x≤b1, the evaluation result is grade four. The grades corresponding to the running state of the grease from bad to good are grade one, grade two, grade three, grade four and grade five in turn.

7. A method of evaluating the operating condition of a grease of a main bearing of a wind turbine generator according to claim 1, characterized in that, Further comprising: After obtaining the evaluation result, updating the influencing factor quantitative data of the main bearing grease to be evaluated to the historical samples, and recalculating the first evaluation threshold or the second evaluation threshold.

8. An apparatus for evaluating the operating condition of a grease of a main bearing of a wind turbine, characterized in that The method comprises the following steps: The first obtaining module is configured to obtain the influencing factor quantitative data of the main bearing grease to be evaluated; The second obtaining module is configured to obtain the influencing factor quantitative data set in the plurality of historical samples of the main bearing grease, The verification module is configured to verify whether the influencing factor quantitative data set conforms to the normal distribution; The first comparison module is configured to compare the impact factor quantization data of the main bearing grease to be evaluated with a first evaluation threshold to obtain an evaluation result; the first evaluation threshold a i = μ + i x σ, i = 1, 2, 3, 4, μ is the mean of the impact factor quantization data set, and σ is the standard deviation of the impact factor quantization data set. The second comparison module is configured to compare the influencing factor quantitative data of the main bearing grease to be evaluated with the second evaluation threshold to obtain an evaluation result; the second evaluation threshold is the influencing factor quantitative data corresponding to a preset value of the cumulative distribution probability, and the second evaluation threshold is greater than the mean value of the influencing factor quantitative data set.

9. The apparatus for evaluating the operating condition of the grease of the main bearing of a wind turbine generator according to claim 8, characterized in that: Further comprising a determination module configured to take the lowest grade in the evaluation result corresponding to each influencing factor as the evaluation result of the running state of the main bearing grease of the wind turbine generator, and the lower the grade, the worse the running state of the grease.

10. The apparatus for evaluating the operating condition of the grease of the main bearing of a wind turbine generator according to claim 8, characterized in that: The application further comprises an updating module for updating the impact factor quantitative data of the main bearing grease to be evaluated into the historical sample after obtaining the evaluation result, and recalculating the first evaluation threshold or the second evaluation threshold.

Images