A method, device, equipment and medium for determining unit equipment anomaly

Abstract

The application discloses a kind of determination method, device, equipment and medium of unit equipment exception, comprising: collecting multiple sets of historical data in first preset time period of unit equipment, and each set of historical data at least includes the first parameter value corresponding to first parameter and the second parameter value corresponding to second parameter;Data cleaning is carried out to multiple sets of historical data, and obtain multiple sets of candidate data;Multiple sets of candidate data are carried out smoothing processing, and generate multiple sets of target data;Equipment exception measurement standard is generated based on all target data;Obtain multiple sets of operating data in second preset time period of target object, and each set of operating data at least includes the third parameter value corresponding to first parameter and the fourth parameter value corresponding to second parameter;Based on equipment exception measurement standard, and each set of operating data at least includes the third parameter value corresponding to first parameter and the fourth parameter value corresponding to second parameter, determine target mean value;According to target mean value, determine whether unit equipment is abnormal.

Description

Technical Field

[0001] This invention relates to the field of wind power technology, specifically to a method, apparatus, equipment, and medium for determining abnormalities in turbine equipment. Background Technology

[0002] In a wind farm, the operating status of the turbine equipment affects the operation of the entire wind farm. Therefore, whether the turbine equipment is operating normally is particularly important.

[0003] Currently, there are many different methods in the industry for determining abnormalities in generator set equipment. Most of them are based on a constant single threshold. However, such methods can lead to a high false alarm rate and a large error, resulting in inaccurate calculations. Furthermore, these methods cannot pinpoint the specific cause of the abnormalities in the generator set equipment, making it difficult to provide accurate troubleshooting and maintenance recommendations for on-site personnel.

[0004] Therefore, how to accurately determine whether the status of the unit equipment is abnormal is an urgent problem to be solved. Summary of the Invention

[0005] Therefore, the technical problem to be solved by the present invention is to overcome the defects of the prior art in that the judgment of abnormality of unit equipment is inaccurate and the cause of abnormality cannot be determined, thereby providing a method, device, equipment and medium for determining abnormality of unit equipment.

[0006] In a first aspect, the present invention provides a method for determining unit equipment anomalies, comprising: collecting multiple sets of historical data of the unit equipment within a first preset time period, each set of historical data including at least a first parameter value corresponding to a first parameter and a second parameter value corresponding to a second parameter; cleaning the multiple sets of historical data to obtain multiple sets of candidate data; smoothing the multiple sets of candidate data to generate multiple sets of target data; generating equipment anomaly measurement standards based on all target data; acquiring multiple sets of operating data of the target object within a second preset time period, each set of operating data including at least a third parameter value corresponding to a first parameter and a fourth parameter value corresponding to a second parameter; determining a target mean based on the equipment anomaly measurement standards and the fact that each set of operating data includes at least the third parameter value corresponding to a first parameter and the fourth parameter value corresponding to a second parameter; and determining whether the unit equipment is abnormal based on the target mean.

[0007] This invention generates multiple sets of target data by cleaning and smoothing multiple sets of historical data collected within a first preset time period. Based on all the target data, it generates equipment anomaly measurement standards. The target mean is determined using these standards and multiple sets of operational data collected within a second preset time period. The relationship between the target mean and a preset threshold is then used to determine whether the unit equipment is abnormal. In this invention, historical data is first cleaned to remove contaminated data, resulting in multiple sets of candidate data with finer granularity. Then, the candidate data is smoothed, addressing the problem of mismatched parameter values ​​within the same set of data due to data lag. This makes the generated target data more objective and closer to real data, thereby improving the objectivity and accuracy of the equipment anomaly measurement standards. After determining the equipment anomaly measurement standards, the target mean is determined based on the difference between the measurement standards for each set of operational data collected within the second preset time period. Since each set of operational data objectively exists during unit operation, the improved accuracy of the equipment anomaly measurement standards indirectly ensures the accuracy of the target mean, thus guaranteeing the accuracy of the result determined based on the target mean and the preset threshold.

[0008] In conjunction with the first aspect, in the first embodiment of the first aspect, smoothing multiple sets of candidate data to generate multiple sets of target data specifically includes: using a sliding window mean method to smooth multiple sets of candidate data to generate multiple sets of target data.

[0009] This method smooths candidate data by taking the average of a sliding window. By acquiring data from a fixed window and smoothing it using the averaging method, it effectively solves the problem of mismatch between parameter values ​​corresponding to multiple parameters caused by data lag. This makes the final target data more consistent with the actual situation and improves the accuracy of equipment anomaly measurement standards.

[0010] In conjunction with the first aspect, in the second embodiment of the first aspect, the target data includes the fifth parameter value corresponding to the first parameter and the sixth parameter value corresponding to the second parameter. The device anomaly measurement standard is generated based on all the target data, specifically including: constructing a baseline curve according to the fifth parameter value corresponding to the first parameter and the sixth parameter value corresponding to the second parameter in each group of target data in all the target data, and using the baseline curve as the device anomaly measurement standard.

[0011] This invention generates multiple sets of target data by smoothing candidate data. Based on the fifth parameter value of the first parameter and the sixth parameter value of the second parameter in all target data, a baseline curve for the first and second parameters is constructed. Since the target data is obtained by smoothing after removing dirty data from historical data, it is relatively objective and has reference value. Therefore, the baseline curve formed based on the target data is also reference value. Furthermore, after obtaining the baseline curve, a relatively objective and accurate reference value for the second parameter corresponding to the third parameter value in each set of operating data can be easily obtained from the curve. This provides a relatively objective standard value for the fourth parameter value corresponding to the second parameter in each set of operating data, ensuring the accuracy of the difference between each set of operating data and the baseline curve. This, in turn, ensures the accuracy of determining whether the unit equipment is abnormal based on the difference corresponding to each set of operating data and a preset threshold.

[0012] In conjunction with the first aspect, in the third embodiment of the first aspect, based on the equipment anomaly measurement standard and the fact that each set of operating data includes at least the third parameter value corresponding to the first parameter and the fourth parameter value corresponding to the second parameter, a target mean is determined. Specifically, this includes: obtaining a reference parameter value corresponding to the second parameter from the benchmark curve based on the benchmark curve and the third parameter value corresponding to the first parameter in the first set of operating data, wherein the first set of operating data is any one of multiple sets of operating data; determining the difference corresponding to the second parameter based on the reference parameter value and the fourth parameter value corresponding to the second parameter in each set of operating data; and determining the target mean based on all the differences.

[0013] This invention obtains a reference value for the second parameter by taking the third parameter value corresponding to the first parameter in the baseline curve and the first set of operating data. Since the reference values ​​of each second parameter provided by the baseline curve are objective and accurate, the target difference corresponding to the second parameter is determined based on the reference parameter value and the fourth parameter value corresponding to the second parameter in each set of operating data. This target difference is the actual difference between all operating data and the baseline curve in the second operating period. Therefore, the comparison between the obtained target difference and the preset threshold can be used to more accurately determine whether the unit equipment is abnormal.

[0014] In conjunction with the first aspect, in the fourth embodiment of the first aspect, determining whether the unit equipment is abnormal based on the target average value specifically includes: determining that the unit equipment is abnormal when the target average value is greater than a preset threshold; or determining that the unit equipment is normal when the target average value is less than or equal to the preset threshold.

[0015] In conjunction with the first aspect, in the fifth embodiment of the first aspect, each set of historical data further includes the parameter value of the third parameter and the parameter value of the fourth parameter; each set of operational data further includes the parameter value of the third parameter and the parameter value of the fourth parameter; after determining the unit equipment abnormality based on the target mean, the method further includes: obtaining a first difference based on the parameter values ​​of the third parameter and the fourth parameter in each set of historical data; constructing a first equipment abnormality cause analysis standard based on all the first differences and all the fourth parameter values ​​of the second parameters; and determining the cause of the unit equipment abnormality based on the first equipment abnormality cause analysis standard and all the fourth parameter values ​​and first differences of the second parameters within a second preset time period.

[0016] In conjunction with the first aspect, in the sixth embodiment of the first aspect, each set of historical data further includes the parameter values ​​of the fifth parameter, the sixth parameter, and the seventh parameter; each set of operational data further includes the parameter values ​​of the fifth parameter, the sixth parameter, and the seventh parameter. After determining the unit equipment anomaly based on the target mean, the method further includes: obtaining a second difference based on the fourth parameter value and the fifth parameter value of the second parameter in each set of historical data; obtaining a third difference based on the sixth parameter value and the seventh parameter value of each set of historical data; constructing a second equipment anomaly cause analysis standard based on all the second differences and all the third differences; and determining the cause of the unit equipment anomaly based on the second equipment anomaly cause analysis standard and all the second differences and all the third differences within a second preset time period.

[0017] This invention acquires historical and operational data related to unit anomalies and uses the above-mentioned method to determine the cause of the unit equipment anomalies using this historical and operational data. This facilitates timely maintenance by staff based on the cause of the anomalies, thereby avoiding the defect of not knowing the cause of unit equipment anomalies. Timely maintenance of unit equipment also indirectly improves the working efficiency of unit equipment.

[0018] Secondly, the present invention provides a device for determining unit equipment anomalies, comprising: a first acquisition module for acquiring multiple sets of historical data of the unit equipment within a first preset time period, each set of historical data including at least a first parameter value corresponding to a first parameter and a second parameter value corresponding to a second parameter; a cleaning module for cleaning the multiple sets of historical data to obtain multiple sets of candidate data; a smoothing module for smoothing the multiple sets of candidate data to generate multiple sets of target data; a generation module for generating equipment anomaly measurement standards based on all target data; a second acquisition module for acquiring multiple sets of operating data of the target object within a second preset time period, each set of operating data including at least a third parameter value corresponding to the first parameter and a fourth parameter value corresponding to the second parameter; a first determination module for determining a target mean based on the equipment anomaly measurement standards and the fact that each set of operating data includes at least the third parameter value corresponding to the first parameter and the fourth parameter value corresponding to the second parameter; and a second determination module for determining whether the unit equipment is abnormal based on the target mean.

[0019] Thirdly, the present invention provides a computer device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory is used to store a computer program, and when the computer program is executed by the processor, the processor performs a method for determining an abnormality of the unit equipment as described in any of the claims of the present invention.

[0020] Fourthly, the present invention provides a computer-readable storage medium for storing computer instructions that, when executed by a processor, implement a method for determining unit equipment anomalies as described in any of the invention's contents. Attached Figure Description

[0021] 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 specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 A flowchart illustrating the method for determining unit equipment anomalies provided in an embodiment of the present invention;

[0023] Figure 2 This is a target scatter plot provided in an embodiment of the present invention;

[0024] Figure 3 The reference curve corresponding to the IGBT temperature and the inlet / outlet pressure difference formed by the target scatter points provided in the embodiments of the present invention;

[0025] Figure 4A flowchart for determining the causes of abnormalities in a water-cooled converter provided in an embodiment of the present invention;

[0026] Figure 5 Connection diagram of the device for determining unit equipment abnormalities provided in an embodiment of the present invention;

[0027] Figure 6 A computer device diagram provided for an embodiment of the present invention. Detailed Implementation

[0028] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] This invention discloses a method for determining unit equipment anomalies, such as... Figure 1 As shown, the method specifically includes:

[0030] Step S1: Collect multiple sets of historical data from the unit equipment within the first preset time period.

[0031] Specifically, each set of historical data corresponds to a historical moment in the first preset time period. Each set of historical data includes at least the first parameter value corresponding to the first parameter and the second parameter value corresponding to the second parameter. The first parameter and the second parameter are key reference parameters that can be used to reflect whether the unit equipment is abnormal. The first parameter value and the second parameter value included in each set of historical data are the parameter values ​​corresponding to the first parameter and the second parameter at the same historical moment.

[0032] Specifically, in one embodiment, if it is necessary to determine whether the water-cooled converter in the unit equipment is abnormal, the parameters included in the multiple sets of historical data collected are at least active power and IGBT (Insulated Gate Bipolar Transistor) temperature, because active power and IGBT temperature are key parameters for determining whether the water-cooled converter is abnormal.

[0033] Specifically, the first preset time period can be the initial stage of the unit equipment's initial use, ensuring that the acquired historical data is from when the unit equipment is in its optimal operating period, thus guaranteeing the accuracy and reference value of the historical data. The first preset time period can be six months or one year, etc. Furthermore, before this step, it is necessary to remove historical data from all historical data collected within the first preset time period that indicates the unit equipment is shut down, under maintenance, or in other abnormal operating states. This ensures that the multiple sets of historical data collected in step S1 are all data from when the unit equipment is in operating state. For example, after acquiring all data for the water-cooled converter for one year, the operating data with active power greater than 100kW is taken as the historical data required by this solution. Here, active power greater than 100kW is a necessary condition for the water-cooled converter to be in operating state. Therefore, based on the reference data of 100kW corresponding to this parameter, all data from non-operating states can be removed.

[0034] Step S2: Perform data cleaning on multiple sets of historical data to obtain multiple sets of candidate data.

[0035] Specifically, after collecting multiple sets of historical data under normal operating conditions, the collected historical data needs to be cleaned to remove dirty data with abrupt changes, resulting in multiple sets of candidate data. For example, if there are 100 sets of historical data, the neighborhood range of each set of historical data is determined with the historical data as the center and the given value as the radius (i.e., the circle determined with the historical data as the center and the given value as the radius). Among them, 86 sets of historical data contain more than 25 sets of other historical data in their respective neighborhood ranges, while the remaining 14 sets of historical data contain less than 10 sets of other historical data in their respective neighborhood ranges. Therefore, these 14 sets of historical data are considered dirty data and should be removed from all historical data, with the remaining 86 sets of historical data identified as candidate data.

[0036] Specifically, methods for cleaning historical data include clustering and binning. In this embodiment, clustering can be used to clean historical data. On the one hand, clustering can clean two-dimensional data; on the other hand, clustering cleans data more finely, resulting in smaller granularity of the candidate data.

[0037] Step S3: Smooth the multiple sets of candidate data to generate multiple sets of target data.

[0038] Specifically, data smoothing methods can include the sliding window averaging method, weighted smoothing method, etc.

[0039] Specifically, due to the lag in data collection, there may be mismatches among multiple parameters in a set of data, such as the current power corresponding to the temperature at the previous moment. Therefore, after cleaning the historical data to obtain multiple sets of candidate data, it is necessary to smooth these multiple sets of candidate data to ensure that the target data generated after smoothing is more in line with the actual situation and more referential.

[0040] Step S4: Generate device anomaly measurement criteria based on all target data.

[0041] Specifically, equipment measurement standards can be a curve fitted based on multiple sets of target data, or a mathematical model or mathematical relationship constructed based on multiple sets of target data.

[0042] Step S5: Obtain multiple sets of running data of the target object within the second preset time period.

[0043] Specifically, each set of running data corresponds to a running moment within the second preset time period, and each set of running data includes at least the third parameter value corresponding to the first parameter and the fourth parameter value corresponding to the second parameter.

[0044] Specifically, the second preset time period can be a period of time in the near future, such as the last day or the last ten days.

[0045] Step S6: Based on the equipment anomaly measurement criteria and the fact that each set of operating data includes at least the third parameter value corresponding to the first parameter and the fourth parameter value corresponding to the second parameter, determine the target mean.

[0046] Specifically, based on the equipment anomaly measurement standard and each set of operating data including at least the third parameter value corresponding to the first parameter and the fourth parameter value corresponding to the second parameter, the gap between each set of operating data and the equipment anomaly measurement standard is determined. Based on the gap corresponding to each set of operating data, a target mean is determined. The target mean is used to reflect the average gap between all operating data and the equipment anomaly measurement standard within the second preset time period.

[0047] Step S7: Determine whether the unit equipment is abnormal based on the target average.

[0048] Specifically, the target mean is compared with a preset threshold. The preset threshold represents the maximum distance between the operating data and the equipment anomaly measurement standard. Therefore, the unit equipment can be determined as abnormal based on the comparison result between the target mean and the preset threshold.

[0049] This invention generates multiple sets of target data by cleaning and smoothing multiple sets of historical data collected within a first preset time period. Based on all the target data, it generates equipment anomaly measurement standards. The target mean is determined using these standards and multiple sets of operational data collected within a second preset time period. Finally, the relationship between the target mean and a preset threshold determines whether the unit equipment is abnormal. In this invention, historical data is first cleaned to remove contaminated data, resulting in multiple sets of candidate data with finer granularity. Then, the candidate data is smoothed, addressing the problem of mismatched parameter values ​​within the same set of data due to data lag. This makes the generated target data more objective and closer to real data, thereby improving the objectivity and accuracy of the equipment anomaly measurement standards. After determining the equipment anomaly measurement standards, the target mean is determined based on the difference between the standards for each set of operational data collected within the second preset time period. Since each set of operational data objectively exists during unit operation, the improved accuracy of the equipment anomaly measurement standards indirectly ensures the accuracy of the target mean, thus guaranteeing the accuracy of the result determined based on the target mean and the preset threshold.

[0050] In the following embodiments, the water-cooled converter of the unit equipment is used as an example.

[0051] In one optional embodiment, smoothing multiple sets of candidate data to generate multiple sets of target data specifically includes: using a sliding window mean method to smooth multiple sets of candidate data to generate multiple sets of target data.

[0052] For example, after cleaning multiple sets of historical data collected from the water-cooled converter, multiple sets of candidate data are obtained, each of which includes power data p corresponding to the useful power. 候i And the temperature data t corresponding to the IGBT temperature at the same moment as the active power. 候i Each set of target data generated from multiple sets of candidate data includes power data P corresponding to the useful power. i And the temperature data T corresponding to the IGBT temperature at the same moment as the active power. i .

[0053] For example, after cleaning the historical data, 10 sets of candidate data were generated, namely: (p 候1 , t 候1 ), (p 候2 , t 候2 ), (p 候3 , t 候3 ), (p 候4 , t 候4 ), (p 候5 , t候5 ), (p 候6 , t 候6 ), (p 候7 , t 候7 ), (p 候8 , t 候8 ), (p 候9 , t 候9 ), (p 候10 , t 候10 In this embodiment, the sliding window size is 3. The process of smoothing multiple sets of candidate data to generate multiple sets of target data is as follows:

[0054] The data in the first sliding window is: (p 候1 , t 候1 ), (p 候2 , t 候2 ), (p 候3 , t 候3 Therefore, the target data generated under this sliding window is (P1, T1), where,

[0055]

[0056]

[0057] The data in the second sliding window is: (p 候2 , t 候2 ), (p 候3 , t 候3 ), (p 候4 , t 候4 Therefore, the target data generated under this sliding window is (P2, T2). The generation method of P2 and T2 is as described above. The following windows also generate the target data corresponding to each sliding window in the same way as described above, so they will not be repeated here.

[0058] Finally, the generated target data sets are: (P1, T1), (P2, T2), (P3, T3), (P4, T4), (P5, T5), (P6, T6), (P7, T7), (P8, T8).

[0059] For example, before smoothing multiple sets of candidate data, it is necessary to clean multiple sets of historical data of the water-cooled converter. In this embodiment, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm is used. The DBSCAN algorithm is a representative density-based clustering algorithm. Unlike partitioning and hierarchical clustering methods, it defines a cluster as the largest set of density-connected points. It can divide regions with sufficiently high density into clusters and can find clusters of arbitrary shapes in noisy spatial databases.

[0060] For example, each set of historical data collected from the water-cooled converter includes power data p corresponding to the useful power. 历i And the temperature data t corresponding to the IGBT temperature at the same moment as the power data. 历i The specific process of cleaning multiple sets of historical data using the DBSCAN algorithm is as follows: First, sort the multiple sets of historical data according to the magnitude of active power in each set; after sorting, calculate the distance l between two adjacent sets of historical data. i Calculate the distance (l) i The mean (μ1), standard deviation (σ1), and distance confidence threshold (R1) of the distance were calculated.

[0061]

[0062]

[0063]

[0064] R1=μ1+3σ1

[0065] Using each set of historical data as the center and R1 as the radius, construct a circle corresponding to each set of historical data as the center, and obtain the number of historical data points (n) within the circle centered on each set of historical data. i ), calculate the amount of data (n) i The mean (μ2), standard deviation (σ2), and confidence threshold (R2) of the data.

[0066]

[0067]

[0068] R² = μ² + 3σ²

[0069] Obtain the number of other historical data contained within a circle centered on each set of historical data. Remove historical data whose number of contained historical data is less than R2 to avoid dirty data when smoothing the generated candidate data, thus achieving a good smoothing effect.

[0070] This method smooths candidate data by taking the average of a sliding window. By acquiring data from a fixed window and smoothing the data using the averaging method, it effectively solves the problem of data mismatch caused by data lag, thus making the final target data more consistent with the actual situation and improving the accuracy of equipment anomaly measurement standards.

[0071] In one optional embodiment, generating a device anomaly measurement standard based on all target data specifically includes: constructing a baseline curve based on the fifth parameter value corresponding to the first parameter and the sixth parameter value corresponding to the second parameter in each group of target data, and using the baseline curve as the device anomaly measurement standard.

[0072] For example, the target data includes the fifth parameter value corresponding to the first parameter and the sixth parameter value corresponding to the second parameter. The fifth parameter value corresponding to the first parameter is used as the horizontal axis and the sixth parameter value corresponding to the second parameter is used as the vertical axis. Then, a baseline curve corresponding to the first parameter and the second parameter is plotted based on the fifth parameter value and the sixth parameter value in each set of target data, and the baseline curve is used as the measurement standard for equipment abnormality.

[0073] For example, as described in the previous embodiment, after smoothing the candidate data to generate multiple sets of target data, the anomaly measurement standard of the water-cooled converter is determined based on the active power and IGBT temperature in the multiple sets of target data. At this time, the active power can be used as the horizontal axis and the IGBT temperature as the vertical axis. According to the generated multiple sets of target data (P1, T1), (P2, T2), (P3, T3), (P4, T4), (P5, T5), (P6, T6), (P7, T7), (P8, T8), a baseline for active power and IGBT temperature is fitted. In this baseline, the temperature value corresponding to each active power value is used as the reference temperature value corresponding to that power value, which is used to provide a standard for the temperature value corresponding to the same power in the operating data.

[0074] This invention generates multiple sets of target data by smoothing candidate data. Based on the fifth parameter value of the first parameter and the sixth parameter value of the second parameter in all target data, a baseline curve for the first and second parameters is constructed. Since the target data is obtained by smoothing after removing dirty data from historical data, it is relatively objective and has reference value. Therefore, the baseline curve formed based on the target data is also reference value. Furthermore, after obtaining the baseline curve, a relatively objective and accurate reference value for the second parameter corresponding to the third parameter value in each set of operating data can be easily obtained from the curve. This provides a relatively objective standard value for the fourth parameter value corresponding to the second parameter in each set of operating data, ensuring the accuracy of the difference between each set of operating data and the baseline curve. This, in turn, ensures the accuracy of determining whether the unit equipment is abnormal based on the difference corresponding to each set of operating data and a preset threshold.

[0075] In one optional embodiment, based on the equipment anomaly measurement standard and the fact that each set of operating data includes at least a third parameter value corresponding to the first parameter and a fourth parameter value corresponding to the second parameter, a target mean is determined. Specifically, this includes: obtaining a reference parameter value corresponding to the second parameter from the benchmark curve based on the benchmark curve and the third parameter value corresponding to the first parameter in the first set of operating data, wherein the first set of operating data is any one of multiple sets of operating data; determining the difference corresponding to the second parameter based on the reference parameter value and the fourth parameter value corresponding to the second parameter in each set of operating data; and determining the target mean based on all the differences.

[0076] For example, in this embodiment, the operating data of the water-cooled converter collected during the second time period includes the power value p corresponding to the active power. 运i And the temperature value t corresponding to the IGBT temperature at the same moment as the power value. 运i The standard for measuring equipment anomalies is a baseline curve formed by active power and temperature.

[0077] First, obtain p from each set of running data from the baseline curve. 运i The corresponding t 参i , t 参i p in the baseline curve 运i The active power corresponds to the reference value of the IGBT temperature during normal operation of the unit equipment, based on the p value in the actual operating data. 运i The corresponding t 运i and the same p in the baseline curve 运i The corresponding t 参i Determine the temperature difference t corresponding to this set of operating data. iThen, after obtaining the temperature difference corresponding to each set of operating data, the interquartile range method can be used to clean the temperature difference corresponding to each set of operating data to remove dirty data in the temperature difference. The specific cleaning process is as follows:

[0078] First, arrange all the temperature differences corresponding to the operating data in ascending order, such as t1, t2, t3, t4, t5, t6, t7, t8, t9, t 10 t 11 Next, the sorted temperature differences are divided into four equal parts. In this embodiment, three of the eleven temperature differences are used to divide the resulting eleven temperature differences into four equal parts. The position of t3 is taken as the first quartile, denoted as Q1; the position of t6 is taken as the second quartile, denoted as Q2; similarly, the position of t9 is taken as the third quartile, denoted as Q3. Then, the difference threshold is determined based on the temperature differences t3, t6, and t9 corresponding to the positions Q1, Q2, and Q3. The determination method is as follows:

[0079] L 阈值 =Q3 + 1.5 * (Q3 - Q1)

[0080] In the formula, Q1 represents the data t3 corresponding to the first quartile; similarly, Q3 represents the data t9 corresponding to the third quartile. Therefore, L... 阈值 = t9 + 1.5·(t9 - t3).

[0081] Finally, after obtaining the difference threshold L 阈值 Then, from all the temperature differences, the temperature difference value t is... i Greater than the difference threshold L 阈值 The temperature difference values ​​are removed, and the mean value of the remaining temperature difference values ​​is determined, which is the target mean value.

[0082] This invention obtains a reference value for the second parameter by taking the third parameter value corresponding to the first parameter in the baseline curve and the first set of operating data. Since the reference values ​​of each second parameter provided by the baseline curve are objective and accurate, the target difference corresponding to the second parameter is determined based on the reference parameter value and the fourth parameter value corresponding to the second parameter in each set of operating data. This target difference is the actual difference between all operating data and the baseline curve in the second operating period. Therefore, the comparison between the obtained target difference and the preset threshold can be used to more accurately determine whether the unit equipment is abnormal.

[0083] In one optional embodiment, determining whether the unit equipment is abnormal based on the target average value specifically includes: determining that the unit equipment is abnormal when the target average value is greater than a preset threshold; or determining that the unit equipment is normal when the target average value is less than or equal to the preset threshold.

[0084] For example, after determining the target mean, the target mean is compared with a preset threshold. If the target mean is less than or equal to the preset threshold, it means that the operating data is within the preset threshold range of normal data, and the unit equipment is normal. If the target mean is greater than the preset threshold, it means that the operating data has exceeded the preset threshold range of normal data, and the unit equipment is abnormal. The preset threshold is obtained based on the historical data of the unit equipment and the equipment abnormality measurement standard.

[0085] For example, when determining whether a water-cooled converter is abnormal, the judgment needs to be made based on the obtained target average value and the preset temperature difference threshold. If the target average value is greater than the temperature difference threshold, the water-cooled converter is determined to be abnormal; if the target average value is less than or equal to the temperature difference threshold, the water-cooled converter is determined to be normal.

[0086] In an optional embodiment, each set of historical data further includes the parameter value of a third parameter and the parameter value of a fourth parameter; each set of operational data further includes the parameter value of a third parameter and the parameter value of a fourth parameter; after determining the unit equipment malfunction based on the target mean, the method further includes: obtaining a first difference based on the parameter values ​​of the third parameter and the fourth parameter in each set of historical data; constructing a first equipment malfunction cause analysis standard based on all the first differences and all the fourth parameter values ​​of the second parameters; and determining the cause of the unit equipment malfunction based on the first equipment malfunction cause analysis standard and all the fourth parameter values ​​and first differences of the second parameters within a second preset time period.

[0087] For example, when a unit equipment malfunction is determined, it is necessary to further determine the cause of the malfunction. In this embodiment, the unit equipment is a water-cooled converter, and the flowchart for determining the malfunction of the water-cooled converter and judging its cause is as follows. Figure 2 As shown, assuming the water-cooled converter is malfunctioning, it is necessary to further determine the cause of the malfunction based on other parameters of the water-cooled converter. Common causes of malfunctions in water-cooled converters include cooling water flow issues, heat dissipation problems, and dust accumulation or aging. The specific steps to determine if the problem is related to cooling water flow are as follows:

[0088] Step S21: Multiple sets of historical data need to be collected within the first preset time period. Each set of historical data includes IGBT temperature value, inlet pressure value, and outlet pressure value. The inlet and outlet pressure difference corresponding to each set of historical data is determined based on the inlet and outlet pressure values ​​in each set of historical data. The IGBT temperature value corresponding to each set of historical data and the inlet and outlet pressure difference at the same moment as the IGBT temperature value are combined into a two-dimensional array. A scatter plot is formed based on the two-dimensional arrays corresponding to all historical data, with IGBT temperature as the horizontal axis and inlet and outlet pressure difference as the vertical axis. The IGBT temperature value, inlet pressure value, and outlet pressure value included in each set of historical data are the specific parameter values ​​of the IGBT temperature and inlet and outlet pressure of the water-cooled converter at the same moment.

[0089] Step S22: Use the DBSCAN algorithm to clean the scatter points in the scatter plot formed by all historical data to obtain candidate scatter points. Then, use the sliding window averaging method to smooth the candidate scatter points to form the target scatter points. The scatter plot formed by the target scatter points is shown below. Figure 3 As shown, the horizontal axis represents IGBT temperature, and the vertical axis represents the pressure difference between the inlet and outlet. Figure 3 The scatter points in the graph are the target scatter points, and a baseline curve corresponding to the IGBT temperature and the inlet / outlet pressure difference is fitted based on the target scatter points. This is the first standard for analyzing the causes of equipment anomalies. Figure 4 As shown.

[0090] Step S23: Collect multiple sets of operating data within the second preset time period. Each set of operating data includes IGBT temperature value, inlet pressure value and outlet pressure value. Determine the inlet and outlet pressure difference, i.e. the first difference value, in each set of operating data based on the inlet and outlet pressure values. Then, form an array with the IGBT temperature value and inlet and outlet pressure difference corresponding to each set of operating data.

[0091] Step S24: Based on the IGBT temperature value in each set of operating data, obtain the reference value of the inlet and outlet pressure difference corresponding to the IGBT temperature value from the first equipment anomaly cause analysis standard corresponding to the IGBT temperature and the inlet and outlet pressure difference. Based on the reference value of the inlet and outlet pressure difference obtained from the first equipment anomaly cause analysis standard corresponding to the IGBT temperature and the inlet and outlet pressure difference corresponding to the IGBT temperature value in the operating data, determine the pressure difference corresponding to each set of operating data. Then, use the interquartile range method to clean the pressure difference difference corresponding to all operating data. Calculate the mean value of all pressure difference differences obtained after cleaning and use this mean value as the target mean value 1 for the pressure difference.

[0092] Step S25: Compare the target average value 1 with the pressure difference threshold 1. If the target average value 1 is greater than the pressure difference threshold 1, the abnormality of the water-cooled converter is determined to be due to a problem with the cooling water flow. If the target average value 1 is less than or equal to the pressure difference threshold 1, the abnormality of the water-cooled converter needs to be determined by combining the judgment results of the heat dissipation of the water-cooled converter.

[0093] In an optional embodiment, each set of historical data further includes the parameter values ​​of the fifth parameter, the sixth parameter, and the seventh parameter; each set of operational data further includes the parameter values ​​of the fifth parameter, the sixth parameter, and the seventh parameter. After determining the unit equipment anomaly based on the target mean, the method further includes: obtaining a second difference based on the fourth parameter value and the fifth parameter value of the second parameter in each set of historical data; obtaining a third difference based on the sixth parameter value and the seventh parameter value of each set of historical data; constructing a second equipment anomaly cause analysis standard based on all the second differences and all the third differences; and determining the cause of the unit equipment anomaly based on the second equipment anomaly cause analysis standard and all the second differences and all the third differences within a second preset time period.

[0094] For example, in an optional embodiment, when it is determined that the water-cooled converter is abnormal, in addition to checking the water flow rate of the water-cooled converter, it is also necessary to check whether its heat dissipation is problematic. The specific steps for this determination are as follows:

[0095] Step S31: Multiple sets of historical data need to be collected within the first preset time period. Each set of historical data includes IGBT temperature value, ambient temperature value, inlet temperature value, and outlet temperature value. Based on the IGBT temperature value and ambient temperature value in each set of historical data, determine the first temperature difference, i.e., the second difference value, for each set of historical data. Based on the inlet temperature value and outlet temperature value in each set of historical data, determine the second temperature difference, i.e., the third difference value, for each set of historical data. Form a two-dimensional array with the first temperature difference and the second temperature difference for each set of historical data. Based on the two-dimensional array corresponding to all historical data, form a scatter plot with the first temperature difference as the horizontal axis and the second temperature difference as the vertical axis. Similarly, the IGBT temperature value, inlet pressure value, and outlet pressure value included in each set of historical data are the specific parameter values ​​of the IGBT temperature and inlet / outlet pressure of the water-cooled converter at the same time.

[0096] Step S32: Use the DBSCAN algorithm to clean the scatter points in the scatter plot formed by the first temperature difference and the second temperature difference corresponding to all historical data to obtain candidate scatter points. Then, use the sliding window averaging method to smooth the candidate scatter points to form target scatter points. Based on the target scatter points, fit the benchmark curve corresponding to the first temperature difference and the second temperature difference, which is the second equipment abnormality cause analysis standard.

[0097] Step S33: Collect multiple sets of operating data within the second preset time period. Each set of operating data includes IGBT temperature value, ambient temperature value, inlet temperature value, and outlet temperature value. Determine the first temperature difference corresponding to each set of operating data based on the IGBT temperature value and ambient temperature value in each set of operating data. Determine the second temperature difference in each set of operating data based on the inlet temperature value and outlet temperature value. Form an array of the first temperature difference and the second temperature difference corresponding to each set of operating data.

[0098] Step S34: Based on the first temperature difference in each set of operating data, obtain the reference value of the second temperature difference corresponding to the first temperature difference from the second equipment abnormality cause analysis standard corresponding to the first temperature difference and the second temperature difference in the operating data. Determine the difference of the second temperature difference in each set of operating data based on the reference value of the second temperature difference and the second temperature difference corresponding to the first temperature difference in the operating data. Clean the difference of the second temperature difference corresponding to all operating data using the interquartile range method. Calculate the mean of all the differences of the second temperature difference obtained after cleaning, which is the target mean 2 corresponding to the second temperature difference.

[0099] Step S35: Compare the target average value 2 with the temperature difference threshold 2. If the target average value 2 is greater than the temperature difference threshold 2, the abnormality of the water-cooled converter is determined to be a problem with heat dissipation. If the target average value 2 is less than or equal to the temperature difference threshold 2, the abnormality of the water-cooled converter needs to be determined by combining the judgment results on the cooling water flow rate.

[0100] When the water-cooled converter malfunctions, if the target mean 1 > preset threshold 1 and the target mean 2 > preset threshold 2, it indicates that the cause of the malfunction is a problem with the cooling water flow and heat dissipation. If the target mean 1 > preset threshold 1 and the target mean 2 ≤ preset threshold 2, it indicates that the cause of the malfunction is a problem with the cooling water flow. If the target mean 1 ≤ preset threshold 1 and the target mean 2 ≤ preset threshold 2, it indicates that the cause of the malfunction is a problem with heat dissipation. If the target mean 1 < preset threshold 1 and the target mean 2 < preset threshold 2, it indicates that the cause of the malfunction is dust accumulation or aging of the water-cooled converter.

[0101] This invention acquires historical and operational data related to unit anomalies and uses the above-mentioned method to determine the cause of the unit equipment anomalies using this historical and operational data. This facilitates timely maintenance by staff based on the cause of the anomalies, thereby avoiding the defect of not knowing the cause of unit equipment anomalies. Timely maintenance of unit equipment also indirectly improves the working efficiency of unit equipment.

[0102] This invention discloses a device for determining unit equipment malfunctions, such as... Figure 5 As shown, it specifically includes the following modules:

[0103] The first acquisition module 51 is used to acquire multiple sets of historical data within a first preset time period of the unit equipment. Each set of historical data includes at least the first parameter value corresponding to the first parameter and the second parameter value corresponding to the second parameter.

[0104] The cleaning module 52 is used to clean multiple sets of historical data and obtain multiple sets of candidate data.

[0105] The smoothing module 53 is used to smooth multiple sets of candidate data to generate multiple sets of target data;

[0106] Generation module 54 is used to generate device anomaly measurement criteria based on all target data;

[0107] The second acquisition module 55 is used to acquire multiple sets of running data of the target object within a second preset time period. Each set of running data includes at least the third parameter value corresponding to the first parameter and the fourth parameter value corresponding to the second parameter.

[0108] The first determining module 56 is used to determine the target mean based on the equipment anomaly measurement standard and each set of operating data includes at least the value of the third parameter corresponding to the first parameter and the value of the fourth parameter corresponding to the second parameter.

[0109] The second determination module 57 is used to determine whether the unit equipment is abnormal based on the target average.

[0110] In one optional embodiment, the smoothing module specifically includes: a smoothing submodule, used to smooth multiple sets of candidate data using a sliding window mean method to generate multiple sets of target data.

[0111] In an optional embodiment, the generation module specifically includes: a construction submodule, used to construct a baseline curve based on the fifth parameter value corresponding to the first parameter in each group of target data in all target data, and the sixth parameter value corresponding to the second parameter, and to use the baseline curve as a standard for measuring equipment anomalies.

[0112] In an optional embodiment, the first determining module specifically includes: an acquisition submodule, configured to acquire a reference parameter value corresponding to a second parameter from the benchmark curve based on the benchmark curve and the third parameter value corresponding to the first parameter in the first set of running data, wherein the first set of running data is any one of multiple sets of running data; the first determining submodule is configured to determine the difference corresponding to the second parameter based on the reference parameter value and the fourth parameter value corresponding to the second parameter in each set of running data; and the second determining submodule is configured to determine the target mean based on all the differences.

[0113] In an optional embodiment, the second determining module specifically includes: a first determining submodule, used to determine that the unit equipment is abnormal when the target average value is greater than a preset threshold; and a second determining submodule, used to determine that the unit equipment is normal when the target average value is less than or equal to the preset threshold.

[0114] In an optional embodiment, after the second determining module, the system further includes: a first obtaining module, configured to obtain a first difference based on the parameter values ​​of the third parameter and the fourth parameter in each set of historical data; a constructing module, configured to construct a first equipment anomaly cause analysis standard based on all the first differences and all the fourth parameter values ​​of the second parameters; and a third determining module, configured to determine the anomaly cause of the unit equipment based on the first equipment anomaly cause analysis standard and the fourth parameter values ​​and first differences of all the second parameters within a second preset time period.

[0115] In an optional embodiment, after the second determining module, the system further includes: a second acquiring module, configured to acquire a second difference based on the fourth parameter value and the fifth parameter value of the second parameter in each set of historical data; a third acquiring module, configured to acquire a third difference based on the sixth parameter value and the seventh parameter value of each set of historical data; a second constructing module, configured to construct a second equipment anomaly cause analysis standard based on all the second differences and all the third differences; and a fourth determining module, configured to determine the anomaly cause of the unit equipment based on the second equipment anomaly cause analysis standard and all the second differences and all the third differences within a second preset time period.

[0116] This embodiment provides a computer device, such as... Figure 6 As shown, the computer device may include at least one processor 61, at least one communication interface 62, at least one communication bus 63, and at least one memory 64. The communication interface 62 may include a display screen and a keyboard; optionally, the communication interface 62 may also include a standard wired interface or a wireless interface. The memory 64 may be high-speed RAM (Random Access Memory) or non-volatile memory, such as at least one disk drive. Optionally, the memory 64 may also be at least one storage device located remotely from the aforementioned processor 61. The processor 61 may be combined with... Figure 6 The described apparatus has an application program stored in memory 64, and the processor 61 calls the program code stored in memory 64 to execute the method for determining unit equipment anomalies in any of the above method embodiments.

[0117] The communication bus 63 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The communication bus 63 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0118] The memory 64 may include volatile memory, such as random-access memory (RAM); the memory may also include non-volatile memory, such as flash memory, hard disk drive (HDD) or solid-state drive (SSD); the memory 64 may also include a combination of the above types of memory.

[0119] The processor 61 can be a central processing unit (CPU), a network processor (NP), or a combination of CPU and NP.

[0120] The processor 61 may further include a hardware chip. This hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof. Optionally, the memory 64 is also used to store program instructions. The processor 61 can invoke the program instructions to implement the method for determining unit equipment anomalies in any embodiment of the present invention.

[0121] This embodiment provides a computer-readable storage medium storing computer-executable instructions that can execute the method for determining equipment malfunctions in any of the above method embodiments. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium may also include combinations of the above types of memory.

[0122] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for determining unit equipment anomalies, characterized in that, include: Collect multiple sets of historical data from the unit equipment within a first preset time period. Each set of historical data includes at least the first parameter value corresponding to the first parameter and the second parameter value corresponding to the second parameter. Data cleaning is performed on multiple sets of historical data to obtain multiple sets of candidate data; The candidate data is smoothed to generate multiple sets of target data. Based on all the target data, generate equipment anomaly measurement criteria; Acquire multiple sets of running data of the target object within a second preset time period, wherein each set of running data includes at least the third parameter value corresponding to the first parameter and the fourth parameter value corresponding to the second parameter. Based on the equipment anomaly measurement criteria, and each set of the operating data includes at least the third parameter value corresponding to the first parameter and the fourth parameter value corresponding to the second parameter, a target mean is determined; Based on the target average, determine whether the unit equipment is abnormal; The target data includes the fifth parameter value corresponding to the first parameter and the sixth parameter value corresponding to the second parameter. The generation of device anomaly measurement criteria based on all the target data specifically includes: Based on the fifth parameter value corresponding to the first parameter in each group of target data in all the target data, and the sixth parameter value corresponding to the second parameter, a benchmark curve is constructed, and the benchmark curve is used as the standard for measuring the equipment anomaly. The determination of the target mean based on the equipment anomaly measurement standard and each set of operational data including at least the third parameter value corresponding to the first parameter and the fourth parameter value corresponding to the second parameter specifically includes: Based on the baseline curve and the third parameter value corresponding to the first parameter in the first set of running data, obtain the reference parameter value corresponding to the second parameter from the baseline curve, wherein the first set of running data is any one of multiple sets of running data; The difference between the second parameter and the reference parameter value and the fourth parameter value corresponding to the second parameter in each set of running data is determined respectively. The target mean is determined based on all the said differences; Each set of historical data also includes the parameter value of the third parameter and the parameter value of the fourth parameter; each set of operational data also includes the parameter value of the third parameter and the parameter value of the fourth parameter; each set of historical data also includes the parameter value of the fifth parameter, the parameter value of the sixth parameter, and the parameter value of the seventh parameter; each set of operational data also includes the parameter value of the fifth parameter, the parameter value of the sixth parameter, and the parameter value of the seventh parameter; after determining the unit equipment abnormality based on the target average, the method further includes: Based on the parameter values ​​of the third parameter and the fourth parameter in each set of historical data, a first difference is obtained; based on all the first differences and all the fourth parameter values ​​of the second parameter, a first equipment anomaly cause analysis standard is constructed; based on the first equipment anomaly cause analysis standard, the fourth parameter values ​​of all the second parameters within the second preset time period, and the first difference, the anomaly cause of the unit equipment is determined; A second difference is obtained based on the fourth and fifth parameter values ​​of the second parameter in each set of historical data; a third difference is obtained based on the sixth and seventh parameter values ​​in each set of historical data; a second equipment anomaly cause analysis standard is constructed based on all the second differences and all the third differences; the anomaly cause of the unit equipment is determined based on the second equipment anomaly cause analysis standard and all the second and third differences within the second preset time period.

2. The method for determining unit equipment anomalies according to claim 1, characterized in that, The step of smoothing multiple sets of candidate data to generate multiple sets of target data specifically includes: The sliding window mean method is used to smooth the multiple sets of candidate data to generate multiple sets of target data.

3. The method for determining unit equipment anomalies according to claim 1 or 2, characterized in that, The step of determining whether the unit equipment is abnormal based on the target average specifically includes: When the target average value is greater than a preset threshold, the unit equipment is determined to be abnormal; Alternatively, if the target average is less than or equal to the preset threshold, then the unit equipment is determined to be normal.

4. A device for determining abnormalities in generator unit equipment, characterized in that, include: The first acquisition module is used to acquire multiple sets of historical data within a first preset time period of the unit equipment. Each set of historical data includes at least the first parameter value corresponding to the first parameter and the second parameter value corresponding to the second parameter. The cleaning module is used to clean multiple sets of historical data and obtain multiple sets of candidate data. A smoothing module is used to smooth multiple sets of candidate data to generate multiple sets of target data; The generation module is used to generate device anomaly measurement criteria based on all the target data; The second acquisition module is used to acquire multiple sets of running data of the target object within a second preset time period. Each set of running data includes at least the third parameter value corresponding to the first parameter and the fourth parameter value corresponding to the second parameter. The first determining module is used to determine a target mean based on the device anomaly measurement standard and each set of the operating data includes at least the third parameter value corresponding to the first parameter and the fourth parameter value corresponding to the second parameter. The second determining module is used to determine whether the unit equipment is abnormal based on the target average value; The target data includes the fifth parameter value corresponding to the first parameter and the sixth parameter value corresponding to the second parameter. The generation module is specifically used to: construct a benchmark curve based on the fifth parameter value corresponding to the first parameter and the sixth parameter value corresponding to the second parameter in each group of target data, and use the benchmark curve as the measurement standard for equipment anomalies. The first determining module is specifically used for: obtaining a reference parameter value corresponding to the second parameter from the benchmark curve based on the benchmark curve and the third parameter value corresponding to the first parameter in the first set of running data, wherein the first set of running data is any one of multiple sets of running data; determining the difference corresponding to the second parameter based on the reference parameter value and the fourth parameter value corresponding to the second parameter in each set of running data; and determining the target mean based on all the differences. Each set of historical data further includes the parameter value of the third parameter and the parameter value of the fourth parameter; each set of operational data further includes the parameter value of the third parameter and the parameter value of the fourth parameter; each set of historical data further includes the parameter value of the fifth parameter, the parameter value of the sixth parameter, and the parameter value of the seventh parameter; each set of operational data further includes the parameter value of the fifth parameter, the parameter value of the sixth parameter, and the parameter value of the seventh parameter; the device further includes: The first anomaly cause determination module is used to obtain a first difference based on the parameter values ​​of the third parameter and the fourth parameter in each set of historical data; construct a first equipment anomaly cause analysis standard based on all the first differences and all the fourth parameter values ​​of the second parameters; and determine the anomaly cause of the unit equipment based on the first equipment anomaly cause analysis standard, the fourth parameter values ​​of all the second parameters within the second preset time period, and the first difference. The second anomaly cause determination module is used to obtain a second difference based on the fourth parameter value and the fifth parameter value of the second parameter in each set of historical data; obtain a third difference based on the sixth parameter value and the seventh parameter value of each set of historical data; construct a second equipment anomaly cause analysis standard based on all the second differences and all the third differences; and determine the anomaly cause of the unit equipment based on the second equipment anomaly cause analysis standard and all the second differences and all the third differences within the second preset time period.

5. A computer device, characterized in that, include: A memory and a processor are communicatively connected, the memory being used to store a computer program, which, when executed by the processor, causes the processor to perform the method for determining unit equipment anomalies as described in any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store computer instructions, which, when executed by a processor, implement the method for determining unit equipment anomalies as described in any one of claims 1 to 3.

Images