Fault detection method and system for fan frequency converter based on big data

Abstract

The invention discloses a fault detection method and system for a fan frequency converter based on big data, and relates to the technical field of fault detection, and the method comprises the steps: arranging a sensor group, and obtaining the real-time operation data of the fan frequency converter; performing entity identification on the historical fault description data to obtain associated parameters of fault types; assigning an initial weight to the associated parameter of each fault type; calculating a real-time deviation ratio, a fault prediction accuracy error and a statistical deviation; calculating a dynamic adjustment coefficient obtained by real-time feedback of each sensor parameter; calculating an error index of each parameter, and optimizing a dynamic adjustment coefficient through a gradient descent algorithm; and calculating a comprehensive fault index by adopting the optimized dynamic adjustment coefficient, and establishing a hierarchical early warning mechanism. The technical problems that potential faults of the frequency converter are difficult to accurately and timely detect in the prior art, so that the operation efficiency of a fan is reduced, even serious consequences are generated, and normal production and operation are influenced are solved.

Description

Technical Field

[0001] This application relates to the technical field of fault detection, and in particular, to a fault detection method and system for a fan frequency converter based on big data. Background Art

[0002] As an important device widely used in many fields such as industry, energy, and construction, the stable operation of a fan plays an irreplaceable role in ensuring production processes, energy supply, and environmental control. The fan frequency converter, as the core device in a wind power generation system, is of great importance. With the continuous development and expansion of the application scope of wind power generation technology, the operating stability and reliability of the fan frequency converter in a wind power station are directly related to the power generation efficiency and safety of the entire system. Therefore, it is particularly important to detect and maintain faults in the fan frequency converter. Traditional methods for detecting faults in fan frequency converters often rely on manual experience, which lacks scientificity and accuracy.

[0003] With the increasing diversity and complexity of the fan operating environment, the existing technology has the technical problem that it is difficult to accurately and timely detect potential faults in the frequency converter, which easily leads to a reduction in the operating efficiency of the fan and even serious consequences, affecting normal production and operation. Summary of the Invention

[0004] The purpose of this application is to provide a fault detection method and system for a fan frequency converter based on big data, so as to solve the technical problem that the existing technology is difficult to accurately and timely detect potential faults in the frequency converter, resulting in a reduction in the operating efficiency of the fan and even serious consequences, affecting normal production and operation.

[0005] In view of the above technical problems, this application provides a fault detection method and system for a fan frequency converter based on big data.

[0006] In the first aspect of the embodiments of this application, a fault detection method for a fan frequency converter based on big data is provided. The method includes:

[0007] Deploy a sensor group to obtain real-time operation data of the fan frequency converter. The sensor group includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, an IGBT status monitoring unit, and a communication status detection unit. The real-time operation data includes input current, output current, input voltage, output voltage, converter temperature, mechanical vibration, IGBT switch status, current shock, communication delay, and data loss rate;

[0008] Based on historical fault description data, apply natural language processing technology to perform entity recognition on the historical fault description data to obtain standardized fault type labels. Analyze each fault type and statistical data to determine sensor parameters closely related to the fault type, and obtain associated parameters of the fault type;

[0009] Assign an initial weight to the associated parameter of each fault type, where the initial weight is used to reflect the importance of the associated parameter in fault judgment;

[0010] Based on historical data and real-time operation data, obtain the real-time monitoring value of the sensor parameter, the data value under normal operation, the historical sampling times, and the collected data value, and calculate the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation;

[0011] According to the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation, calculate the dynamic adjustment coefficient obtained by the real-time feedback of each sensor parameter. The sensor parameters include parameters such as input current, output current, input voltage, output voltage, inverter temperature, mechanical vibration, IGBT switch state, current impact, communication delay, and data loss rate;

[0012] According to the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation of each parameter, calculate the error index of each parameter, and optimize the value of the dynamic adjustment coefficient through the gradient descent algorithm;

[0013] Adopt the optimized dynamic adjustment coefficient to calculate the comprehensive fault index in real time and establish a hierarchical early warning mechanism.

[0014] In the second aspect of the embodiments of the present application, a fault detection system for a wind turbine inverter based on big data is provided. The system includes:

[0015] A sensor group layout module, which is used to layout a sensor group and obtain the real-time operation data of the wind turbine inverter. The sensor group includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, an IGBT status monitoring unit, and a communication status detection unit. The real-time operation data includes input current, output current, input voltage, output voltage, inverter temperature, mechanical vibration, IGBT switch state, current impact, communication delay, and data loss rate;

[0016] A fault type associated parameter obtaining module, which is used to perform entity recognition on historical fault description data based on historical fault description data by applying natural language processing technology to obtain standardized fault type labels, analyze each fault type and statistical data, determine sensor parameters closely related to the fault type, and obtain associated parameters of the fault type;

[0017] An importance judgment module, which is used to assign an initial weight to the associated parameters of each fault type, and the initial weight is used to reflect the importance of the associated parameters in fault judgment;

[0018] A deviation calculation module, which is used to obtain the real-time monitoring value of the sensor parameter, the data value under the normal operation state, the historical sampling times and the collected data value based on the historical data and the real-time operation data, and calculate the real-time deviation ratio, the fault prediction accuracy error and the statistical deviation;

[0019] A dynamic adjustment coefficient calculation module, which is used to calculate the dynamic adjustment coefficient obtained by the real-time feedback of each sensor parameter according to the real-time deviation ratio, the fault prediction accuracy error and the statistical deviation, and the sensor parameters include parameters such as input current, output current, input voltage, output voltage, frequency converter temperature, mechanical vibration, IGBT switch state, current impact, communication delay, and data loss rate;

[0020] An error index calculation module, which is used to calculate the error index of each parameter according to the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation of each parameter, and optimize the value of the dynamic adjustment coefficient through the gradient descent algorithm;

[0021] A hierarchical early warning mechanism establishment module, which is used to calculate the comprehensive fault index in real time by using the optimized dynamic adjustment coefficient and establish a hierarchical early warning mechanism.

[0022] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0023] Arrange a sensor group to obtain real-time operation data of the fan frequency converter. The sensor group includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, an IGBT status monitoring unit, and a communication status detection unit. The real-time operation data includes input current, output current, input voltage, output voltage, frequency converter temperature, mechanical vibration, IGBT switch status, current impact, communication delay, and data loss rate. Based on historical fault description data, apply natural language processing technology to perform entity recognition on the historical fault description data to obtain standardized fault type labels. Analyze each fault type and statistical data to determine the sensor parameters closely related to the fault type and obtain the associated parameters of the fault type. Assign an initial weight to the associated parameters of each fault type, and the initial weight is used to reflect the importance of the associated parameters in fault judgment. Based on historical data and real-time operation data, obtain the real-time monitoring values, data values under normal operation status, historical sampling times, and collected data values of sensor parameters, and calculate the real-time deviation ratio, fault prediction accuracy error, and statistical deviation. According to the real-time deviation ratio, fault prediction accuracy error, and statistical deviation, calculate the dynamic adjustment coefficient obtained by real-time feedback of each sensor parameter. The sensor parameters include parameters of input current, output current, input voltage, output voltage, frequency converter temperature, mechanical vibration, IGBT switch status, current impact, communication delay, and data loss rate. Calculate the error index of each parameter according to the real-time deviation ratio, fault prediction accuracy error, and statistical deviation of each parameter, and optimize the value of the dynamic adjustment coefficient through the gradient descent algorithm. Use the optimized dynamic adjustment coefficient to calculate the comprehensive fault index in real time and establish a hierarchical early warning mechanism. This solves the technical problem that it is difficult to accurately and timely detect potential faults of the frequency converter in the prior art, resulting in a reduction in the operating efficiency of the fan or even serious consequences, affecting normal production and operation.

[0024] The above description is only an overview of the technical solution of this application. In order to be able to more clearly clarify the technical means of this application, it can be implemented in accordance with the content of the specification. And in order to make the above and other purposes, features, and advantages of this application more obvious and understandable, the specific embodiments of this application are hereinafter specifically exemplified. Brief Description of the Drawings

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the operations in the front or below do not necessarily need to be executed precisely in sequence. On the contrary, according to needs, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0026] Figure 1Schematic flowchart of a fault detection method for a fan frequency converter based on big data provided by an embodiment of the present application;

[0027] Figure 2 Schematic structural diagram of a fault detection system for a fan frequency converter based on big data provided by an embodiment of the present application.

[0028] Description of reference numerals: Sensor group layout module 10, Fault type correlation parameter acquisition module 20, Importance judgment module 30, Deviation calculation module 40, Dynamic adjustment coefficient calculation module 50, Error index calculation module 60, Hierarchical early warning mechanism establishment module 70. Detailed implementation manners

[0029] By providing a fault detection method and system for a fan frequency converter based on big data, the present application solves the technical problem in the prior art that it is difficult to accurately and timely detect potential faults of the frequency converter, resulting in a decrease in the operating efficiency of the fan or even serious consequences, affecting normal production and operation.

[0030] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0031] It should be noted that the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server including a series of steps or units is not necessarily limited to those clearly listed steps or units, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices.

[0032] Embodiment 1, as Figure 1 shown, the present application provides a fault detection method for a fan frequency converter based on big data, wherein the method includes:

[0033] Layout a sensor group to obtain real-time operation data of the fan frequency converter. The sensor group includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, an IGBT status monitoring unit, and a communication status detection unit. The real-time operation data includes input current, output current, input voltage, output voltage, frequency converter temperature, mechanical vibration, IGBT switch status, current impact, communication delay, and data loss rate;

[0034] Specifically, a sensor group is arranged at key positions of the fan frequency converter to obtain real-time operation data of the fan frequency converter. Among them, the sensor group includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, an IGBT status monitoring unit, and a communication status detection unit; the real-time operation data includes input current, output current, input voltage, output voltage, frequency converter temperature, mechanical vibration, IGBT switch status, current impact, communication delay, and data loss rate.

[0035] Further, for the arrangement of the sensor group to obtain the real-time operation data of the fan frequency converter, the sensor group includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, an IGBT status monitoring unit, and a communication status detection unit, which includes:

[0036] The current sensor is used to monitor input current parameters and output current parameters;

[0037] The voltage sensor is used to monitor input voltage parameters and output voltage parameters;

[0038] The temperature sensor is used to monitor the frequency converter temperature parameter;

[0039] The vibration sensor is used to monitor mechanical vibration parameters;

[0040] The IGBT status monitoring unit is used to monitor IGBT switch status parameters and current impact parameters;

[0041] The communication status detection unit is used to monitor communication delay parameters and data loss rate.

[0042] Specifically, a sensor group is arranged to obtain the real-time operation data of the fan frequency converter. The sensor group includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, an IGBT status monitoring unit, and a communication status detection unit. Among them, the current sensor is used to monitor input current parameters and output current parameters; the voltage sensor is used to monitor input voltage parameters and output voltage parameters; the temperature sensor is used to monitor the frequency converter temperature parameter to prevent overheating; the vibration sensor is used to monitor mechanical vibration parameters to prevent mechanical failures; the IGBT status monitoring unit is used to monitor IGBT switch status parameters and current impact parameters to prevent switch failures; the communication status detection unit is used to monitor communication delay parameters and data loss rate to ensure the reliability and timeliness of data transmission.

[0043] Based on historical fault description data, apply natural language processing technology to perform entity recognition on the historical fault description data to obtain standardized fault type labels, analyze each fault type and statistical data, determine the sensor parameters closely related to the fault type, and obtain the associated parameters of the fault type;

[0044] Further, based on the historical fault description data, apply natural language processing technology to perform entity recognition on the historical fault description data to obtain standardized fault type labels, analyze each fault type and the statistical data, determine the sensor parameters closely related to the fault type, and obtain the associated parameters of the fault type, including:

[0045] Collect a large amount of historical fault description data from multiple data sources such as fault reports, maintenance logs, and work order records. The fault description data includes the name of the faulty device, the fault phenomenon, the cause of the fault, and the repair process;

[0046] Use natural language processing technology to preprocess and perform entity recognition on the historical fault description data;

[0047] Perform semantic clustering on the extracted entities;

[0048] According to the clustering results, classify faults with similar descriptions into the same category and generate standardized fault type labels;

[0049] Analyze each fault type and the statistical data, and evaluate the correlation between each sensor parameter and the occurrence of the fault;

[0050] Determine a set of closely related sensor parameters for each fault type to form a mapping relationship between the fault type and the associated parameters;

[0051] Store the analyzed mapping relationship in a data table. The data table includes the fault type and the sensor parameters corresponding to the fault type to obtain the associated parameters of the fault type.

[0052] Specifically, collect a large amount of historical fault description data from multiple data sources (such as fault reports, maintenance logs, work order records, etc.). These data should contain key information such as the name of the faulty device, the fault phenomenon, the cause of the fault, and the repair process;

[0053] Preprocess the collected historical fault description data, including removing noise, word segmentation, part-of-speech tagging, etc., and use the named entity recognition (NER) method in natural language processing technology to extract key entities (such as device names, fault phenomena, etc.) from the historical fault description data;

[0054] Use text similarity calculation methods (such as cosine similarity, Jaccard similarity, etc.) to calculate the similarity between entities, and apply clustering algorithms (such as K-means, hierarchical clustering, etc.) to cluster the entities;

[0055] According to the clustering results, classify faults with similar descriptions into the same category to form standardized fault type labels;

[0056] Collect sensor parameter data related to faulty devices and use statistical methods (such as correlation analysis, chi-square test, etc.) to evaluate the correlation between sensor parameters and fault types;

[0057] According to the analysis results, screen out the sensor parameters closely related to each fault type to form a mapping relationship between fault types and associated parameters;

[0058] Design the data table structure, including the fault type and the sensor parameter fields corresponding to this fault type, store the analyzed mapping relationship data in the form of a data table, and obtain the associated parameters of the fault type.

[0059] Assign an initial weight to the associated parameters of each fault type, and the initial weight is used to reflect the importance of the associated parameters in fault judgment;

[0060] Furthermore, the step of assigning an initial weight to the associated parameters of each fault type, where the initial weight is used to reflect the importance of the associated parameters in fault judgment, includes:

[0061] Use statistical methods to identify the outliers of each parameter, and correspond the outliers of each sensor parameter with the standardized fault type labels to form a set of data pairs;

[0062] Use the Pearson correlation coefficient to calculate the correlation coefficient between each sensor parameter outlier sequence and the standardized fault type label sequence;

[0063] Combine expert experience and historical data to assign an initial weight to the associated parameters of each fault type;

[0064] Compare and sort the correlation coefficients of all sensor parameters, and determine the initial weight according to the magnitude of the correlation coefficient. The initial weight is used to reflect the importance of the associated parameters in fault judgment.

[0065] Specifically, use statistical methods (such as the 3σ principle, box plot, etc.) to identify the outliers of each sensor parameter. These outliers may represent the abnormal state or fault precursor of the device. Correspond the identified outliers of sensor parameters with the standardized fault type labels to form a set of data pairs. Each data pair contains the outlier and its corresponding fault type label;

[0066] Calculate the Pearson correlation coefficient between each sensor parameter outlier sequence and the standardized fault type label sequence. The Pearson correlation coefficient measures the linear correlation degree between two variables, and its value ranges from -1 to 1. Among them, a positive correlation coefficient indicates a positive correlation between the two variables, a negative correlation coefficient indicates a negative correlation, and a correlation coefficient close to 0 indicates no obvious linear relationship between the two variables;

[0067] Domain experts evaluate the importance of the associated parameters for each fault type based on experience and knowledge. Using the fault cases and parameter changes in historical data, analyze the actual role of each parameter in fault judgment. Combining the expert evaluation and the analysis results of historical data, assign an initial weight to the associated parameters for each fault type;

[0068] Compare and sort the correlation coefficients of all sensor parameters. Determine the initial weight of each parameter according to the magnitude of the correlation coefficient. The weight can accurately reflect the importance of each parameter in fault judgment. Parameters with larger correlation coefficients will be assigned higher weights because they have a stronger linear relationship with the fault type label and are therefore more important in fault judgment.

[0069] Based on historical data and real-time operation data, obtain the real-time monitoring values of sensor parameters, the data values under normal operation conditions, the historical sampling times, and the collected data values, and calculate the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation;

[0070] Specifically, extract the monitoring value of the current sensor parameter and the data value under normal operation conditions from the real-time operation data, and calculate the real-time deviation ratio through the real-time deviation ratio calculation formula; predict the occurrence of faults based on historical data, calculate the fault prediction accuracy error using the mean square error formula, and evaluate the accuracy of the prediction; obtain the historical sampling times and the historical collected data values from the historical data, and calculate the mean value of each parameter in the historical data through the mean value calculation formula; then calculate the statistical deviation through the statistical deviation calculation formula and based on the calculated mean value.

[0071] Furthermore, the obtaining of the real-time monitoring values of sensor parameters, the data values under normal operation conditions, the historical sampling times, and the collected data values, and the calculation of the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation based on historical data and real-time operation data include:

[0072] Based on historical data and real-time operation data, obtain the real-time monitoring values of sensor parameters and the data values under normal operation conditions, and calculate the real-time deviation ratio;

[0073] The real-time deviation ratio calculation formula is:

[0074]

[0075] where d i is the real-time deviation ratio, i is the i-th sensor parameter, P i is the real-time monitoring value of the i-th parameter, and P i,norm is the data value of parameter i under normal operation conditions;

[0076] Calculate the fault prediction accuracy error e using the mean square error formula i ;

[0077] Based on historical data, calculate the mean value of each parameter to obtain the statistical deviation s i ;

[0078] The mean value calculation formula is:

[0079]

[0080] Where, μ i is the mean value of parameter P i in the historical data, N is the sampling times of the historical data, and P i (j) is parameter P i the data value collected at the j-th time;

[0081] The statistical deviation calculation formula is:

[0082]

[0083] Where, s i is the statistical deviation of parameter P i , N is the sampling times of the historical data, μ i is the mean value of parameter P i in the historical data, and (P i (j) -μ i ) is the deviation degree of the single measurement value from the mean value.

[0084] According to the real-time deviation ratio, fault prediction accuracy error, and statistical deviation, calculate the dynamic adjustment coefficient obtained from the real-time feedback of each sensor parameter, where the sensor parameters include parameters such as input current, output current, input voltage, output voltage, inverter temperature, mechanical vibration, IGBT switch state, current impact, communication delay, and data loss rate;

[0085] Specifically, based on the three indicators of the real-time deviation ratio, fault prediction accuracy error, and statistical deviation calculated previously, calculate the dynamic adjustment coefficient obtained from the real-time feedback of each sensor parameter. The sensor parameters include parameters such as input current, output current, input voltage, output voltage, inverter temperature, mechanical vibration, IGBT switch state, current impact, communication delay, and data loss rate;

[0086] Furthermore, the calculating the dynamic adjustment coefficient obtained from the real-time feedback of each sensor parameter according to the real-time deviation ratio, fault prediction accuracy error, and statistical deviation includes:

[0087] Adaptive adjustment is performed according to the real-time deviation ratio, the error of the fault prediction accuracy, and the statistical deviation, and the dynamic adjustment coefficient obtained by the real-time feedback of each sensor parameter is calculated. The calculation formula is as follows:

[0088] β i (t + 1)=β i (t)+λ1×d i +λ2×e i +λ3×s i ;

[0089] Among them, β i (t + 1) is the dynamic adjustment coefficient obtained by the real-time feedback at time t + 1, λ1, λ2, and λ3 are preset adjustment factors, d i is the real-time deviation ratio, e i is the error of the fault prediction accuracy, s i is the statistical deviation, and the range of β i is 0.5 ≤ β i ≤ 1.5.

[0090] Specifically, when calculating the dynamic adjustment coefficient obtained by the real-time feedback of each sensor parameter, we comprehensively consider three key factors: the real-time deviation ratio, the error of the fault prediction accuracy, and the statistical deviation. This adaptive adjustment process aims to ensure that the system can dynamically respond to changes in sensor parameters, thereby optimizing the monitoring efficiency, improving the accuracy of fault warning, and enhancing the overall reliability of the system;

[0091] The calculation formula for the dynamic adjustment coefficient obtained by the real-time feedback of the sensor parameter is β i (t + 1)=β i (t)+λ1×d i +λ2×e i +λ3×s i ;

[0092] Among them, β i (t + 1) is the dynamic adjustment coefficient obtained by the real-time feedback at time t + 1, λ1, λ2, and λ3 are preset adjustment factors, d i is the real-time deviation ratio, e i is the error of the fault prediction accuracy, s i is the statistical deviation, and the range of β i is 0.5 ≤ β i ≤ 1.5.

[0093] According to the real-time deviation ratio, the error of the fault prediction accuracy, and the statistical deviation of each parameter, calculate the error index of each parameter, and optimize the value of the dynamic adjustment coefficient through the gradient descent algorithm;

[0094] Further, calculating the error index of each parameter according to the real-time deviation ratio, fault prediction accuracy error, and statistical deviation of each parameter, and optimizing the value of the dynamic adjustment coefficient through the gradient descent algorithm, including:

[0095] Obtain the real-time deviation ratio, fault prediction accuracy error, and statistical deviation of each parameter;

[0096] Combining the real-time deviation ratio, fault prediction accuracy error, and statistical deviation, calculate an initial dynamic adjustment coefficient for each sensor parameter;

[0097] Calculate the error index of each sensor parameter, and optimize the value of the dynamic adjustment coefficient through the gradient descent algorithm.

[0098] Specifically, based on the previous calculation formulas of the real-time deviation ratio, fault prediction accuracy error, and statistical deviation, obtain the real-time deviation ratio, fault prediction accuracy error, and statistical deviation of each parameter (including input current parameter, output current parameter, input voltage parameter, output voltage parameter, frequency converter temperature parameter, mechanical vibration parameter, IGBT switch state parameter, current impact parameter, communication delay parameter, data loss rate);

[0099] Combining the real-time deviation ratio, fault prediction accuracy error, and statistical deviation, calculate an initial dynamic adjustment coefficient for each sensor parameter. This coefficient reflects the degree of attention or adjustment strength of the system to each sensor parameter in the current state;

[0100] Then, calculate the error index of each sensor parameter, and optimize the value of the dynamic adjustment coefficient through the gradient descent algorithm. The gradient descent algorithm is an iterative optimization algorithm that updates the value of the coefficient according to the gradient of the objective function (here refers to the error index) with respect to the dynamic adjustment coefficient.

[0101] Further, the calculating the error index of each sensor parameter and optimizing the value of the dynamic adjustment coefficient through the gradient descent algorithm includes:

[0102] Calculate the error index of each sensor parameter, and the calculation formula is:

[0103] E i =|P i real -P i norml |;

[0104] Where, E i is the error index of sensor parameter i, P i real is the current measured value of sensor parameter i, P i normlis the data value of the sensor parameter i in the normal operating state;

[0105] The gradient descent algorithm is used to optimize the dynamic adjustment coefficient, and the calculation formula of the loss function is:

[0106] W i (t+1) = W i (t) - η × W i × E i 2 ;

[0107] Among them, W i (t+1) is the dynamic adjustment coefficient after the (t + 1)-th iteration, W i (t) is the dynamic adjustment coefficient after the t-th iteration, and η is the learning rate.

[0108] Specifically, first, calculate the error index of each sensor parameter according to the formula. The calculation formula is E i = |P i real - P i norml |. Among them, E i is the error index of the sensor parameter i, P i real is the current measured value of the sensor parameter i, and P i norml is the data value of the sensor parameter i in the normal operating state;

[0109] Then, use the gradient descent algorithm to optimize the dynamic adjustment coefficient. The calculation formula of the loss function is W i (t+1) = W i (t) - η × W i × E i 2 。Among them, W i (t+1) is the dynamic adjustment coefficient after the (t + 1)-th iteration, W i (t) is the dynamic adjustment coefficient after the t-th iteration, and η is the learning rate. The error index gradient of each sensor parameter with respect to its respective dynamic adjustment coefficient is calculated according to this step, and then the gradient descent algorithm is used to update the value of the dynamic adjustment coefficient to achieve the optimization purpose.

[0110] The optimized dynamic adjustment coefficient is used to calculate the comprehensive fault index in real time, and a hierarchical early warning mechanism is established.

[0111] Specifically, according to the dynamically adjusted coefficients optimized by the gradient descent algorithm before, these coefficients reflect the importance and adjustment strength of different sensor parameters in the current system state. The comprehensive fault index is calculated in real time through a calculation formula, and then a hierarchical early warning mechanism is established. We divide the fault early warning into four levels, and each level corresponds to different levels of urgency and response measures. This can improve the efficiency of fault handling, thereby minimizing the impact of faults on production operations. Through such a hierarchical early warning mechanism, we can achieve precise monitoring of the equipment operation status, ensuring the stable operation and continuous optimization of the system.

[0112] Furthermore, the adoption of the optimized dynamically adjusted coefficients to calculate the comprehensive fault index in real time and establish a hierarchical early warning mechanism includes:

[0113] Construct a real-time fault index calculation formula based on the fault type, associated parameters, initial weights, and dynamically adjusted coefficients, and calculate the comprehensive fault index;

[0114] Calculate the real-time fault index, and the real-time fault index calculation formula is:

[0115] α i * = α i × β i ;

[0116] where α i * is the real-time fault index, α i is the initial weight of the i-th parameter, and β i is the dynamically adjusted coefficient obtained through real-time feedback;

[0117] Calculate the comprehensive fault index, and the comprehensive fault index calculation formula is:

[0118]

[0119] where F is the comprehensive fault index, α i * is the real-time fault index, P i is the real-time monitoring value of the i-th parameter, and P i,norm is the data value of parameter i in the normal operation state;

[0120] Adopt the optimized dynamically adjusted coefficient W i Calculate the optimized comprehensive fault index, and the calculation formula is:

[0121]

[0122] where F o is the optimized comprehensive fault index, n is the total number of sensors, and Wi is the optimized dynamic adjustment coefficient, P i is the real-time monitoring value of the i-th parameter, P i,norm is the data value of parameter i under normal operating conditions;

[0123] Based on the optimized comprehensive fault index F o , a hierarchical early warning mechanism is established, and the fault early warning is divided into four levels;

[0124] When F o < F low , it is judged as the normal level;

[0125] When F low ≤ F o < F medium , it is judged as the minor abnormality level and enters the observation state;

[0126] When F medium ≤ F o < F high , it is judged as the serious abnormality level, and the intelligent management system recommends inspection and maintenance;

[0127] When F o ≥ F high , it is judged as the emergency fault level, and the intelligent management system starts to shut down and sends an alarm.

[0128] Specifically, based on the previously obtained fault type, associated parameters, initial weights, and dynamic adjustment coefficients, a real-time fault index calculation formula is constructed to calculate the real-time comprehensive fault index. The real-time fault index calculation formula is:

[0129] α i * = α i × β i ;

[0130] Among them, α i * is the real-time fault index, α i is the initial weight of the i-th parameter, and β i is the dynamic adjustment coefficient obtained by real-time feedback;

[0131] Then, based on the previously calculated real-time fault index, the comprehensive fault index is calculated. The comprehensive fault index calculation formula is:

[0132]

[0133] Among them, F is the comprehensive fault index, α i * is the real-time fault index, P i is the real-time monitoring value of the i-th parameter, Pi,norm is the data value of parameter i under normal operating conditions;

[0134] The optimized comprehensive fault index is calculated using the optimized dynamic adjustment coefficient, and the calculation formula is:

[0135]

[0136] where, F o is the optimized comprehensive fault index, n is the total number of sensors, W i is the optimized dynamic adjustment coefficient, P i is the real-time monitoring value of the i-th parameter, P i,norm is the data value of parameter i under normal operating conditions;

[0137] Based on the optimized comprehensive fault index, a hierarchical early warning mechanism is established. This mechanism aims to achieve timely identification and response to potential problems by carefully dividing the severity of fault warnings. Specifically, the levels are divided into normal level, minor anomaly level, severe anomaly level, and emergency fault level. Among them, the normal level means that the optimized comprehensive fault index is less than the minor anomaly threshold, the system runs smoothly, and no measures need to be taken; the minor anomaly level means that the optimized comprehensive fault index has risen slightly but is still within an acceptable range, and at this time, it enters the observation state; the severe anomaly level means that the optimized comprehensive fault index reaches or exceeds the severe anomaly threshold, and at this time, the intelligent management system should issue a prompt, suggesting that the operator immediately conduct inspection and maintenance, determine the potential fault source, and take necessary corrective measures; the emergency fault level means that the optimized comprehensive fault index has risen sharply, reaching or exceeding the emergency fault threshold, and at this time, the system should issue an emergency warning signal, the intelligent management system starts to stop running, and sends an alarm to prevent the fault from further expanding and causing serious consequences. The specific judgment is as follows:

[0138] When F o < F low , it is judged as the normal level;

[0139] When F low ≤ F o < F medium , it is judged as the minor anomaly level and enters the observation state;

[0140] When F medium ≤ F o < F high , it is judged as the severe anomaly level, and the intelligent management system suggests inspection and maintenance;

[0141] When F o ≥ F high , it is judged as the emergency fault level, and the intelligent management system starts to shut down and send an alarm.

[0142] In summary, the embodiments of the present application at least have the following technical effects:

[0143] Deploy a sensor group to obtain real-time operation data of the fan frequency converter. The sensor group includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, an IGBT status monitoring unit, and a communication status detection unit. The real-time operation data includes input current, output current, input voltage, output voltage, frequency converter temperature, mechanical vibration, IGBT switch status, current impact, communication delay, and data loss rate. Based on historical fault description data, apply natural language processing technology to perform entity recognition on the historical fault description data to obtain standardized fault type tags. Analyze each fault type and statistical data to determine the sensor parameters closely related to the fault type, and obtain the associated parameters of the fault type. Assign an initial weight to the associated parameters of each fault type, and the initial weight is used to reflect the importance of the associated parameters in fault judgment. Based on historical data and real-time operation data, obtain the real-time monitoring values of sensor parameters, the data values under normal operation conditions, the historical sampling times, and the collected data values, and calculate the real-time deviation ratio, fault prediction accuracy error, and statistical deviation. According to the real-time deviation ratio, fault prediction accuracy error, and statistical deviation, calculate the dynamic adjustment coefficient obtained by real-time feedback of each sensor parameter. The sensor parameters include the parameters of input current, output current, input voltage, output voltage, frequency converter temperature, mechanical vibration, IGBT switch status, current impact, communication delay, and data loss rate. Calculate the error index of each parameter according to the real-time deviation ratio, fault prediction accuracy error, and statistical deviation of each parameter, and optimize the value of the dynamic adjustment coefficient through the gradient descent algorithm. Adopt the optimized dynamic adjustment coefficient to calculate the comprehensive fault index in real time and establish a hierarchical early warning mechanism.

[0144] Embodiment 2, based on the same inventive concept as a fault detection method for a fan frequency converter based on big data in the foregoing embodiment, as Figure 2 shown, the present application provides a fault detection system for a fan frequency converter based on big data. The system and method embodiments in the present application are based on the same inventive concept. Among them, the system includes:

[0145] A sensor group deployment module 10, which is used to deploy a sensor group to obtain real-time operation data of the fan frequency converter. The sensor group includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, an IGBT status monitoring unit, and a communication status detection unit. The real-time operation data includes input current, output current, input voltage, output voltage, frequency converter temperature, mechanical vibration, IGBT switch status, current impact, communication delay, and data loss rate;

[0146] A fault type associated parameter acquisition module 20, which is configured to, based on historical fault description data, apply natural language processing technology to perform entity recognition on the historical fault description data to obtain a standardized fault type label, analyze each fault type and statistical data, determine sensor parameters closely related to the fault type, and obtain associated parameters of the fault type;

[0147] An importance judgment module 30, which is configured to assign an initial weight to the associated parameter of each fault type, and the initial weight is used to reflect the importance of the associated parameter in fault judgment;

[0148] A deviation calculation module 40, which is configured to, based on historical data and real-time operation data, obtain the real-time monitoring value of the sensor parameter, the data value under the normal operation state, the historical sampling times, and the collected data value, and calculate the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation;

[0149] A dynamic adjustment coefficient calculation module 50, which is configured to calculate a dynamic adjustment coefficient obtained by real-time feedback of each sensor parameter according to the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation, and the sensor parameters include parameters such as input current, output current, input voltage, output voltage, inverter temperature, mechanical vibration, IGBT switch state, current impact, communication delay, and data loss rate;

[0150] An error index calculation module 60, which is configured to calculate an error index of each parameter according to the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation of each parameter, and optimize the value of the dynamic adjustment coefficient through a gradient descent algorithm;

[0151] A hierarchical early warning mechanism establishment module 70, which is configured to use the optimized dynamic adjustment coefficient to calculate a comprehensive fault index in real time and establish a hierarchical early warning mechanism.

[0152] It should be noted that the above sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily require the specific order or continuous order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0153] The above are only the preferred embodiments of the present application, and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included within the protection scope of the present application.

[0154] This specification and the drawings are merely exemplary illustrations of the present application and are considered to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A fault detection method for a fan frequency converter based on big data, characterized in that, The method includes: Deploying a sensor group to obtain real-time operation data of the fan frequency converter. The sensor group includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, an IGBT status monitoring unit, and a communication status detection unit. The real-time operation data includes input current, output current, input voltage, output voltage, frequency converter temperature, mechanical vibration, IGBT switch status, current impact, communication delay, and data loss rate; Based on historical fault description data, applying natural language processing technology to perform entity recognition on the historical fault description data to obtain standardized fault type labels, analyzing each fault type and statistical data to determine sensor parameters closely related to the fault type, and obtaining associated parameters of the fault type; Assigning an initial weight to the associated parameters of each fault type, where the initial weight is used to reflect the importance of the associated parameters in fault judgment; Based on historical data and real-time operation data, obtaining real-time monitoring values, data values under normal operation conditions, historical sampling times, and collected data values of sensor parameters, and calculating real-time deviation ratios, fault prediction accuracy errors, and statistical deviations; According to the real-time deviation ratios, fault prediction accuracy errors, and statistical deviations, calculating dynamic adjustment coefficients obtained by real-time feedback of each sensor parameter. The sensor parameters include parameters of input current, output current, input voltage, output voltage, frequency converter temperature, mechanical vibration, IGBT switch status, current impact, communication delay, and data loss rate; According to the real-time deviation ratios, fault prediction accuracy errors, and statistical deviations of each parameter, calculating the error index of each parameter, and optimizing the value of the dynamic adjustment coefficient through the gradient descent algorithm; Adopting the optimized dynamic adjustment coefficient to calculate the comprehensive fault index in real time and establishing a hierarchical early warning mechanism.

2. The fault detection method of a fan frequency converter based on big data according to claim 1, characterized in that, The deployment of the sensor group to obtain real-time operation data of the fan frequency converter. The sensor group includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, an IGBT status monitoring unit, and a communication status detection unit includes: The current sensor is used to monitor input current parameters and output current parameters; The voltage sensor is used to monitor input voltage parameters and output voltage parameters; The temperature sensor is used to monitor the frequency converter temperature parameter; The vibration sensor is used to monitor mechanical vibration parameters; The IGBT status monitoring unit is used to monitor IGBT switch status parameters and current impact parameters; The communication status detection unit is used to monitor communication delay parameters and data loss rate.

3. A fault detection method for a fan frequency converter based on big data according to claim 1, characterized in that, The entity recognition of historical fault description data based on historical fault description data by applying natural language processing technology to obtain standardized fault type labels, analyzing each fault type and statistical data to determine sensor parameters closely related to the fault type, and obtaining associated parameters of the fault type includes: Collecting a large amount of historical fault description data from multiple data sources such as fault reports, maintenance logs, and work order records. The fault description data includes the name of the faulty equipment, fault phenomenon, fault cause, and repair process; Preprocess historical fault description data and perform entity recognition using natural language processing technology; Perform semantic clustering on the extracted entities; According to the clustering results, classify faults with similar descriptions into the same category and generate standardized fault type labels; Analyze each fault type and statistical data to evaluate the correlation between each sensor parameter and the occurrence of faults; Determine a set of closely related sensor parameters for each fault type to form a mapping relationship between the fault type and the associated parameters; Store the analyzed mapping relationship in a data table, where the data table includes the fault type and the sensor parameters corresponding to the fault type, and obtain the associated parameters of the fault type.

4. The fault detection method of a fan frequency converter based on big data according to claim 1, wherein Assign an initial weight to the associated parameters of each fault type, and the initial weight is used to reflect the importance of the associated parameters in fault judgment, including: Use statistical methods to identify the outliers of each parameter, and correspond the outliers of each sensor parameter with the standardized fault type labels to form a set of data pairs; Use the Pearson correlation coefficient to calculate the correlation coefficient between each sequence of sensor parameter outliers and the sequence of standardized fault type labels; Combine expert experience and historical data to assign an initial weight to the associated parameters of each fault type; Compare and sort the correlation coefficients of all sensor parameters, and determine the initial weight according to the magnitude of the correlation coefficient. The initial weight is used to reflect the importance of the associated parameters in fault judgment.

5. The fault detection method of a fan frequency converter based on big data according to claim 1, characterized in that, Based on historical data and real-time operation data, obtain the real-time monitoring values of sensor parameters, the data values under normal operation conditions, the historical sampling times, and the collected data values, and calculate the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation, including: Based on historical data and real-time operation data, obtain the real-time monitoring values of sensor parameters and the data values under normal operation conditions, and calculate the real-time deviation ratio; The formula for calculating the real-time deviation ratio is: where d i is the real-time deviation ratio, i is the i-th sensor parameter, and P i is the real-time monitoring value of the i-th parameter, and P i,norm is the data value of parameter i under normal operating conditions; Calculate the fault prediction accuracy error e using the mean squared error formula i ; Based on historical data, calculate the mean of each parameter to obtain the statistical deviation s i ; The formula for calculating the mean value is: Among them, μ i is the mean value of parameter P i in historical data, N is the number of sampling times of historical data, P i (j) is parameter P i at the data value collected at the j-th time; The formula for calculating the statistical deviation is: Among them, s i is the statistical deviation of parameter P i , N is the sampling times of historical data, and μ i is the mean value of parameter P i in the historical data. (P i (j) - μ i ) is the degree of deviation between a single measurement value and the mean value.

6. The fault detection method of a fan frequency converter based on big data according to claim 1, characterized in that, According to the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation, calculate the dynamic adjustment coefficient obtained by the real-time feedback of each sensor parameter, including: Perform adaptive adjustment according to the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation, and calculate the dynamic adjustment coefficient obtained by the real-time feedback of each sensor parameter. The calculation formula is: β i (t + 1) = β i (t) + λ1 × d i + λ2 × e i + λ3 × s i ; Among them, β i (t + 1) is the dynamic adjustment coefficient obtained by real-time feedback at time t + 1, λ1, λ2, and λ3 are preset adjustment factors, d i is the real-time deviation ratio, e i is the fault prediction accuracy error, s i is the statistical deviation, β i ranges from 0.5 ≤ β i ≤ 1.

5.

7. The fault detection method of a fan frequency converter based on big data according to claim 1, characterized in that, According to the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation of each parameter, calculate the error index of each parameter, and optimize the value of the dynamic adjustment coefficient through the gradient descent algorithm, including: Obtain the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation of each parameter; Combine the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation to calculate an initial dynamic adjustment coefficient for each sensor parameter; Calculate the error index of each sensor parameter, and optimize the value of the dynamic adjustment coefficient through the gradient descent algorithm.

8. The fault detection method of a fan frequency converter based on big data according to claim 7, characterized in that, The calculation of the error index of each sensor parameter and the optimization of the value of the dynamic adjustment coefficient through the gradient descent algorithm include: Calculate the error index of each sensor parameter. The calculation formula is: Among them, E i is the error index of the sensor parameter i, and P i real is the current measured value of the sensor parameter i, and P i norml is the data value of the sensor parameter i under normal operating conditions; Use the gradient descent algorithm to optimize the dynamic adjustment coefficient. The calculation formula of the loss function is: W i (t+1) = W i (t) - η × W i × E i 2 ; Among them, W i (t+1) is the dynamic adjustment coefficient after the (t + 1)-th iteration, and W i (t) is the dynamic adjustment coefficient after the t-th iteration, and η is the learning rate.

9. The fault detection method of a fan frequency converter based on big data according to claim 1, characterized in that Adopting the optimized dynamic adjustment coefficient, calculating the comprehensive fault index in real time, and establishing a hierarchical early warning mechanism, including: Constructing a real-time fault index calculation formula according to the fault type, associated parameters, initial weight, and dynamic adjustment coefficient, and calculating the comprehensive fault index; Calculating the real-time fault index, and the real-time fault index calculation formula is: α i * ＝α i ×β i ； Among them, α i * is the real-time fault index, and α i is the initial weight of the i-th parameter, and β i is the dynamic adjustment coefficient obtained from real-time feedback; Calculating the comprehensive fault index, and the comprehensive fault index calculation formula is: Among them, F is the comprehensive fault index, and α i * is the real-time fault index, and P i is the real-time monitoring value of the i-th parameter, and P i,norm is the data value of parameter i under normal operating conditions; Adopt the optimized dynamic adjustment coefficient W i Calculate the optimized comprehensive fault index, and the calculation formula is: Among them, F o is the optimized comprehensive fault index, n is the total number of sensors, W i is the optimized dynamic adjustment coefficient, P i is the real-time monitoring value of the i-th parameter, P i,norm is the data value of parameter i under normal operating conditions; Based on the optimized comprehensive fault index F o , a hierarchical early warning mechanism is established, and the fault early warning is divided into four levels; When F o <F low it is judged as the normal level; When F low ≤ F o < F medium it is judged as a minor abnormal level and enters the observation state; When F medium ≤ F o < F high it is judged as a severe abnormal level, and the intelligent management system recommends inspection and maintenance; When F o ≥ F high , it is judged as the emergency failure level, and the intelligent management system starts to shut down and sends an alarm.

10. A fault detection system for a fan frequency converter based on big data, characterized in that, For implementing the steps of the fault detection method of a wind turbine frequency converter based on big data according to any one of claims 1 to 9, the system includes: A sensor group layout module, which is used to layout a sensor group to obtain the real-time operation data of the wind turbine frequency converter. The sensor group includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, an IGBT status monitoring unit, and a communication status detection unit. The real-time operation data includes input current, output current, input voltage, output voltage, frequency converter temperature, mechanical vibration, IGBT switch status, current impact, communication delay, and data loss rate; A fault type associated parameter acquisition module, which is used to perform entity recognition on the historical fault description data by applying natural language processing technology based on the historical fault description data, obtain a standardized fault type label, analyze each fault type and statistical data, determine the sensor parameters closely related to the fault type, and obtain the associated parameters of the fault type; An importance judgment module, which is used to assign an initial weight to the associated parameters of each fault type, and the initial weight is used to reflect the importance of the associated parameters in fault judgment; A deviation calculation module, which is used to obtain the real-time monitoring value of the sensor parameter, the data value under the normal operation state, the historical sampling times, and the collected data value based on the historical data and the real-time operation data, and calculate the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation; A dynamic adjustment coefficient calculation module, which is used to calculate the dynamic adjustment coefficient obtained by the real-time feedback of each sensor parameter according to the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation. The sensor parameters include the parameters of input current, output current, input voltage, output voltage, frequency converter temperature, mechanical vibration, IGBT switch status, current impact, communication delay, and data loss rate; An error index calculation module, which is used to calculate the error index of each parameter according to the real-time deviation ratio, the fault prediction accuracy error, and the statistical deviation of each parameter, and optimize the value of the dynamic adjustment coefficient through the gradient descent algorithm; A hierarchical early warning mechanism establishment module, which is used to adopt the optimized dynamic adjustment coefficient, calculate the comprehensive fault index in real time, and establish a hierarchical early warning mechanism.

Images