Intelligent device for detecting performance of frequency converter

Abstract

The invention relates to the technical field of frequency converters, in particular to an intelligent device for detecting the performance of a frequency converter, which is provided with a preprocessing module, an abnormal influence module, a performance analysis module, a detector and a controller, and calculates a feature superposition coefficient based on a determined environmental gradient influence characterization value and a flow interference characterization value. The method comprises the following steps of: determining a plurality of to-be-detected areas, setting a label for each to-be-detected area, responding to a set strong area influence label, determining an abrupt change influence characterization value, calculating a performance abnormity characterization value in combination with a characteristic superposition coefficient, and determining to call a detector for positioning a detection area and detecting the detection area; the method is used for controlling detection starting time and detection frequency of the detector. According to the method, the frequency converter is subjected to regionalization analysis, the influence of the environment data, the vibration data and the instantaneous voltage on the detection accuracy of the frequency converter is comprehensively analyzed, the area needing to be detected is determined in a targeted mode, and the detection efficiency and the detection accuracy are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of frequency converters, and particularly relates to an intelligent device for performance detection of a frequency converter. BACKGROUND

[0002] A frequency converter is a power control device for controlling an alternating current motor by changing the frequency of the motor's power supply. As the role of frequency converters in industrial automation and energy saving becomes increasingly prominent, reliable detection of their performance becomes crucial. However, traditional detection methods highly depend on connecting actual motors and mechanical loads for testing, leading to complex test environment setup, high cost, low efficiency, safety risks, and difficulty in accurately controlling and repeating load characteristics, making dynamic performance evaluation difficult, functional testing (such as protection mechanisms) incomplete, and automation level low, which severely restricts the flexibility of on-site testing. Moreover, environmental factors can also affect the performance of frequency converters, leading to errors in performance detection. Therefore, there is an urgent need to develop a new type of dedicated detection device that is safe, efficient, and accurate, and integrates automated measurement and analysis functions, to overcome the fundamental limitations of traditional methods.

[0003] Chinese Patent Publication No. CN119492947A discloses a frequency converter performance detection device, which includes a detection box with a heat-conducting sheet inside, the heat-conducting sheet forms a sealed cold-hot channel at the edge of the inner cavity of the detection box; a damping shock absorber is installed at the bottom of the detection box; a vibration generator is provided at the lower part of the detection box to provide a vibration source for the detection box; a humidifier is connected to the inner cavity of the detection box; and a cold-hot generator is connected to the cold-hot channel. The invention sets up a vibration generator, a humidifier, and a cold-hot generator, and adapts to the detection box, achieving multifunctional simulation of vibration, high humidity, high temperature, and low temperature environments in the external environment, realizing stability detection of frequency converters in extreme environments, and setting a cold-hot channel in a circulating flow state at the inner edge of the detection box, which covers the detection box with cold-hot air flowing on multiple sides, improving the uniformity of cold-hot air distribution.

[0004] Chinese patent publication No. CN112130010A discloses a static frequency converter SFC performance detection system under large current working condition, comprising: SFC device and its control system, isolation transformer, power distribution cabinet incoming line circuit breaker QF1, isolation transformer incoming line circuit breaker QF2, QF3, voltage transformer PT, current transformer CT1, CT2; the isolation transformer incoming line circuit breaker QF2, QF3 has a soft starting device; the SFC device comprises a rectifier NB, an electric reactor Ld and an inverter MB. The above detection system solves the defects that the SFC can only perform small capacity synchronous motor small current performance verification before the static frequency converter is put into operation on the engineering site, and cannot perform large current operation performance verification; only the existing conventional detection system isolation transformer, power distribution system and corresponding control strategy are used, without configuring large capacity synchronous motor and excitation system, the running, control and protection characteristics of SFC under large current working condition can be verified.

[0005] However, the prior art still has the following problems,

[0006] In actual situations, when detecting the frequency converter, full domain detection or discrete single point detection is mostly performed. Full domain detection causes excessive consumption of resources due to redundant data in low risk areas. Single point detection causes coverage blind area due to lack of spatial gradient perception, which increases the false judgment rate of multi-physical field coupling faults. The fault spreads to the monitoring point to trigger an alarm, which causes detection errors and reduces the detection accuracy. SUMMARY

[0007] Therefore, the present application provides an intelligent device for performance detection of a frequency converter to solve the problem that in actual situations, when detecting the frequency converter, full domain detection or discrete single point detection is mostly performed. Full domain detection causes excessive consumption of resources due to redundant data in low risk areas. Single point detection causes coverage blind area due to lack of spatial gradient perception, which increases the false judgment rate of multi-physical field coupling faults. The fault spreads to the monitoring point to trigger an alarm, which causes detection errors and reduces the detection accuracy.

[0008] To achieve the above purpose, the present application provides an intelligent device for performance detection of a frequency converter, which comprises:

[0009] A preprocessing module is used to obtain environmental data of a frequency converter operating space, analyze the environmental data to determine an environmental gradient influence representation value, obtain vibration data of the frequency converter and instantaneous voltage data of key nodes of the frequency converter;

[0010] An abnormal area module is connected with the preprocessing module and is used to analyze the vibration data to determine a flow disturbance representation value, calculate a feature superposition coefficient in combination with the environmental gradient influence representation value, confirm a plurality of to-be-detected areas, and set labels for each to-be-detected area;

[0011] a performance analysis module, connected with the abnormal area module, for analyzing the transient voltage data to determine a mutation influence characteristic value, calculating a performance abnormality characteristic value based on the mutation influence characteristic value and a characteristic superposition coefficient, and judging whether to call a detector;

[0012] a detector, connected with the performance analysis module, for locating a detection area and detecting the detection area;

[0013] a controller, connected with the performance analysis module and the detector, for controlling a start detection time and a detection frequency of the detector based on the performance abnormality characteristic value;

[0014] wherein the space environment data includes an environment temperature and an environment humidity, and the key nodes include an input side power grid voltage mutation node and an output side motor terminal voltage mutation node.

[0015] Further, the preprocessing module determines an environment gradient influence characteristic value, including,

[0016] constructing a three-dimensional coordinate system and placing the environment data in the three-dimensional coordinate system;

[0017] determining a ratio of the environment temperature to a reference environment temperature as a temperature influence factor;

[0018] determining a ratio of the environment humidity to a reference environment humidity as a humidity influence factor;

[0019] determining a mean value of a sum of the temperature influence factor and the humidity influence factor as the environment gradient influence characteristic value.

[0020] Further, the abnormal area module determines a flow disturbance characteristic value, including,

[0021] determining a vibration frequency of the frequency converter and a corresponding amplitude;

[0022] determining a ratio of the vibration frequency to a reference vibration frequency as a frequency influence factor;

[0023] determining a ratio of a variance of the amplitude to a reference amplitude variance as an amplitude influence factor;

[0024] determining a mean value of a sum of the frequency influence factor and the amplitude influence factor as the flow disturbance characteristic value.

[0025] Further, the abnormal area module calculates a characteristic superposition coefficient, including,

[0026] determining a ratio of the flow disturbance characteristic value to a reference flow disturbance characteristic value as a first influence factor;

[0027] determining a ratio of the ambient gradient influence characteristic value to a reference ambient gradient influence characteristic value as a second influence factor;

[0028] determining a weighted sum of the first influence factor and the second influence factor as a feature superposition coefficient.

[0029] Further, the abnormal region module confirms a plurality of to-be-detected regions, sets a label for each of the to-be-detected regions, and includes,

[0030] determining the feature superposition coefficient in the three-dimensional coordinate system;

[0031] If the feature superposition coefficient is greater than or equal to a feature superposition coefficient threshold value, the feature superposition coefficients corresponding to consecutive values greater than or equal to the feature superposition coefficient threshold value are divided into the same region, and a strong region influence label is set for the region;

[0032] If the feature superposition coefficient is less than the feature superposition coefficient threshold value, the feature superposition coefficients corresponding to consecutive values less than the feature superposition coefficient threshold value are divided into the same region, and a weak region influence label is set for the region.

[0033] Further, the performance analysis module analyzes the transient voltage data to determine a mutation influence characteristic value, including,

[0034] calculating a difference between the input side power grid voltage mutation node and the output side motor terminal voltage mutation node;

[0035] determining a ratio of the difference to a reference difference as a mutation influence characteristic value.

[0036] Further, the performance analysis module calculates a performance anomaly characteristic value, including,

[0037] determining a ratio of the mutation influence characteristic value to a reference mutation influence characteristic value as a mutation influence factor;

[0038] determining a ratio of the feature superposition coefficient to a reference feature superposition coefficient as a feature influence factor;

[0039] determining a weighted sum of the mutation influence factor and the feature influence factor as a performance anomaly characteristic value.

[0040] Further, the performance analysis module determines whether to call a detector, wherein,

[0041] If the performance anomaly characteristic value is greater than or equal to a performance anomaly characteristic value threshold value, the detector is called;

[0042] If the performance anomaly representation value is less than the performance anomaly representation value threshold, the detector is not called, and the detection frequency is adjusted.

[0043] Further, the detection region corresponds to a region in which the performance anomaly representation value is greater than or equal to the performance anomaly representation value threshold.

[0044] Further, the controller controls the start detection time and the detection frequency of the detector based on the performance anomaly representation value, including,

[0045] determining the time at which the detector is called as the start detection time;

[0046] determining that the detection frequency is positively correlated with the performance anomaly representation value.

[0047] Compared with the prior art, the present application sets a pretreatment module, an anomaly influence module, a performance analysis module, a detector, and a controller, calculates a feature superposition coefficient based on the determined environmental gradient influence representation value and the flow disturbance representation value, confirms a plurality of detection regions, sets a label for each of the detection regions, determines a mutation influence representation value in response to the set strong region influence label, calculates a performance anomaly representation value in combination with the feature superposition coefficient, determines to call the detector, locates the detection region, detects the detection region, and controls the start detection time and the detection frequency of the detector. The present application analyzes the frequency converter in a regional manner, comprehensively analyzes the influence of environmental data, vibration data, and instantaneous voltage on the detection accuracy of the frequency converter, and determines the region that needs to be detected, thereby improving the detection efficiency and the detection accuracy.

[0048] In particular, the environmental gradient influence representation value is determined by analyzing the environmental temperature and the environmental humidity, which provides a data basis for subsequent calculation of the feature superposition coefficient and regional division. In actual situations, different operating units generate different degrees of heat during the operation of the frequency converter, causing non-uniform temperature rise of each operating unit of the frequency converter, which damages the performance of the frequency converter. At the same time, the humidity distribution of the operating space is also uneven. High humidity environments (> 60% RH) are prone to condensation short circuits, insulation deterioration, and metal corrosion. Low humidity environments (< 40% RH) are prone to static electricity accumulation and accelerated organic material embrittlement. Further, these two uneven distributions cause the internal failure risk of the frequency converter to present a spatial gradient feature. At this time, if the past global detection method is used, data redundancy will be caused due to excessive sampling of low-risk regions, resulting in excessive consumption of calculation, storage, and communication resources. If the discrete single-point detection method is used, it will be difficult to effectively cover the high-risk region due to insufficient spatial resolution, resulting in the omission of key failure signs. Based on this, the present application considers dividing the operating space of the frequency converter into regions and analyzing the environmental data as a division condition to improve the performance detection efficiency and accuracy of the frequency converter.

[0049] Especially, by analyzing the vibration data of the frequency converter, the characteristic superposition coefficient is calculated in combination with the environmental gradient influence characteristic value to divide the operation space into regions and set labels. In the operation process of the frequency converter, each operation unit will generate vibration spectra with different characteristics due to the effects of electromagnetic force, switching noise, mechanical resonance, and cooling fan excitation. The difference in vibration frequency affects the aerodynamic characteristics of the micro-environment inside and around the frequency converter, leading to uneven air flow field distribution in the operation space, forming a low flow rate environment in some areas, affecting the normal use of the frequency converter. Further, the vibration data in this area will further strengthen the deviation of temperature and humidity, increase the damage probability of the frequency converter, and affect the performance detection of the frequency converter. Based on this, the vibration data is also used as a regional division condition, and is coupled with the environmental data for analysis to calculate the characteristic superposition coefficient to divide the operation space into regions and set labels for the detection area, providing a data basis for subsequent detection, and improving the performance detection efficiency and accuracy of the frequency converter.

[0050] Especially, in response to the setting result of the strong regional influence label, the transient voltage data of the frequency converter is analyzed to determine the mutation influence characteristic value, and the performance abnormality characteristic value is calculated in combination with the characteristic superposition coefficient to call the detector. In actual situations, the performance detection of the frequency converter is mostly periodic point detection, or responsive detection based on the results of global detection and discrete single-point detection. It can be understood that both periodic detection and responsive detection have a time lag, and discrete sampling cannot continuously capture the damage process, leading to the continuous accumulation of hidden failures before the threshold alarm. More seriously, the conservative setting of the alarm threshold makes the system cross the damage inflection point when triggered, and the multi-physical field coupling effect further causes irreversible performance degradation. Further, the prior art detects the entire frequency converter to determine the abnormal position, which is low in efficiency. Based on this, the present application considers analyzing multi-source data only in the region where the strong regional influence label is set, and comprehensively analyzing the characteristic superposition coefficient and the influence of the transient voltage of the frequency converter on the performance of the region to quickly locate the abnormal region of the frequency converter, thereby improving the performance detection efficiency and accuracy of the frequency converter.

[0051] Especially, the present application sets a controller to regulate the detector according to the analysis result of the performance analysis module, so as to keep the detection of the frequency converter timely responsive and improve the performance detection efficiency and accuracy of the frequency converter. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 The structural schematic diagram of the intelligent device for performance detection of the frequency converter of the present application embodiment;

[0053] Figure 2 The logic block diagram of the present application embodiment for confirming a plurality of detection areas and setting labels for each of the detection areas;

[0054] Figure 3 This is a logic block diagram illustrating whether to invoke the detector in an embodiment of the invention. Detailed Implementation

[0055] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0056] It should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the term "connection" should be interpreted broadly. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0057] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0058] Please see Figure 1 , Figure 1 This is a schematic diagram of the structure of an intelligent device for performance testing of a frequency converter according to an embodiment of the invention. The intelligent device for performance testing of a frequency converter of the present invention includes:

[0059] The preprocessing module is used to acquire environmental data of the inverter's operating space, analyze the environmental data to determine the environmental gradient influence characterization value, acquire the inverter's vibration data and the instantaneous voltage data of the inverter's key nodes;

[0060] An abnormal region module, which is connected to the preprocessing module, is used to analyze the vibration data to determine the flow disturbance characterization value, calculate the feature superposition coefficient by combining the environmental gradient influence characterization value, identify several areas to be detected, and set labels for each area to be detected.

[0061] The performance analysis module, which is connected to the abnormal region module, analyzes the instantaneous voltage data in response to the set strong region influence label to determine the sudden change influence characterization value, calculates the performance anomaly characterization value based on the sudden change influence characterization value and the feature superposition coefficient, and determines whether to call the detector.

[0062] A detector, connected to the performance analysis module, is used to locate the detection area and perform detection on the detection area;

[0063] A controller, connected to the performance analysis module and the detector, is used to control the start detection time and detection frequency of the detector based on the performance anomaly characterization value.

[0064] The spatial environment data includes ambient temperature and ambient humidity, and the key nodes include the input-side power grid voltage mutation node and the output-side motor terminal voltage mutation node.

[0065] Specifically, there are no restrictions on the method of obtaining ambient temperature. For example, it can be an infrared camera set up inside the operating space to detect the temperature distribution. Of course, other devices can also be used, as long as they can detect the temperature data of the operating space. This will not be elaborated further.

[0066] Specifically, there is no limitation on the method of obtaining ambient humidity. For example, it can be done by setting up a network based on point sensors to scan the ambient humidity. Of course, those skilled in the art can also use other methods, as long as it can ensure that the humidity data of the operating space can be detected. This will not be elaborated further.

[0067] Specifically, there are no restrictions on the method of acquiring vibration data. For example, vibration data can be collected by an infrared camera. Of course, other devices can also be used, as long as they can detect the vibration data of the frequency converter. This will not be elaborated further.

[0068] Specifically, there are no restrictions on the method of acquiring instantaneous voltage data. For example, fiber optic isolated voltage sensors can be set up at key nodes, or other devices can be used, as long as they can detect the instantaneous voltage data of the frequency converter. This will not be elaborated further.

[0069] Specifically, the preprocessing module determines the characterization values ​​of the environmental gradient impact, including,

[0070] Used to construct a three-dimensional coordinate system, and to place the spatial environment data within the three-dimensional coordinate system;

[0071] The ratio of the ambient temperature to the reference ambient temperature is used to determine the temperature influence factor.

[0072] The ratio of the ambient humidity to the reference ambient humidity is used to determine the humidity influence factor;

[0073] The mean of the sum of the temperature influence factor and the humidity influence factor is used to determine the environmental gradient influence characterization value.

[0074] Specifically, in implementation, the operating space of the frequency converter is regarded as a cube, the three axes of the three-dimensional coordinate system are the three sides of the cube, and the origin of the three-dimensional coordinate system is the intersection of the three sides of the cube.

[0075] Specifically, the reference ambient temperature is calculated in advance by obtaining several temperature values ​​of the frequency converter during normal operation and determining the average value of each temperature value as the reference ambient temperature.

[0076] Specifically, the reference ambient humidity is calculated in advance by obtaining several humidity values ​​during the normal operation of the frequency converter and determining the average value of each humidity value as the reference ambient humidity.

[0077] Specifically, by analyzing ambient temperature and humidity, the influence of environmental gradients is determined, providing a data foundation for subsequent calculation of feature superposition coefficients and regional division. In practice, during inverter operation, different operating units generate varying degrees of heat, leading to abnormal non-uniform temperature rises in each unit and impairing inverter performance. Simultaneously, the humidity distribution in the operating space is uneven. High humidity environments (>60% RH) easily cause condensation short circuits, insulation degradation, and metal corrosion; low humidity environments (<40% RH) easily accumulate static electricity and accelerate the embrittlement of organic materials, further... These two uneven distributions can lead to spatial gradient characteristics in the internal fault risk of the frequency converter. In this case, if the traditional full-domain detection method is used, data redundancy will occur due to oversampling of low-risk areas, resulting in excessive consumption of computing, storage and communication resources. If the discrete single-point detection method is used, it will be difficult to effectively cover high-risk areas due to insufficient spatial resolution, resulting in the omission of key fault symptoms. Based on this, the present invention considers dividing the operating space of the frequency converter into regions and using environmental data as the division condition for analysis to improve the efficiency and accuracy of frequency converter performance detection.

[0078] The abnormal region module determines the flow disturbance characterization values, including,

[0079] Used to determine the vibration frequency and corresponding amplitude of the frequency converter;

[0080] The frequency influence factor is used to determine the ratio of the vibration frequency to the reference vibration frequency.

[0081] The amplitude influence factor is used to determine the ratio of the amplitude variance to the reference amplitude variance.

[0082] The mean of the sum of the frequency influence factor and the amplitude influence factor is used to determine the flow disturbance characterization value.

[0083] Specifically, the reference vibration frequency is calculated in advance. Several vibration frequencies of the frequency converter during normal operation are obtained in advance, and the average value of each vibration frequency is determined as the reference vibration frequency.

[0084] Specifically, the reference amplitude variance is calculated in advance. Several amplitude variances of the frequency converter during normal operation are obtained in advance, and the average value of each amplitude variance is determined as the reference amplitude variance.

[0085] The abnormal region module calculates the feature overlay coefficients, including:

[0086] The first influencing factor is used to determine the ratio of the flow disturbance characterization value to the baseline flow disturbance characterization value.

[0087] The second influencing factor is used to determine the ratio of the environmental gradient influence characterization value to the baseline environmental gradient influence characterization value.

[0088] The weighted sum of the first influence factor and the second influence factor is used to determine the feature superposition coefficient.

[0089] Specifically, the baseline flow interference characterization value is calculated in advance. Several flow interference characterization values ​​are obtained in advance during the normal operation of the frequency converter, and the average value of each flow interference characterization value is determined as the baseline flow interference characterization value.

[0090] Specifically, the baseline environmental gradient influence characterization value is calculated in advance. Several environmental gradient influence characterization values ​​during the normal operation of the frequency converter are obtained in advance, and the average value of each environmental gradient influence characterization value is determined as the baseline environmental gradient influence characterization value.

[0091] Specifically, the sum of the weight coefficients of the first and second influencing factors is 1. When adjusting the weights, considering that environmental factors have a greater impact on inverter performance than vibration, the second influencing factor is given a higher weight. The weight coefficient of the first influencing factor is set to 0.4 and the weight coefficient of the second influencing factor is set to 0.6.

[0092] Specifically, by analyzing the vibration data of the frequency converter and combining it with the characteristic superposition coefficient of the environmental gradient influence value, the operating space is divided into regions and labeled. During the operation of the frequency converter, each operating unit will generate vibration spectra with different characteristics due to electromagnetic force, switching noise, mechanical resonance, and cooling fan excitation. The differences in these vibration frequencies affect the aerodynamic characteristics of the internal and surrounding microenvironment of the frequency converter, resulting in uneven airflow distribution in the operating space and the formation of low-velocity environments in some areas, affecting the normal operation of the frequency converter. Furthermore, the vibration data in these areas will amplify the deviation of temperature and humidity, increasing the probability of damage to the frequency converter and affecting the performance testing of the frequency converter. Based on this, this invention also uses vibration data as a region division condition and performs coupled analysis with environmental data to calculate the characteristic superposition coefficient, thereby dividing the operating space into regions and labeling the areas to be tested. This provides a data basis for whether to conduct subsequent testing, improving the efficiency and accuracy of frequency converter performance testing.

[0093] Please see Figure 2 , Figure 2 This is a logic block diagram illustrating the process of identifying several regions to be detected and assigning labels to each of the regions to be detected, as described in an embodiment of the invention. Specifically, the abnormal region module identifies several regions to be detected and assigns labels to each of the regions to be detected, including:

[0094] Used to place the feature superposition coefficients within the three-dimensional coordinate system;

[0095] If the feature superposition coefficient is greater than or equal to the feature superposition coefficient threshold, then the consecutive feature superposition coefficients that are greater than or equal to the feature superposition coefficient threshold are divided into the same region, and a strong region influence label is set for the region.

[0096] If the feature superposition coefficient is less than the feature superposition coefficient threshold, then consecutive feature superposition coefficients less than the feature superposition coefficient threshold are divided into the same region, and a weak region influence label is set for the region.

[0097] Specifically, the feature superposition coefficient threshold characterizes a boundary of regional changes in the frequency converter under the influence of multiple factors. As pre-calculated data, several feature superposition coefficients of the frequency converter during abnormal operation are obtained in advance, and the product of the mean of each feature superposition coefficient and the influence coefficient is determined as the feature superposition coefficient threshold. In practice, in order to improve the calculation accuracy, the influence coefficient is determined to be 0.9.

[0098] It is understandable that within a three-dimensional coordinate system, there are several regions that strongly influence the label and regions that weakly influence the label. Regions that strongly influence the label may be connected to regions that weakly influence the label.

[0099] Specifically, the performance analysis module analyzes the instantaneous voltage data to determine the characterization values ​​of the impact of sudden changes, including,

[0100] Used to calculate the difference between the voltage abrupt change node on the input side of the power grid and the voltage abrupt change node on the output side of the motor end;

[0101] The ratio of the difference to the baseline difference is used to determine the characterization value of the mutation effect.

[0102] Specifically, the benchmark difference is calculated in advance. Several differences are obtained in advance during the normal operation of the frequency converter, and the average value of each difference is determined as the benchmark difference.

[0103] Specifically, the performance analysis module calculates performance anomaly characteristic values, including,

[0104] The ratio of the mutation impact characterization value to the baseline mutation impact characterization value is used to determine the mutation impact factor;

[0105] The ratio of the feature superposition coefficient to the benchmark feature superposition coefficient is used to determine the feature influence factor;

[0106] The weighted sum of the mutation impact factor and the characteristic impact factor is used to determine the performance anomaly characterization value.

[0107] Specifically, the baseline mutation impact characterization value is calculated in advance. Several mutation impact characterization values ​​during the normal operation of the frequency converter are obtained in advance, and the average value of each mutation impact characterization value is determined as the baseline mutation impact characterization value.

[0108] Specifically, the benchmark feature superposition coefficient is the feature superposition coefficient corresponding to the benchmark environmental gradient influence characterization value and the benchmark flow disturbance characterization value.

[0109] Specifically, the sum of the weighting coefficients of the mutation impact factor and the characteristic impact factor is 1. When adjusting the weighting coefficients, considering that the pressure difference can lead to a faster reduction in inverter performance, the weighting coefficient of the mutation impact factor is set to 0.6 and the weighting coefficient of the characteristic impact factor is set to 0.4.

[0110] Specifically, in response to the setting results of the strong regional influence tag, the instantaneous voltage data of the frequency converter is analyzed to determine the characteristic value of the sudden change effect. Combined with the feature superposition coefficient, the performance anomaly characteristic value is calculated, and the detector is called. In practice, the performance detection of frequency converters is mostly periodic point inspection, or responsive detection based on the results of global detection and discrete single-point detection. It is understandable that both periodic detection and responsive detection have time delays. Discrete sampling cannot continuously capture the damage process, causing latent failures to continue to accumulate before the threshold alarm. More seriously, the conservative setting of the alarm threshold means that the system has already crossed the damage inflection point when it is triggered, and the multi-physics coupling effect further causes irreversible performance degradation. Furthermore, the existing technology detects the entire frequency converter and determines the abnormal location during the detection, which is inefficient. Based on this, the present invention considers analyzing multi-source data only for the area where the strong regional influence tag is set, comprehensively analyzing the influence of the feature superposition coefficient and the instantaneous voltage of the frequency converter on the performance of the area, quickly locating the abnormal area of ​​the frequency converter, and improving the efficiency and accuracy of frequency converter performance detection.

[0111] Please see Figure 3 , Figure 3 This is a logic block diagram illustrating whether to invoke the detector according to an embodiment of the invention. Specifically, the performance analysis module determines whether to invoke the detector, wherein...

[0112] If the performance anomaly characterization value is greater than or equal to the performance anomaly characterization value threshold, then the detector is invoked;

[0113] If the performance anomaly characterization value is less than the performance anomaly characterization value threshold, the detector will not be invoked, and the detection frequency will be adjusted.

[0114] Specifically, the performance anomaly characterization threshold represents a boundary where the frequency converter malfunctions. It is pre-calculated data, in which several performance anomaly characterization values ​​of the frequency converter during abnormal operation are obtained in advance, and the product of each performance anomaly characterization value and the performance coefficient is determined as the performance anomaly characterization threshold. In practice, in order to improve the calculation accuracy, the performance coefficient is determined to be 0.9.

[0115] Specifically, the detection area is the region corresponding to the performance anomaly characterization value being greater than or equal to the performance anomaly characterization value threshold.

[0116] Specifically, the controller controls the detector's start detection time and detection frequency based on the performance anomaly characterization values, including:

[0117] The time when the detector is invoked is used to determine the start detection time;

[0118] This is used to determine that the detection frequency is positively correlated with the performance anomaly characterization value.

[0119] It is understandable that a reference detection frequency exists, which is the frequency set by the project and can be obtained from the relevant files for installing the frequency converter.

[0120] Specifically,

[0121] If the performance anomaly characterization value is less than the performance anomaly characterization value threshold but greater than 0.7 times the performance anomaly characterization value threshold, then the detection frequency is determined to be twice the benchmark detection frequency.

[0122] If the performance anomaly characterization value is less than or equal to 0.7 times the performance anomaly characterization value threshold and greater than or equal to 0.5 times the performance anomaly characterization value threshold, then the detection frequency is determined to be 1.5 times the benchmark detection frequency.

[0123] If the performance anomaly characterization value is less than 0.5 times the performance anomaly characterization value threshold, the benchmark detection frequency remains unchanged.

[0124] It is understandable that the detection frequency is a positive integer, and the decimal part is rounded up during the calculation.

[0125] Specifically, the present invention includes a controller to adjust the detector based on the analysis results of the performance analysis module, so as to maintain timely response to the detection of the frequency converter and improve the efficiency and accuracy of frequency converter performance detection.

[0126] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0127] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An intelligent device for performance testing of frequency converters, characterized in that, include: The preprocessing module is used to acquire environmental data of the inverter's operating space, analyze the environmental data to determine the environmental gradient influence characterization value, acquire the inverter's vibration data and the instantaneous voltage data of the inverter's key nodes; An abnormal region module, which is connected to the preprocessing module, is used to analyze the vibration data to determine the flow disturbance characterization value, calculate the feature superposition coefficient by combining the environmental gradient influence characterization value, identify several areas to be detected, and set labels for each area to be detected. The performance analysis module, which is connected to the abnormal region module, analyzes the instantaneous voltage data in response to the set strong region influence label to determine the sudden change influence characterization value, calculates the performance anomaly characterization value based on the sudden change influence characterization value and the feature superposition coefficient, and determines whether to call the detector. A detector, connected to the performance analysis module, is used to locate the detection area and perform detection on the detection area; A controller, connected to the performance analysis module and the detector, is used to control the start detection time and detection frequency of the detector based on the performance anomaly characterization value. The spatial environment data includes ambient temperature and ambient humidity, and the key nodes include the input-side power grid voltage mutation node and the output-side motor terminal voltage mutation node.

2. The intelligent device for performance testing of a frequency converter according to claim 1, characterized in that, The preprocessing module determines the environmental gradient impact characterization values, including: Used to construct a three-dimensional coordinate system, and to place the environmental data within the three-dimensional coordinate system; The ratio of the ambient temperature to the reference ambient temperature is used to determine the temperature influence factor. The ratio of the ambient humidity to the reference ambient humidity is used to determine the humidity influence factor; The mean of the sum of the temperature influence factor and the humidity influence factor is used to determine the environmental gradient influence characterization value.

3. The intelligent device for performance testing of a frequency converter according to claim 1, characterized in that, The abnormal region module determines the flow disturbance characterization values, including, Used to determine the vibration frequency and corresponding amplitude of the frequency converter; The frequency influence factor is used to determine the ratio of the vibration frequency to the reference vibration frequency. The amplitude influence factor is used to determine the ratio of the amplitude variance to the reference amplitude variance. The mean of the sum of the frequency influence factor and the amplitude influence factor is used to determine the flow disturbance characterization value.

4. The intelligent device for performance testing of a frequency converter according to claim 1, characterized in that, The abnormal region module calculates the feature superposition coefficient, including: The first influencing factor is used to determine the ratio of the flow disturbance characterization value to the baseline flow disturbance characterization value. The second influencing factor is used to determine the ratio of the environmental gradient influence characterization value to the baseline environmental gradient influence characterization value. The weighted sum of the first influence factor and the second influence factor is used to determine the feature superposition coefficient.

5. The intelligent device for performance testing of a frequency converter according to claim 2, characterized in that, The abnormal region module identifies several regions to be detected and assigns labels to each region, including... Used to place the feature superposition coefficients within the three-dimensional coordinate system; If the feature superposition coefficient is greater than or equal to the feature superposition coefficient threshold, then the consecutive feature superposition coefficients that are greater than or equal to the feature superposition coefficient threshold are divided into the same region, and a strong region influence label is set for the region. If the feature superposition coefficient is less than the feature superposition coefficient threshold, then consecutive feature superposition coefficients less than the feature superposition coefficient threshold are divided into the same region, and a weak region influence label is set for the region.

6. The intelligent device for performance testing of a frequency converter according to claim 1, characterized in that, The performance analysis module analyzes the instantaneous voltage data to determine the characterization values ​​of the impact of sudden changes, including: Used to calculate the difference between the voltage abrupt change node on the input side of the power grid and the voltage abrupt change node on the output side of the motor end; The ratio of the difference to the baseline difference is used to determine the characterization value of the mutation effect.

7. The intelligent device for performance testing of a frequency converter according to claim 1, characterized in that, The performance analysis module calculates performance anomaly characterization values, including: The ratio of the mutation impact characterization value to the baseline mutation impact characterization value is used to determine the mutation impact factor; The ratio of the feature superposition coefficient to the benchmark feature superposition coefficient is used to determine the feature influence factor; The weighted sum of the mutation impact factor and the characteristic impact factor is used to determine the performance anomaly characterization value.

8. The intelligent device for performance testing of a frequency converter according to claim 1, characterized in that, The performance analysis module determines whether to invoke the detector, wherein... If the performance anomaly characterization value is greater than or equal to the performance anomaly characterization value threshold, then the detector is invoked; If the performance anomaly characterization value is less than the performance anomaly characterization value threshold, the detector will not be invoked, and the detection frequency will be adjusted.

9. The intelligent device for performance testing of a frequency converter according to claim 1, characterized in that, The detection area is the region corresponding to the performance anomaly characterization value being greater than or equal to the performance anomaly characterization value threshold.

10. The intelligent device for performance testing of a frequency converter according to claim 1, characterized in that, The controller controls the start detection time and detection frequency of the detector based on the performance anomaly characterization value, including: The time when the detector is invoked is used to determine the start detection time; This is used to determine that the detection frequency is positively correlated with the performance anomaly characterization value.

Images