Electronic device and method for diagnosing motor performance

Abstract

An electronic device for setting a test pattern for diagnosing motor performance according to one embodiment of the present invention may comprise a processor configured to set a target frequency of a motor, divide a test period of the motor into an acceleration period, a constant-speed period, and a deceleration period on the basis of the target frequency, set a frequency ramp rate over time in the acceleration period and the deceleration period, and set a test time for the constant-speed period.

Description

Electronic device and method for diagnosing motor performance

[0001] The present invention relates to an electronic device and method for diagnosing the performance of a motor.

[0002] An inverter is a device connected to a motor that controls its operation. If a malfunction occurs in the motor, it can affect the inverter connected to it. If it is determined that the inverter may be damaged due to a motor malfunction, the inverter's operation can be stopped via a trip, thereby halting motor control itself.

[0003] Generally, motors require a high current when first starting and operate at a constant speed after reaching the target speed following an acceleration phase. After operation according to commands, they pass through a deceleration phase and come to a stop. Additionally, in the case of a bidirectional motor, it can operate in both forward and reverse directions.

[0004] Motor performance tests, which are executed to determine whether a motor is malfunctioning, are performed using patterns used for motor operation in the actual field. However, since there are limitations in the range for determining motor malfunctions, the performance of the model is degraded. Furthermore, problems can only be identified through actual operation, which has an adverse effect on materials and productivity.

[0005] Meanwhile, with the commercialization of AI models, there is an increasing number of attempts to use them to determine motor abnormalities. However, despite the acceleration and deceleration sections being the most critical for motor damage and the fastest time to identify abnormalities, currently developed training models have limitations in that they are trained using only data from constant speed sections.

[0006] Therefore, it is necessary to derive an optimal pattern to predict the motor state more accurately without limiting model performance.

[0007] Furthermore, it is necessary to design a learning model capable of predicting the motor's condition even during acceleration and deceleration and providing diagnostic information to enable proactive maintenance.

[0008] The objective of the present invention is to provide an electronic device and method that further minimize damage to the motor and can quickly detect abnormalities.

[0009] The object of the present invention is to an electronic device and method for setting a test pattern for diagnosing the performance of an optimal motor.

[0010] An electronic device for diagnosing the performance of a motor according to an embodiment of the present invention for solving the above technical problem comprises: a memory in which at least one instruction is stored; and a processor for executing said instruction. The processor collects control data for each of a predefined acceleration section, constant speed section, and deceleration section of the motor, and identifies whether there is an abnormality in said motor by inputting information corresponding to said control data into a plurality of models and obtaining output information, wherein said plurality of models may be learned for each of said acceleration section, said constant speed section, and said deceleration section.

[0011] In one embodiment of the present invention, the processor can determine a frequency change rate corresponding to the acceleration section and the deceleration section, determine a test time corresponding to the constant speed section to generate a test pattern, and test the motor according to the test pattern to collect the control data.

[0012] In one embodiment of the present invention, the processor can determine the maximum frequency of the test pattern based on the maximum operating frequency of the motor.

[0013] In one embodiment of the present invention, the processor calculates a range of frequency change rates over time in the acceleration and deceleration sections based on the maximum operating frequency of the motor and the shortest starting time and maximum starting time according to the DC link voltage, and the frequency change rate in the test pattern may be included in the range of frequency change rates.

[0014] In one embodiment of the present invention, the plurality of models may include a first plurality of models learned to determine whether there is a motor abnormality for each of two or more rate of change intervals in the acceleration interval, a second model learned to determine whether there is a motor abnormality in the constant speed interval, and a third plurality of models learned to determine whether there is a motor abnormality for each of two or more rate of change intervals in the deceleration interval.

[0015] In one embodiment of the present invention, the two or more rate of change intervals may be set based on a first frequency rate of change according to the shortest starting time according to the DC link voltage and a second frequency rate of change according to the maximum starting time according to the DC link voltage.

[0016] In one embodiment of the present invention, the processor can identify whether there is an abnormality in the motor in the acceleration section by inputting the control data to a model among the first plurality of models that corresponds to the rate of change of the frequency of the acceleration section.

[0017] In one embodiment of the present invention, the processor can identify whether there is an abnormality in the motor in the deceleration section by inputting the control data to a model among the third plurality of models that corresponds to the frequency change rate of the deceleration section.

[0018] In one embodiment of the present invention, the processor can diagnose the performance of the motor by determining whether there is a motor abnormality in the acceleration section, constant speed section, and deceleration section.

[0019] In one embodiment of the present invention, the control data may include current data and frequency data of the motor.

[0020] In addition, a method for diagnosing the performance of a motor according to an embodiment of the present invention for solving the above technical problem comprises: a step of collecting control data for each of the predefined acceleration section, constant speed section, and deceleration section of the motor; and a step of identifying whether there is an abnormality in the model by inputting information corresponding to the control data into a plurality of models and obtaining output information, wherein the plurality of models may be learned for each of the acceleration section, the constant speed section, and the deceleration section.

[0021] In one embodiment of the present invention, the method may further include the steps of: determining a frequency change rate corresponding to the acceleration section and the deceleration section, and determining a test time corresponding to the constant speed section to generate a test pattern; and testing the motor according to the test pattern to collect the control data.

[0022] In one embodiment of the present invention, the method may further include the step of determining the maximum frequency of the test pattern based on the maximum operating frequency of the motor.

[0023] In one embodiment of the present invention, the method further includes the step of calculating a range of frequency change rates over time in the acceleration and deceleration sections based on the maximum operating frequency of the motor and the shortest starting time and maximum starting time according to the DC link voltage, and the frequency change rate in the test pattern may be included in the range of frequency change rates.

[0024] In one embodiment of the present invention, the plurality of models may include a first plurality of models learned to determine whether there is a motor abnormality for each of two or more rate of change intervals in the acceleration interval, a second model learned to determine whether there is a motor abnormality in the constant speed interval, and a third plurality of models learned to determine whether there is a motor abnormality for each of two or more rate of change intervals in the deceleration interval.

[0025] In one embodiment of the present invention, the two or more rate of change intervals may be set based on a first frequency rate of change according to the shortest starting time according to the DC link voltage and a second frequency rate of change according to the maximum starting time according to the DC link voltage.

[0026] In one embodiment of the present invention, the method may further include the step of inputting the control data into a model among the first plurality of models corresponding to the frequency change rate of the acceleration section to identify whether there is an abnormality in the motor during the acceleration section.

[0027] In one embodiment of the present invention, the method may further include the step of identifying whether there is an abnormality in the motor in the deceleration section by inputting the control data into a model among the third plurality of models that corresponds to the frequency change rate of the deceleration section.

[0028] In one embodiment of the present invention, the method may further include a step of determining the performance of the motor based on whether there is a motor abnormality in the acceleration section, constant speed section, and deceleration section.

[0029] In one embodiment of the present invention, the control data may include current data and frequency data of the motor.

[0030] According to one embodiment of the present invention, the motor operating range can be changed more widely to diagnose the motor's performance more accurately and quickly.

[0031] According to one embodiment of the present invention, by distinguishing between test patterns before, after, and during operation, the cost of handling failures that may occur during actual operation can be reduced.

[0032] According to one embodiment of the present invention, the test pattern can also be utilized as a pattern for deriving the optimal starting value of the motor.

[0033] According to one embodiment of the present invention, since motor abnormalities can be identified during the acceleration / deceleration phase occurring simultaneously with motor startup, rapid detection of motor abnormalities is possible, thereby minimizing motor damage.

[0034] According to one embodiment of the present invention, since the model is developed and applied by dividing it into sections, the model can be trained more rigorously and precisely. In particular, as the greater the rate of change of frequency in the acceleration / deceleration section, the greater the heat / stress applied to the motor, the section (time point) of motor burnout can be precisely identified.

[0035] According to one embodiment of the present invention, since the site where the motor is started necessarily has an acceleration / deceleration section, it is expected to be applicable to various sites such as robots and automobiles.

[0036] FIG. 1 is a schematic diagram illustrating a motor performance diagnostic system according to one embodiment of the present invention.

[0037] FIG. 2 is a block diagram illustrating the configuration of an electronic device according to one embodiment of the present invention.

[0038] FIG. 3 is a diagram illustrating the operation flowchart of an electronic device according to one embodiment of the present invention.

[0039] FIG. 4 is a diagram illustrating a test section according to the frequency of a motor according to one embodiment of the present invention.

[0040] FIG. 5 is a diagram illustrating test pattern setting parameters according to one embodiment of the present invention.

[0041] FIG. 6 is a drawing illustrating a test pattern according to one embodiment of the present invention.

[0042] FIG. 7 is a drawing illustrating the operation of an electronic device according to the first embodiment of the present invention.

[0043] FIG. 8 is a drawing illustrating the operation of an electronic device according to a second embodiment of the present invention.

[0044] FIG. 9 is a diagram illustrating the operation flowchart of an electronic device according to one embodiment of the present invention.

[0045] FIG. 10 is a diagram illustrating the learning process of a model according to one embodiment of the present invention.

[0046] FIG. 11 is a drawing illustrating the operation of an electronic device according to one embodiment of the present invention.

[0047] Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings. The detailed description disclosed below, together with the accompanying drawings, is intended to describe exemplary embodiments of the present invention and is not intended to represent the only embodiment in which the present invention can be practiced. In order to clearly explain the present invention in the drawings, parts unrelated to the description may be omitted, and the same reference numerals may be used for identical or similar components throughout the specification.

[0048] FIG. 1 is a schematic diagram illustrating a motor performance diagnostic system according to one embodiment of the present invention.

[0049] The motor performance diagnostic system (1) of FIG. 1 may include an inverter (10), a motor (20), and an electronic device (100).

[0050] An inverter (10) is a device connected to a motor (20) to control the operation of the motor (20). The inverter (10) and the motor (20) can be connected to each other through various methods such as single-phase connection, three-phase connection, and pulse width modulation (PWM), and the inverter (10) may include a configuration for collecting control data according to the start of the motor (20). At this time, the control data may include current data, frequency data, etc. of the motor (20) as time-series data.

[0051] The electronic device (100) is a device that diagnoses the performance of the motor (20) using control data or sets a test pattern for diagnosing the performance of the motor, and can be implemented as a computer, PLC (Programmable Logic Controller), server, smartphone, tablet PC, smart pad, laptop, etc. In addition, unlike what is shown in FIG. 1, the electronic device (100) can be implemented as a component of the inverter (10), and the method of implementation of the electronic device (100) is not limited to any one. However, for the convenience of the following explanation, the inverter (10) and the electronic device (100) are assumed to be separate devices.

[0052] In the present invention, a method is proposed for setting a test pattern for motor diagnosis having various motor operating ranges to diagnose the performance of the motor (20) more accurately and quickly. In addition, a method is proposed to reduce the cost of handling failures that may occur during actual operation by diagnosing the performance of the motor (20) through the pattern before and after operation.

[0053] In addition, a method for diagnosing the performance of the motor (20) is proposed by utilizing a model trained to diagnose the performance of the motor (20) based on control data in the acceleration and deceleration sections.

[0054] In addition, the present invention proposes a method of utilizing a learned model to diagnose the performance of a motor (20) based on control data, and in particular, a method of diagnosing the performance of a motor (20) by subdividing it into multiple models based on the rate of change of frequency over time even within the acceleration and deceleration sections.

[0055] According to one embodiment of the present invention, it is possible to determine in advance whether there is an abnormality in the motor through a learned model while using the motor (20) in real time.

[0056] Hereinafter, the configuration and operation of an electronic device (100) according to one embodiment of the present invention will be described in detail with reference to the drawings.

[0057] FIG. 2 is a block diagram illustrating the configuration of an electronic device according to one embodiment of the present invention.

[0058] An electronic device (100) according to one embodiment of the present invention may include an input unit (110), a communication unit (120), a display unit (130), a storage unit (140), and a processor (150).

[0059] The input unit (110) generates input data in response to user input of the electronic device (100). For example, the user input may be a user input requesting the operation of a program for diagnosing motor performance, a user input setting the target frequency of the motor, a user input setting the rate of change of frequency over time, a user input setting the test time of a constant speed section, a user input preprocessing control data, a user input setting the rate of change section, etc. In addition to this, it may be applied without limitation if it is a user input necessary to identify whether there is an abnormality in the motor (20), set a test pattern for performance diagnosis, or diagnose the performance of the motor (20).

[0060] The input unit (110) includes at least one input means. The input unit (110) may include a keyboard, a key pad, a dome switch, a touch panel, a touch key, a mouse, a menu button, etc.

[0061] The communication unit (120) can communicate with external devices such as a server, an inverter (10) to transmit and receive control data, information about the test section, a target frequency, a frequency change rate, a test time per section, and multiple learning models.

[0062] The communication unit (120) can communicate with external devices such as a server, an inverter (10) to transmit and receive control data, information about the starting section, frequency change rate, multiple learning models, etc.

[0063] To this end, the communication unit (120) can perform wireless communication such as 5G (5th generation communication), LTE-A (long term evolution-advanced), LTE (long term evolution), Wi-Fi (wireless fidelity), Bluetooth, or wired communication such as LAN (local area network), WAN (Wide Area Network), and power line communication.

[0064] The display unit (130) displays display data according to the operation of the electronic device (100). The display unit (130) may display a screen for setting the target frequency of the motor, a screen for setting the rate of change of frequency over time in the acceleration and deceleration sections of the test section, a screen for setting the test time of the test section, etc. Each screen may be implemented in the form of a Graphic User Interface (GUI). In addition, it may display a screen illustrating the frequency change over time, a screen displaying the test section, a screen displaying the starting section, a screen displaying the abnormality and performance of the motor, and other screens receiving user input.

[0065] The display unit (130) includes a liquid crystal display (LCD), a light emitting diode (LED) display, an organic light emitting diode (OLED) display, a micro electro mechanical systems (MEMS) display, and an electronic paper display. The display unit (130) can be combined with the input unit (110) to be implemented as a touch screen.

[0066] The storage unit (140) stores operation programs of the electronic device (100). The storage unit (140) includes storage with non-volatile properties that can preserve data (information) regardless of whether power is provided, and memory with volatile properties in which data to be processed by the processor (150) is loaded and data cannot be preserved if power is not provided. Storage includes flash memory, hard-disc drive (HDD), solid-state drive (SSD), and ROM (Read Only Memory), and memory includes buffer and RAM (Random Access Memory).

[0067] The storage unit (140) can store control data, information about test sections, target frequency, frequency change rate, test time per section, multiple learning models, etc. The storage unit (140) can store calculation programs, etc., necessary during the process of setting the motor target frequency, classifying test sections, setting the frequency change rate, setting the test time, collecting control data, calculating the frequency change rate, learning models, identifying abnormalities in the motor, and diagnosing the performance of the motor.

[0068] Additionally, the storage unit (140) can store control data, information on the starting section, frequency change rate, and multiple learning models. The storage unit (140) can store calculation programs, etc., necessary for the process of collecting control data, classifying the starting section, calculating the frequency change rate, learning models, identifying whether there is an abnormality in the motor, and diagnosing the performance of the motor.

[0069] The processor (150) can control at least one other component (e.g., hardware or software component) of the electronic device (100) by executing software such as a program, and can perform various data processing or operations.

[0070] A processor (150) according to one embodiment of the present invention may set a target frequency of a motor, divide a test section of the motor into an acceleration section, a constant speed section, and a deceleration section based on the target frequency of the motor, set a rate of change of frequency over time in the acceleration section and the deceleration section of the motor, and set a test time for the constant speed section.

[0071] The processor (150) can test the motor according to the test pattern to collect control data, and input the control data into a plurality of models trained to determine whether the motor is abnormal according to the test section to identify whether the motor is abnormal. At this time, the processor (150) may train a plurality of models trained to determine whether the motor is abnormal, or receive and store a plurality of previously trained and built models from an external source and use them, and is not limited to either one.

[0072] A processor (150) according to one embodiment of the present invention collects control data according to the start of the motor, divides the start section of the motor into an acceleration section, a constant speed section, and a deceleration section based on the target frequency of the motor, calculates the rate of change of frequency over time in the acceleration section and the deceleration section based on the control data, and inputs the control data to a plurality of models trained to determine whether the motor is abnormal according to the start section and the rate of change, thereby identifying whether the motor is abnormal.

[0073] At this time, the processor (150) may train a plurality of models to determine whether there is an abnormality in the motor, or receive and store a plurality of previously trained and built models from the outside and use them, and is not limited to either one.

[0074] Meanwhile, the processor (150) may perform at least some of the data analysis, processing, and result information generation for performing the above operations using at least one of machine learning, neural network, or deep learning algorithms as a rule-based or artificial intelligence algorithm. Examples of neural networks may include models such as CNN (Convolutional Neural Network), DNN (Deep Neural Network), and RNN (Recurrent Neural Network).

[0075] FIG. 3 is a diagram illustrating the operation flowchart of an electronic device according to one embodiment of the present invention.

[0076] According to one embodiment of the present invention, the processor (150) can set the target frequency of the motor (20) (S10).

[0077] For example, the processor (150) can set a target frequency of the motor (20) based on the maximum operating frequency of the motor (20). The maximum operating frequency of the motor (20) refers to the frequency at which the motor (20) can operate to the maximum. For example, if the maximum operating frequency of the motor (20) is 50Hz, the target frequency of the motor (20) can be set in a range between 0Hz and 50Hz. At this time, the target frequency can be set in frequency units based on the maximum operating frequency of the motor (20), or set as n% of the maximum operating frequency (n is a real number from 0 to 100).

[0078] At this time, the processor (150) can set a target frequency using an artificial intelligence model trained to design an optimal test pattern. The optimal test pattern may be a pattern that can best detect anomalies or a pattern that can reduce test costs the most. The importance of the optimal test pattern may change depending on the factors used to determine failure prediction. For example, when predicting failure based on current, a pattern that increases the amount of change in current may be the optimal pattern. As another example, when predicting failure based on temperature, a pattern that increases the amount of change in temperature may be the optimal pattern.

[0079] The processor (150) can generate a test pattern by using an artificial intelligence model to adjust the operating frequency from n% of the maximum operating frequency to the maximum operating frequency, and can set an optimal target frequency based on the generated test pattern.

[0080] As another example, the processor (150) can receive user input to set the target frequency of the motor (20).

[0081] Test patterns for performance diagnosis can be divided into test patterns used before / after operation and test patterns used during operation. In the case of test patterns used before / after operation, the target frequency can be set by extending the range up to the maximum operating frequency. Test patterns used during operation can be set with the maximum frequency of the motor (20) used in the actual field as the target frequency. For example, a motor with a maximum operating frequency of 50Hz can be used at a maximum of 30Hz in the actual field.

[0082] According to one embodiment of the present invention, the processor (150) can divide the test section of the motor (20) into an acceleration section, a constant speed section, and a deceleration section based on the target frequency of the motor (20) (S20).

[0083] The test section of the motor (20) refers to a series of sections in which the motor (20) operates (stop → operation → stop). Specifically, the test section may include an acceleration section in which the motor (20) accelerates from a stopped state until it reaches a target frequency, a constant speed section in which it operates at a constant speed for a required time after reaching the target frequency, and a deceleration section in which it operates at a deceleration speed to return to a stopped state after the required time.

[0084] When the motor (20) operates in both directions, it will operate in the order of acceleration, constant speed, and deceleration sections in the reverse direction after the deceleration section. An example of dividing the test sections of the motor (20) is shown in FIG. 4.

[0085] According to one embodiment of the present invention, the processor (150) can set the rate of change of frequency over time in the acceleration and deceleration sections of the motor (20) (S30).

[0086] The processor (150) can calculate the range of the rate of change of frequency over time in the acceleration and deceleration sections based on the motor's target frequency and the shortest and maximum start times according to the DC link voltage.

[0087] DC link voltage refers to the basic DC voltage that the inverter (10) uses to generate AC voltage. The speed and torque of the motor are generally closely related to the DC link voltage. As the DC link voltage increases, the range of voltage and frequency supplied to the motor increases, so higher speed and output can be produced.

[0088] The DC link voltage rating can be determined according to the capacity of the inverter (10). For example, a 200V class inverter must have a DC link voltage of 410Vdc or less, and a 400V class inverter must have a DC link voltage of 820Vdc or less. The inverter (10) can be operated within a range where the DC link voltage is not exceeded. Therefore, the shortest starting time according to the DC link voltage may mean the time when the frequency change rate is highest within a range where the DC link voltage is not exceeded. The maximum starting time may be the maximum acceleration / deceleration time suitable for the actual field or the maximum time that can be set by the inverter (10) itself.

[0089] The processor (150) can set the rate of change of frequency over time using the time and target frequency of each acceleration and deceleration section. This can be expressed as the following mathematical formula 1.

[0090]

[0091] The processor (150) can set the test time for the acceleration section and the deceleration section. In this case, the processor (150) can calculate the frequency change rate based on the test time and the target frequency. The test time has a range from the shortest start time to the maximum start time.

[0092] However, in addition to this, the processor (150) can set the frequency change rate and calculate the test time by substituting the target frequency into the above mathematical formula 1. The frequency change rate can be set within the range of the previously calculated frequency change rate.

[0093] At this time, as described above, the processor (150) can set the rate of change of frequency over time in the acceleration and deceleration sections of the motor (20) using an artificial intelligence model trained to design an optimal test pattern. The processor (150) can likewise generate a test pattern by adjusting the test time within the range of the shortest start time to the maximum start time using an artificial intelligence model, and can set the test time in the acceleration and deceleration sections based on the generated test pattern.

[0094] According to one embodiment of the present invention, the processor (150) can set a test time for a constant speed section (S40).

[0095] The constant speed section is a section that operates at a constant frequency after reaching the target frequency. The processor (150) can receive user input to set the test time for the constant speed section.

[0096] At this time, as described above, the processor (150) can set the test time of the constant speed section using an artificial intelligence model trained to design an optimal test pattern. The processor (150) can likewise generate a test pattern while adjusting the test time using an artificial intelligence model, and set the optimal test time of the constant speed section based on the generated test pattern.

[0097] According to one embodiment of the present invention, the motor operating range can be changed more widely to diagnose the motor's performance more accurately and quickly.

[0098] According to one embodiment of the present invention, by distinguishing between test patterns before, after, and during operation, the cost of handling failures that may occur during actual operation can be reduced.

[0099] According to one embodiment of the present invention, an optimal starting value for a motor can be derived by referring to a result value (e.g., torque output, current, etc.) derived through a test pattern, and this can be reflected in a motor operation command.

[0100] FIG. 4 is a diagram illustrating a test section according to the frequency of a motor according to one embodiment of the present invention.

[0101] FIG. 4 illustrates an example of dividing a test section based on the frequency of the motor (20), as previously described in relation to S20 of FIG. 3. In FIG. 4, the explanation assumes that the motor (20) is a bidirectional motor.

[0102] The target frequency of the motor (20) is 50Hz, and based on the target frequency, the test section of the motor (20) in the forward direction (FWD) can be divided into an acceleration section (①), a constant speed section (②), and a deceleration section (③).

[0103] Afterwards, even in the reverse direction (REV), the test section of the motor (20) can be divided into an acceleration section (④), a constant speed section (⑤), and a deceleration section (⑥) based on the target frequency.

[0104] Setting forward and reverse directions allows for securing a wider variety of test patterns and is advantageous in terms of increasing test cases. On the other hand, if the motor operates in only one direction during operation, the configuration regarding testing in the other direction will need to be determined based on the specific site conditions.

[0105] FIG. 5 is a diagram illustrating test pattern setting parameters according to one embodiment of the present invention.

[0106] As previously described in relation to FIG. 3, the test pattern for performance diagnosis can be divided into a test pattern used before / after operation and a test pattern used during operation. That is, the electronic device (100) can set a test pattern for performance diagnosis of the motor corresponding to two or more performance diagnosis points. Meanwhile, when testing before / after operation, waste of material input can be prevented or the operation itself can be not affected, and productivity can be increased by taking measures before the next operation based on the results before / after operation.

[0107] Test pattern setting parameters may include a target frequency, a frequency change rate, a test direction, and a test time. First, the electronic device (100) may set the target frequency based on the maximum operating frequency of the motor or through user input. The electronic device (100) may set the frequency change rate by considering the target frequency, the shortest starting time and the maximum starting time according to the DC link voltage.

[0108] The electronic device (100) can set a test pattern for at least one of the forward direction and the reverse direction when the motor is driven in both directions.

[0109] The electronic device (100) can set test times for acceleration and deceleration sections based on the shortest start time and maximum start time according to the DC link voltage. The electronic device (100) can set test times for constant speed sections through user input.

[0110] FIG. 6 is a drawing illustrating a test pattern according to one embodiment of the present invention.

[0111] As an example of a test pattern, a test pattern for a site where the motor is operated at a maximum operating frequency of 50 Hz and a site where the motor is operated at 80% of the maximum operating frequency is illustrated.

[0112] At this time, each section (sections ① to ⑥) is considered to correspond to the sections of FIG. 4. A test pattern can be completed by setting the frequency change rate for the acceleration section (sections ① and ④) and the deceleration section (sections ③ and ⑥), and setting the test time for the constant speed section (sections ② and ⑤).

[0113] FIG. 7 is a drawing illustrating the operation of an electronic device according to the first embodiment of the present invention.

[0114] The electronic device (100) can collect control data by testing the motor according to a test pattern. The electronic device (100) can select a test pattern to be performed from among the pre-existing test patterns (e.g., first test pattern, second test pattern, third test pattern, etc.).

[0115] The control data may include current data and frequency data of the motor (20). In this case, if the motor (20) is connected to the inverter (10) in three phases, the current data may include current data for each phase (e.g., U, V, W phases). The electronic device (100) may collect control data from the inverter (10), but the collection path is not limited to any one path, such as through a server.

[0116] The electronic device (100) can identify whether there is an abnormality in the motor by inputting control data into a model trained to determine whether there is an abnormality in the motor.

[0117] The abnormality of the motor can be identified by inputting control data into multiple models trained to determine the abnormality of the motor according to the test section.

[0118] FIG. 8 is a drawing illustrating the operation of an electronic device according to a second embodiment of the present invention.

[0119] Unlike what was previously shown in Fig. 7, a case in which a separate model exists to determine whether there is an anomaly in each section will be explained.

[0120] The electronic device (100) can collect control data by testing the motor according to a test pattern. The electronic device (100) can select a test pattern to be performed from among the pre-existing test patterns (e.g., first test pattern, second test pattern, third test pattern, etc.).

[0121] The electronic device (100) can identify whether there is an abnormality in the motor by inputting control data into a plurality of models trained to determine whether there is an abnormality in the motor according to the test section.

[0122] Multiple models may include a first model trained to determine whether there is a motor abnormality in the acceleration section, a second model trained to determine whether there is a motor abnormality in the constant speed section, and a third model trained to determine whether there is a motor abnormality in the deceleration section. Multiple models may be generated for the reverse section as well as the forward section. The process of generating the models can be done using known artificial intelligence technologies such as deep learning and machine learning, so a description of specific details is omitted.

[0123] Meanwhile, the electronic device (100) can determine the overall performance of the motor (20) based on the results of each model.

[0124] There are various criteria for judging performance, and for example, it can be identified based on the importance of the section where the malfunction occurred, the proportion of the section where the malfunction occurred, etc. In addition, there are various ways to display performance. For example, a figure showing 85% of the total performance out of 100% could be displayed, or the repair status of the motor or inverter, the risk level, etc., could be indicated based on that figure.

[0125] According to one embodiment of the present invention, when determining whether there is an abnormality in the motor for each section, the time at which the abnormality occurs can be clearly distinguished, and the cause can be clearly distinguished by analyzing the control data at that time.

[0126] FIG. 9 is a diagram illustrating the operation flowchart of an electronic device according to one embodiment of the present invention.

[0127] A processor (150) according to one embodiment of the present invention can collect control data according to the start of the motor (20) (S10).

[0128] The control data may include current data and frequency data of the motor (20). In this case, if the motor (20) is connected to the inverter (10) in three phases, the current data may include current data for each phase (e.g., U, V, W phases).

[0129] The processor (150) can collect control data from the inverter (10), but the collection path is not limited to any one, such as through a server.

[0130] A processor (150) according to one embodiment of the present invention can divide the starting section of the motor (20) into an acceleration section, a constant speed section, and a deceleration section based on the target frequency of the motor (20) (S20).

[0131] The starting section of the motor (20) refers to a series of sections in which the motor (20) operates (stop → operation → stop). Specifically, the starting section may include an acceleration section in which the motor (20) accelerates from a stopped state until it reaches a target frequency (corresponding to a target speed), a constant speed section in which it operates at a constant speed for a required time after reaching the target frequency, and a deceleration section in which it operates at a deceleration speed to return to a stopped state after the required time.

[0132] When the motor (20) operates in both directions, it will operate in the order of acceleration, constant speed, and deceleration sections in the reverse direction after the deceleration section. An example of dividing the starting sections of the motor (20) is shown in FIG. 4.

[0133] A processor (150) according to one embodiment of the present invention can calculate the rate of change of frequency over time in acceleration and deceleration sections based on control data (S30).

[0134] The processor (150) can use the rate of change of frequency over time in the acceleration and deceleration sections by using the time and target frequency for each section. The equation for the rate of change of frequency is the same as Equation 1 described above.

[0135] A processor (150) according to one embodiment of the present invention can identify whether there is an abnormality in the motor (20) by inputting control data to a plurality of models trained to determine whether there is an abnormality in the motor (20) according to the starting section and the rate of change (S40).

[0136] Multiple models may include a first multiple model trained to determine whether there is a motor malfunction for each of two or more rate of change intervals in the acceleration section, a second model trained to determine whether there is a motor malfunction in the constant speed section, and a third multiple model trained to determine whether there is a motor malfunction for each of two or more rate of change intervals in the deceleration section. That is, since the performance of the motor (20) may vary depending on the magnitude of the frequency rate of change over time even within the acceleration and deceleration sections, the models can be trained individually to reflect this. The process of training multiple models is explained with reference to FIG. 10.

[0137] The processor (150) divides the control data of the motor (20) acquired in real time according to the starting section and can select which model to use in each section based on the frequency change rate of each section. This process is explained in detail with reference to FIG. 11.

[0138] The processor (150) can identify whether there is an abnormality in the motor (20) in the acceleration section and deceleration section by inputting control data (this is limited to control data of the corresponding section) to the selected model.

[0139] Specifically, the processor (150) can identify whether there is an abnormality in the motor during the acceleration section by inputting control data to the model corresponding to the frequency change rate of the acceleration section among the first plurality of models. Similarly, the processor (150) can identify whether there is an abnormality in the motor during the deceleration section by inputting control data to the model corresponding to the frequency change rate of the deceleration section among the third plurality of models.

[0140] Meanwhile, it is obvious that the processor (150) can identify whether there is an abnormality in the motor (20) by using a second model that is trained to determine whether there is an abnormality in the constant speed section even in the constant speed section.

[0141] According to one embodiment of the present invention, since motor abnormalities can be identified during the acceleration / deceleration phase occurring simultaneously with motor startup, rapid detection of motor abnormalities is possible, thereby minimizing motor damage.

[0142] According to one embodiment of the present invention, since the model is developed and applied by dividing it into sections, the model can be trained more rigorously and precisely. In particular, as the greater the rate of change of frequency in the acceleration / deceleration section, the greater the heat / stress applied to the motor, the section (time point) of motor burnout can be precisely identified.

[0143] According to one embodiment of the present invention, since the site where the motor is operated necessarily has acceleration and deceleration sections, it is expected that it can be applied to various sites such as robots and automobiles.

[0144] FIG. 4 is a diagram illustrating the starting section according to the frequency of a motor according to one embodiment of the present invention.

[0145] FIG. 4 illustrates an example of dividing the starting section based on the frequency of the motor (20), as previously described in relation to S21 of FIG. 3. In this case, FIG. 4 is described assuming that the motor (20) is a bidirectional motor.

[0146] The target frequency of the motor (20) is 50Hz, and based on the target frequency, the starting section of the motor (20) in the forward direction can be divided into an acceleration section (①), a constant speed section (②), and a deceleration section (③).

[0147] Afterwards, even in the reverse direction, the starting section of the motor (20) based on the target frequency can be divided into an acceleration section (④), a constant speed section (⑤), and a deceleration section (⑥).

[0148] FIG. 10 is a diagram illustrating the learning process of a model according to one embodiment of the present invention.

[0149] FIG. 10 illustrates the process of an electronic device (100) learning a plurality of models, as previously described in relation to S41 of FIG. 9. Likewise, the motor (20) is described as being a bidirectional motor.

[0150] Control data (in particular, current data, hereinafter referred to as learning data) can be collected, and the collected learning data can be divided according to the starting section. At this time, the learning data of the acceleration section (①, ④) and the deceleration section (③, ⑥) can be further subdivided according to two or more rate of change sections.

[0151] Two or more rate of change intervals can be set based on the target frequency of the motor (20) and the shortest starting time and maximum starting time according to the DC link voltage.

[0152] The target frequency of the motor (20) may be the maximum frequency used in the actual field or the setting parameter (Max Freq) of the inverter (10).

[0153] The DC link voltage rating can be determined according to the capacity of the inverter (10). For example, a 200V class inverter must have a DC link voltage of 410Vdc or less, and a 400V class inverter must have a DC link voltage of 820Vdc or less. The inverter (10) can be operated within a range where the DC link voltage is not exceeded. Therefore, the shortest starting time according to the DC link voltage may mean the time when the frequency change rate is highest within a range where the DC link voltage is not exceeded. The maximum starting time may be the maximum acceleration / deceleration time suitable for the actual field or the maximum time that can be set by the inverter (10) itself.

[0154] Two or more rate of change intervals can be set based on the first frequency rate of change according to the shortest start time and the second frequency rate of change according to the maximum start time.

[0155] For example, let's examine the process of subdividing the rate of change intervals of the FWD acceleration section.

[0156] When the first frequency change rate according to the shortest start time is a4 and the second frequency change rate according to the maximum start time is a1 (where a4 > a1), the change rate interval can be divided into two or more based on the value between a1 and a4.

[0157] As illustrated in FIG. 10, when setting the rate of change interval to three, for example, the interval between a1 and a4 can be divided into a section where the frequency rate of change is a1~a2, a section where it is a2~a3, and a section where it is a3~a4, based on the 1 / 3 point (e.g., a2) and the 2 / 3 point (e.g., a3).

[0158] In addition, when setting two rate of change intervals, for example, based on a point between a1 and a4 (e.g., a2), the frequency rate of change can be divided into an interval from a1 to a2 and an interval from a2 to a4. In this case, the number of rate of change intervals or the method of setting the criteria for dividing the rate of change intervals is not limited to any one.

[0159] The processor (150) can divide the collected training data according to the operating interval and further subdivide it according to two or more rate of change intervals. The processor (150) can train a model for each rate of change interval within each acceleration interval using the subdivided training data. At this time, there may be various methods for training the model, and the processor (150) can train a model to determine whether there is an abnormality in the motor (20) using the training data.

[0160] If an anomaly occurs in any section, the collected control data may exhibit the following examples. For instance, the pattern of reaching the target frequency based on the commanded acceleration / deceleration time may change (the time to reach the actual frequency increases, or the arrival of the target frequency is delayed). Alternatively, the current waveform may show a pattern different from the existing one, a phase difference may occur in the three-phase current, the sum of the three-phase currents may differ from the existing one, the maximum / minimum / median / average values ​​of a cycle may change, significant noise may occur in the current data, or the DC link voltage value may differ from the existing one (decreasing or increasing compared to the existing one).

[0161] The processor (150) can train the model to determine whether the motor (20) is abnormal through abnormal aspects of the control data. For example, the processor (150) may perform at least some of the data analysis, processing, and result information generation for performing the above operations using at least one of machine learning, neural network, or deep learning algorithms as a rule-based or artificial intelligence algorithm. Examples of neural networks may include models such as CNN (Convolutional Neural Network), DNN (Deep Neural Network), and RNN (Recurrent Neural Network).

[0162] Referring to FIG. 10, the processor (150) can generate a ①-1 model using the training data of the first rate of change interval in the ① FWD acceleration interval. The processor (150) can generate a ①-2 model using the training data of the second rate of change interval in the ① FWD acceleration interval. The processor (150) can generate a ①-3 model using the training data of the third rate of change interval in the ① FWD acceleration interval.

[0163] The same applies to other driving sections below, and Figure 10 is as follows when all acceleration / deceleration sections are created based on three rate of change sections.

[0164] ① The first plurality of models corresponding to the FWD acceleration section are ①-1 Model, ①-2 Model, and ①-2 Model; ② The second model corresponding to the FWD constant speed section is ② Model; and ③ The second plurality of models corresponding to the FWD deceleration section are ③-1 Model, ③-2 Model, and ③-3 Model.

[0165] Likewise, the first plurality of models corresponding to the ④ REV acceleration section are ④-1 Model, ④-2 Model, and ④-2 Model, the second model corresponding to the ⑤ REV constant speed section is ⑤ Model, and the second plurality of models corresponding to the ⑥ REV deceleration section are ⑥-1 Model, ⑥-2 Model, and ⑥-3 Model.

[0166] In this way, by distinguishing the rate of change intervals in the acceleration and deceleration sections, the impact of motor damage can be approached sensitively and relatively.

[0167] If rapid acceleration and deceleration are set during motor startup, significant heat, overcurrent, and vibration may be generated in the motor, which can negatively affect its lifespan. Therefore, in models with a high rate of change in acceleration and deceleration, the model's sensitivity can be set high to detect an anomaly even when the values ​​in the control data differ only slightly from the existing values. Conversely, in models with a low rate of change in acceleration and deceleration, the model's sensitivity can be set to be less sensitive.

[0168] For example, if an abnormality occurs when the phase difference of the three-phase current exceeds a first threshold value, the model with a high rate of change in acceleration / deceleration can be adjusted to have a threshold value lower than the first threshold value, and the model with a low rate of change in acceleration / deceleration can be adjusted to have a certain margin around the first threshold value.

[0169] FIG. 11 is a drawing illustrating the operation of an electronic device according to one embodiment of the present invention.

[0170] Figure 11 illustrates the process of utilizing a model trained according to the operation interval and the rate of change interval. In this case, the division of the interval and the model trained accordingly are adopted exactly as described above with reference to Figure 10.

[0171] The processor (150) can collect control data according to the start of the motor (20) and calculate the rate of change of frequency in each acceleration section and deceleration section.

[0172] The processor (150) can identify the corresponding rate of change interval for the frequency rate of change calculated in each acceleration and deceleration interval. For example, it is assumed that the frequency rate of change (r1) calculated in the FWD acceleration interval belongs to the first rate of change interval among the three rate of change intervals.

[0173] The processor (150) can select a ①-1 Model corresponding to the first rate of change section of the ① FWD acceleration section and input control data of the ① FWD acceleration section into the ①-1 Model to identify whether there is an abnormality in the motor (20) in the ① FWD acceleration section.

[0174] Likewise, the frequency change rate for sections ② to ⑥ is calculated to select a corresponding model, and based on the control data in the corresponding section, it will be possible to identify whether there is an abnormality in the motor (20).

[0175] As a result of checking for abnormalities in the motor (20) in each section, it was identified that there were abnormalities in the motor (20) in the ① FWD acceleration section and the ⑥ REV deceleration section.

[0176] The processor can determine the performance of the motor (20) based on whether there is a motor abnormality in the acceleration section, constant speed section, and deceleration section.

[0177] The criteria for judgment may vary, and for example, it may be identified based on the importance of the section where the abnormality occurred, the proportion of the section where the abnormality occurred, etc. In addition, there may be various ways to display performance. For example, as shown in FIG. 11, a figure may be displayed indicating that 85% of the total performance is being achieved, or the repair status of the motor or inverter, the risk level, etc., may be indicated based on that figure.

[0178] According to one embodiment of the present invention, when determining whether there is an abnormality in the motor for each section, the time at which the abnormality occurs can be clearly distinguished, and the cause can be clearly distinguished by analyzing the control data at that time.

Claims

1. Memory in which at least one instruction is stored; and A processor that executes the above instructions; including The above processor is, Collect control data for each of the predefined acceleration, constant speed, and deceleration sections of the motor, and Information corresponding to the above control data is input into a plurality of models to obtain output information, thereby identifying whether there is an abnormality in the motor, An electronic device for diagnosing the performance of a motor, wherein the plurality of models are each learned for the acceleration section, the constant speed section, and the deceleration section.

2. In Paragraph 1, The above processor is, Determine the rate of change of frequency corresponding to the acceleration section and the deceleration section, determine the test time corresponding to the constant speed section, and generate a test pattern. An electronic device for diagnosing the performance of a motor by testing the motor according to the above test pattern and collecting the above control data.

3. In Paragraph 2, The above processor is, An electronic device for diagnosing the performance of a motor that determines the maximum frequency of the test pattern based on the maximum operating frequency of the motor.

4. In Paragraph 2, The above processor is, Calculate the range of the rate of change of frequency over time in the acceleration and deceleration sections based on the maximum operating frequency of the above motor and the shortest and maximum starting times according to the DC link voltage, and An electronic device for diagnosing the performance of a motor that includes the frequency change rate in the above test pattern within the range of the above frequency change rate.

5. In Paragraph 1, The above multiple models are, An electronic device for diagnosing the performance of a motor, comprising a first plurality of models trained to determine whether there is a motor abnormality for each of two or more rate of change intervals in the acceleration interval, a second model trained to determine whether there is a motor abnormality in the constant speed interval, and a third plurality of models trained to determine whether there is a motor abnormality for each of two or more rate of change intervals in the deceleration interval.

6. In Paragraph 5, The above 2 or more rate of change intervals are, An electronic device for diagnosing the performance of a motor based on a first frequency change rate according to the shortest starting time according to the DC link voltage and a second frequency change rate according to the maximum starting time according to the DC link voltage.

7. In Paragraph 5, The above processor is, An electronic device for diagnosing motor performance by inputting control data into a model corresponding to the frequency change rate of the acceleration section among the first plurality of models above to identify whether there is an abnormality in the motor in the acceleration section.

8. In Paragraph 5, The above processor is, An electronic device for diagnosing motor performance by inputting control data into a model corresponding to the frequency change rate of the deceleration section among the third plurality of models above to identify whether there is an abnormality in the motor in the deceleration section.

9. In Paragraph 1, The above processor is, An electronic device for diagnosing the performance of a motor, which determines the performance of the motor based on whether there is a motor abnormality in the acceleration section, constant speed section, and deceleration section.

10. In Paragraph 1, The above control data is an electronic device for diagnosing the performance of a motor, including current data and frequency data of the motor.

11. A method for diagnosing the performance of a motor performed by an electronic device, A step of collecting control data for each of the predefined acceleration, constant speed, and deceleration sections of the motor; and The method includes the step of inputting information corresponding to the above control data into a plurality of models to obtain output information and identifying whether there is an abnormality in the models. A method for diagnosing the performance of a motor, wherein the plurality of models are each learned for the acceleration section, the constant speed section, and the deceleration section.

12. In Paragraph 11, A step of determining a frequency change rate corresponding to the acceleration section and the deceleration section, and determining a test time corresponding to the constant speed section to generate a test pattern; and A method for diagnosing the performance of a motor, further comprising the step of testing the motor according to the above test pattern and collecting the above control data.

13. In Paragraph 12, A method for diagnosing the performance of a motor, further comprising the step of determining the maximum frequency of the test pattern based on the maximum operating frequency of the motor.

14. In Paragraph 12, The method further includes a step of calculating the range of the rate of change of frequency over time in the acceleration and deceleration sections based on the maximum operating frequency of the motor, and the shortest starting time and maximum starting time according to the DC link voltage. A method for diagnosing the performance of a motor that includes the frequency change rate in the above test pattern within the range of the above frequency change rate.

15. In Paragraph 11, The above multiple models are, A method for diagnosing the performance of a motor, comprising: a first plurality of models trained to determine whether there is a motor abnormality for each of two or more rate of change intervals in the acceleration interval; a second model trained to determine whether there is a motor abnormality in the constant speed interval; and a third plurality of models trained to determine whether there is a motor abnormality for each of two or more rate of change intervals in the deceleration interval.

16. In Paragraph 15, The above 2 or more rate of change intervals are, A method for diagnosing the performance of a motor based on a first frequency change rate according to the shortest starting time according to the DC link voltage and a second frequency change rate according to the maximum starting time according to the DC link voltage.

17. In Paragraph 15, A method for diagnosing the performance of a motor, further comprising the step of inputting the control data into a model corresponding to the frequency change rate of the acceleration section among the first plurality of models to identify whether there is an abnormality in the motor in the acceleration section.

18. In Paragraph 15, A method for diagnosing motor performance, further comprising the step of inputting the control data into a model corresponding to the frequency change rate of the deceleration section among the third plurality of models to identify whether there is an abnormality in the motor during the deceleration section.

19. In Paragraph 11, A method for diagnosing the performance of a motor, further comprising the step of determining the performance of the motor based on whether there is a motor abnormality in the acceleration section, constant speed section, and deceleration section.

20. In Paragraph 11, The above control data is a method for diagnosing the performance of a motor, including current data and frequency data of the motor.

Images