System and method for monitoring motor winding degradation

Abstract

The invention relates to a system and method for monitoring motor winding degradation. Systems (100) and methods are provided for monitoring motor winding degradation of an electric machine having an electric motor (106) driven by a variable speed drive (VSD) (102). A sensor of the VSD (102) detects a measurement of the phase current and / or phase voltage while driving the electric motor (106). Each measurement value (114) is processed using a filtering module (116) to calculate a symmetric component (118). The resulting symmetric component (118) is used to identify a motor winding state (120) of one or more motor windings.

Description

Technical Field

[0001] This disclosure relates to systems, methods, and apparatus for evaluating the insulation condition of electric machine windings. Background Technology

[0002] Three-phase squirrel-cage induction motors are common electric motors used in commercial applications. Three-phase synchronous motors are less common, but still widely used in commercial applications. These motors are used in a variety of industries and implemented in a wide range of applications, including compressors, vacuum cleaners, fans, and pump systems. Any rotating electric motor (e.g., the motor in an air compressor setup) consists of two main parts: a stationary stator and a rotating rotor. The stator, connected to a three-phase mains power supply, generates a rotating magnetic field that passes through the air gap. In an induction motor, this rotating magnetic field induces a current in the squirrel cage on the rotor, thus generating a magnetic field. In a synchronous motor, a magnetic field is generated on the rotor by permanent magnets or by three-phase windings connected to a three-phase voltage source via slip rings. The interaction between the magnetic fields of the stator and rotor generates torque, causing the rotor shaft to rotate.

[0003] While such motors are advantageous due to their simple structure and low cost, failures caused by thermal, mechanical, and environmental stresses are unavoidable. Many of these failures occur due to insulation degradation in the motor windings. For example, thermal stress caused by increased winding temperature shortens lifespan and damages insulation. Exposure to moisture and particulate matter also compromises insulation integrity. Mechanical vibrations of motor components can even cause high-quality insulation to deteriorate over time. Because electric motors play a vital role in a wide range of industrial and commercial applications, unexpected failures in these devices can lead to downtime, resulting in economic losses, or in some cases, even personal injury.

[0004] Generally, motor fault diagnosis focuses on detecting faults in motor components (i.e., stator, rotor, and bearings). Techniques for detecting these faults include resistance testing, high-potential testing, surge testing, and partial discharge testing. However, performing such tests requires expensive specialized equipment, which may not be cost-effective in assessing the condition of the motor windings during normal operation. Alternative techniques require storing three-phase current values ​​in memory for extended periods and applying time-intensive computational techniques such as Fourier analysis and current characteristic analysis to evaluate quasi-sinusoidal and / or pulse-width modulation time waveforms.

[0005] Therefore, an improved motor winding fault diagnosis system is needed to obtain real-time monitoring of the operating electric motor, including motor winding insulation, in order to better adapt to motor control and mitigate dangerous situations. Summary of the Invention

[0006] The inventors of this disclosure have developed a novel technique for using a variable speed drive (VSD) to drive a motor and monitor the condition of the motor windings. The disclosed technique is generally applicable to the driving of multiphase electric motors.

[0007] This disclosure relates to a system and method for monitoring winding degradation of a multiphase AC electric motor driven by a variable speed drive. By advantageously using sensors from the same variable speed drive responsible for controlling the speed and torque of the electric motor, external or remote sensors for evaluating symmetry component values ​​are eliminated, where the negative symmetry component, known to indicate potential motor winding problems. Therefore, the system can be simplified, using fewer components and thus reducing costs.

[0008] In this embodiment, the system updates the estimated symmetrical components of the three-phase current in real time based on combining previous estimates of the symmetrical components with the most recently measured three-phase current using a mathematical model or filtering module. In other words, the system uses a mathematical model or filtering module to update the symmetrical components on an instantaneous basis based on immediately preceding measurements. The system's symmetrical components (including positive and negative sequences) are obtained simultaneously using the same algorithm. Furthermore, the symmetrical components of the three-phase current and three-phase voltage can be calculated to obtain at least four symmetrical components (including positive and negative sequences), or six symmetrical components (including zero, positive, and negative sequences). Advantageously, particularly in embodiments where the controller is part of the drive, the system can be used to simultaneously monitor problems in multiple components. In a drive, problems or abnormal readings may also be observed in the applied voltage due to some control over the current; in applications where the motor is directly applied to a strong and stable voltage grid, the voltage is constant, and problems may not be observed in the symmetrical components of the voltage.

[0009] For each newly received phase current value measurement, the symmetry component is updated. This approach avoids the need to store and process a series of three-phase measurements acquired over a long period at a relatively high sampling rate (e.g., a 50Hz waveform sampled at 2kHz for 10 cycles means storing 400 values ​​per phase and requiring more advanced and time-intensive signal processing techniques on the entire 400x3 data matrix). Furthermore, this approach allows the symmetry component to be updated at the same rate as the measurement; therefore, faster control actions can be taken. In other words, the system can bypass storing multiple phase current values ​​in memory to obtain current measurement samples and update the symmetry component in real time. The calculated symmetry component includes both negative and positive sequence currents.

[0010] The system uses a filtering module to obtain the symmetric component. Exemplary filtering modules include an extended Kalman filter (EKF) and / or a second-order generalized integrator (SOGI) filter. In an exemplary embodiment, the tracking characteristics obtained from the filtering module are enhanced by selecting appropriate parameters to smooth subsequently obtained values. In embodiments employing an extended Kalman filter, the system can provide an estimator for the applied frequency.

[0011] Based on a symmetrical component (e.g., a negative sequence current), the system can provide the status of one or more motor windings. If the system identifies a significant deviation (e.g., a deviation exceeding a predetermined threshold) between the symmetrical component and a balanced three-phase current system (i.e., three sine waves with the same frequency and amplitude but each offset by 120°), the motor winding status can indicate an insulation problem. For example, in response to an indication of an insulation problem where the deviation exceeds a predetermined threshold, a signal can be returned to the transmission drive, allowing the motor machine to be safely paused or shut down to permit repair and maintenance. Furthermore, the system can classify the deviation level into different fault types and / or winding degradation levels.

[0012] This invention is provided to introduce some concepts in a simplified form, which will be further described in the following "Detailed Description". These and other features, aspects, and advantages of this disclosure will be better understood from the following description, appended claims, and drawings. Attached Figure Description

[0013] To illustrate how the advantages and features of the systems and methods described herein can be obtained, the embodiments briefly described above will be described in more detail with reference to the specific embodiments shown in the accompanying drawings. It should be understood that these drawings depict only typical embodiments of the systems and methods described herein and should not be considered as limiting their scope. Certain systems and methods will be described and explained with additional specificity and detail using the accompanying drawings, in which:

[0014] Figure 1 The diagram illustrates a system with a variable speed drive to both drive an electric motor and monitor motor winding degradation.

[0015] Figure 2 The diagram illustrates the controller architecture of the variable speed drive.

[0016] Figure 3 The illustration shows exemplary graphical representations of positive and negative sequences generated in real time over a short time range by the controller and using a filtering module.

[0017] Figure 4 The illustration shows exemplary graphical representations of positive and negative sequences generated over a long period of time using a filtering module at regularly spaced time intervals.

[0018] Figure 5 The diagram illustrates a scheme for using an extended Kalman filter (EKF) to process measurements obtained from a driver in order to calculate one or more symmetric components and frequencies.

[0019] Figure 6 The diagram illustrates a scheme for processing measurements obtained from a driver using a second-order generalized integrator (SOGI) filter to compute one or more symmetric components.

[0020] Figure 7 The illustration shows a method for indicating the state of motor windings using batch processing or real-time streaming methods.

[0021] Figure 8 The diagram illustrates a method for classifying the state of motor windings and generating actionable notifications.

[0022] definition

[0023] To facilitate understanding of the disclosed embodiments of the methods and system components, it is necessary to describe several terms.

[0024] The term "compressor" refers to a machine that draws low-pressure gas from an auxiliary storage device as a raw input and then outputs high-pressure gas for storage or feeding to other processes. The terms "compressor" and "compressor element" are not intended to limit the scope and may refer to displacement compressors and / or dynamic compressors (turbo compressors) and / or individual components of a compressor.

[0025] The term "computer storage medium" refers to the physical storage medium that stores computer-executable instructions and / or data structures. Storage media (such as digital data carriers) include computer hardware such as random access memory (RAM), read-only memory (ROM), electrically erasable programmable ROM (EEPROM), solid-state drives (SSDs), flash memory, phase-change memory (PCM), optical disc storage devices, and disk storage devices.

[0026] The term "controller" generally refers to a collection of sensors and electrical components, i.e., a computerized command terminal for regulating various compressor components. A compressor controller includes at least one main processing unit with a graphical interface and is adapted to monitor instrumentation for various compressor parts (e.g., motors, rotors, filters, bearings, valves, pressure sensors, temperature sensors). A single controller can be arranged to monitor instrumentation for multiple compressors. An exemplary compressor controller operates to control safe start-up and shutdown processes, provide real-time information from compressor instrumentation, adjust the power output of one or more motors, stabilize compressor operation, control processing variables, alert and warn the operator of problems, and / or initiate automatic shutdown in the event of an unsafe condition.

[0027] The term "network" refers to one or more data links that enable the wired or wireless transmission of electronic data between computer systems and / or modules and / or other electronic devices.

[0028] The term "processor" or "processing unit" refers to one or more devices, circuits, and / or processing cores configured to process data (such as computer program instructions), and includes personal computers, computing units, desktop computers, laptop computers, message processors, handheld devices, multiprocessor systems, microprocessor-based or programmable consumer electronics, network PCs, minicomputers, mainframe computers, mobile phones, PDAs, tablet computers, pagers, routers, switches, etc. Unless otherwise stated, references to a first processor may also apply to a second processor, and vice versa.

[0029] The term "service" refers to an automated program that performs different actions based on input. As used herein, the terms "executable module," "executable component," "component," "module," "service," or "engine" may refer to a hardware processing unit, or to a software object, routine, or method that can be executed on a two-component system and / or a software update system.

[0030] The term "software" generally refers to computer-executable instructions, code, data, applications, programs, program modules, etc., stored on or in any form or type of computer-readable medium configured to store computer-executable instructions, etc., in a manner accessible to a computing device.

[0031] The term “symmetric component” usually refers to a set of three components: (1) zero sequence (or same pole) component, (2) positive sequence (or direct) component, and (3) negative sequence (or indirect) component.

[0032] The term "variable speed drive" (VSD) generally refers to a drive that provides stepless variable control over the operating speed and / or torque of an electric motor. For clarity and without loss of generality, the term "variable speed drive" as used herein refers to the following types of drives, including but not limited to: adjustable speed drives, adjustable frequency drives, adjustable frequency inverters, variable frequency inverters, variable frequency drives, and converter units.

[0033] As used herein, references to any type of machine learning or artificial intelligence can include any type of machine learning algorithm or device, convolutional neural networks, multilayer neural networks, recurrent neural networks, recurrent neural networks, deep neural networks, decision tree models (e.g., decision trees, random forests, and gradient boosting trees), linear regression models, logistic regression models, support vector machines (SVMs), artificial intelligence devices, or any other type of intelligent computing system. Any amount of training data can be used (and may be improved in the future) to train the machine learning algorithm to dynamically perform the disclosed operations.

[0034] When elements are introduced in the appended claims, the articles “a,” “an,” “the,” and “the” are intended to indicate the presence of one or more of these elements. The terms “comprising,” “including,” and “having” are intended to be inclusive and mean that additional elements may be present in addition to those listed. Detailed Implementation

[0035] Figure 1 An exemplary system 100 for monitoring motor winding degradation is illustrated. The system includes a variable speed drive (VSD) 102 electrically connected to a power grid 104 and an electric motor 106. The power grid 104 supplies power to the VSD, which drives at least one electric motor 106. In a preferred embodiment, the power grid 104 is a three-phase power grid. The electric motor 106 may drive a corresponding compressor, pump, or fan.

[0036] In an embodiment, VSD 102 includes at least one rectifier, a DC bus, and an inverter unit. The inverter of VSD 102 can be a current source inverter (CSI) or a voltage source inverter (VSI). The disclosed filtering techniques can be applied to the reference current (CSI) or voltage (VSI) and the resulting voltage or current. If a winding fault occurs, an increment or deviation can be seen in at least the voltage, current, or both measurements. In other words, the symmetrical components of the three-phase current and / or three-phase voltage can be calculated. When calculating the symmetrical component of the current, the system uses the motor current measured by sensors in VSD 102. When calculating the symmetrical component of the motor voltage, the system uses the instantaneous reference voltage calculated by the driver.

[0037] In one embodiment, VSD 102 is arranged to match motor input requirements with demand output fluctuations. VSD 102 includes two or more sensors 112 to detect phase current and / or phase voltage. In another embodiment, VSD 102 includes VSI and is part of the compressor system.

[0038] VSD 102 is communicatively connected to controller 108. Controller 108 typically includes at least one processor and at least one hardware storage device (e.g., storage device 122) storing computer-executable instructions that the processor performs various tasks. In embodiments, controller 108 includes a start / stop switch to initiate / deactivate processes for operating a connected electric motor 106. These processes can be started and stopped locally or remotely. In embodiments where controller 108 is part of a compressor having an electric motor 106 for operating compressor components, controller 108 includes features such as pressure regulation, operating status monitoring and registration, data retrieval (e.g., for monitoring service intervals and storing available data in “shutdown” events), remote control and monitoring for centralized compressor management (e.g., digital monitoring), and relay monitoring. Controller 108 can be connected to display 111 or a graphical user interface to enable functions such as displaying operating status, delivery air pressure, and drive motor revolutions per minute (rpm); reading air delivery values; indicating compressor status with compressor load, temperature, and pressure values; and so on. The display 111 may be part of the VSD 102 (i.e., physically connected to the VSD 102), or it may be part of a remotely connected computing unit (i.e., a smartphone, tablet, or computer). The controller 108 preferably enables application control (e.g., using Bluetooth on a smartphone, tablet, or computer). (Connected) and monitored via VSD 102 for electric motor 106.

[0039] In embodiments where controller 108 is integrated into VSD 102, the interface between controller 108 and VSD 102 can be accessed via a connection to network 110. Therefore, no physical interaction with the electric motor 106 or VSD 102 is required to operate controller 108. In embodiments where controller 108 is part of VSD 102, multiple electric motors 106 (including those without an integrated controller) can be controlled and their status monitored via a network connection using controller 108 integrated with VSD 102.

[0040] Network 110 is communicatively coupled to controller 108 of VSD 102. Network 110 facilitates transmission and distribution from VSD 102 (see...). Figure 2The information is obtained from sensor 112 of VSD 102. In an alternative embodiment of system 100, controller 108 is connected to service 113 as an alternative or supplement to connection to network 110. Service 113 may include typical components of a physical server, such as a motherboard, processor, memory storage device, one or more hard drives, power supply, and network adapter or external network connectivity. Service 113 may store one or more software update packages to controller 108. In an embodiment, service 113 is physically connected to controller 108 to manually perform further processing of the information obtained from sensor 112 of VSD 102.

[0041] Figure 2 The illustration depicts the general operation of controller 108 for processing measured current and / or voltage values ​​114. Controller 108 is arranged to receive measured values ​​114 from two or more sensors 112 and / or instantaneous values ​​from two or more reference signals in VSD 102 (e.g., in a voltage source inverter (VSI), the measured current value is "sensed," while the reference voltage value is "applied"). Each measured or resulting (current and / or voltage) value 114 is processed by filtering module 116 to calculate one or more symmetrical components 118. Those skilled in the art will recognize that value 114 can generally refer to multiple values ​​measured simultaneously; that is, measured value 114 can be interpreted as part of a set of values ​​corresponding to the measured or resulting current and / or voltage. Filtering module 116 is typically a computational platform for updating the estimated symmetrical components 118 based on the latest measured value 114. Controller 108 updates the resulting symmetrical components 118 with each newly incoming measured current and / or voltage value 114. In other words, based on the immediately preceding measurement 114, system 100 uses a mathematical model or filtering module 116 to update the symmetry component 118 in real time. In a preferred embodiment, the filtering module 116 updates the estimated symmetry component 118 for each measured or obtained (current and / or voltage) value 114 in real time, i.e., calculating 100 to 1,000,000 samples (preferably at least 1,000, and more preferably at least 5,000 samples) of the symmetry component 118 of the electric machine per second. In an embodiment, 250 to 200,000 samples of the symmetry component 118 of the electric machine (including electric motor 106) are calculated per second. The filtering module 116 includes an algorithm for calculating the symmetry component 118.

[0042] In an embodiment, the filtering module 116 includes an extended Kalman filter (EKF) as a simplified real-time streaming method that recursively updates the symmetric component 118 with its input measurement 114 at each sampling occurrence. Compared to existing batch signal analysis methods (e.g., Fast Fourier Transform (FFT)), the EKF provides a simplified method for calculating the symmetric component 118 for monitoring insulation problems. The EKF allows the controller 108 to indicate the motor winding state 120 based on the calculated symmetric component 118. Using the EKF, the filtering module 116 provides at least 1,000, preferably at least 2,000, and more preferably at least 5,000 readings per second (i.e., in real time) of the symmetric component 118, which are used to assess the trend of the positive sequence 126, and more preferably, the trend of the negative sequence 128.

[0043] In one embodiment, the filtering module 116 includes a second-order generalized integrator (SOGI) filter as a real-time streaming method that recursively updates the symmetric component 118 using its input measurement 114. Compared to existing batch signal analysis methods (e.g., FFT), the SOGI filter provides a simplified approach and further delivers better steady-state results using a single input parameter. In another embodiment, the filtering module 116 includes additional input parameters for smoothing subsequent measurements 114 to produce smooth tracking features and reduce the impact of signal noise.

[0044] Figure 3 The illustration shows a graphical representation 124 of the positive sequence 126 and the negative sequence 128 of the symmetric component 118, calculated in real time. The graphical representation 124 can be generated in real time as part of a motor winding status 120 indicator. The motor winding status 120 can be generated on the display of the VSD 102 or transmitted to another service having a display and communicating with the network 110. (E.g., via the graphical representation 124) The real-time generation of the motor winding status 120 indicates the overall "health" status of the motor windings and can be used to remind the user of necessary maintenance or parts replacement.

[0045] Figure 3 The real-time sequence of the calculated symmetric component 118 is shown, while Figure 4 A calculated symmetric component 118 is depicted per second, per minute, or per day to illustrate the evolution of the symmetric component 118 when the winding becomes unbalanced due to a fault. Figure 4The illustration shows a graphical representation 130 of a series of calculated positive sequences 126 and negative sequences 128 spaced regularly over a long time period (i.e., seconds, minutes, or days), with a fault reading 132 observed at the negative sequence 128. Graphical representation 130 illustrates that the negative sequence 128 provides a fault reading 132 when the symmetry of the motor windings is damaged. This information is provided as part of the motor winding condition 120. Other information, such as compressor load, temperature, and pressure values, can also be evaluated to some extent using measurements 114 of the symmetry component 118. For example, a low positive sequence current indicates a low compressor load; however, a correct assessment of flow, pressure, and / or temperature levels based on measurements 114 of the symmetry component 118 may require empirical formulas.

[0046] Over time, the collection of measurements of the symmetry component 118 (particularly the negative sequence 128) can provide an accurate motor winding condition 120, allowing the user to observe and determine whether or what action should be taken. For a specific type and model of motor winding, a predetermined time period can be defined as a reference time period. The initial sample or reference time period can be determined, for example, when the electric motor 106 (i.e., with initial motor windings) is first put into service, and can be based on readings of the symmetry component 118 that lead to winding failure or degradation. During the operation of the electric motor 106, the symmetry component 118 collected within a given time period can be compared periodically or even continuously with the symmetry component 118 within the reference time period to predict or anticipate potential motor winding degradation. When the estimated symmetry component 118 begins to deviate over time, the motor winding is degrading, and an actionable notification can be issued to the user for tracking or investigation of the electric motor 106. Advantageously, providing real-time updated calculations of the symmetry component 118 allows the user to take immediate action to shut down and repair the electric motor 106 and its corresponding motor windings. In addition, damage caused by this can be avoided (e.g., when the motor windings burn out and catch fire or when performance loss occurs before the actual failure).

[0047] The reference symmetry component 118 and the estimated or real-time symmetry component 118 may also depend on parameters such as operating pressure or other operating and motor conditions. In an embodiment, the motor winding state 120 is modeled (e.g., using a machine learning algorithm in a cloud environment or network 110) as a relationship between the reference symmetry component 118 and the motor condition via a polynomial or other suitable type of fitted curve (e.g., neural network regression) to generate a ground truth. During the operational life of the electric motor 106, the generated model can be updated based on the estimated and updated negative sequence 128 of the three-phase current measured each time. The difference between the updated model and the ground truth increases over time, establishing a reliable metric for calculating the level of motor winding degradation. When using symmetry component analysis with both the positive sequence 126 and the negative sequence 128, deviations can be identified within the controller 108 and / or other processing devices (e.g., cloud environment, network, service), and thus it may be possible to obtain a link or correlation between parameter deviations and fault types.

[0048] As noted above, the negative sequence 128 can be reported to the user via the motor winding status 120 indicator through the controller 108. Furthermore, the data representing the motor winding status 120 can be sent to another device for further processing. For example, the motor winding status 120 can be a value between 0 and 1, where 0 corresponds to a faulty motor winding and 1 corresponds to a new and healthy motor winding. Alternatively, the health status can be reported as "healthy," "degraded," or "faulty."

[0049] Figure 5 The diagram illustrates how controller 108 can process measured value 114. Measured value 114 is obtained from a three-phase power system and fed into filter module 116. In an embodiment, filter module 116 includes an extended Kalman filter algorithm to process input value 114. The corresponding state or symmetric component 118 is then calculated and output to the workspace. This disclosure is not limited to the illustrated configuration and scheme of the extended Kalman filter, and those skilled in the art will recognize that alternative implementations of the filter can be utilized.

[0050] about Figure 5 The diagram illustrates an implementation of EKF, and for illustrative purposes, the following elements and formulas are provided. For the EKF implementation, the predicted (prior) state estimate at iteration k is defined as...

[0051]

[0052] The covariance of the predicted (prior) state estimate is defined as

[0053]

[0054] Innovation is defined as

[0055]

[0056] Innovation covariance is defined as

[0057]

[0058] (Near-optimal) Kalman gain is defined as

[0059]

[0060] The updated (posterior) state estimate at the k-th iteration is defined as

[0061]

[0062] The updated (posterior) state estimate covariance is defined as

[0063] P k|k =(IK k H k )P k|k-1

[0064] about Figure 5 The above can be interpreted as follows: x represents "state" or state estimate (5x1 vector); z represents "current" or measured / obtained current / voltage value (3x1 vector); P represents "stateCov" or state estimate covariance (5x5 matrix); R represents "measCov" or measurement covariance (3x3 matrix); and Q represents "modelCov" or model covariance (5x5 matrix).

[0065] Furthermore, the state estimate x is defined as

[0066]

[0067] Where θ represents the angle of the current / voltage phasor dynamics, and the other four quantities represent estimates of the positive (+) and negative (-) sequence components of the current / voltage in the alpha (α)-beta (β) reference frame. These four quantities are expressed by their three corresponding quantities (i) via the following exemplary formula. a i b i c Related:

[0068]

[0069] The state covariance matrix P is iteratively updated starting from the identity matrix at the time of the first iteration.

[0070] The measurement covariance R is defined as

[0071]

[0072] Where sigma_f(σ f ) and sigma_i(σ i The values ​​) represent the phasor rotation speed and the noise / error of the measured / obtained current / voltage, respectively. These are typically found in the application's dynamic specifications (sigma_f) and sensor specifications (sigma_i).

[0073] The model covariance Q is defined as

[0074]

[0075] Where lambda(λ) and sigma_f,Ts These are two independent smoothing factors that need to be predetermined by the engineer. This smoothing process is typical for Kalman filters and allows engineers to weigh the importance of the model for measurement.

[0076] Referring to the illustrated embodiment, the EKF scheme includes several other matrices and functions, as described below. Matrix F is called the state transition matrix and is implemented as follows:

[0077]

[0078] The function f is called the transition model, and this transition matrix is ​​used:

[0079]

[0080] Matrix H is called the measurement or observation matrix and is implemented as follows:

[0081]

[0082] The function h is called the observation model, and this observation matrix is ​​used:

[0083]

[0084] The inventors have also conceived of similar schemes and methods, and those skilled in the art will understand. Figure 5 The system architecture observed and the fundamental principles taught by the above formulas.

[0085] Figure 6The diagram illustrates an embodiment of a system for calculating the symmetrical components of a three-phase signal using a second-order generalized integrator (SOGI) filter. The parameter kSOGI is set to be modifiable and is typically set to a value close to 1. The SOGI filter system has two additional parameters: a fixed sampling time Ts and the estimated frequency of the signal fEstim. The estimated frequency of the signal fEstim is set to the frequency applied by the variable speed drive. These parameters can also be used in embodiments employing the EKF model. The SOGI scheme is robust and allows for slight deviations from reality in fEstim, thus eliminating the risk of divergence. The outputs of the two integrators are two three-phase signals offset from each other by 90°. The results allow for the computation of the sequence using the following matrix multiplication:

[0086]

[0087] EKF updates the symmetric components using only the last sample, while the SOGI method uses the last three samples (see, for example, [link to documentation]). Figure 6 China and Israel -1 The symmetric components are updated using a three-delay block (as indicated). The number of samples used is still significantly less than the number of samples required by conventional FFT methods in batch processing. Furthermore, this disclosure is not limited to the configuration and scheme of the SOGI filter depicted, and those skilled in the art will recognize that alternative implementations of the filter can be utilized.

[0088] Figure 7 The illustration shows an example of how system 100 can monitor the state of motor windings using a batch processing method or a real-time streaming method. While driving the electric motor, the inverter receives current measurements from a three-phase power supply to the electric motor 106. In this embodiment, a batch of measured current values ​​is stored in the controller's memory or storage device. This batch of measured current values ​​may include a set of data points acquired over a preset time period (such as 200 milliseconds, 2 seconds, or 20 time periods). The batch processing method records the current and / or voltage for a period of time (e.g., 20 time periods) before applying signal processing techniques to the entire data matrix, thus minimizing the impact of measurement noise on the overall estimated symmetric components. Given the motor's operating speed and number of pole pairs, all signal vectors are filtered at the applied electrical frequency using an FFT algorithm (producing a set of phasors for current and / or voltage), and then a symmetric component Fortescue transform is applied (producing positive, negative, and zero sequences of current and / or voltage). Based on the calculated negative sequence values, the controller analyzes whether the symmetric components exceed a predetermined threshold. If the threshold is exceeded, the controller indicates a motor winding fault and / or generates an operable notification to the user—otherwise, the process restarts. In one embodiment, the process automatically restarts after indicating a motor winding fault or generating an operable notification. In an alternative embodiment, the process also includes shutting down the electric motor.

[0089] Figure 7 The process also describes a step where the controller bypasses storing a batch of measured current values ​​to generate a complete data matrix before applying signal analysis. Alternatively, each measurement is input into a filter model using a streaming or real-time method. As noted above, the filter model may include an extended Kalman filter, an SOGI filter, or other similar filtering methods. The filtering module calculates the symmetric component of each measurement, and the controller analyzes whether any sequence of values ​​of the symmetric component exceeds a predetermined threshold. If the threshold is exceeded, the controller indicates a motor winding fault and / or generates an actionable notification to the user. Otherwise, the process restarts. In one embodiment, the process automatically restarts after indicating a motor winding fault or generating an actionable notification. In an alternative embodiment, the process also includes shutting down the electric motor.

[0090] Figure 8 The illustration depicts a method for classifying trends in symmetrical components as motor winding faults. While driving the electric motor, controller 108 receives current measurements from a three-phase power supply to the electric motor 106. Each current measurement is input to a filtering module in real time. The filtering module calculates a negative sequence for each measurement. In one embodiment, controller 108 sends the trend of the negative sequence values ​​(i.e., a set of sequence values ​​over a longer time period, such as 5 seconds, 5 minutes, etc.) to a service (or network) for further processing. The service assesses whether the trend deviates from predetermined criteria, including predefined values, limits, etc. If the trend does not deviate from the reference criteria, the processing restarts, and additional measurements are acquired and processed. If it is determined that the trend has deviated from the reference criteria, the service (or cloud environment in the network) uses a machine learning algorithm to classify the trend as being associated with a certain type of motor winding fault. Motor winding fault types may indicate contamination, wear, vibration, or partial discharge (i.e., the effects of voltage surges). Motor winding fault types can also be broadly classified as "degraded" or "faulty." Based on the classification trends, the system can generate actionable notifications to remind users that identified motors need to be maintained, replaced, or turned off.

[0091] It will be understood that while the terms “first,” “second,” etc., may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used only to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the invention. As used herein, the term “and / or” includes any combination of one or more items listed in the associated list. It should further be understood that relational terms such as “first” and “second” are used only to distinguish one entity from another and do not necessarily require or imply any such relationship or order between these entities.

[0092] It should be understood that while many features and advantages of various embodiments of the present disclosure, as well as details of the structure and function of various embodiments thereof, have been outlined in the foregoing description, this detailed description is merely illustrative, and changes in detail may be made within the scope of the principles of the present disclosure, particularly in terms of component structure and arrangement, and its scope covers the full range indicated by the broad general meaning of the terms expressed in the appended claims.

Claims

1. A method for monitoring motor winding degradation of an electric machine having an electric motor (106) driven by a variable speed drive (VSD) (102), the method comprising: measuring phase current values (114) directly from at least two sensors (112) of the VSD (102); updating, in real-time, symmetrical components (118) of the electric machine using a filter module (116) based on the immediately preceding measured phase current values (114); and generating a motor winding condition (120) based on the symmetrical components (118); wherein the VSD (102) is used both to drive the electric motor (106) and to monitor the motor winding condition (120).

2. The method of claim 1, wherein the symmetrical components are computed for each new incoming phase current value measurement.

3. The method of claim 1, wherein the step of storing a plurality of subsequent phase current values in a memory to obtain a batch of current measurement samples is bypassed to compute the symmetrical components of the electric machine in real-time.

4. The method of claim 1, wherein updating the symmetrical components (118) of the electric machine is further based on an instantaneous set voltage applied on the VSD.

5. The method of claim 1, wherein 250 to 200,000 samples of the symmetrical components of the electric machine are computed per second.

6. The method of claim 1, wherein the filter is an extended Kalman filter (EKF).

7. The method of claim 6, wherein the step of updating, in real-time, the symmetrical components of the electric machine further comprises using the EKF to provide a frequency estimator.

8. The method of claim 1, wherein the filter comprises a second order generalized integrator (SOGI) filter.

9. The method of claim 1, wherein the motor winding condition indicates an insulation problem based on a deviation of the symmetrical components from a balanced three-phase current system.

10. The method of claim 9, further comprising the step of providing a signal to the VSD in response to an indication of an insulation problem where the deviation exceeds a predetermined threshold.

11. A system (100) for monitoring motor winding degradation of an electric machine, the system (100) comprising: an electric motor (106); a power supply grid (104); and a variable speed drive (VSD) (102) for driving the electric motor (106); wherein the VSD (102) is connected to a controller (108) having one or more storage devices (122) storing instructions executable to cause the controller (108) to: measure phase current values (114) using at least two sensors (112) of the VSD (102); update, in real-time, symmetrical components (118) of the electric machine using a filter module (116) based on the immediately preceding measured phase current values (114); and generate a motor winding condition (120) based on the symmetrical components (118).

12. The system of claim 11, wherein the symmetrical components are computed for each new incoming phase current value measurement.

13. The system of claim 11, wherein the instructions are further executable to cause the controller (108) to update the symmetrical components (118) of the electric machine based on an instantaneous set voltage applied on the VSD.

14. The system of claim 11, wherein 250 to 200,000 samples of the symmetrical components of the electric machine are computed per second.

15. The system of claim 11, wherein the filter is an extended Kalman filter (EKF).

16. The system of claim 11, wherein the filter comprises a second order generalized integrator (SOGI) filter.

17. The system of claim 11, wherein the motor winding condition is indicative of an insulation problem based on a deviation of the symmetrical components from a balanced three-phase current system.

18. The system of claim 17, wherein the instructions are further executable by the controller to provide a signal to the VSD in response to an indication of the insulation problem where the deviation exceeds a predetermined threshold.

19. The system of claim 11, wherein the VSD is connected with the electric motor current and is configured to automatically adjust an operating speed in real time.

20. A method for monitoring motor winding degradation of an electric machine having a variable speed drive (VSD) (102) and an electric motor (106), the method comprising: measuring phase current values (114) directly from at least two sensors (112) of the VSD (102); updating, in real time, symmetrical components (118) of the electric machine using an extended Kalman filter (EKF) or a second order generalized integrator (SOGI) filtering module (116) based on the immediately preceding measured phase current values (114); wherein a new estimate of the symmetrical components (118) is generated upon each new incoming measurement obtained using the at least two sensors (112) of the VSD (102).

Images