Monitoring health of power system

Abstract

A health-monitoring method (figures 3A, 3B) for a power system having a multiphase motor (M), the method comprising: acquiring (202) a multiphase current signal (CS); applying (204) an empirical mode

Description

FIELD The present invention relates to a method for monitoring a health of a power system, a health monitoring system, a power system including a health monitoring system, and an aircraft including a power system. BACKGROUND Power systems generally utilize components, such as, power sources, inverters, electric motors, control systems, and the like. Power systems that are utilized in certain safety and mission critical applications, such as aerospace, may need to ensure reliability and availability of such components, especially electric motors. In some aircrafts, multiple identical high power density propulsion motors may be utilized in a multi-lane architecture and it may be critical to ensure reliability of such motors. Permanent magnet electrical motors are better choice for aerospace applications. However, faults occurring in such motors may lead to significant damage to the motor itself and the associated power system. Thus, monitoring and maintenance of power systems is critical for long-term and efficient working of power systems. A bearing is commonly utilized in an electric motor for supporting a rotating shaft of the electric motor and may be considered as one or the most critical components of the electric motor. Bearing faults constitute a significant portion of motor faults, and therefore, such faults are a key reason for downtime of the electric motor. Thus, an accurate detection of bearing faults at an early stage is critical to achieve condition-based maintenance to ensure a secure and reliable operation of the electric motor, thereby reducing loss of revenue. Present health monitoring systems employ sensors, such as vibration sensors, for detecting bearing faults. Addition of sensors and associated hardware may need to be carefully investigated at least for aerospace applications due to space and weight constraints. Some health monitoring systems may detect bearing faults by monitoring a stator current of the electric motor and determining characteristic fault frequencies. However, such techniques may still lack accuracy and reliability in fault detection. Therefore, the industry recognizes the importance of a versatile and integrated approach to health monitoring of power systems that may operate with existing system configurations and may accurately determine bearing faults for providing comprehensive diagnostics. SUMMARY In accordance with a first aspect of the present disclosure, a method for monitoring a health of a power system is disclosed. The power system comprises at least one multiphase motor with a bearing and at least one inverter driving the at least one multiphase motor. The method comprises: a) acquiring a multiphase current signal supplied by the at least one inverter to the at least one multiphase motor, the multiphase current signal has a supply frequency and a plurality of phase currents; b) applying an Empirical Mode Decomposition (EMD) technique to each phase current of the plurality of phase currents of the multiphase current signal to extract a plurality of Intrinsic Mode Functions (IMFs) corresponding to each phase current; c) obtaining a residual signal after extraction of the plurality of IMFs from the multiphase current signal; d) calculating a skewness of each IMF from the plurality of IMFs; e) determining one or more faulty IMFs from the plurality of IMFs that have the skewness outside of a predetermined skewness range; f) reconstructing a new signal with the one or more faulty IMFs and the residual signal; g) transforming the new signal from a time domain to a frequency domain to generate a transformed spectrum; h) calculating a plurality of characteristic frequencies of the bearing based at least on the supply frequency of the multiphase current signal, a geometry of the bearing, and a plurality of integer values, each characteristic frequency from the plurality of characteristic frequencies corresponds to a respective integer value from the plurality of integer values; i) selecting a plurality of frequency windows around the plurality of characteristic frequencies of the bearing, such that each frequency window from the plurality of frequency windows is around a respective characteristic frequency from the plurality of characteristic frequencies, each frequency window has a predetermined frequency range, and the transformed spectrum comprises a plurality of amplitudes in each frequency window; j) calculating an average value of the plurality of amplitudes of the transformed spectrum in each frequency window; k) calculating a maximum value from the plurality of amplitudes of the transformed spectrum in each frequency window; I) determining a characteristic frequency ratio for each frequency window as a ratio of the maximum value to the average value; and m) determining that the at least one multiphase motor has a bearing fault if the characteristic frequency ratio for each frequency window is above a predetermined threshold. The method according to the present disclosure may enable incipient detection of the bearing fault in the at least one multiphase motor with improved accuracy and reliability. The proposed method involves applying the EMD technique to extract the plurality of IMFs corresponding to each phase current. The method further comprises calculating the skewness of each IMF and determining the one or more faulty IMFs having the skewness outside of the predetermined skewness range. Thus, only the one or more faulty IMFs from the plurality of IMFs may be selected for further processing. This may enhance an accuracy of detecting the bearing fault while reducing a computational effort. In other words, the multiphase current signal is processed in such a way that only a fault information is utilized for further processing, thereby reducing computational time and effort. Therefore, the proposed method provides a cost-effective and reliable technique for determining the bearing fault as compared to conventional techniques. The selection of the plurality of frequency windows around the plurality of characteristic frequencies may eliminate a risk of any deviation in a theoretical and an actual characteristic frequency value. Further, a magnitude of a fault indicator is amplified as compared to traditional approaches. Accordingly, the proposed method may prevent failure of the at least one multiphase motor and the power system, thereby reducing downtime and overall maintenance costs. The method may detect the bearing fault by monitoring only the multiphase current signal (using existing sensor data) and without requiring use of additional sensors and associated hardware. In other words, the method may leverage available current measurements. The proposed method may be applied to any type of electric motor, e.g., permanent magnet machines, induction machines, wound field synchronous machines, 6-phase or 12-phase machines, etc. Further, the method may be applied to existing health monitoring systems for detecting the bearing fault in real time. The incipient bearing fault detection along with other equipment health monitoring (EHM) parameters may improve an accuracy of the health monitoring systems and aid in designing an ideal maintenance scheme for different use cases. Further developments of the invention can be found in the dependent claims and show particularly advantageous possibilities to realize above-described concept in light of the object of the disclosure and regarding further advantages. In a further development, the at least one multiphase motor comprises a first multiphase motor having a first bearing and a second multiphase motor having a second bearing. The at least one inverter comprises a first inverter driving the first multiphase motor and a second inverter driving the second multiphase motor. The method further comprises determining the characteristic frequency ratio for each frequency window for each of the first multiphase motor and the second multiphase motor as per steps a) to I) of the method of the first aspect. The method further includes determining a health index ratio as a ratio of the characteristic frequency ratio of the first multiphase motor to the characteristic frequency ratio of the second multiphase motor for each frequency window. The method further comprises determining that the first multiphase motor has a bearing fault if the health index ratio is greater than one by a predetermined tolerance value for each frequency window. The method further comprises determining that the second multiphase motor has a bearing fault if the health index ratio is less than one by the predetermined tolerance value for each frequency window. The method may facilitate monitoring of the health of the power system having the first multiphase motor and the second multiphase motor, e.g., in a multi-lane architecture. The health index may be formulated based on the characteristic frequency ratio of the first multiphase motor and the characteristic frequency ratio of the second multiphase motor for each frequency window. Thus, the bearing faults in the first multiphase motor and the second multiphase motor may be determined simultaneously, thereby allowing preventive maintenance and repair / replacement activities to be performed. The method may be extended to power systems having multiple motors arranged in multiple lanes, and the bearing fault may be determined in real time in any particular lane. This technique may not require any specific threshold definition as lane data may be utilized as threshold. In some cases, EHM parameters that are calculated for mechanical components for a set of operating points (i.e., speed, torque, etc.) may be compared with other lanes to assess degradation and potential faults in a lane. In a further development, the predetermined tolerance value is 0.1. This may allow accurate determination of the bearing fault in the first multiphase motor and the second multiphase motor. In a further development, the predetermined skewness range is from -0.35 to +0.35. This may allow the one or more faulty IMFs to be determined from the plurality of IMFs. In other words, any IMF having the skewness outside the predetermined range may be considered as the faulty IMF. In a further development, the predetermined frequency range for each frequency window is 5 hertz (Hz). This predetermined frequency range may be utilized as a reference for plurality of amplitudes in each frequency window, thereby allowing calculation of the average value of the plurality of amplitudes and the maximum value from the plurality of amplitudes. In a further development, the plurality of characteristic frequencies of the bearing is calculated as: Fb = Fs + KF0 where Fs is the supply frequency, K is the integer value, and Fo is an outer race frequency calculated as: Nb (BD \1 Fn = — Fr 1 — — cos a 0 2 L \PD / J where Nb is a total number of balls, Fr is a shaft rotational frequency, BD is a ball diameter of the balls, PD is a pitch diameter, and a is a contact angle. The above equations may allow determination of the plurality of characteristic frequencies of the bearing, thereby facilitating determination of the bearing fault. In a further development, the average value of the plurality of amplitudes of the transformed signal spectrum in each frequency window is calculated as: Ne-1 Aea=^y A^ i=0 where Ne is a number of spectral lines in the transformed spectrum, f is the ith spectral line in the transformed spectrum, i is an integer value between 0 and Ne-1, and Ae is the amplitude. The above equation may allow determination of the average value of the plurality of amplitudes in each frequency window, thereby facilitating determination of the bearing fault. In a further development, the maximum value from the plurality of amplitudes of the transformed spectrum in each frequency window is calculated as: Ne — 1 Aed = N~^ maX ^Ae ” A^’Ae ~Af + ............] e i = 0 where Ne is a number of spectral lines in the transformed spectrum, Ae is the amplitude, i is an integer value between 0 and Ne-1, fss is a frequency resolution, and Af is a frequency interval for searching the plurality of characteristic frequencies. The above equation may allow determination of the maximum value from the plurality of amplitudes in each frequency window, thereby facilitating determination of the bearing fault. In a further development, the new signal is transformed using Fast Fourier Transform (FFT). The FFT may allow the new signal to be transformed from the time domain to the frequency domain. In a further development, the method comprises providing a DC supply to the at least one inverter. The inverter may receive the DC supply for driving the at least one multiphase motor. In a further development, the multiphase current signal is a three-phase current signal. Thus, the method may be utilized for monitoring the health of the power system having a three-phase motor. According to a second aspect, a health monitoring system is disclosed. The health monitoring system comprises a processor communicably coupled to the at least one inverter. The processor is configured to perform the method of the first aspect. According to a third aspect, a power system is disclosed. The power system comprises at one multiphase motor, at least one inverter configured to drive the at least one multiphase motor, and the health monitoring system of the second aspect. The processor is communicably coupled to the at least one inverter. In a further development, the at least one multiphase motor is a permanent magnet synchronous motor. According to a fourth aspect, an aircraft comprising the power system of the third aspect is disclosed. It shall be understood that the method for monitoring the health of the power system according to the first aspect of the disclosure, the health monitoring system according to the second aspect of the disclosure, the power system according to the third aspect of the disclosure, and the aircraft according to the fourth aspect of the disclosure may comprise identical or similar developments, in particular as described in the dependent claims. Therefore, a development of one aspect of the disclosure is also applicable to another aspect of the disclosure. These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter. The embodiments of the invention are described in the following on the basis of the drawings. The latter is not necessarily intended to represent the embodiments to scale. Drawings are, where useful for explanation, shown in schematized and / or slightly distorted form. With regard to additions to the teachings immediately recognizable from the drawings, reference is made to the relevant state-of-the-art. It should be borne in mind that numerous modifications and changes can be made to the form and detail of an embodiment without deviating from the general idea of the disclosure. The features of the disclosure in the description, in the drawings and in the claims may be essential for a further development of the invention either individually or in any combination. In addition, all combinations of at least two of the features disclosed in the description, drawings and / or claims fall within the scope of the disclosure. The general idea of the disclosure is not limited to the exact form or detail of the preferred embodiment shown and described below, or to an object which would be limited in comparison to the object claimed in the claims. For specified design ranges, values within the specified limits are also disclosed as limit values and thus arbitrarily applicable and claimable. BRIEF DESCRIPTION OF THE DRAWINGS Further advantages, features and details of the invention result from the following description of the preferred embodiments as well as from the drawings, which show in: FIG. 1 is a schematic block diagram of a power system, according to an embodiment of the present disclosure; FIG. 2 is a schematic block diagram of a health monitoring system, according to an embodiment of the present disclosure; FIGS. 3A and 3B is a flow chart illustrating a method for monitoring a health of the power system, according to an embodiment of the present disclosure; FIG. 4 is an exemplary graph illustrating a variation of a characteristic frequency ratio with a plurality of integer values for a healthy bearing and a faulty bearing, according to an embodiment of the present disclosure; FIG. 5 is a schematic block diagram of the power system, according to another embodiment of the present disclosure; and FIG. 6 is a flow chart illustrating a method for monitoring a health of the power system of FIG. 5, according to an embodiment of the present disclosure. DETAILED DESCRIPTION Aspects and embodiments of the present disclosure will now be discussed with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. FIG. 1 shows a schematic block diagram of a power system 100, according to an embodiment of the present disclosure. The power system 100 comprises at least one multiphase motor M having a bearing 110 and at least one inverter 104 driving the at least one multiphase motor M. The term “at least one multiphase motor M” is interchangeably referred to herein as “the multiphase motor M”. Further, the term “at least one inverter 104” is interchangeably referred to herein as “the inverter 104”. In some embodiments, the bearing 110 may be used to support a rotating shaft (e.g., a rotor) of the multiphase motor M. The power system 100 may be utilized in a variety of applications, including transportation (aerospace, ships, land vehicles, etc.), petroleum, metallurgy, chemical industry, wind power, and the like. Specifically, an aircraft (not shown) comprises the power system 100. For example, the power system 100 may be utilized for propulsion of the aircraft. In some embodiments, the at least one multiphase motor M is a permanent magnet synchronous motor, i.e., including a permanent-magnet rotor. However, other types of motors (e.g., asynchronous or induction motors) may also be utilized in some other embodiments. The multiphase motor M is driven by an alternating current (AC). In some embodiments, the power system 100 further comprises a power source 102 for providing a DC supply 108 to the inverter 104. The inverter 104 may transform the DC supply 108 to an AC supply for driving the multiphase motor M. Subsequently, the inverter 104 supplies a multiphase current signal CS to the multiphase motor M. In some examples, the multiphase motor M may be three-phase motor M. Alternatively, in some other embodiments, a direct current (DC) motor may also be utilized with the power system 100. The power system 100 further comprises a multiphase current measurement device 106. The multiphase current measurement device 106 may be configured to measure the multiphase current signal CS supplied by the at least one inverter 104 to the at least one multiphase motor M. In some cases, the multiphase current measurement device 106 may be in the form of a current sensor disposed between the inverter 104 and the multiphase motor M. In some embodiments, the multiphase current measurement device 106 may be a part of the inverter 104. In some embodiments, the multiphase current signal CS is a three-phase current signal. For example, the inverter 104 provides a three-phase current to the multiphase motor M, i.e., the three-phase motor. In some embodiments, the multiphase current signal CS (i.e., the three-phase current signal) has a supply frequency F and a plurality of phase currents Ci, C2, C3. In some other embodiments, the multiphase current signal CS may include any number of the phase currents. The power system 100 further comprises a health monitoring system 112 for monitoring a health of the power system 100. The health monitoring system 112 comprises a processor 114. The processor 114 is communicably coupled to the at least one inverter 104. Specifically, the processor 114 is communicably coupled to the multiphase current measurement device 106. The processor 114 is configured to receive the multiphase current signal CS from the multiphase current measurement device 106. In some embodiments, the health monitoring system 112 may monitor the health of the power system 100 by monitoring the multiphase current signal CS supplied by the inverter 104 to the multiphase motor M. Specifically, the multiphase current signal CS may be analysed to determine faults in the multiphase motor M. More specifically, the multiphase current signal CS may be analysed to determine a bearing fault associated with the bearing 110 of the multiphase motor M. For example, in case of the bearing fault, there may be fluctuations in the multiphase current signal CS supplied by the inverter 104 to the multiphase motor M. In some examples, the processor 114 may include one or more devices, circuits, and / or processing cores configured to process data, such as computer program instructions. The processor 114 may include, for example, one or more of a general-purpose processor (e.g., ARM-based processor), a Digital Signal Processor (DSP), a Programmable Logic Device (PLD), an Application-Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), etc. FIG. 2 shows a schematic block diagram of the health monitoring system 112, according to an embodiment of the present disclosure. FIGS. 3A and 3B show a flow chart illustrating a method 200 for monitoring a heath of the power system 100 (shown in FIG. 1). The processor 114 of the health monitoring system 112 of FIGS. 1 and 2 is configured to perform the steps of the method 200. The method 200 will be set forth by way of FIGS. 2, 3A and 3B. Referring to FIGS. 1-3B, in some embodiments, the method 200 comprises providing the DC supply 108 to the at least one inverter 104. At step 202, the method 200 further comprises acquiring the multiphase current signal CS supplied by the at least one inverter 104 to the at least one multiphase motor M. The multiphase current signal CS has the supply frequency F and the plurality of phase currents Ci, C2, C3. In some embodiments, the multiphase current signal CS may be acquired by the processor 114 from the multiphase current measurement device 106 or the inverter 104. At step 204, the method 200 further comprises applying an empirical mode decomposition (EMD) technique 116 to each phase current Ci, C2, C3 of the plurality of phase currents Ci, C2, C3 to extract a plurality of intrinsic mode functions (IMFs) 118 corresponding to each phase current Ci, C2, C3. In other words, the plurality of IMFs 118 may be extracted corresponding to the phase current Ci, the phase current C2, and the phase current C3 separately. The EMD technique 116 may be employed to extract weak impact components in the multiphase current signal CS. EMD technique is a known non-linear decomposition process utilized to analyze and represent non-stationary real world signals. In general, EMD technique decomposes a time domain signal into the IMFs, e.g., IMF1, IMF2, IMF3, and so on, which are simple harmonic functions collected through an iterative process. The iterative process (known as sifting) eliminates most of signal anomalies and makes a signal wave profile more symmetric. This enables further processing to decompose a bandwidth of interest. The resulting IMFs represent different oscillation scales in the signal, with the first IMF capturing the highest-frequency oscillations and the last IMF capturing the lowest-frequency oscillations. A frequency content embedded in the processed IMFs reflects a physical meaning of the underlying frequencies. A final residual signal, obtained after all the IMFs have been extracted, represents a trend of the original signal. At step 206, the method 200 further comprises obtaining a residual signal 128 after extraction of the plurality of IMFs 118 corresponding to each phase current Ci, C2, C3 of the multiphase current signal CS. At step 208, the method 200 further comprises calculating a skewness 120 of each IMF 118 from the plurality of IMFs 118. In some embodiments, the skewness 120 of each IMF 118 may be indicative of an asymmetry of a probability density distribution of the corresponding phase current Ci, C2, C3 of the multiphase current signal CS. At step 210, the method 200 further comprises determining one or more faulty IMFs 124 from the plurality of IMFs 118 that have the skewness 120 outside of a predetermined skewness range 122. In some embodiments, if one or more IMFs 118 from the plurality of IMFs 118 contain a shock component caused due to the bearing fault or damage, the shock component may increase a probability density of the corresponding phase current Ci, C2, C3 with a larger amplitude, and the corresponding skewness 120 may be outside the predetermined skewness range 122. This may allow the one or more faulty IMFs 124 to be determined from the plurality of IMFs 118. In some embodiments, the predetermined skewness range 122 is from -0.35 to +0.35. At step 212, the method 200 further comprises reconstructing a new signal 126 with the one or more faulty IMFs 124 and the residual signal 128. Thus, the new signal 126 may comprise only the one or more faulty IMFs 124. The new signal 126 may be further processed for determining the bearing fault. At step 214, the method 200 further comprises transforming the new signal 126 from a time domain TD to a frequency domain FD to generate a transformed spectrum 130. In some embodiments, the new signal 126 is transformed using Fast Fourier Transform (FFT) 129. Simply stated, FFT 129 converts waveform data in a time domain to a frequency domain. FFT 129 accomplishes this by breaking down an original time-based waveform into a series of sinusoidal terms, each with a unique magnitude, frequency, and phase. This process, in effect, converts the waveform in the time domain (that is difficult to describe mathematically) into a more manageable series of sinusoidal functions that when added together, exactly reproduce the original waveform. Plotting an amplitude of each sinusoidal term versus its frequency creates a transformed spectrum, which is the response of the original waveform in the frequency domain. At step 216, the method 200 further comprises calculating a plurality of characteristic frequencies Fb of the bearing 110 based at least on the supply frequency F of the multiphase current signal CS, a geometry 140 of the bearing 110, and a plurality of integer values 142. Each characteristic frequency Fb from the plurality of characteristic frequencies Fb corresponds to a respective integer value 142 from the plurality of integer values 142. In some embodiments, each characteristic frequency Fb of the bearing 110 may represent a fundamental natural frequency of the bearing 110 for the given supply frequency F and the integer value 142. In some embodiments, the plurality of characteristic frequencies Fb of the bearing 110 is calculated according to the equation: Fb = Fs + KF0 where Fs is the supply frequency F, K is the integer value 142, and Fo is an outer race frequency calculated according to the equation: Nb (BD \1 Fn = — Fr 1 — — cos a 0 2 L \PD / J Where Nb is a total number of balls of the bearing 110, Fr is a shaft rotational frequency, BD is a ball diameter of the balls, PD is a pitch diameter, and a is a contact angle. At step 218, the method 200 further comprises selecting a plurality of frequency windows 144 around the plurality of characteristic frequencies Fb of the bearing 110, such that each frequency window 144 from the plurality of frequency windows 144 is around a respective characteristic frequency Fb from the plurality of characteristic frequencies Fb. Each frequency window 144 has a predetermined frequency range 138. The transformed spectrum 130 comprises a plurality of amplitudes 134 in each frequency window 144. In some embodiments, the predetermined frequency range for each frequency window 144 is 5 hertz (Hz). In some embodiments, the predetermined frequency range 138 may be utilized as a reference for the plurality of amplitudes 134 in each frequency window 144. At step 220, the method 200 further comprises calculating an average value Aea of the plurality of amplitudes 134 of the transformed spectrum 130 in each frequency window 144. In other words, the average value Aea represents an average of magnitudes of the plurality of amplitudes 134 in each frequency window 144. In some embodiments, the average value Aea of the plurality of amplitudes 134 of the transformed spectrum 130 in each frequency window 144 is calculated according to the equation: Ne-1 A^=^~^Ae(fd e 1=0 where Ne is a number of spectral lines in the transformed spectrum 130, f is the ith spectral line in the transformed spectrum 130, i is an integer value between 0 and Ne-1, and Ae is the amplitude 134. At step 222, the method further comprises calculating a maximum value Aed from the plurality of amplitudes 134 of the transformed spectrum 130 in each frequency window 144. In other words, the maximum value Aed represents a maximum magnitude from the plurality of amplitudes 134 considering each frequency window 144. In some embodiments, the maximum value Aed from the plurality of amplitudes 134 of the transformed spectrum 130 in each frequency window 144 is calculated according to the equation: Ne-1 maX lAe^fd ~ Af^Ae(lfd ~ Af + fs^’............] e i=0 where Ne is the number of the spectral lines in the transformed spectrum 130, Ae is the amplitude 134, i is the integer value between 0 and Ne-1, fss is a frequency resolution, and Af is a frequency interval for searching the plurality of characteristic frequencies Fb. At step 224, the method 200 further comprises determining a characteristic frequency ratio 150 for each frequency window 144 as a ratio of the maximum value Aed to the average value Aea. In some embodiments, a value of the characteristic frequency ratio 150 may be utilized to determine whether there is a bearing fault and a severity of the bearing fault. At step 226, the method 200 further comprises determining that the at least one multiphase motor M has a bearing fault 154 if the characteristic frequency ratio 150 for each frequency window 144 is above a predetermined threshold 152 (also shown in FIG. 4). Thus, the bearing fault 154 associated with the at least one multiphase motor M may be determined based on the characteristic frequency ratio 150. The predetermined threshold 152 may vary based on application requirements. The method 200 of the present disclosure may enable incipient detection of the bearing fault 154 in the multiphase motor M with improved accuracy and reliability. The method 200 involves applying the EMD technique 116 to extract the plurality of IMFs 118 corresponding to each phase current Ci, C2, C3. The method 200 further comprises calculating the skewness 120 of each IMF 118 and determining the one or more faulty IMFs 124 having the skewness 120 outside of the predetermined skewness range 122. Thus, only the one or more faulty IMFs 124 may be selected for further processing. This may enhance an accuracy of detecting the bearing fault 154 while reducing a computational effort. In other words, the multiphase current signal CS may be processed in such a way that only a fault information (i.e., the one or more faulty IMFs 124) is utilized for further processing, thereby reducing computational time and effort. Therefore, the method 200 provides a cost-effective and reliable technique for determining the bearing fault 154 as compared to conventional techniques. The selection of the plurality of frequency windows 144 around the plurality of characteristic frequencies Fb may eliminate a risk of any deviation in a theoretical and an actual characteristic frequency value. Accordingly, the method 200 may prevent failure of the multiphase motor M and the power system 100, thereby reducing downtime and overall maintenance costs. The method 200 may detect the bearing fault 154 by monitoring only the multiphase current signal CS (using existing sensor data) and without requiring use of additional sensors and associated hardware. The method 200 may be applied to any type of electric motor, e.g., permanent magnet machines, induction machines, wound field synchronous machines, 6-phase or 12-phase machines, etc. Further, the method 200 may be applied to existing health monitoring systems for detecting the bearing fault 154 in real time. The incipient bearing fault detection along with other equipment health monitoring (EHM) parameters may improve an accuracy of the health monitoring systems and aid in designing an ideal maintenance scheme for different use cases. FIG. 4 shows an exemplary graph 300 illustrating a variation of the characteristic frequency ratio 150 with the plurality of integer values 142 for a healthy bearing and a faulty bearing. The characteristic frequency ratio 150 is shown along the vertical axis or ordinate of the graph 300 and the plurality of integer values 142 are shown along the horizontal axis or abscissa of the graph 300. As shown in FIG. 4, the characteristic frequency ratio 150 of the faulty bearing is above the predetermined threshold 152 for the plurality of integer values 142. Accordingly, the characteristic frequency ratio 150 may be utilized for determining faulty bearings associated with an electric motor. FIG. 5 shows a schematic block diagram of a power system 400, according to another embodiment of the present disclosure. The power system 400 may be substantially similar to the power system 100 shown in FIG. 1, and same components in this embodiment are referred to by same reference numerals and differences between the embodiments are discussed. Specifically, in the power system 400, the at least one multiphase motor M comprises a first multiphase motor M1 having a first bearing 110a and a second multiphase motor M2 having a second bearing 110b. In addition, the at least one inverter 104 comprises a first inverter 104a driving the first multiphase motor M1 and a second inverter 104b driving the second multiphase motor M2. The power system 400 further comprises a first power source 102a for providing a first DC supply 108a to the first inverter 104a, and a second power source 102b for providing a second DC supply 108b to the second inverter 104b. Alternatively, in some other embodiments, the first inverter 104a and the second inverter 104b may be powered by a single power source. The first inverter 104a supplies a first multiphase current signal CSi to the first multiphase motor M1 and the second inverter 104b supplies a second multiphase current signal CS2 to the second multiphase motor M2. The first multiphase current signal CS1 (i.e., a three-phase current signal) has a supply frequency F1 and a plurality of phase currents Cn, C12, C13. Similarly, the second multiphase current signal CS2 (i.e., a three-phase current signal) has a supply frequency F2 and a plurality of phase currents C21, C22, C23. In some embodiments, the power system 400 further comprises a first multiphase current measurement device 106a and a second multiphase current measurement device 106b. The first multiphase current measurement device 106a is configured to measure the first multiphase current signal CS1 supplied by the first inverter 104a to the first multiphase motor M1. Similarly, the second multiphase current measurement device 106b is configured to measure the second multiphase current signal CS2 supplied by the second inverter 104b to the second multiphase motor M2. The health monitoring system 112 monitors a health of the power system 400. The processor 114 of the health monitoring system 112 is communicably coupled to the first inverter 104a and the second inverter 104b. Specifically, the processor 114 is communicably coupled to the first multiphase current measurement device 106a for acquiring the first multiphase current signal CS1. Further, the processor 114 is communicably coupled to the second multiphase current measurement device 106b for acquiring the second multiphase current signal CS2. FIG. 6 shows a schematic block diagram of a method 500 for monitoring the health of the power system 400. The processor 114 of the health monitoring system 112 is configured to perform the steps of the method 500. The method 500 will be set forth by way of FIGS. 5 and 6. In some embodiments, the method 500 may be an extension of the method 200 shown in FIGS. 3A and 3B. Referring to FIGS. 2, 3A, 3B, 5 and 6, at step 502, the method 500 comprises determining the characteristic frequency ratio 150a, 150b for each frequency window 144 for each of the first multiphase motor M1 and the second multiphase motor M2, respectively, as per steps of the method 200. At step 504, the method 500 further comprises determining a health index ratio 420 as a ratio of the characteristic frequency ratio 150a of the first multiphase motor M1 to the characteristic frequency ratio 150b of the second multiphase motor M2 for each frequency window 144. At step 506, the method 500 further comprises determining that the first multiphase motor M1 has a bearing fault 422 if the health index ratio 420 is greater than one by a predetermined tolerance value 430 for each frequency window 144. In some embodiments, the predetermined tolerance value 430 is 0.1. However, the predetermined tolerance value 430 may vary based on application requirements. At step 508, the method 500 further comprises determining that the second multiphase motor M2 has a bearing fault 424 if the health index ratio 420 is less than one by the predetermined tolerance value 430 for each frequency window 144. If the health index ratio 420 is not greater than one by the predetermined tolerance value 430 for each frequency window 144 or the health index ratio 420 is not less than one by the predetermined tolerance value 430 for each frequency window 144, then both the first multiphase motor M1 and the second multiphase motor M2 do not have any bearing fault. The method 500 may facilitate monitoring of the health of the power system 400 having the first multiphase motor M1 and the second multiphase motor M2, e.g., in a multilane architecture. The health index ratio 420 may be formulated based on the characteristic frequency ratio 150a of the first multiphase motor M1 and the characteristic frequency ratio 150b of the second multiphase motor M2 for each frequency window 144. Thus, the bearing faults 422, 424 in the first multiphase motor M1 and the second multiphase motor M2, respectively, may be determined simultaneously, thereby allowing preventive maintenance and repair / replacement activities to be performed. The method 500 may be extended to power systems having multiple motors arranged in multiple lanes, and the bearing fault may be determined in real time in any particular lane. This technique may not require any specific threshold definition as lane data may be utilized as threshold. In some cases, EHM parameters that are calculated for mechanical components for a set of operating points (i.e., speed, torque, etc.) may be compared with other lanes to assess degradation and potential faults in a lane. Referring to FIGS. 1-6, the method 200, 500 of the present disclosure may facilitate monitoring of the health of the power system 100, 400. Specifically, the method 200, 500 may enable incipient detection of the bearing fault 154, 422, 424 in the at least one multiphase motor M, M1, M2 with improved accuracy and reliability. The method 200, 500 involves selecting only the one or more faulty IMFs 124 from the plurality of IMFs 118 for further processing. This may enhance an accuracy of detecting the bearing fault 154, 422, 424 while reducing a computational effort. In other words, the multiphase current signal CS, CSi, CS2 is processed in such a way that only a fault information is utilized for further processing, thereby reducing computational time and effort. Therefore, the proposed method provides a cost-effective and reliable technique for determining the bearing fault as compared to conventional techniques. The selection of the plurality of frequency windows 144 around the plurality of characteristic frequencies Fb may eliminate a risk of any deviation in a theoretical and an actual characteristic frequency value. Accordingly, the method 200, 500 may prevent failure of the at least one multiphase motor M, M1, M2 and the power system 100, 400, thereby reducing downtime and overall maintenance costs. The method 200, 500 may detect the bearing fault 154, 422, 424 by monitoring only the multiphase current signal CS, CS1, CS2 (using existing sensor data) and without requiring use of additional sensors and associated hardware. The method 200, 500 may be applied to any type of electric motor, e.g., permanent magnet machines, induction machines, wound field synchronous machines, 6-phase or 12-phase machines, etc. Further, the method 200, 500 may be applied to existing health monitoring systems for detecting the bearing fault 154, 422, 424 in real time. The incipient bearing fault detection along with other equipment health monitoring (EHM) parameters may improve an accuracy of the health monitoring systems and aid in designing an ideal maintenance scheme for different use cases. It should be understood that steps of the method 200, 500 are not necessarily presented in any particular order and that performance of some or all the steps in an alternative order(s) is possible and is contemplated. The steps have been presented in the demonstrated order for ease of description and illustration. Further, it should be understood that steps can be added, omitted and / or performed simultaneously without departing from the scope of the appended claims. Moreover, it should also be understood that the illustrated methods 200, 500 can be ended at any time. It should be noted that the present method 200, 500 may be applied to other electrical assets of the power system 100, 400 for determining the health of the power system 100, 400. The term “processor,” as used herein may refer to any of the foregoing structure or any other structure suitable for implementation of the techniques described herein. In addition, in some aspects, the functionality described herein may be provided within dedicated software modules or hardware modules configured for performing the techniques of this disclosure. Even if implemented in software, the techniques may use hardware such as a processor to execute the software, and a memory to store the software. In any such cases, the computers described herein may define a specific machine that is capable of executing the specific functions described herein. Also, the techniques could be fully implemented in one or more circuits or logic elements, which could also be considered a processor. Although specific embodiments have been illustrated and described herein, it will be appreciated by those of ordinary skill in the art that a variety of alternate and / or equivalent implementations can be substituted for the specific embodiments shown and described without departing from the scope of the present disclosure. This application is intended to cover any adaptations or variations of the specific embodiments discussed herein. Therefore, it is intended that this disclosure be limited only by the claims and the equivalents thereof. List of reference signs (part of the description) 100 Power System 102 Power Source 102a First Power Source 102b Second Power Source 104 Inverter 104a First Inverter 104b Second Inverter 106 Multiphase Current Measurement Device 106a First Multiphase Current Measurement Device 106b Second Multiphase Current Measurement Device 108 DC Supply 108a First DC Supply 108b Second DC Supply 110 Bearing 110a First Bearing 110b Second Bearing 112 Health Monitoring System 114 Processor 116 Empirical Mode Decomposition (E 118 IMFs 120 Skewness 122 Predetermined Skewness Range 5 124 Faulty IMFs 126 New Signal 128 Residual Signal 129 Fast Fourier Transform 130 Transformed Spectrum 10 134 Amplitude 138 Predetermined Frequency Range 140 Geometry 142 Integer Values 144 Frequency Windows 15 150 Characteristic Frequency Ratio 150a Characteristic Frequency Ratio 150b Characteristic Frequency Ratio 152 Predetermined Threshold 154 Bearing Fault 20 200 Method 202 Step 204 Step 206 Step 208 Step 25 210 Step 212 Step 214 Step 216 Step 218 Step 30 220 Step 222 Step 224 Step 226 Step 300 Graph 35 400 Power System 420 Health Index Ratio 422 Bearing Fault Technique 424 Bearing Fault 430 Predetermined Tolerance Value 500 Method 502 Step 5 504 Step 506 Step 508 Step Aea Average Value Aed Maximum Value 10 cs Multiphase Current Signal Cl Phase Current c2 Phase Current c3 Phase Current CSi First Multiphase Current Signal 15 CS2 Second Multiphase Current Signal Cl 1 Phase Current Ci2 Phase Current C13 Phase Current C2i Phase Current 20 C22 Phase Current C23 Phase Current Fb Characteristics Frequency F Supply Frequency FD Frequency Domain 25 M Multiphase Motor M1 First Multiphase Motor M2 Second Multiphase Motor TD Time Domain

Claims

1. A method (200, 500) for monitoring a health of a power system (100, 400) having at least one multiphase motor (M) with a bearing (110) and at least one inverter (104) driving the at least one multiphase motor (M), the method (200, 500) comprising the steps of:a) acquiring (202) a multiphase current signal (CS) supplied by the at least one inverter (104) to the at least one multiphase motor (M), the multiphase current signal (CS) having a supply frequency (F) and a plurality of phase currents (Ci, C2> C3);b) applying (204) an Empirical Mode Decomposition (EMD) technique (116) to each phase current (Ci, C2, C3) of the plurality of phase currents (Ci, C2, C3) of the multiphase current signal (CS) to extract a plurality of Intrinsic Mode Functions (IMFs) (118) corresponding to each phase current (Ci, C2, C3);c) obtaining (206) a residual signal (128) after extraction of the plurality of IMFs (118) corresponding to each phase current (Ci, C2, C3) of the multiphase current signal (CS);d) calculating (208) a skewness (120) of each IMF (118) from the plurality of IMFs (118);e) determining (210) one or more faulty IMFs (124) from the plurality of IMFs (118) that have the skewness (120) outside of a predetermined skewness range (122);f) reconstructing (212) a new signal (126) with the one or more faulty IMFs (124) and the residual signal (128);g) transforming (214) the new signal (126) from a time domain (TD) to a frequency domain (FD) to generate a transformed spectrum (130);h) calculating (216) a plurality of characteristic frequencies (Fb) of the bearing (110) based at least on the supply frequency (F) of the multiphase current signal (CS), a geometry (140) of the bearing (110), and a plurality of integer values (142), wherein each characteristic frequency (Fb) from the plurality of characteristic frequencies (Fb) corresponds to a respective integer value (142) from the plurality of integer values (142);i) selecting (218) a plurality of frequency windows (144) around the plurality of characteristic frequencies (Fb) of the bearing (110), such that each frequency window (144) from the plurality of frequency windows (144) is around a respective characteristic frequency (Fb) from the plurality of characteristic frequencies (Fb), wherein each frequency window (144) has a predeterminedfrequency range (138), and wherein the transformed spectrum (130) comprises a plurality of amplitudes (134) in each frequency window (144);j) calculating (220) an average value (Aea) of the plurality of amplitudes (134) of the transformed spectrum (130) in each frequency window (144);k) calculating (222) a maximum value (Aetj) from the plurality of amplitudes (134) of the transformed spectrum (130) in each frequency window (144);I) determining (224) a characteristic frequency ratio (150) for each frequency window (144) as a ratio of the maximum value (Aed) to the average value (Aea); andm) determining (226) that the at least one multiphase motor (M) has a bearing fault (154) if the characteristic frequency ratio (150) for each frequency window (144) is above a predetermined threshold (152).

2. The method (500) of claim 1, wherein the at least one multiphase motor (M) comprises a first multiphase motor (M1) having a first bearing (110a) and a second multiphase motor (M2) having a second bearing (110b), wherein the at least one inverter (104) comprises a first inverter (104a) driving the first multiphase motor (M1) and a second inverter (104b) driving the second multiphase motor (M2), the method (500) further comprising:determining (502) the characteristic frequency ratio (150a, 150b) for each frequency window (144) for each of the first multiphase motor (M1) and the second multiphase motor (M2) as per steps a) to I) of claim 1;determining (504) a health index ratio (420) as a ratio of the characteristic frequency ratio (150a) of the first multiphase motor (M1) to the characteristic frequency ratio (150b) of the second multiphase motor (M2) for each frequency window (144);determining (506) that the first multiphase motor (M1) has a bearing fault (422) if the health index ratio (420) is greater than one by a predetermined tolerance value (430) for each frequency window (144); anddetermining (508) that the second multiphase motor (M2) has a bearing fault (424) if the health index ratio (420) is less than one by the predetermined tolerance value (430) for each frequency window (144).

3. The method (500) of claim 2, wherein the predetermined tolerance value (430) is 0.1.

4. The method (200, 500) of any preceding claim, wherein the predetermined skewness range (122) is from -0.35 to +0.35.

5. The method (200, 500) of any preceding claim, wherein the predetermined frequency range (138) for each frequency window (144) is 5 hertz (Hz).

6. The method (200, 500) of any preceding claim, wherein the plurality of characteristic frequencies (Fb) of the bearing (110) is calculated as:Fb = Fs ± KF0where Fs is the supply frequency (F), K is an integer value (142), and Fo is an outer race frequency calculated as:Nb (FD \1Fo = — Fr 1 — \ ~pQ cos a |where Nb is a total number of balls of the bearing (110), Fr is a shaft rotational frequency, BD is a ball diameter of the balls, PD is a pitch diameter, and a is a contact angle.7.8.The method (200, 500) of any preceding claim, wherein the average value (Aea) of the plurality of amplitudes (134) of the transformed spectrum (130) in each frequency window (144) is calculated as:Ne-1Aea^T A^ Np i—ii = 0where Ne is a number of spectral lines in the transformed spectrum (130), fj is the ith spectral line in the transformed spectrum (130), i is an integer value between 0 and Ne-1, and Ae is the amplitude (134).The method (200, 500) of any preceding claim, wherein the maximum value (Aed) from the plurality of amplitudes (134) of the transformed spectrum (130) in each frequency window (144) is calculated as:Ne — 1Aed=^^ max lAeWd - Af),Ae(ifd — Af + fss).............] e i=0where Ne is a number of spectral lines in the transformed spectrum, Ae is the amplitude (134), i is an integer value between 0 and Ne-1, fss is a frequencyresolution and Af is a frequency interval for searching the plurality of characteristic frequencies (Fb).

9. The method (200, 500) of any preceding claim, wherein the new signal (126) is transformed using Fast Fourier Transform (FFT) 129.

10. The method (200, 500) of any preceding claim, further comprising providing a DC supply (108) to the at least one inverter (104).

11. The method (200, 500) of any preceding claim, wherein the multiphase current signal (CS) is a three-phase current signal.

12. A health monitoring system (112) comprising a processor (114) communicably coupled to the at least one inverter (104, 104a, 104b), wherein the processor (114) is configured to perform the method (200, 500) of any preceding claim.

13. A power system (100, 400) comprising:atone multiphase motor(M, M1, M2);at least one inverter (104, 104a, 104b) configured to drive the at least one multiphase motor (M, M1, M2); andthe health monitoring system (112) of claim 12, wherein the processor (114) is communicably coupled to the at least one inverter (104, 104a, 104b).

14. The power system (100,400) of claim 13, wherein the at least one multiphase motor (M, M1, M2) is a permanent magnet synchronous motor.

15. An aircraft including the power system (100, 400) of claim 13 or 14.

Images