Condition monitoring for a mechanical drive train coupled with electric machines

Abstract

Method for monitoring the functionality of a mechanical drive train, the method comprising: Receiving voltage and current signals from at least one phase of an electric machine coupled to the mechanical drive train; Representing the electrical machine, which has a non-sinusoidal flux distribution, as a combination of several sinusoidally distributed virtual electrical machines corresponding to a harmonic order, based on the obtained voltage and current signals; Determining a torque profile or spectrum that is assigned to one or more combinations of the sinusoidally distributed virtual electrical machines; and Detecting the presence of an anomaly in the mechanical drivetrain based on the torque profile or spectrum.

Description

GENERAL TECHNICAL BACKGROUND

[0001] The invention relates generally to condition monitoring and in particular to a method and a system for detecting an anomaly in a mechanical drive train coupled with electric machines.

[0002] A power generation unit typically consists of a mechanical drivetrain and an electrical machine. Such an electrical machine generally includes a generator or an electric motor, which can be synchronous or asynchronous. Furthermore, the mechanical drivetrain includes reciprocating engines, gas turbines, wind turbines, or compressors. Problems or anomalies are generally detected during the operation of mechanical drivetrains. Non-limiting examples of such anomalies include a broken cylinder in a multi-cylinder piston engine or a defective cylinder causing significant fuel loss. Therefore, the functionality of such a power generation unit is usually monitored using a multi-sensor system installed in the mechanical drivetrain.Currently, the sensor system includes dedicated sensors for measuring vibration, temperature, and pressure to detect any anomalies in the mechanical drivetrain. Such sensor systems are prone to failure and require frequent maintenance. The sensors also require frequent calibration to obtain accurate measurements, further increasing the complexity of performing anomaly detection. Furthermore, these sensor systems incur additional costs for the power generation unit.

[0003] EP 1 541 869 A1 describes a method and a system for monitoring the functionality of a mechanical drive train with a compressor that includes a motor. The method comprises receiving voltage and current signals from the motor and detecting the presence of an anomaly in the mechanical drive train, e.g., fluctuations in motor temperature, based on algebraic manipulation of the received voltage and current signals.

[0004] US 6,756,757 B2 discloses a system and method for controlling a rotating electromagnetic machine, for example, a permanent magnet motor or a switched reluctance motor, comprising a stator with a plurality of phase windings and a rotor. A drive is connected to the phase windings to supply current to the windings. The control system includes an estimator connectable to the machine, which receives signals representing the phase winding voltage and the rotor position and outputs parameter estimates for an electrical model of the machine based on the received signals. A torque model receives the parameter estimates from the estimator to estimate the torque for associated rotor position-phase current combinations of the machine. A controller outputs a control signal to the drive in response to a torque request and rotor position signals, as well as the torque model.

[0005] US 2010 / 0 169 030 A1 discloses a device embedded in a power distribution enclosure that enables analysis of the states of electromechanical machines or their drive or output units. The analysis uses operating voltages and currents supplied to or originating from the electromechanical machines. Because these voltages and currents are available at the enclosure, wiring or other means of communication to sensors on the electromechanical machines or on the drive or output units are neither required nor provided.

[0006] US 6,590,362 B2 discloses a method and system for the early detection of incipient faults in an electric motor. First, current and voltage values ​​for one or more phases of the electric motor are measured during motor operation. Then, a current predictor based on a neural network is used to generate a set of current predictions using the measured voltage values ​​and an estimate of the motor's rotational speed. Next, the set of current predictions is combined with the measured current values ​​and used together to calculate a set of fault indicators. Finally, a comparison of the fault indicators with a set of predefined baseline values ​​determines whether an impending electrical, mechanical, and / or electromechanical fault is present.

[0007] There is still a need for improvement in the accurate detection of anomalies in a mechanical drive train of a power generation unit. SHORT DESCRIPTION

[0008] According to one embodiment of the invention, a method for monitoring the functionality of a mechanical drive train is provided. The method includes obtaining voltage and current signals from at least one phase of an electric machine coupled to the mechanical drive train. The method also includes representing the electric machine, which has a non-sinusoidal flux distribution, as a combination of several sinusoidally distributed virtual electric machines corresponding to a harmonic order, based on the obtained voltage and current signals. The method further includes determining a torque profile or spectrum associated with one or more combinations of the sinusoidally distributed virtual electric machines.Finally, the procedure includes detecting the presence of an anomaly in the mechanical drivetrain based on the torque profile or spectrum.

[0009] According to one embodiment of the invention, a system for monitoring the functionality of a mechanical drive train is provided. The system includes a device module that is connected to an electric machine which is coupled to the mechanical drive train.The device module is configured to measure characteristic properties of the machine and includes a memory containing commands for obtaining voltage and current signals from at least one phase of an electrical machine coupled to the mechanical drive train, for representing the electrical machine, which has a non-sinusoidal flux distribution, as a combination of several sinusoidally distributed virtual electrical machines corresponding to a harmonic order based on the obtained voltage and current signals, for determining a torque profile or spectrum associated with one or more combinations of the sinusoidally distributed virtual electrical machines, and for detecting the presence of an anomaly in the mechanical drive train based on the torque profile or spectrum. DRAWINGS

[0010] These and other features, aspects and advantages of the present invention will be more easily understood if the following detailed description is read with reference to the accompanying drawings, in which the same reference numerals in all drawings denote the same parts, wherein: Fig. Figure 1 shows a block diagram of a system for determining an anomaly in a mechanical drive train of a power conversion unit according to an embodiment of the present invention. Fig. Figure 2 shows flux linkage vectors of nth harmonic order determined by a non-sinusoidally distributed electrical machine according to an embodiment of the present invention. Fig. Figure 3 shows a representation of the several virtual electrical machines having different instantaneous speeds and connected to the same shaft according to an embodiment of the present invention. Fig. Figure 4 shows the different harmonic flux linkage vectors which rotate with different electrical angular velocities according to an embodiment of the present invention. Fig. Figure 5 shows an amplitude modulation due to load torque pulsations, which leads to oscillations in the electrical angular velocity of a stator flux linkage vector in a non-sinusoidally distributed electrical machine according to an embodiment of the present invention. Fig. Figure 6 shows a fundamental oscillation which, according to an embodiment of the present invention, has been identified from a frequency spectrum of an estimated torque in order to further determine an exact electrical angular velocity w e * to estimate or calculate. Fig. Figure 7 shows a non-limiting example of a torque profile generated for a single-cylinder piston engine connected to an electric machine according to an embodiment of the present invention. Fig. Figure 8 shows a non-limiting example of a representation of a superposition of multi-cylinder torque profiles according to an embodiment of the present invention. Fig. Figure 9 shows a non-limiting example of a resulting torque profile of a fault-free multi-cylinder piston engine running according to a specific firing order according to an embodiment of the present invention. Fig. Figure 10 shows another non-restrictive example of a torque profile of a faulty multi-cylinder piston engine running according to a given firing order. Fig. Figure 11 shows a non-restrictive example of a filtered torque profile representing phase information of a drive end and a non-drive end of a faulty cylinder. Fig. Figure 12 shows a flowchart for a method 400 for monitoring the functionality of a mechanical drive train in a power generation unit according to an embodiment of the present invention. DETAILED DESCRIPTION

[0011] When elements of different embodiments of the present invention are introduced, the articles "one," "the," and "this" mean that one or more elements are present. The expressions "comprise," "include," and "have" are not exclusive and mean that other elements besides those listed may also be present. Any examples of operating parameters do not exclude other parameters of the disclosed embodiments.

[0012] Fig. Figure 1 is a block diagram of a system 10 for determining an anomaly in a mechanical drive train 12 of a power generation unit according to an embodiment of the present invention. The system 10 includes an electric machine 14 connected to the mechanical drive train 12 via a shaft 16. Generally, such an electric machine 14 includes a generator or a motor, which may be synchronous or asynchronous. In one embodiment, the mechanical drive train 12 includes a piston engine with one or more cylinders, which are fired using a defined firing order. In another embodiment, the mechanical drive train 12 may also include a compressor, a pump, or another mechanical system driven by a motor.In a non-restrictive example, such a power generation unit is used in a diesel locomotive, wherein the mechanical drive train is a 12-cylinder reciprocating diesel engine connected via a shaft to an electric generator. In another embodiment, the mechanical drive train includes a pump. The electric generator sends electrical power to a traction motor located on each axle, which drives the wheels of the diesel locomotive.

[0013] To control and monitor the electric machine 14, a device module 18, for example a relay, a measuring device, or another suitable device, is connected to the electric machine 14. Note that the device module 18 may include components of a computer or be a computer itself. As shown, the device module 18 includes, for example, a processor 20, a memory 22, and a display 24. The display 24 includes a visual and / or audible display function. The memory 22 includes any suitable volatile memory, non-volatile memory, or a combination thereof. The memory 22 stores all parameters, algorithms, or other data for monitoring and controlling the power generation unit and also allows the processor 24 to access this data.It should be noted that embodiments of the invention are not limited to any particular processor for performing the processing tasks of the invention. The term "processor," as used herein, is intended to denote any machine capable of performing the calculations or computations necessary to accomplish the tasks of the invention. The term "processor" is intended to denote any machine capable of receiving structured input and processing the input according to predefined rules to produce an output. It should also be noted that, as the person skilled in the art will understand, the processor may be equipped with a combination of hardware and software to perform the tasks of the invention.

[0014] The device module 18 monitors various parameters of the electric machine 14. In a non-restrictive example, the device module 18 is connected to various monitoring components, such as sensors, transformers, etc., in the electric machine 14. The monitoring components serve to monitor current, voltage, or any other parameters. As shown by lines 26, the device module 18 receives three-phase stator currents (I a , I b , I c ) from the electric machine 14, which is connected to the mechanical drive train 12. As indicated by lines 28, the device 18 receives three-phase stator voltages (V a , V b , V c) from the electric machine 14, which is connected to the mechanical drive train 12. Note that various signal processing components may be included in or between the electric machine 14 and the device module 18, for example, signal conditioners, amplifiers, filters, etc. The device module 18 also includes a switching mechanism to control the operation of the power generation unit. As explained in more detail below, the device module 18 can switch off the mechanical drive train 12 via a switch in response to an anomaly.

[0015] Furthermore, the memory 22 of the device module 18 contains several instructions for executing a torque calculation algorithm to determine exact torque values ​​and subsequently the anomaly in the mechanical drive train 12. In one embodiment, the instructions in memory 24 can record or receive three-phase stator current signals 26 (I a , I b and I c ) and the three-phase stator voltages 28 (V a , V b and V c) from the electric machine 14, which is connected to the mechanical drive train 12 via a shaft 16. In another embodiment, the instructions involve representing the electric machine 14, which has a non-sinusoidal flux distribution, as a combination of several virtual electric machines with sinusoidal distributions according to a higher harmonic order, based on the obtained voltage and current signals. Note that the non-sinusoidal flux distribution in the electric machine 14 is caused by non-ideal winding distributions in slots and teeth, pronounced poles, an unbalanced power supply or load impedance connected to the stator terminals of the electric machine, or by any geometric irregularities in the electric machine 14. In general, a torque calculation equation is given by Te=IM[P2φSIS*] wherein IS* a complex conjugate of a current vector I S is and φ S the resulting flux chaining vector is, where the resulting flux chaining vector φ S furthermore, from the stator voltages and stator currents of the electric machine as φS=∫[VS−ISRS]dt=ISLS+φf. is calculated, where Vs is the stator voltage, I s the stator current is, R s the electrical resistance is, L s the inductance of the stator circuit is and φ f The flux linkage is created solely by the magnetic voltage (MMF) of the rotor.

[0016] However, due to the non-sinusoidal flux distribution in the electrical machine 14, several harmonics are present in the flux linkage vector φ. S and the current vectors IS* available, such as; φS=|φS|ejωet=∑n(φfnejnωe*t+InLSejn(ωe*t+α)) IS=∑nInejn(ωe*t+α) wherein ω e * is the electric angular velocity of both the flux and the current harmonics, and α is the phase shift of the current vector with respect to the other vectors. ω e is the electric angular velocity of the resulting flux linkage vector. The flux linkage φ f , which is determined solely by the rotor MMF, can be represented as φf=∑n(φfnejnωe*t)

[0017] Therefore, the flux chaining vectors of nth harmonic order established by the non-sinusoidal electric machine can be represented as in Fig. 2. Thus, it becomes clear that the stator flux linkage is of nth harmonic order φ. Sn with an electric angular velocity nω e * rotates, just like the nth harmonic flux chain originating from the rotor φ fn is being set up.

[0018] An interaction between the harmonics of the current and the flux generates ripple components in the calculated torque, which cannot be accurately calculated using conventional methods. Therefore, the present invention incorporates the concept of multiple virtual machines to solve this problem, as described in Fig. 3 is shown.

[0019] Fig. Figure 3 shows a representation 40 of several sinusoidally distributed virtual electrical machines 42, connected to the same shaft and having different numbers of poles, according to one embodiment of the invention. The rotors of the several virtual electrical machines rotate at the same mechanical speed ω. m , but generate flux chaining vectors with different electric angular velocities ω e *, 5ω e *, 7ω e *...etc. The angular velocity ω eThe instantaneous velocity of the resulting flux chaining vector, as observed from the terminals, is the angular velocity ω. m is the instantaneous mechanical speed of the shaft to which all virtual machines are connected, and the angular velocity ω e * is the instantaneous velocity of the flux chain vector generated by the virtual machine corresponding to the fundamental frequency and denoted by ω m is related as ωe*=P2ωm where P is the number of poles of the electric machine.

[0020] Thus, a first method for determining the mechanical torque T can be used. m expressed on the wave as Tm=Pinωm=P2[Re(VSIS*)ωe*] in which P in The input power is and Re is the real part of the multiplied V. S and IS* is. At any time, ω can e* by resolving the resulting flux chaining vector observed at the terminals into several harmonics rotating at corresponding harmonic velocities, as shown in Figure 50 of Fig. Figure 4 shows this resolution. This resolution can be achieved using the information about the ratio of the amplitudes of φ. fn This can be done for different values ​​of n. This information can be extracted by recording the ratio of the harmonic voltages at the stator terminals during an open-circuit condition.

[0021] Note that the non-sinusoidal flux distribution in the electrical machine 14 (which is in Fig. (as shown in Figure 1) is also caused by load pulsations introduced by the mechanical drive train 12. Due to these load pulsations, each harmonic of the flux linkage vector is amplitude-modulated. Assume that the mechanical angular velocity of the rotor is constant at ω m , then sidebands appear on each side of the fundamental frequency component in the resulting flux. Thus, the electric angular velocity of the stator flux linkage of nth harmonic order φ is Sn no longer nω e * This phenomenon is described in Fig. 5 explained. To obtain a correct instantaneous velocity of the flux chaining vector ω e * To calculate, the fundamental frequency component of the flux linkage is therefore taken into account.

[0022] Furthermore, determining the exact instantaneous velocity ω includes e* firstly, performing a Fast Fourier Transform (FFT) on a real or imaginary part of the resulting flow chain vector φ S Secondly, since all harmonics exhibit a similar amplitude modulation pattern, some harmonic of the nth order is selected, and the sidebands are discarded, as shown in Figure 60 of Fig. Figure 6 shows that harmonic 62 is selected and the remaining sidebands are discarded. Third, all remaining frequencies are considered zero, and a time-domain vector is reconstructed from the modified FFT. Finally, the velocity of the resulting vector is given as the exact estimate of the corresponding instantaneous velocity nω. e * considered. The speed of each virtual machine is thus determined using the input power P inextracted to determine exact torque values ​​corresponding to each of the virtual machines.

[0023] In another method for determining the exact electric velocity ω e * the flow chain φ will be first f , which is built up by the rotor MMF, is determined by φf=φS−ISLS.

[0024] Secondly, an FFT is performed, and every nth harmonic is extracted, with all other frequencies considered as zero, similar to the procedure described in Fig. 6 is described. The time domain vector is then reconstructed, and its velocity is given as the exact estimate of the corresponding electric velocity nω. e * considered. The speed of each individual virtual machine is thus extracted to determine the torque from the input power P. in to determine.

[0025] In a second method for calculating the torque, individual torque profiles are calculated for each of the harmonic electrical machines. As in Fig. As shown in Figure 3, the nth-order machine has a number of poles nP. Thus, the total torque is composed of the individual torque profiles as follows: Tm=∑nTmn=Im[P2(∑nbφfn)IS*]

[0026] In one embodiment, the instructions in memory include 24 (in Fig. (Figure 1) determines a torque profile associated with one or more combinations of sinusoidally distributed virtual electrical machines. To determine the torque profile precisely, the instructions further include determining a first torque profile associated with a sinusoidally distributed virtual electrical machine corresponding to the fundamental frequency. The instructions also include determining a second torque profile associated with a sinusoidally distributed virtual electrical machine corresponding to the second harmonic order, and several subsequent torque profiles associated with subsequent harmonic orders of sinusoidally distributed virtual electrical machines. Furthermore, the instructions include combining the first torque profile, the second torque profile, and the subsequent torque profiles to determine the torque profile precisely.Finally, the system detects 10 (in . Fig. (1 shown) the anomaly in the mechanical drivetrain by comparing the torque profile with a set of expected torque profiles based on a malfunction in one or more components of the mechanical drivetrain. In one embodiment, the method for detecting the anomaly involves determining the presence of a broken cylinder using information about the firing order of the engine or the presence of a defective cylinder with significant fuel loss.

[0027] A non-limiting example of a torque profile 100 generated for a single-cylinder piston engine (primary drive) connected to an electric machine is given in Fig. Figure 7 is shown. The y-axis of the diagram shows the torque in Newton-meter units. The x-axis of the diagram shows the crankshaft angle of the engine cylinder in degrees of rotation. The torque profile 100 shows several phases (compression 102, power stroke 104, exhaust 106, intake 108) of a complete rotational cycle of the single-cylinder piston engine.

[0028] Similarly, it shows Fig. Figure 8 shows a non-restrictive example of a representation 120 of a superposition of multi-cylinder torque profiles 122. The y-axis of the diagram shows the torque in newton-meter units. The x-axis of the diagram shows the crank angle of the engine cylinder in degrees of rotation. The torque profiles 122 clearly show the firing order of the corresponding multiple engine cylinders of the mechanical drivetrain. In this embodiment, the firing order information 124 for the mechanical drivetrain (12-cylinder piston engine) is provided to minimize vibration and achieve smooth running for a long fatigue life of the engine.

[0029] Furthermore, it shows Fig. Figure 9 shows a non-restrictive example of a resulting torque profile 150 of a fault-free multi-cylinder piston engine running according to a specific firing order. The y-axis of the graph shows the torque in Newton-meter units. The x-axis of the graph represents time in seconds. Fig. Figure 10 shows another non-restrictive example of a torque profile 200 of a faulty multi-cylinder piston engine running according to a given firing order. The torque profile shows a defect 202 in one of the engine cylinders. As discussed above, such a defect could involve a broken cylinder or fuel loss from a defective cylinder of the piston engine. Such a defect leads to a pronounced set of harmonics in the torque spectrum in a sequence such as: ωrmN,2ωrmN,…3ωrmN,ωrm,… where N is the number of cylinders and ω rmThe mechanical speed of the engine is the determining factor. This also leads to a fundamental error frequency. Note that torque harmonics occur at all multiples of half the operating speed of a four-stroke engine and all multiples of the operating speed of a two-stroke engine.

[0030] In one embodiment, the processor calculates 20 (as in Fig. (1 shown) a basic error frequency in the torque profile for the faulty engine using information about the number of revolutions of the mechanical drivetrain for complete ignition of all cylinders. System 10 (as shown in Fig. (as shown in 1) also determines the presence and magnitude of sideband frequency components in the torque spectrum around the fundamental error frequency at intervals of the calculated fundamental error frequency. The magnitude of the sideband frequency indicates the severity of the anomaly. Furthermore, the system includes 10 (as shown in Fig. (Figure 1) shows the detection of the anomaly by filtering the third torque profile to extract the magnitude of the calculated fundamental fault frequency. Furthermore, the defective engine cylinder is identified using phase information extracted from the filtered torque profile.

[0031] Fig. Figure 11 shows a non-restrictive example of a filtered torque profile 300, which represents phase information of a defective cylinder 302 at a drive end and a defective cylinder 304 at a non-drive end. As shown, the characteristic information about the peak values ​​of the torque profile corresponding to the defective cylinder 302 at the drive end (off-cylinder) is recorded at the corresponding time. Using this timing information and a mapping to the firing order information of the mechanical primary drive, the defective cylinder is determined. Similarly, due to the time delay that occurs in the firing order of the mechanical drivetrain, the waveform of the defective cylinder 304 at the non-drive end shows a shift of approximately 180 degrees with respect to the waveform of the defective cylinder at the drive end 302.If any other cylinder is defective, there will be some other phase shift that can be clearly attributed to the location of the defective cylinder.

[0032] Fig.Figure 12 is a flowchart of a method 400 for monitoring the functionality of a mechanical drive train in a power conversion unit. The method includes, in step 402, obtaining voltage and current signals from at least one phase of an electric machine coupled to the mechanical drive train. The method also includes, in step 404, representing the electric machine, which has a non-sinusoidal flux distribution, as a combination of several harmonically ordered, sinusoidally distributed virtual electric machines based on the acquired voltage and current signals. Furthermore, in step 406, the method includes determining a torque profile associated with one or more combinations of the sinusoidally distributed virtual electric machines.The precise determination of the torque profile further includes steps in which a first torque profile is determined, which is assigned to a sinusoidally distributed virtual electric machine corresponding to the fundamental frequency. The procedure also includes determining a second torque profile, which is assigned to a sinusoidally distributed virtual electric machine corresponding to the second harmonic order, and likewise includes determining several subsequent torque profiles, which are assigned to sinusoidally distributed virtual electric machines corresponding to subsequent harmonic orders. The torque profile discussed in step 406 is finally generated by combining the first torque profile, the second torque profile, and the subsequent torque profiles.

[0033] Finally, in step 408, the procedure involves detecting the presence of an anomaly in the mechanical drivetrain based on the torque profile. Detecting the presence of the anomaly in the mechanical drivetrain further involves calculating a frequency spectrum from the torque profile, determining the magnitude and phase of the multiple frequency components in the torque spectrum, and comparing the frequency spectrum to an expected spectrum from a functioning drivetrain to determine the nature of the defect or anomaly. Note that the magnitude of multiple frequency components in the torque spectrum indicates the severity of the mechanical drivetrain anomaly, while the phase of multiple frequency components in the torque spectrum indicates the location of the mechanical drivetrain anomaly.The process also includes the modern extraction of the frequency components by appropriately filtering the torque spectrum.

[0034] Advantageously, the present method enables the processing of information from an electrical machine for the rapid and trouble-free detection of anomalies in a mechanical drivetrain, such as a broken cylinder, fuel loss in a defective cylinder, etc. Furthermore, when used with one or more computers and / or machines, the aforementioned algorithm creates an online monitoring function for systems (e.g., piston engines) and allows the user to plan the shutdown process and maintenance of the piston engine with defective cylinders as anomalies in advance.

[0035] Furthermore, the person skilled in the art recognizes the interchangeability of various features from different embodiments. Likewise, the various described process steps and features, as well as other known equivalents for such processes and features, can be mixed and combined by a person skilled in the art to construct additional systems and techniques in accordance with the principles of this disclosure. It should be clarified that not all of the objectives or advantages described above necessarily have to be achieved according to a particular embodiment. For example, the person skilled in the art will recognize that the systems and techniques described herein can be implemented or carried out in a manner that achieves or optimizes one advantage or group of advantages taught herein without necessarily having to achieve other objectives or advantages taught or proposed herein.

[0036] Although only certain features of the invention have been presented and described herein, a person skilled in the art will think of numerous modifications and alterations. Therefore, it should be understood that the appended claims are intended to cover all such modifications and alterations that fall within the true scope of the invention.

Claims

[1] A method for monitoring the functionality of a mechanical drive train, the method comprising: Obtaining voltage and current signals from at least one phase of an electrical machine coupled to the mechanical drive train; Representing the electrical machine having a non-sinusoidal flux distribution as a combination of a plurality of harmonic-order sinusoidally distributed virtual electrical machines based on the obtained voltage and current signals; Determining a torque profile or spectrum associated with one or more combinations of the sinusoidally distributed virtual electric machines; and Detecting the presence of an anomaly in the mechanical drivetrain based on the torque profile or spectrum. [2] The method of claim 1, wherein the anomaly is detected by comparing the torque profile to a set of expected torque profiles based on a malfunction in one or more components of the mechanical driveline. [3] The method of claim 1, further comprising: Determining a first torque profile associated with a sinusoidally distributed electric machine corresponding to a fundamental frequency; Determining a second torque profile associated with a second harmonic order sinusoidally distributed virtual electric machine; Determining a plurality of subsequent torque profiles associated with subsequent harmonic orders corresponding to sinusoidally distributed virtual electric machines; and Combining the first torque profile, the second torque profile, and subsequent torque profiles to accurately determine the torque profile. [4] The method of claim 3, wherein the combining further comprises multiplying appropriate gain and phase factors by the first torque profile, the second torque profile, and the subsequent torque profiles corresponding to the fundamental frequency, second, and subsequent higher harmonic order sinusoidally distributed virtual electric machines. [5] The method of claim 1, wherein the non-sinusoidal flux line distribution in the electric machine is caused due to non-ideal winding distributions in slots, teeth, salient poles, an unbalanced power supply or load impedance connected to the stator terminals of the electric machine, or due to any geometric irregularities in the electric machine. [6] The method of claim 1, wherein the non-sinusoidal flux distribution in the electric machine is caused by pulsations introduced by the mechanical drive train connected to the electric machine. [7] The method of claim 1, wherein the mechanical drivetrain has a plurality of periodicities and a plurality of frequency components associated with the malfunction in one or more components of the mechanical drivetrain. [8] The method of claim 7, further comprising detecting the anomaly in the mechanical drive train based on the plurality of periodicities and the plurality of frequency components. [9] The method of claim 8, wherein the frequency components are calculated using information about the multiple periodicities of a malfunction in one or more components of the mechanical drive train. [10] The method of claim 8, wherein detecting the presence of the anomaly in the mechanical drive train further comprises: Calculating a frequency spectrum from the torque profile; Determining a magnitude and a phase of the plurality of frequency components in the torque spectrum; and Comparing the frequency spectrum with an expected spectrum from a fully functional powertrain to determine the anomaly. [11] The method of claim 10, wherein the magnitude of the plurality of frequency components in the torque spectrum indicates the severity of the mechanical driveline anomaly. [12] The method of claim 10, wherein the phase of the plurality of frequency components in the torque spectrum indicates the location of the mechanical driveline anomaly. [13] The method of claim 8, wherein detecting the anomaly further comprises temporally extracting the frequency components by appropriately filtering the torque spectrum. [14] The method of claim 1, further comprising determining a corresponding angular velocity for each harmonic order sinusoidal distributed virtual electric machine. [15] The method of claim 14, further comprising determining the anomaly in the mechanical drive train based on an input power and the corresponding angular velocity of the harmonic order sinusoidal distributed virtual electric machine. [16] The method of claim 1, further comprising measuring a plurality of machine parameters from the electric machine, the machine parameters including a stator inductance, a stator resistance, and a leakage inductance. [17] A system for monitoring the functionality of a mechanical drive train, the system comprising: a device module in communication with an electric machine coupled to the mechanical drivetrain and the device module configured to measure characteristics of the electric machine, the device comprising a memory, the memory comprising instructions to: Obtaining voltage and current signals from at least one phase of the electrical machine coupled to the mechanical drive train; Representing the electrical machine having a non-sinusoidal flux distribution as a combination of a plurality of harmonic-order sinusoidally distributed virtual electrical machines based on the obtained voltage and current signals; Determining a torque profile or spectrum associated with one or more combinations of the sinusoidally distributed virtual electric machines; and Detecting the presence of an anomaly in the mechanical driveline based on the torque profile or spectrum.

Images