Patent Application
20180038714
The present disclosure generally relates to a control system, and more particularly relates to determining a position of a moveable device within a system.
Complex systems, such as aircraft, automobiles, spacecraft or other complex machinery include numerous parts which move. Aircraft, for example, have rudders, flaps, valves, landing gear, and numerous other parts which move. The position of the moveable parts within the complex system often needs to be known for the accurate operation of the complex system.
Accordingly, it is desirable to provide improved systems and methods for determining a position of a moveable device, which more accurately and efficiently determines the position. Furthermore, other desirable features and characteristics of the present invention will become apparent from the subsequent detailed description and the appended claims, taken in conjunction with the accompanying drawings and the foregoing technical field and background
In one embodiment, for example, a system for determining a position of a moveable device is provided. The system may include, but is not limited to, a variable displacement transformer, including but is not limited to, a shaft mechanically coupled to the moveable device, a primary coil arranged on a first side of the shaft, a first secondary coil arranged on a second side of the shaft, and a second secondary coil arranged on the second side of the shaft, a digital to analog converter electrically connected to the primary coil of the variable displacement transformer, the digital to analog converter configured to provide a sine wave signal to the primary coil of the variable displacement transformer having a predetermined frequency and a predetermined amplitude, and a processor electrically coupled to the first secondary coil and the second secondary coil, the processor configured to calculate, for each sample corresponding to a voltage induced by the primary coil onto the first secondary coil and a voltage induced by the primary coil onto the second secondary coil during a frequency period of the alternating current signal, Fourier Transform components by multiplying the voltage induced by the primary coil onto the first secondary coil by a first predetermined value and storing the results in a first accumulator, multiplying the voltage induced by the primary coil onto the first secondary coil by a second predetermined value and storing the results in a second accumulator, multiplying the voltage induced by the primary coil onto the second secondary coil by the first predetermined value and storing the results in a third accumulator, and multiplying the voltage induced by the primary coil onto the second secondary coil by the second predetermined value and storing the results in a fourth accumulator, the first predetermined value and the second predetermined value being unique to each frequency period of the sine wave signal, and determining, when a predetermined number of samples have been calculated, a sum of the values of the first buffer, a sum of the values of the second buffer, a sum of the values of the third buffer and a sum of the values of the fourth buffer, and determining the magnitude of the first secondary coil signal by calculating the square root of the sum of the squares of the first and second accumulators and storing the results in a first magnitude buffer, determining the magnitude of the second secondary coil signal by calculating the square root of the sum of the squares of the third and fourth accumulators and storing the results in a second magnitude buffer, and determining the position of the movable device based upon determined magnitude of the first secondary coil signal and the magnitude of the second secondary coil signal.
In another embodiment, for example, a method for determining a position of a moveable device is provided. The method may include, but is not limited to, generating, by an digital to analog converter, a sine wave signal having a predetermined frequency and a predetermined amplitude, outputting, by the digital to analog converter, the generated sine way signal to a primary coil of a variable displacement transformer, the variable displacement transformer comprising a shaft mechanically coupled to the movable device, outputting, by a first secondary coil of the variable displacement transformer, a voltage induced by the primary coil on the first secondary coil to a processor, outputting, by a second secondary coil of the variable displacement transformer, a voltage induced by the primary coil on the second secondary coil to a processor, calculating, by the processor, for each sample corresponding to a voltage induced by the primary coil onto the first secondary coil and a voltage induced by the primary coil onto the second secondary coil during a frequency period of the alternating current signal, Fourier Transform components by multiplying the voltage induced by the primary coil onto the first secondary coil by a first predetermined value and storing the results in a first accumulator, multiplying the voltage induced by the primary coil onto the first secondary coil by a second predetermined value and storing the results in a second accumulator, multiplying the voltage induced by the primary coil onto the second secondary coil by the first predetermined value and storing the results in a third accumulator, and multiplying the voltage induced by the primary coil onto the second secondary coil by the second predetermined value and storing the results in a fourth accumulator, the first predetermined value and the second predetermined value being unique to each frequency period of the sine wave signal, and determining, when a predetermined number of samples have been calculated, a sum of the values of the first buffer, a sum of the values of the second buffer, a sum of the values of the third buffer and a sum of the values of the fourth buffer, and determining, by the processor, the magnitude of the first secondary coil signal by calculating the square root of the sum of the squares of the first and second accumulators and storing the results in a first magnitude buffer, determining, by the processor, the magnitude of the second secondary coil signal by calculating the square root of the sum of the squares of the third and fourth accumulators and storing the results in a second magnitude buffer, and determining, by the processor, the position of the movable device based upon determined magnitude of the first secondary coil signal and the magnitude of the second secondary coil signal
The detailed description will hereinafter be described in conjunction with the following drawing figures, wherein like numerals denote like elements, and wherein:
The following detailed description is merely exemplary in nature and is not intended to limit the invention or the application and uses of the invention. As used herein, the word “exemplary” means “serving as an example, instance, or illustration.” Thus, any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments. All of the embodiments described herein are exemplary embodiments provided to enable persons skilled in the art to make or use the invention and not to limit the scope of the invention which is defined by the claims. Furthermore, there is no intention to be bound by any expressed or implied theory presented in the preceding technical field, background, brief summary, or the following detailed description.
The moveable device 110 may be any component of a vehicle, industrial, commercial or residential system which moves. In an aircraft, for example, the moveable device 110 may be any of various control surfaces such as a flap, an aileron, a rudder, a flaperon, a valve, any of various pilot inceptors including a control stick, a trim wheel, a rudder pedal and a steering yoke, any of various engine controls including a bleed valve, a thrust reverser and an inlet control vane, or the like.
The system 100 further includes a variable displacement transformer (VDT) 120. The VDT 120 may be a linear variable displacement transformer (LVDT) or a rotary variable displacement transformer (RVDT), depending upon whether the moveable device 110 uses linear motion or rotary motion. The VDT 120 converts a position of the moveable device 110 into a proportional electrical signal containing a polarity, which indicates a direction relative to a zero reference, and an amplitude representative of a distance from the zero reference.
In one embodiment, for example, the VDT 120 may include a primary coil 122. The primary coil 122 receives an input an alternating current (AC) signal from a Digital to Analog Converter (DAC) 130, as discussed in further detail below. The VDT 120 may further include two secondary coils 124 and 126 and a shaft 128. The shaft 128 is mechanically coupled to the movable device 110 in a manner that enables the VDT to accurately determine the position of the moveable device. The shaft 128 is positioned between the coils 122, 124 and 126. When the moveable device 110 is moved, the shaft 128 moves a corresponding amount, either rotationally in the case of a RVDT or linearly in the case of a LVDT. When the primary coil 122 receives the input AC signal from the DAC 130, the primary coil 122 induces an AC sine wave current on the secondary coils 124 and 126 proportional to the position of the shaft 128. Accordingly, the secondary coils 124 and 126 output an AC sine wave voltage which can accurately be converted into a position of the moveable device 110, as discussed in further detail below.
The DAC 130 may be composed of an analog filter that allows a digital version of the sine wave signal to be generated within the processor. There are several methods for generating a sine wave signal digitally. Two advantageous ways to generate the sine wave signal in the processor include pulse width modulation and pulse density modulation. In pulse width modulation the amplitude of the resulting analog signal is determined by the duty cycle of the digital signal into the analog filter. In pulse density modulation the periods of the digital pulse are fixed and the amplitude is determined by the quantity of the ‘logic one’ outputs versus the quantity of the ‘logic zero’ outputs. Because the fundamental frequency used in pulse density modulation can be higher than for a pulse width signal for a given processor clock and analog filer combination, pulse density modulation can produce a higher fidelity analog sine wave signal. In both modulation schemes, the serial stream of bits used for the signal generation can be calculated prior to circuit operation and stored in memory 180. The system 100, however, does not require either of these sine wave generation methods, and can work with other methods of providing a sine wave generation to the VDT sensor 120.
Each secondary coil 124 and 126 is coupled to an anti-aliasing filter 140. The anti-aliasing filters 140 prevent unwanted electrical noise from causing aliasing errors in the conversion of the signal data into position information. These anti-aliasing filters can be any of a number of filter types including passive filters built with combinations of resistors capacitors and inductors or a subset of these, active filters built with a combination of the passive elements previously mentioned and amplifiers. Anti-aliasing filters are typically low pass filters because the aliasing threats to the signal conditioning are typically higher that the signals of interest. The anti-aliasing filters 140 can also be bandpass or band rejection filters when the threats are such that these filters remove the unwanted aliasing frequency components.
The anti-aliasing filters 140 are coupled to the analog to digital converters (ADCs) 150. In one embodiment, for example, the ADCs 150 may be delta-sigma, ADCs which are well suited to the signal processing because they provide a good combination of DC accuracies and high resolution. Another benefit of delta-sigma ADCs is that the serial stream of conversion data from the modulator front end allows for efficient, high order, digital low pass filter implementations. In one embodiment, for example, the delta-sigma, analog to digital converters may include a digital low pass filter 160 included on the same device. The digital filters 160 can provide very effective aliasing noise rejection because the transition from the pass band to the rejection band is typically very steep with rejection levels of 100 decibels (dB) or more and with band transitions of hundreds of dB per decade of frequency change. The phase performance of the digital filters is typically very linear and predictable due to the digital nature of the calculations. However, in other embodiments, separate digital filters 160 may be coupled to the output of the ADCs 150.
As seen in
The processor 170 determines the position of the movable device 110 by processing the data output from the ADCs 150, and filtered by the digital filters 160 as discussed in further detail below. In one embodiment, for example, the processor may also control a frequency of ADCs 150 conversions and/or the number of ADC 150 conversions to use for each discrete Fourier transform (DFT) calculated in the processor 170 and/or the frequency of the DAC 130 to modify DFT transfer characteristics or to modify a signal-to-noise ratio, as discussed in further detail below. In the embodiment illustrated in
The system 100 may further include a memory 180 communicatively coupled to the processor 170. The memory may 180 store non-transitory computer readable instructions for implementing the system 100 as discussed in further detail below. The memory 180 may also be used to store pre-calculated data and/or operate as an accumulator as discussed in further detail below.
VDT sensors, such as the VDT 120, are typically made using various metals including ferrites, copper and stainless steel. Metals have significant thermal coefficients of expansion (TCEs). The TCEs of the various metals frequently do not match, which results in variations in the mechanical alignments of the moving parts of the VDT 120. Due to this phenomena, the performance of the VDT 120 as a position measurement device is compromised. Instead of having a linear relationship between the displacement of the shaft and the ratio of the secondary winding voltages, a more complex relationship results over temperature variations. To compensate for this phenomena, the system 100 further includes a temperature sensor 190 mounted on the VDT 120 and an analog to digital converter 195 coupled to the output of the temperature sensor 190. The temperature of the VDT 120 is measured by the temperature sensor 190. The Temperature sensor 190 could be of various types including a platinum resistance temperature device (RTD), a thermocouple, or a thermistor, or the like. The VDT sensor 120 behavior as a function of temperature can be determined prior to installation by careful measurements. These measurements provide a characterization of a given VDT 120 that allows a more accurate position measurement to be made when the deviations due to temperature are used to adjust the results of the position calculation based on VDT temperature and the characterization stored in the memory 180.
Because the modulation of the sine wave signal and the demodulation of the sensor signals are done in the same processor 170 using the same clock source for both, there is no opportunity for frequency or phase drift between the these components which provides for greater accuracy.
Upon receiving the signal corresponding to a sample of the VDT 120, the processor 170 then calculates the DFT for the sample by correlation, using the data received from the ADCs 150 and digital filters 160 and stores time samples and the DFT results in the memory 180. (Step 320). The data output from the ADCs 150 and the digital filters 160 is a set of sequential time domain data that is input to the processor 170. The data contains frequency and amplitude information about the position of the movable device 110 over the period of time when the signals from 124 and 126 were sampled. The DFT converts the data received from the digital filter 160, from the time domain into the frequency domain. The correlation method of performing the DFT reduces the amount of math to be performed by the processor 170 when the magnitude of signals is desired for only one frequency when compare to the other methods of calculating a DFT, which necessarily requires calculations for multiple frequency components.
In one embodiment, for example, the processor 170 may use Equation 1 to perform the DFT by correlation.
ReX[k] and ImX[k] are the two dimensional, rectangular, vector components of the magnitude of the secondary signals from the secondary coils 124 and 126. In the correlation method of calculating the DFT, “i” is effectively incremented through N values to determine the ReX and ImX vector components of each secondary coil's voltage magnitude. The ‘k’ in the equation is chosen to align with the frequency of excitation of the VDT; the frequency of the DAC 130. The value of x[i] corresponds to the digitized voltage level input to the processor 170 from the ADCs 150 and digital filters 160 at each sampled point in time i. The value of N corresponds to the number of samples in the correlation DFT calculation. N may be selected to be any integer value only when the for the correlation DFT calculation. N may be selected such that N times the sample interval is equal to an integer multiple of the number of the sensor sine wave cycles received from the secondary coils 124 and 126 for the correlation method. Therefore, the processor 170 may select N based upon an optimal frequency for the sensor 120 and for the ADC 150 and for the filters 140 and 160. For example, it may be determined that the best frequency to operate the VDT sensor has been determined to be 1800 Hz for reasons other than the DFT calculation such as optimal sensor operation and system timing constraints. Additionally, in this example, the optimal rate at which the ADC 150 operates may be determined to be 126,000 data samples per second, again for reasons other than the DFT calculations such as for rejecting aliasing threats. In this example appropriate values of N would be 70, 140, 210, 280 or even 2870, because these values of N divided by the sample data rate equals an integer number of 1800 Hz sine wave periods. As discussed in further detail below, when the FFT method of calculating the DFT is used, N must be chosen to be an integer power of two such as 32, 64, 128 or 256.
One benefit of the method 300 is that the method 300 does not require that N be an integer power of two to perform the DFT by correlation. In contrast, typical full Fourier transform calculations require the number of samples N to be an integer power of two. For the example above, the optimal sampling rate determined to be 126,000 data samples per second and the sensor frequency of 1800 Hz sine wave, does not result in a value of N that equals an integer power of two. Either the Sine wave frequency needs to change or the sample rate would need to change to satisfy the ‘integer power of two’ requirement, which would cause the implementation to use less optimal settings for the sensor and the analog to digital converter. By allowing N to be any integer value, the precision of the position determined by the system 100 and the overall performance of the system 100 can be optimized. In other words, rather than requiring the operations of the sensor 120 and the analog to digital converter 150 to be less than optimal for non-Fourier transform calculation reasons, the correlation process is more accommodating than other DFT methods.
Another example of the benefit of the correlation method over the other methods of DFT calculation is demonstrated when one considers the DFT's ability to attenuate unwanted signals.
Yet another benefit of the method 300 is that the values of the cosine and sine function can be calculated in advance and stored in the memory 180, for example, in a look-up table, as they are fixed values. As seen above,
from the real and imaginary components of equation 1 are values which can be calculated in advance as each contain variables which are known in advance. When compared to the FFT method, the number of sine and cosine values that need to be generated and stored is significantly less in the correlation method, thereby requiring a less memory usage. Furthermore, unlike the FFT equations, the processor 170 can perform incremental data calculations as the samples from the analog to digital converter arrive, before having all of the data samples, which produces the result sooner, reducing latency. In other words, the processor 170 can multiply each received value x[i] by the respective pre-calculated cosine and sine values stored in the memory 180 and then accumulate the DFT results, ReX[k] and ImX[k], in two memory 180 locations. Because all of the sampled x[i] values are multiplied by the factors above and the results of the multiplications are accumulated immediately after each individual multiplication, the results are available one multiplication and one addition after the final x[i] is received by the processor 170.
The processor 170 continues to process the data from the VDT 120 and perform the DFT calculation by correlation described above until all of the N samples of the DFT have been calculated. (Step 330).
The processor 170 then calculates a magnitude of the real and imaginary components of the DFT by correlation for each of the secondary windings 124 and 126 of the VDT 120. (Step 340). In one embodiment, for example, the processor may calculate the magnitude according to Equation 2:
Magwinding=√{square root over ((ReX[k]2+ImX[k]2))} Equation 2
The processor 170 may then determine the position of the moveable device 110 by performing the calculation described in Equation 3. (Step 350).
Where MagA is the magnitude of the data from secondary winding 124 and MagB is the magnitude of the data from secondary winding 126 calculated in Step 340.
The calculated position of the moveable device is then used by the processor 170 or otherwise output to another relevant system which requires the position information.
In one embodiment, for example, the processor may immediately return to step 310 to initiate a next position determination. However, in another embodiment, for example, the processor may determine if a new frame has started. (Step 360). In this embodiment, the system 100 may utilize a a fixed update frame rate and then maintain that frame rate by only calculating the position of the moveable device 110 once per each frame. This is done for many practical reasons. One of the main reasons is that some system stability analysis is determined based on predictable update frame rates. Aircraft systems, for example, are dependent on multiple position sensors and other sensors. Another consideration in stability has to do with the latency of the data. Knowing when the samples are taken with respect to the other control calculations also aides in the stability of a system.
As discussed above, the VDT 120 may be made using various metals having different coefficients of expansion (TCEs). Because the components of the VDT 120 expand and contract at different temperatures, the mechanical alignments of the moving parts of the VDT 120 may be compromised.
The processor 170 may then compensate the position calculated in Step 350 according to the determined temperature in Step 510. (Step 520). As discussed above, the behavior of the VDT 120 as a function of temperature can be determined prior to use by taking measurements of the VDT 120 at various temperatures. These measurements provide a temperature characterization of a given VDT 120 that allows a more accurate position measurement to be made. This temperature characterization can be stored in the memory 180. Accordingly, the processor 170 in Step 520 modifies the position of the moveable device 110 calculated in Step 350 by adjusting the position according to the temperature characterization of the VDT 120.
In one embodiment, for example, the processor 170 may modify the determined position of the moveable device via the method described in
In one embodiment, for example, the processor 170 may also compensate the system 100 for vibration.
Once enough samples are received from the VDT 120, the processor 170 performs a FFT on the data. (Step 630). By performing a FFT of the data from the VDT 120 the processor 170 can detect a vibration of the VDT 120 and determine a frequency of the vibration which may be corrupting the position measurement. By transforming the time domain data output from the VDT 120 into a frequency domain utilizing a FFT calculation, the processor 170 can determine which one or more frequency bins of the FFT include a vibration component.
The processor 170 then determines if any FFT bin magnitudes are greater than current correction limits. (Step 640). In other words, the processor 170 determines if the magnitude of a frequency bin of the FFT calculation corresponding to a vibration frequency is greater than a current correction limit. In one embodiment, for example, a correction limit may be chosen to be, for example, 2% of the full scale amplitude of the position sensor.
When a magnitude of one or more frequency bins is greater than the current correction level limit, the processor 170 determines if the largest bin is aligned with the current configuration. (Step 650). In other words, the processor 170 determines if the number of samples N from step 330 alignes a notch of the correlation calculation with the largest bin. Adjusting N is tha easiest way to move the notches of greater attenuation. However, setting other than the number N can also be part of the configuration. For example, a frequency of sensor excitation could alternatively be modified to move the attenuation notches.
When the largest bin is not aligned with the current configuration (from Step 650) or when the magnitude of a bin is less than or equal to the current correction level limit (from Step 640), the processor 170 selects a configuration stored in the memory 180 that best attenuates the frequency of vibration of the largest bin from the FFT calculation of Step 630. (Step 660). In other words, the frequency and amplitude information of unwanted signal components determined by the FFT performed in the processor 170 can be used to select from a set of VDT and DFT configurations stored in the memory 180 to achieve an optimal filter. The configurations would include optional combinations of sine wage generator frequencies, the value N for the correlation DFT calculation and the frequency of ADC 150 conversions.
When the largest bin is aligned with the current configuration (from Step 650), or after selecting a stored configuration in Step 660 the processor 170 then returns to Step 610 to begin a next vibration correction determination and optimal configuration selection. In one embodiment, for example, the method 600 could be performed once per each frame of the correlation calculation of the method 300 as each aircraft or other vehicle housing the moveable device 110 is operating. As the correlation method allows for more choices for the value of N and for the number of samples used in each DFT calculation, the ability of the system to adapt to threat conditions is increased above that for DFT methods that use only correlation or only FFTs to determine the position of the moveable device 110. However, in other embodiments, for example, the method 600 could be performed periodically, upon request of a user of the system 100 or even just once for each aircraft type certificate.
While at least one exemplary embodiment has been presented in the foregoing detailed description of the invention, it should be appreciated that a vast number of variations exist. It should also be appreciated that the exemplary embodiment or exemplary embodiments are only examples, and are not intended to limit the scope, applicability, or configuration of the invention in any way. Rather, the foregoing detailed description will provide those skilled in the art with a convenient road map for implementing an exemplary embodiment of the invention. It being understood that various changes may be made in the function and arrangement of elements described in an exemplary embodiment without departing from the scope of the invention as set forth in the appended claims.
