FIELD OF THE INVENTION
The invention relates generally to tire monitoring systems for collecting measured tire parameter data during vehicle operation and, more particularly, to a system and method for estimating tire inflation pressure and wear state based upon such measurements.
BACKGROUND OF THE INVENTION
Vehicle-mounted tires may be monitored by tire pressure monitoring systems (TPMS) which measure tire parameters such as pressure and temperature during vehicle operation. Data from TPMS tire-equipped systems is used to ascertain the status of a tire based on measured tire parameters and alert the driver of conditions, such as low tire pressure or leakage, which may require remedial maintenance. Sensors within each tire are typically installed in a green or uncured tire and then subject to cure at high temperatures. The high temperature and pressure can damage the sensor. Furthermore, additional cost is typically associated with mounting the sensor in the tire. It is generally desired to have a tire sensor that is durable enough to sustain 60 million cycles. Further, the location of the sensor makes it extremely difficult to replace if the sensor stops functioning.
Other factors such as tire wear state are important considerations for vehicle operation and safety. It is accordingly further desirable to measure tire wear state and communicate wear state to vehicle systems such as braking and stability control systems in conjunction with the measured tire parameters of pressure and temperature.
SUMMARY OF THE INVENTION
According to one aspect of the invention, a tire property estimation system includes a sensor for measuring vertical acceleration and longitudinal acceleration of the tire. The sensor is preferably mounted on the wheel or rim, but may also be mounted elsewhere. The system includes mode means for determining the tire vertical mode frequency and torsional frequency; the system used tire identification for generating tire-specific frequency mode coefficients using tire-specific identification data. A tire wear estimation is made based upon the tire inflation pressure data, the vertical mode frequency data, and the tire-specific frequency mode coefficients. An indirect inflation pressure estimation is made based upon the tire wear pressure data, the torsional mode frequency data, and the tire-specific frequency mode coefficients.
Definitions
“ANN” or “Artificial Neural Network” is an adaptive tool for non-linear statistical data modeling that changes its structure based on external or internal information that flows through a network during a learning phase. ANN neural networks are non-linear statistical data modeling tools used to model complex relationships between inputs and outputs or to find patterns in data.
“Aspect ratio” of the tire means the ratio of its section height (SH) to its section width (SW) multiplied by 100 percent for expression as a percentage.
“Asymmetric tread” means a tread that has a tread pattern not symmetrical about the center plane or equatorial plane EP of the tire.
“Axial” and “axially” means lines or directions that are parallel to the axis of rotation of the tire.
“Chafer” is a narrow strip of material placed around the outside of a tire bead to protect the cord plies from wearing and cutting against the rim and distribute the flexing above the rim.
“Circumferential” means lines or directions extending along the perimeter of the surface of the annular tread perpendicular to the axial direction.
“Equatorial Centerplane (CP)” means the plane perpendicular to the tire's axis of rotation and passing through the center of the tread.
“Footprint” means the contact patch or area of contact created by the tire tread with a flat surface as the tire rotates or rolls.
“Inboard side” means the side of the tire nearest the vehicle when the tire is mounted on a wheel and the wheel is mounted on the vehicle.
“Kalman Filter” is a set of mathematical equations that implement a predictor-corrector type estimator that is optimal in the sense that it minimizes the estimated error covariance when some presumed conditions are met.
“Lateral” means an axial direction.
“Lateral edges” means a line tangent to the axially outermost tread contact patch or footprint as measured under normal load and tire inflation, the lines being parallel to the equatorial centerplane.
“Luenberger Observer” is a state observer or estimation model. A “state observer” is a system that provide an estimate of the internal state of a given real system, from measurements of the input and output of the real system. It is typically computer-implemented, and provides the basis of many practical applications.
“MSE” is an abbreviation for Mean square error, the error between and a measured signal and an estimated signal which the Kalman Filter minimizes.
“Net contact area” means the total area of ground contacting tread elements between the lateral edges around the entire circumference of the tread divided by the gross area of the entire tread between the lateral edges.
“Piezoelectric Film Sensor” a device in the form of a film body that uses the piezoelectric effect actuated by a bending of the film body to measure pressure, acceleration, strain or force by converting them to an electrical charge.
“PSD” is Power Spectral Density (a technical name synonymous with FFT (Fast Fourier Transform).
“Radial” and “radially” means directions radially toward or away from the axis of rotation of the tire.
BRIEF DESCRIPTION OF THE DRAWINGS
The invention will be described by way of example and with reference to the accompanying drawings in which:
FIG. 1 is a tire property estimation system schematic diagram.
FIG. 2 is a logic flow diagram for determining a correlation model of tire vertical mode as a function of inflation pressure and tread depth.
FIG. 3 is a logic flow diagram for determining a correlation model of tire torsional mode as a function of inflation pressure and tread depth.
FIG. 4 is a schematic illustrating the model of a tire shown as a spring mass damper system.
FIG. 5 is a frequency graph showing an amplitude vs. frequency of the vertical acceleration signal, and particularly illustrating wheel hop and tire vertical mode.
FIG. 6 is a frequency graph showing an amplitude vs. frequency comparison of the vertical mode at different inflation pressure levels.
FIG. 7 is a frequency graph showing amplitude vs. frequency comparison of the vertical mode of a tire inflated to 42 psi at four different tread depths.
FIG. 8 is a tire vertical mode frequency vs. tire inflation pressure graph showing the vertical mode frequency curves at four different tread depths.
FIG. 9 is a three dimensional graph showing the “goodness” of model fit, showing tire vertical mode frequency vs. tread depth and pressure.
FIG. 10 is a tire vertical mode frequency vs tire inflation pressure showing “goodness” of model fit.
FIG. 11 is a tire vertical mode frequency vs tire tread depth showing “goodness” of model fit.
FIG. 12 is a schematic illustrating the model of a tire shown as a torsional spring mass damper system.
FIG. 13 is a frequency graph showing an amplitude vs. frequency of the longitudinal acceleration signal, and particularly illustrating first torsional mode and second torsional mode.
FIG. 14 is a frequency graph showing an amplitude vs. frequency comparison of the longitudinal acceleration signal of the first torsional mode at different inflation pressure levels.
FIG. 15 is a frequency graph showing cumulative probability vs. frequency of the longitudinal mode at different inflation pressure levels.
FIG. 16 is a frequency graph showing multiple curves of tire first torsional mode vs. tire inflation pressure at four different tread depths.
FIG. 17 is a frequency graph showing an amplitude vs. frequency comparison of the longitudinal acceleration signal of the second torsional mode at different inflation pressure levels.
FIG. 18 is a frequency graph showing multiple curves of tire second torsional mode vs. tire inflation pressure at four different tread depths.
FIG. 19 is a three dimensional graph showing the “goodness” of model fit, showing tire torsional mode frequency vs. tread depth and pressure.
FIG. 20 illustrates a graph of frequency vs. tire inflation pressure for multiple curves of varying tread depth both in the vertical mode and torsional mode.
FIG. 21 illustrates how the output from the system may be communicated to a user via a cell phone network.
DETAILED DESCRIPTION OF THE INVENTION
Referring to FIG. 1, a tire property estimation system 10 is shown based on spectral analysis of the tire vibration signal. Such a system is useful in advising a vehicle owner on when to change tires and may be used to provide a driver with information on the tire inflation pressure and wear state of each tire. Tire properties generally change as a function of tire wear and temperature.
A representative tire 12 is typically mounted to a vehicle and includes a ground-engaging tread region 14 that wears over time. A first input to tire property estimation system 10 is tire identification or tire ID 20. The tire ID 20 may be input to the tire property estimation system 10 by a tag on the tire, or input by a user, or may be supplied by communication with a tire pressure monitoring system module (TPMS). The tire ID 20 includes tire size and tire model.
The tire property estimation system 10 employs an experimentally determined method to determine a real time estimate of tire inflation pressure and tire tread depth. The system includes a sensor module 30 mounted on the wheel 22 or tire 12 of a vehicle. The sensor module 30 preferably includes a tri-axial accelerometer. The sensor module 30 may also be mounted on a lug nut, valve stem, rim, axle, hub, unsprung mass of the vehicle or vehicle suspension strut. It is preferred that the sensor module 30 be located on the wheel in an accessible position. The sensor module 30 includes a micro-processor, communication means such as blue tooth, or other wireless communication means, and data storage means. One sensor module 30 suitable for use with the invention is a Texas Instruments CC 2650 wireless MCU, sold by Texas Instruments.
As shown in FIG. 4, a wheel of a vehicle can be modeled as a spring mass damper system. For the tire vertical mode, the following equation holds:
Vertical mode frequency ω=square root of k/m
where “k” is the tire carcass, inflation pressure dependent and m is the mass of the tire belt and tread mass. The dependence of tire wear (reduction in mass of the tread) and tire vertical mode of a rotating tire forms the basis for a correlation model between the tire wear state and the tire vertical mode frequency.
In order to determine a correlation model for a tire vertical mode frequency as a function of pressure and tread depth, the following steps are used to determine the correlation model. See FIG. 2 which describes the steps. A first input signal 32 to the tire property estimation system 10 is preferably a vertical acceleration signal Az measured by the sensor 30. A second input to the system is tire ID. The system input may also optionally include wheel speed signal, as described in more detail, below.
FIG. 5 illustrates an exemplary vertical acceleration signal Az 32 shown in the frequency domain. The Az signal 32 has been processed in the frequency domain. As shown in FIG. 5, the signal illustrates the signal peak at 18 Hz known as wheel hop. Wheel hop is typically representative around 12-15 Hz and is the vibration of the tire belt and rim. The wheel hop vertical mode depends mainly on suspension spring properties supporting the test wheel and tire and the overall tire stiffness. The Az signal 32 also shows a second signal peek at 90 Hz known as tire vertical mode. The tire vertical mode is typically in the frequency range of 90-100 HZ, and is representative of the tire belt vertical vibration. Provertical mode extractions 28 are made, extracting tire vertical modes. The Az signal is processed by the microprocessor of the sensor module 30 using a bandpass filter in the 50-100 hz range to isolate the tire vertical mode. The tire vertical mode was determined by means of an FFT (Fast Fourier Transform) analysis. The FFT analysis, conventionally used as a signal processing tool, yields tire vibration modes including the vertical mode represented in the subject graphs. As used herein, FFT is an algorithmic tool which operates by decomposing an N point time domain signal into N time domain signals each composed of a single point. The second step is to calculate the N frequency spectra corresponding to these N time domain signals. Lastly, the N spectra are synthesized into a single frequency spectrum. FIG. 6 illustrates the Az signal which has been processed as described. The peak of each curve 36 is then selected as representative of the tire vertical mode data point.
FIG. 2 illustrates the steps to generate a correlation model of vertical mode as a function of pressure and tread depth. In order to generate a correlation model for a specific tire, a series of experiments is conducted as shown in FIGS. 6 and 7. As shown in FIG. 6, a series of vertical acceleration measurements Az are made of the specific tire over a range of inflation pressures, 18, 24, 30, 36, and 42 psi, step 40. These measurements are conducted at a known tread depth, in this case 2/32 in. The measurements are repeated at additional tread depths, step 50. The peak of the signal curve is then selected as representative of the tire vertical mode for each pressure and tread depth measured, (step 60). Step 70 is illustrated in FIG. 7, wherein a series of vertical acceleration measurements Az is made over a range of tread depths for a constant pressure. The peak of each signal curve is then selected to be representative of the tire vertical mode 36. FIG. 8 illustrates the tire vertical mode curves for each tread depth measured, showing the variation of the tire vertical mode with tire inflation pressure, and illustrating a linear relationship. FIG. 8 further illustrates for the exemplary data set that there is an approximate 10 hz shift in resonance frequency for the 10/32 tread depth curve to the 2/32 tread depth curve. There is an approximate 20 Hz shift in resonance frequency for a given tread depth due to variation in tire inflation pressure.
An exemplary test matrix of measurements used to determine the correlation model of vertical mode is shown in FIG. 9. The data is curve fit to generate a single equation that is representative of the data. FIG. 9 also illustrates a three dimensional plot of the correlation model of tire vertical mode vs pressure and tread depth. For the data set shown in FIG. 9, the correlation model equation is:
Vertical mode frequency=Poo+P10*Pressure+P01*tread depth (1)
Wherein Poo is 72.87, p10 is 0.8011, and p01 is −39.41
The correlation model as described above, is unique to a model and size of a tire. The steps are repeated for each tire as desired, and the coefficients are stored in a data matrix that correspond to the tire ID. FIGS. 10 and 11 illustrate validity of the correlation model. FIG. 10 illustrates the goodness of fit for tire vertical mode frequency as a function of tire inflation pressure, illustrating a linear relationship and goodness of fit of 0.997. FIG. 11 illustrates a goodness of fit for tire vertical mode frequency as a function of tread depth, with a correlation coefficient of 0.992. Thus, the spring mass damper model has been validated by experimental data.
FIG. 12 illustrates that the tire may also be modeled as a torsional spring mass damper system, wherein Cθ is the torsional stiffness, and Ja is the moment of inertia of the axle. The equation is represented below:
natural frequency Fn=1/2π√{square root over ((Cθ)}/Ja) (2)
As equation 1 illustrates two unknowns, Pressure and tread depth, a second equation is utilized to solve for the two unknowns. The second equation is a correlation model of the torsional mode which is a function of pressure and tread depth. As shown in
of the tire vertical modes to correlate the influence of the tire wear state (depth level of the tire tread 14) on the tire vertical mode by using spectral analysis methods.
Application 32 of a correlation model is made between the tire wear state and the tire vertical mode using the tire specific models developed from TPMS-facilitated tire identification information.
In order to determine the torsional mode correlation model, the longitudinal acceleration Ax signal 34 is analyzed as described in FIG. 3. FIG. 13 illustrates a typical FFT longitudinal acceleration Ax signal which has been processed using spectral analysis in the frequency domain. As shown in FIG. 13, the first peak is the longitudinal suspension mode, the section peak is the first torsional in phase mode, and the third peak is the second torsional mode. Preferably, the first or second torsional mode is utilized to determine the torsional correlation model.
The first step 100 to determine the torsional mode correlation model is to measure the longitudinal acceleration signal Ax over a range of known inflation pressures for a known tread depth. FIG. 14 illustrates that the FFT acceleration signal Ax has been processed using a bandpass filter of 35-60 Hz and Fourier transform to extract the first torsional mode. The first torsional mode 105 may be selected at the peak of the curve or as shown in FIG. 15, the first torsional mode may be the frequency at which a normalized magnitude takes a value in the range of 0.2-0.9. In this example, the torsional mode frequency step 110 is selected at a value of 0.5. Next, the steps described above are repeated for a range of tread depths (step 120). FIG. 16 illustrates the first torsional mode for curves of varying tread depths, over a range of inflation pressures. FIG. 16 further illustrates an approximate 2 hz variation due to inflation pressure, and an approximate 2 Hz shift in resonance frequency due to tread depth.
FIG. 17 illustrates curves of the second torsional mode frequency versus frequency for various inflation pressures. FIG. 18 illustrates the second torsional mode for curves of varying tread depths, over a range of inflation pressures. FIG. 18 further illustrates an approximate 6-7 hz shift in resonance frequency due to inflation pressure, and an approximate 3-5 Hz shift in resonance frequency due to tread depth. Thus the second torsional mode is preferred as it has greater sensitivity.
An exemplary test matrix of measurements used to determine the correlation model of torsional mode is shown in FIG. 19. The data is curve fit to generate a single equation that is representative of the data. FIG. 19 also illustrates a three dimensional plot of the correlation model of tire torsional mode vs pressure and tread depth. For the data set shown in FIG. 19, the correlation model equation is:
Torsional mode frequency=Poo+P10*Pressure+P01*tread depth (3)
Wherein Poo is 74.85, p10 is 0.5586, and p01 is −17.15
The torsional correlation model as described above, is unique to a model and size of a tire. The steps are repeated for each tire as desired, and the coefficients are stored in a data matrix that correspond to the tire ID.
FIG. 1 illustrates how the vertical mode and torsional mode correlation models are utilized in a real time system. First, the tire ID is input into the system. Next, the sensor module 30 senses the vertical acceleration Az 32 and the longitudinal acceleration Ax 34. Then the signals are processed by the microprocessor of the sensor module to filter, and perform the FFT analysis to extract the vertical mode and torsional mode, preferably the second torsional mode. Next, the vertical mode and the torsional mode is input into the correlation models equations 1 and 3. The microprocessor solves the equations to determine pressure and tread depth step 230, and then communicates the pressure and tread depth output to a user, and/or to a cellular phone, and/or to the vehicle.
Variations in the present invention are possible in light of the description of it provided herein. While certain representative embodiments and details have been shown for the purpose of illustrating the subject invention, it will be apparent to those skilled in this art that various changes and modifications can be made therein without departing from the scope of the subject invention. It is, therefore, to be understood that changes can be made in the particular embodiments described which will be within the full intended scope of the invention as defined by the following appended claims.
Patent Application
20180154707
