Patent Application
20140018999
Various embodiments of the present invention pertain to apparatus and methods for diagnosing the status of a multidegree of freedom system, and some embodiments pertain to analysis of wheeled vehicles.
Wheeled vehicles, such as the high mobility multi-purpose wheeled vehicle (HMMWV), are subjected to a wide range operational loading conditions. These vehicles are expected to function on various terrains from sand dunes to mountainous regions to highways. Also, vehicles are subjected to other varying conditions such as changes in payload, changes in tire pressure, and other factors. As a result, mechanical faults can occur in the wheel ends, suspension, and frame. The occurrence of faults in ground vehicles leads to high operation and support costs. In fact, the U.S. Department of Defense spent approximately ⅗th of its 500 billion dollar budget in 2006 on operation and support costs. The most commonly used maintenance techniques are (a) run-to-failure maintenance, which prescribes maintenance only after a failure occurs, and (b) preventive maintenance, which prescribes that service be conducted routinely to avert failure. However, the run-to-failure approach can result in increased maintenance costs because an entire subsystem (suspension) may need to be replaced instead of just one damaged component (tie bolt) if failure takes place. Preventive maintenance can be expensive because it is based on reliability predictions of the average time to failure for a fleet of vehicles, and such predictions can be conservative. For example, preventive maintenance is often carried out when convenient, so healthy as well as damaged system components may be replaced during these maintenance actions leading to part shortages and higher inventory costs.
Condition-based maintenance (CBM) is an approach that makes maintenance decisions based on the condition or health of an individual vehicle and its components. CBM is meant to reduce unnecessary maintenance while ensuring that proactive maintenance is conducted when needed to prevent failure. This method aims to increase the availability of vehicles at a lower cost. However,
There are two difficulties with this onboard condition monitoring approach. First, the number of datasets required to develop a library of possible healthy signatures extracted from an N-dimensional sensor suite on a vehicle given M terrains on which that vehicle can operate is of order MN. For example, 6 sensors over 10 terrains would require that one million datasets be used to establish a fully populated reference set for fault detection. If 240 datasets were acquired each day on average, then it would take 11 years to develop this library of healthy signatures for each individual vehicle. This large number of datasets would be needed to characterize the normal operational response of the vehicle due to the non-stationary nature of the loading. Second, many vehicles are not equipped with sensors nor the acquisition systems to acquire, process, and store data; therefore, to implement condition-based maintenance, vehicles would need to be equipped with instrumentation leading to high costs.
What is needed are methods, apparatus, and systems that improve the detection of faults in a vehicle. Various embodiments of the present invention do this in novel and unobvious ways.
Various embodiments of the present invention pertain to improved methods for detecting worn or faulted components on a vehicle.
Various other embodiments of the present invention pertain to simplified methods of testing a vehicle.
One aspect of the present invention pertains to an apparatus for analyzing a vehicle. Some embodiments include a first separable segment of driving surface, said first segment having a shape adapted and configured to locally elevate a vehicle driven over said first segment. Other embodiments further include a second separable segment of driving surface, said second segment having a top surface adapted and configured to be driven on by a wheeled vehicle, said second segment being located proximate to said first segment. Yet other embodiments include a motion sensor providing a signal corresponding to motion of said second segment. Still other embodiments include a computer receiving the signal and having software that analyzes the signal.
Another aspect of the present invention pertains to a method for analyzing a vehicle on a roadway. Some embodiments further include providing a portable segment of driving surface located proximate to a local elevational change in the roadway, and a sensor providing a signal corresponding to the response of the segment. Other embodiments include driving the vehicle first over the elevational change at a vehicle velocity sufficient to cause vehicle response. Yet other embodiments include driving the responding vehicle over the segment, recording the signal during said driving over the segment, and correcting the recorded signal for the response of the vehicle.
Yet another aspect of the present invention pertains to an apparatus for analyzing a wheeled vehicle having a wheelbase and a front track. Some embodiments include a first panel having a substantially flat top surface adapted and configured to be driven on by the vehicle, said first panel having a length less than the wheelbase. Other embodiments further include a second panel having a substantially flat top surface adapted and configured to be driven on by the vehicle, said second panel having a length less than the wheelbase. Yet other embodiments further include a first motion sensor attached to said first panel and providing a signal corresponding to motion of said first panel. Still other embodiments include a second motion sensor attached to said second panel and providing a signal corresponding to motion of said second panel;
It will be appreciated that the various apparatus and methods described in this summary, as well as elsewhere in this application, can be expressed as a large number of different combinations and subcombinations. All such useful, novel, and inventive combinations and subcombinations are contemplated herein, it being recognized that the explicit expression of each of these combinations is unnecessary.
Some of the figures shown herein may include dimensions. Further, some of the figures shown herein may have been created from scaled drawings or from photographs that are scalable. It is understood that such dimensions, or the relative scaling within a figure, are by way of example, and not to be construed as limiting.
For the purposes of promoting an understanding of the principles of the invention, reference will now be made to the embodiments illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated device, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates. At least one embodiment of the present invention will be described and shown, and this application may show and/or describe other embodiments of the present invention. It is understood that any reference to “the invention” is a reference to an embodiment of a family of inventions, with no single embodiment including an apparatus, process, or composition that should be included in all embodiments, unless otherwise stated. Further, although there may be discussion with regards to “advantages” provided by some embodiments of the present invention, it is understood that yet other embodiments may not include those same advantages, or may include yet different advantages. Any advantages described herein are not to be construed as limiting to any of the claims.
The use of an N-series prefix for an element number (NXX.XX) refers to an element that is the same as the non-prefixed element (XX.XX), except as shown and described thereafter The usage of words indicating preference, such as “preferably,” refers to features and aspects that are present in at least one embodiment, but which are optional for some embodiments. As an example, an element 1020.1 would be the same as element 20.1, except for those different features of element 1020.1 shown and described. Further, common elements and common features of related elements are drawn in the same manner in different figures, and/or use the same symbology in different figures. As such, it is not necessary to describe the features of 1020.1 and 20.1 that are the same, since these common features are apparent to a person of ordinary skill in the related field of technology. This description convention also applies to the use of prime (′), double prime (″), and triple prime (′″) suffixed element numbers. Therefore, it is not necessary to describe the features of 20.1, 20.1′, 20.1″, and 20.1′″ that are the same, since these common features are apparent to persons of ordinary skill in the related field of technology.
Although various specific quantities (spatial dimensions, temperatures, pressures, times, force, resistance, current, voltage, concentrations, wavelengths, frequencies, heat transfer coefficients, dimensionless parameters, etc.) may be stated herein, such specific quantities are presented as examples only, and further, unless otherwise noted, are approximate values, and should be considered as if the word “about” prefaced each quantity. Further, with discussion pertaining to a specific composition of matter, that description is by example only, and does not limit the applicability of other species of that composition, nor does it limit the applicability of other compositions unrelated to the cited composition.
Incorporated herein by reference is U.S. patent application Ser. No. 61/098,995, filed Sep. 22, 2008, titled INSTRUMENTED CLEAT.
What will be shown and described herein, along with various embodiments of the present invention, is discussion of one or more tests that were performed. It is understood that such examples are by way of example only, and are not to be construed as being limitations on any embodiment of the present invention.
Current maintenance schedules for ground vehicles are determined based on reliability predictions (e.g., mean time to failure) of a population of vehicles under anticipated operational loads; however, vehicles that experience component damage often lie in the tails of the reliability distribution for a given platform. For example, a certain group of military vehicles may be deployed to operate on a harsh terrain that is particularly taxing on the mechanical components in the suspensions or frames in those vehicles. Operation & support costs for military weapon systems accounted for approximately ⅗th of the $500B Department of Defense budget in 2006. In an effort to ensure readiness and decrease these costs for ground vehicle fleets, condition-monitoring technologies are being developed to enable maintenance decisions about individual vehicles.
Some embodiments of the present invention pertain to an apparatus that is an off-board vehicle damage detection system that can be used to enable CBM for a fleet of military wheeled vehicles. As an individual vehicle traverses a rubberized cleat, the cleat excites the vehicle dynamic response through an impulse delivered by the cleat to the wheels. The measured response of the cleat due to interactions between the wheel and a trailing edge plate aft of the cleat is then compared to a reference response to identify anomalies. The anomalies in the response are then used to detect, locate, and classify faults to enable CBM.
Dynamics-based condition monitoring is used because vibrations are a passive source of response data, which are global functions of the mechanical loading and properties of the vehicle. A common way of detecting faults in mechanical equipment, such as the suspension and chassis of a ground vehicle, that is used by many other researchers and companies is to compare measured operational vibrations with onboard sensors to a reference (or healthy) signature to detect anomalies. However, many vehicles are not equipped with sensors nor the acquisition systems to acquire, process, and store data; therefore, to implement condition monitoring, one must overcome the economic and technical barriers associated with equipping ground vehicles to continuously monitor the response.
Current maintenance schedules for ground vehicles determined based on reliability predictions of a population of vehicles under anticipated operational loads can lead to unnecessary maintenance and, in some cases, in-field failures depending on differences in the usage of individual vehicles. Condition-based maintenance is scheduled instead according to the condition of each vehicle to reduce the risk of failure and maintenance costs. However, on-board instrumentation for acquiring, processing, and storing operational data is expensive, and this data is also difficult to analyze due to variations in loading.
An instrumented diagnostic system for diagnosing mechanical faults in ground vehicle wheel ends and suspensions is shown and described in various embodiments of the present invention. A cleat 50 excites the vehicle's dynamic response through an impulse delivered to the vehicle's front and back tires. The response of an instrumented plate 56 is then recorded while in contact with the vehicle's tires using motion sensors 60. The measured dynamic response is compared to a reference response, and anomalies that correspond to vehicle faults are then detected.
An instrumented diagnostic cleat is proposed in this work for fault identification as illustrated in
Various embodiments of the present invention include correcting the measured response data, such as ignoring or modifying data pertaining to vehicle chassis modes of vibration in the frequency range below 10 Hz and natural frequencies in the free dynamic response of the cleat above 10 Hz. Tire and suspension faults are simulated in a high mobility multi-purpose wheeled vehicle and the faults are detected. Tire faults are simulated by decreasing the pressure within each tire below the manufacturer recommended level, whereas suspension faults are simulated by disconnecting each damper to mimic the effects of broken damper. The data indicates that the faults and locations of the faults are identified with 90% confidence in 7 out of 8 fault cases. Various other embodiments include correcting the measurements to compensate for changes in vehicle speed.
Various embodiments include a measurement and analysis system for diagnosing faults in the wheels and suspensions of a wheeled ground vehicle. Resilient cleats 50 were placed in front of two instrumented plates 56 to record the dynamic response of the vehicle 20 as it traversed the cleats. It is shown using modal impact testing together with complex mode indicator function analysis and operational mode analysis that resonance frequencies of the vehicle were dominant below 10 Hz while resonance frequencies of the plates were dominant above 10 Hz. It was also shown that the boundary conditions of the two plates, which were rested on elastomeric pads 57, were somewhat different leading to differences in the X, Y, and Z directional acceleration measurements on the two plates.
Tire faults and suspension faults were both simulated in the vehicle, and these faults were detected and located with 90% confidence. A relatively simple fault index including the integral of the discrete Fourier transform was used to perform the diagnostic analysis. Variations in the fault index were observed due to changes in vehicle speed, but these variations could be reduced by estimating the vehicle speed and then utilizing this estimate to adjust the fault index accordingly. Broad frequency ranges were selected to provide more robust and less sensitive diagnostic performance.
The diagnostic capability due to tire faults in the form of reductions in tire air pressure was examined using the cleat and plate system 68. The prescribed healthy front and rear tire pressures were 20 and 22 psi, respectively. To simulate a tire fault in one tire at a time, the air pressure in the tire of interest was reduced to 10 psi. Thirty datasets were then acquired for the faulty tire condition. A comparison of the mean baseline datasets and the datasets for the driver front faulty tire were plotted in
The instrumented diagnostic cleat in some embodiments includes two rubber cleats 50 (or changes in elevation), or speed bumps, that are placed in front of two instrumented plates 56. The two rubber cleats 50 are spaced 0.91 m apart to align the center of the rubber cleats 50 with the center of the HMMWV tires on each side of the vehicle. The rubber cleats are preferably each held in place with four bolts, in some embodiments of the present invention. However, it understood that in yet other embodiments the cleats 50 are not attached to the surface of roadway 22.
Each metal plate 56 is 2.1 m long, 0.91 m wide, and 1.27 cm thick. The plate thickness was chosen to withstand a point load force equivalent to the force exerted on the plate by the HMMWV's tires without any plastic deformation to prevent the possibility of the plates warping or deforming during testing. Also, the plate length is preferably the length that permits one tire at a time to be in contact with the plate. In some embodiments, the plates have a length that is less than about the wheelbase of the vehicle being tested (the wheel base being the distance from the centerline of a front wheel to the centerline of a rear wheel). It is preferred that the front tire be out of contact with the plate 56 before the rear tire comes into contact with plate 56. Further, it is preferred that the plates have a length that is at least as long as the circumference of a tire.
It is preferred that plates 56 be fabricated from a material that is stiff, such as steel or aluminum. However, the present invention is not constrained to metallic plates, and includes plates fabricated from any material. Preferably, the plate material, and further the geometry of the plate (including length, width, and thickness) be selected such that the primary vibratory responses of the plate are sufficiently high so as to not interfere with the collection of the plate's forced response from contact with the tires.
The metal plates are preferably supported on rubber elastomeric pads 57 to isolate the plates from extraneous ground vibrations. Each resilient pad 57 is 2.2 m long, 1.0 m wide, and 5.0 cm thick. The metal plates are preferably not restrained or attached to the pads in any way. As shown in
The metal plates 56 are instrumented with four (1000 mV/g) tri-axial accelerometers 60. Two accelerometers are preferably installed on each plate along the inner edge of the plate closest to the center of the vehicle. The two accelerometers 60 are installed 0.76 m and 1.5 m away from the rubber FIG. on each plate. The accelerometers 60 are mounted with super glue and given a minimum of two hours to cure before testing. The accelerometers 60 are numbered as shown in
Although reference has been made to the use of accelerometers 60 for the generation of signals corresponding to vibratory motion of the plates in terms of acceleration, it is understood that yet other embodiments of the present invention are not so constrained. Yet other embodiments contemplate the use of motion sensors 60 that provide output signals corresponding to any type of movement of the plate, including displacement, velocity, acceleration, or further time derivatives. Further, although reference has been made to the use of tri-axial accelerometers, it is understood that the motion sensors 50 can be single or dual dimensional in terms of response, and further that each plate can have a mixture of motion sensors, some which are uni-axial, others of which are tri-axial, some of which measure acceleration, and others of which measure displacement, as examples.
The measured data is segmented into two parts corresponding to the response of the plates while the front tires are in contact with the plates and while the rear tires are in contact with the plates. If a vehicle travels at 5 mph, one of the tires is in contact with the instrumented plates for 0.97 sec resulting in a 1.03 Hz frequency resolution in the spectra. In various embodiments of the present invention the software of computer 70 includes an algorithm for separating front wheel response from rear wheel response. In some embodiments, this algorithm includes calculation of vehicle velocity based on the timing of impacts to a plate 56, whereas in other embodiments the algorithm looks at the nature of the time response, with the response of the plates to the front tires diminishing as tires roll toward the front edge of a plate and the response of the plate to the rear tire being substantially larger in magnitude as the rear tire leaves the cleat 50 and first rolls onto plate 56.
The vehicle is instrumented with sensors in order to better understand how the diagnostic cleat is responding, with and without faults present in the vehicle. The onboard sensors are not required to implement the instrumented diagnostic cleat for fault detection. The HMMWV is instrumented with two (100 mV/g tri-axial) accelerometers and ten (100 mV/g) single axis accelerometers. The two tri-axial accelerometers are installed on the upper and lower control arms as close as possible to the ball joint for the passenger front suspension. Likewise, two single axis accelerometers are installed on the upper and lower control arms as close as possible to the ball joint for each of the driver front, driver rear, and passenger rear suspensions as shown in
Modal impact testing was conducted on the instrumented metal plates in order to determine how the free response of the plates influences the measurements when the vehicle is driven over the diagnostic cleat. It was believed that resonance frequencies of the plates would include responses when the vehicle was driven over the cleat. The response of each plate is measured using four tri-axial accelerometers as previously mentioned. A mini sledge impact hammer instrumented with a load cell is used to apply an impulsive force at various locations on the plate as illustrated in
The plates are impacted while the vehicle 20 front tires are resting on the plates in three locations. The three tire positions correspond to the center of the HMMWV's front tires resting on the metal plates at 0, 0.89, and 1.78 m from the rubber cleats. In yet other embodiments, the baseline response at the plate is measured with a static load applied on the plate, and including those embodiments in which the static weight is representative of the weight supported by a wheel of a vehicle. Further, it is anticipated that this response spectrum of the loaded plate 56 can be a function of weight, such that the correction applied to the responses made during testing of a vehicle include modifying the correction for the weight of the vehicle. In some embodiments, the plate modal response spectrum (which can subsequently be used to correct motion data during vehicle testing) is prepared with a simple application of load to the surface of the plate, whereas in other embodiments the load is transferred into the plate through a resilient interface, such as an elastomeric interface.
The auto power spectra for the average force at each impact location exhibited a 1 dB rolloff by approximately 400 Hz. Frequency response functions between the input forces and the acceleration responses are estimated using the H1 estimator. These functions are valid between 1 and 400 Hz due to the rolloff of the modal impact forces. The analysis that is described here is limited to the frequency range from 1 to 100 Hz. However, it is understood that other embodiments of the present invention are not so constrained, and include frequency response functions not limited to any particular frequency range.
The complex mode indicator function (CMIF) is used here to identify the modes of vibration that comprise the measured frequency response functions of the driver and passenger side plates of the diagnostic cleat. The CMIF calculates the singular value decomposition of the frequency response function matrix as expressed below:
[H(ω)]=[U(ω)][Σ(ω)][V(ω)]H (1)
where [H(ω)] is the frequency response function matrix with No rows and Ni columns (with No equal to the number of response measurements, 3, and Ni is the number of modal impacts, 12, on each plate), [U(ω)] is the left singular vector matrix, [Σ(ω)] is the singular value matrix, and [V(ω)] is the right singular vector matrix. The CMIF used in some embodiments is a plot of the singular values of the imaginary part of the frequency response function matrix versus frequency, and the peaks in these singular value curves denote the resonance frequencies.
These plots reveal that there are peaks in vibration of the driver's side plate at 19, 34, 57, and 92 Hz and peaks in vibration of the passenger's side plate at 23, 38, 57 (heavily damped), and 68 Hz. The passenger side plate has a much larger CMIF magnitude, which is also reflected in the measurements of the plate forced response. The differences in the two plate modes of vibration are due to the boundary conditions provided by the roadway and the elastomeric pads.
The vehicle 20 is driven over the diagnostic cleat 50 and plate 56 multiple times at 5 mph in order to acquire measurements for analyzing the response of the vehicle. The speed of the vehicle is controlled by the driver with the speedometer gauge. The tire pressure is set to a recommended tire pressure, which is inscribed on each tire. The recommended tire pressures are 20 psi for the front wheels and 22 psi for the rear wheels. The data collected in this experiment was collected within one day with one driver to minimize the variability due to the weather and the driver. The data acquisition system is set to collect 4.5 sec of data from the vehicle sensors at a sample rate of 2048 samples per second. Although what has been shown and described is a method in which the vehicle is driven multiple times over the cleat on one day with a particular driver, it is understood that the invention is not so constrained, and includes those embodiments in which data is recorded and analyzed from a single traversing of the vehicle over the cleat and plate, and further without limitation as to when the data is collected or who is driving the vehicle.
Method 100 further includes exciting 120 of the vehicle with an impulse-type input. As previously described, in some embodiments this input can be provided by electrical or hydraulic actuation means. However, preferably, the excitation 120 of the vehicle 20 is provided by driving the vehicle over a change in roadway elevation, which can be a dip or a bump. Preferably, the change in elevation is a cleat 50. Further, in yet other embodiments, the locating 110 of the vehicle is performed by driving the vehicle onto the plate after having driven the vehicle over the cleat 50.
Method 100 further includes recording 130 the motion of the plate. The motion can be measured with any means for measuring motion, including as examples, a laser velocimeter, piezoelectric accelerometers, or displacement transducers. Preferably, the motion is measured using an accelerometer 60 coupled to plate 56. The motion of the plate is recorded while the vehicle is still responding to the input provided by excitation 120.
The recorded motion data includes correction 140 the motion data for predetermined vehicle responses. This correction can be in any format, including the time domain or frequency domain. Further, in some embodiments the data is corrected in the order domain, which is preferably established by the length of time for a tire to traverse plate 56 from one end to the other end. Preferably, the plate has a length that is less than the wheelbase of the vehicle, so that only one tire is on a plate at any particular moment. Algorithms for correcting the motion data are described further in method 200 and shown in
The recorded motion data includes correction 150 the motion data for predetermined plate responses. This correction can be in any format, including the time domain or frequency domain. Further, in some embodiments the data is corrected in the order domain, which is established by the length of time for a tire to traverse plate 56 from one end to the other end. Preferably, the plate has a length that is less than the wheel base of the plate, so that only one tire is on a plate at any particular moment. The algorithm for correcting the motion data is described further in method 300 and shown in
Method 100 further includes preparing 160 a fault index from the corrected data. This fault index can be prepared in any of the time, frequency, or order domains. Further, the fault index can be of any type.
Method 100 further includes identifying 170 a condition of the vehicle from the fault index. The condition is preferably identified by comparing the fault index to a historical and/or predetermined database. As one example, a predetermined database can include a range of fault indices that correspond to a known fault that was induced in a vehicle during a test. As another example, a historic database can include responses from this same vehicle, or from vehicles of similar type, measured over a period of time.
Method 200 includes the act 210 of providing a vehicle in a known condition. This known condition can be a vehicle with no known faults, or can be a vehicle with one or more faults either purposefully introduced into the vehicle or arising from use or improper manufacture of the vehicle. In some embodiments, the vehicle is a new vehicle that is in the final stages of assembly by an OEM. In yet other embodiments, the vehicle is in the final stages of preparation at a repair depot or garage.
Method 200 further includes the act 200 of exciting the vehicle with an impulse-type input. As noted in method 100, this input can be provided by actuating means, or provided by driving the vehicle over an elevational change. The input used with act 220 is similar to the input provided by act 120 of method 100.
Method 200 further includes the act 230 of recording the vehicle response data. Subsequently, method 200 preferably includes the act 240 of preparing a correction algorithm that is applied to the motion data in method 100 to minimize the influence of vehicle responses, including rigid body vehicle responses such as rolling, pitching, and jounce and rebound (vertical translation). In one particular vehicle as shown in
This correction algorithm can be of any type that minimizes the effect of standard or routine vehicle responses from the plate motion data, yet does not eliminate or significantly minimize the plate response data indicative of vehicle faults. In some embodiments, the algorithm is a simple low frequency cutoff. In yet other embodiments the algorithm includes a high pass filter intended to roll-off the lower, rigid body modes of vehicle vibration. In yet other embodiments the algorithm includes truncation of the plate motion data in the time domain. Still further examples are provided herein. In still other embodiments, the motion data is presented to the system user, and the user is instructed as to what time periods or frequency bans to ignore or de-emphasize.
Method 300 includes the act 310 of providing a plate in a known condition. This known condition can be a plate with no known faults, or can be a plate with one or more faults either purposeful introduced into the plate or arising from use or improper manufacture of the plate. In some embodiments, the plate is a new plate that is in the final stages of assembly by an OEM. In yet other embodiments, the plate is in the final stages of preparation at a repair depot or garage.
Method 300 further includes the act 320 of exciting the plate with an impulse-type input, such as a hammer as used in modal testing. It is further preferred that the plate be supported on a resilient isolator 57. Still further, in some embodiments the plate modal data is acquired when the plate is loaded statically, as by a tire of a vehicle to be tested.
Method 300 further includes the act 330 of recording the plate response data. Subsequently, method 300 preferably includes the act 340 of preparing a correction algorithm that is applied to the motion data in method 100 to minimize the influence of plate modal responses from the plate response taken during vehicle testing. This correction algorithm can be of any type that minimizes the effect of standard or routine plate responses from the plate motion data, yet does not eliminate or significantly minimize the plate response data indicative of vehicle faults. In some embodiments, the algorithm is a simple high frequency cutoff. In yet other embodiments the algorithm includes a low pass filter intended to roll-off the higher modes of plate vibration. In yet other embodiments the algorithm includes truncation of the plate motion data in the time domain. Still further examples are provided herein. In still other embodiments, the motion data is presented to the system user, and the user is instructed as to what time periods or frequency bans to ignore or de-emphasize.
The average measured responses from the accelerometers that are mounted on the lower control arms are plotted in
In order to avoid minimize the vehicle response spectra that are created by the successive excitations by the front and rear wheels, the measured time response in
The modes of vibration of the vehicle are estimated by analyzing the operational deflection shapes. An operational deflection shape is defined as the dynamic deflection of the vehicle under the front wheel excitation at a particular frequency. The operational deflection shapes are a weighted sum of the modes of vibration of the vehicle. Unlike the modal deflection shapes, the operational deflection shapes are dependent on the excitations; for example, if the data for a rear wheel crossing is instead analyzed, the shapes are somewhat different than those described here. Unlike a frequency response function, which is calculated by measuring both the excitation and response of the vehicle as in the modal impact tests, the operational data used in some embodiments includes the vehicle response measurements.
The operational deflection shapes of the vehicle are determined with the operational deflection shape frequency response function, or ODS FRF, which is calculated using response data acquired during the front wheel crossing of the cleat,
where GXX(ω) is the auto power spectrum of the measured response variable and GXY(ω) is the cross power spectra between the measured response and the reference measurement. The reference is one of the measured responses. The ODS FRF determines the magnitude and phase of the deflection at each measurement location and the ODS FRFs have peaks at natural frequencies of the system. A vehicle measurement grid is constructed that graphically represents the location of each accelerometer that is used. Then the product of the magnitude and phase of the ODS FRF at a specific frequency for each accelerometer is represented on this grid to visualize the operational deflection shapes at that frequency.
Using this technique, the modes of the vehicle are identified by studying the animations of the operational deflection shapes. The ODS results at the peaks indicated in
The instrumented diagnostic cleat plates exhibit resonance frequencies above 10 Hz and the vehicle chassis and suspension exhibit resonance frequencies below 10 Hz. Experimental data can be used to understand the accelerometer responses of the two plates in the diagnostic cleat. The HMMWV (without faults) is driven over the diagnostic cleat 50 times at 5 mph. 2.28 seconds of data are sampled at 3200 samples per second, and the measured response is divided into two equally sized segments of length 1.14 seconds, which span the response due to the front wheel excitation and the response to the rear wheel excitation, respectively. The scaled DFT of the measured responses due to a front wheel excitation and rear wheel excitation are then calculated for each accelerometer in each direction.
The magnitude of the average scaled DFT for the front and rear wheel excitations for accelerometer 4 on the passenger side plate in the X, Y, and Z directions are plotted in
Both plates exhibit peaks in the X and Y directions at approximately 55 Hz for the front wheel excitation case. For the rear wheel excitation case, the responses of both plates display peaks at 57 Hz, which is a natural frequency of the driver's side plate. In the Y direction, both plates also display peaks near 20 Hz, which is also close to a natural frequency of both plates. The measured response of the driver's side and passenger's side plates may be similar in the X direction because X corresponds to the forward direction of the vehicle. Based on these results for the X and Y directions, it is also evident that the vehicle modes of vibration dominate the response of the plates at low frequency below 10 Hz, which can be seen in the plots in
One difference in the data for the driver and passenger side plates occurs in the Z direction (vertical). The passenger side plate response was larger than the driver side plate response. This difference in the amplitude of response is attributed to differences in the boundary condition on the elastomeric pad on which the plate rests. The responses corresponding to the front and rear wheel excitations are similar for the Z direction on both plates. In some embodiments, the corrections applied to the data measured from sensors 56 can be different as applied to one side of the vehicle versus the other side of the vehicle. In some cases, the corrections applied can be different based on the plate response data (such as the data from sensor 56 recorded at higher frequencies). Further, the fault index applied to the right side or left side can differ from one another based on the measured responses of the plates.
Tire faults were simulated in the vehicle by decreasing the air pressure. This method of simulating faults in the tire leads to changes in the tire stiffness and damping by changing the degree to which the air and sidewall contribute to the forces supplied by the tire patch. The HMMWV that is used for testing simulated tire faults as four tires whose maximum tire pressure is 30 psi. These tires are also runflats that can operated at 0 psi because a belt within the tire prevents the vehicle from riding on the wheel rim when the tire pressure goes to zero. The pressure within the tires is set to a value between 0 psi and 30 psi. The tire pressure fluctuates somewhat around the set pressure during testing by ±1 psi due to temperature changes within the tire and the ambient environment. The assumed nominal tire pressures are 20 psi for the front wheels and 22 psi for the rear wheels.
To conduct the first set of tire fault tests, the tire pressure is set to 5, 10, 15, 20, 25, and 30 psi within the passenger front tire. Pressures above 20 psi correspond to overinflated tires while pressures low the 20 psi correspond to underinflated tires. The tire pressures in the remaining tires are set to the nominal tire pressures. The power spectral density is determined for accelerometer 4 in the Z-direction while the passenger front tire is in contact with the diagnostic cleat. The resulting average power spectral densities are plotted in
As the tire pressure is decreased in the passenger front tire, the magnitude of the response of the passenger side plate decreases between 10 and 20 Hz because the effectiveness of the tire decreases with tire pressure. The integral of the magnitude of the scaled DFT is estimated as follows to form a fault index (FI),
where Δf is the frequency resolution, a is the lower frequency limit, and b is the upper frequency limit. The fault index from 10 to 20 Hz is determined for each run and plotted in
Tests are also conducted with tire faults, one at a time, in the driver's front, passenger's front, driver's rear, and passenger's rear tires. For each fault case, the tire containing the fault is filled to 10 psi and the remaining three tires are set to their nominal pressures. The fault index is then calculated from 20 to 40 Hz for each run using accelerometer 4 in the Z-direction for the passenger front and rear tire excitation measurements. The fault index results are plotted in
The fault index calculated from the magnitude of the scaled DFT from 20 to 80 Hz for each run is determined for accelerometer 1 in the Y-direction for the driver front and rear tire excitation measurements and the results are plotted in
Front suspension faults are observed in the cleat response when either of the front vehicle tires is in contact with the diagnostic cleat, particularly when the front tire opposite to the side containing the suspension fault is simulated is in contact with the diagnostic cleat. For example, a driver front suspension fault has the greatest effect on the measured cleat response when the passenger front tire is in contact with the diagnostic cleat. These responses in the opposite side of the vehicle from which the suspension faults are simulated occur because a disabled damper causes the vehicle to roll excessively.
The fault index in some embodiments is calculated to identify the changes in the cleat response due to suspension faults. The fault index calculated from the magnitude of the scaled DFT from 5 to 35 Hz is determined for accelerometer 1 in the Y direction for each data set for the passenger front and rear tire excitation measurements, and the results are plotted in
Likewise, the fault index calculated from the magnitude of the scaled DFT from 5 to 30 Hz is determined for accelerometer 4 in the Y direction for each data set for the passenger front and rear tire excitation measurements, and the results are plotted in
A fault index according to yet another embodiment was extracted from the measured data for each dataset by calculating the sum of the spectral magnitudes for sensors 3 and 4 in the vertical direction across all three of these frequency ranges after the front wheels traversed the speed bump. The resulting fault indices were plotted in
Although what has been shown and described are specific examples of various fault indices, it is understood that various other embodiments of the present invention contemplate other types of indices, and still other embodiments do not include calculation of any fault index. As one example, in some embodiments of the present invention the substantially raw accelerometer data, especially displayed on a graphical user interface 78, may provide sufficient information for an operator to identify a fault in the vehicle, or a condition of the vehicle. In still other embodiments, there is a fault index that includes phase angle information.
Some embodiments compensate for the variability due to temperature and humidity. Some embodiments compensate for the angle at which the axles of the vehicle cross the cleat. Some embodiments compensate for the vehicle speed. The speed is difficult for the driver to control and is subject to error given the approximate nature of the speedometer. All of the datasets have been taken using a nominal vehicle speed of 5 mph; however, there are small variations around this speed.
In one embodiment, the vehicle speed is estimated by calculating the time that passes between the instant when the front wheels first contact the instrumented plate and the instant when the rear wheels leave the plate. This elapsed time is used in some embodiments together with the vehicle wheelbase length and cleat length to estimate the average speed of the vehicle throughout the measurement process. In some embodiments, vehicle information such as wheelbase length and track width are inputs provided by an operator, especially through graphical user interface 78. Yet other embodiments include an additional sensor (such as a motion detector coupled to a cleat 50) to provide vehicle velocity data.
The wheels are preferably offset from the centerline 56.1 of plate 56 to excite a larger number of modes of vibration in the plate. Many of these mode shapes over a wide frequency range have symmetric shapes that have a nodes (points of no deflection when the plate is excited) along the geometric centerline 56.1 of each plate. If the wheels are driven down the center of the plate, these symmetric modes will not be excited which will reduce the response of the two plates. To avoid this, the plates are positioned so the wheels are off center and excite a larger number of modes of vibration in each plate. There can be two speed bumps 50R and 50L, sitting side by side as shown in
Arranged forward of cleat 120 in one embodiment are right and left members 156, each of which is isolated from the roadway by a corresponding mat 157 (shown in crosshatch). Preferably, mats 157 are sufficiently resilient to reduce the transmissibility of roadway vibratory motion into the plates 156.
Each plate 156 includes at least one sensor 160 placed proximate to an edge of the corresponding plate 156. In the embodiment shown in
Further, it is preferred that each of the plates 156 are aligned relative to the left and right tracks of the vehicle. System 168 (which for the sake of clarity does not show the computer or display or cabling) includes a first plate 156L which is registered toward the right such that the left wheel path 120.2L extends generally to the left of centerline 156.1L. Therefore, the left tire of the vehicle is spaced apart from the center of the corresponding plate.
Arranged forward of cleat 220 in one embodiment are right and left members 256, both of which is isolated from the roadway by a single mat 257 (shown in crosshatch). Preferably, mat 257 is sufficiently resilient to reduce the transmissibility of roadway vibratory motion into the plates 256.
Each plate 256 includes at least one sensor 260 placed proximate to an edge of the corresponding plate 256. In the embodiment shown in
Further, it is preferred that each of the plates 256 are aligned relative to the left and right tracks of the vehicle. System 268 (which for the sake of clarity does not show the computer or display or cabling) includes a first plate 256L which is registered toward the left such that the left wheel path 220.2L extends generally to the right of centerline 256.1L. Therefore, the left tire of the vehicle is spaced apart from the center of the corresponding plate.
trends observed in the data. The subscript “s” refers to the sprung mass of the vehicle, the subscript “u” refers to the unsprung mass of the vehicle, and the subscript “p” refers to properties of the plate.
A fault index according to another embodiment can be calculated based on a cross correlation function between different axes of measurement and/or different wheels, such, as one example, as the Z and Y measurements on opposite wheels:
Fault index=RMS[Rzy(τ)]
Various aspects of different embodiments of the present invention are expressed in paragraphs X1, X2, and X3, as follows:
X1. One aspect of the present invention pertains to an apparatus for analyzing a vehicle. The method preferably includes a first portable segment of driving surface, said first segment having a top surface adapted and configured to be driven on by a wheeled vehicle, said first segment having a cross-sectional shape adapted and configured to locally elevate a vehicle driven over said first segment. The apparatus preferably includes a second separable segment of driving surface, said second segment having a top surface adapted and configured to be driven on by a wheeled vehicle, said second segment being located proximate to said first segment. The apparatus preferably includes a motion sensor providing a signal corresponding to motion of said second segment. The apparatus preferably includes a computer receiving the signal and having software that analyzes the signal.
X2. Another aspect of the present invention pertains to a method for analyzing a vehicle on a roadway. The method preferably includes providing a segment of driving surface located proximate to a local elevational change in the roadway, and a sensor providing a signal corresponding to the spatial response of the segment. The method preferably includes driving the vehicle first over the elevational change. The method preferably includes driving the responding vehicle over the segment. The method preferably includes recording the signal during said driving over the segment. The method preferably includes correcting the recorded signal for the response of the vehicle.
X3. Another aspect of the present invention pertains to an apparatus for analyzing a wheeled vehicle having a wheelbase and a front track. The apparatus preferably includes a first member adapted and configured to be driven on by the vehicle, said first panel having a width less than the front track and a length less than the wheelbase. The apparatus preferably includes a second panel having a substantially flat top surface adapted and configured to be driven on by the vehicle. The apparatus preferably includes a first motion sensor attached to said first panel and providing a signal corresponding to motion of said first panel. The method preferably includes a second motion sensor attached to said second panel and providing a signal corresponding to motion of said second panel.
Yet other embodiments pertain to any of the previous statements X1, X2, or X3, which are combined with one or more of the following other aspects:
Wherein the motion sensor is attached to said second segment.
Wherein the motion sensor is one of a displacement sensor, a velocity sensor, or an acceleration sensor.
Wherein said second segment has a substantially flat top surface.
Wherein said second segment is not attached to the roadway.
Which further comprises means for isolating said second segment from the roadway.
Wherein said second segment is supported from the roadway by resilient material.
Wherein the resilient material is a layer of an elastomeric material
Wherein the resilient material comprises a plurality of spaced apart elastomeric members.
Which further comprises calculating a velocity of the vehicle driving over the segment from the signal, and said correcting based on the velocity over the segment. which further comprises determining a condition of the vehicle from the corrected signal.
Wherein the condition is a tire with low air pressure or a worn shock absorber. which further comprises analyzing the corrected signal and recommending maintenance to the vehicle from said analyzing.
Which further comprises preparing a fault index from the corrected signal.
Wherein said preparing is from the signal in the time domain or frequency domain.
Wherein the response is pitching or rolling of the vehicle.
Wherein the vehicle response is a fundamental mode of rigid body motion.
Wherein the response of the vehicle is below a frequency, and said correcting is by removing frequency content of the signal below the frequency.
Wherein the segment responds in a vibratory mode above a frequency, and said correcting is by removing frequency content of the signal above the frequency.
Wherein said correcting is with a bandpass filter.
Which further comprises not attaching the segment to the roadway.
Which further comprises supporting the segment on the road with a resilient material.
Which further comprises isolating the segment from responses of the roadway.
Wherein the elevational change is a rise in the level of the roadway and the portable segment is substantially flat.
Wherein the elevational change is one of an asphalt or concrete speed bump. wherein the elevational change is a reduction in the level of the roadway.
Wherein the elevational change is a second portable segment of roadway.
Wherein said first panel and said second panel are placed side by side on a roadway with a gap between the interior edges of said first and second panels.
Wherein the length of said first panel is greater than the circumference of a tire of the vehicle.
Which further comprises a third motion sensor attached to said first panel and providing a signal corresponding to motion of said first panel, said third sensor being spaced apart from said first sensor along the length of said first panel.
A fourth motion sensor attached to said second panel and providing a signal corresponding to motion of said second panel;
Wherein the first panel is fabricated from one of aluminum or steel.
Which further comprises a first isolating member adapted and configured to be placed between said first panel and a roadway.
Wherein said isolating member is a resilient pad of about the same length and width as said first panel.
Wherein said first panel has an edge, and said first sensor is placed proximate to the edge of said first panel, and said second panel has an edge, and said second sensor is placed proximate to the edge of said second panel.
Wherein the fault index is a cross correlation of response data along different axes of measurement.
Wherein the fault index is a cross correlation of response data of different wheels.
While the inventions have been illustrated and described in detail in the drawings and foregoing description, the same is to be considered as illustrative and not restrictive in character, it being understood that only certain embodiments have been shown and described and that all changes and modifications that come within the spirit of the invention are desired to be protected.
This application claims the benefit of priority to U.S. Provisional Patent Application Ser. No. 61/454,800, filed Mar. 21, 2011 and U.S. Provisional Patent Application Ser. No. 61/602,407, filed Feb. 23, 2012, both of which are incorporated herein by reference.
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/US2012/029954 | 3/21/2012 | WO | 00 | 9/18/2013 |
| Number | Date | Country | |
|---|---|---|---|
| 61454800 | Mar 2011 | US | |
| 61602407 | Feb 2012 | US |
