METHOD FOR DIAGNOSING A COMPONENT OF A VEHICLE

• US 20090062977A1
• Filed: 07/30/2008
• Published: 03/05/2009
• Est. Priority Date: 07/30/2007
• Status: Active Grant

• Current Assignee: S.A.T.E. - Systems And Advanced Technologies Engineering Srl
• Original Assignee: S.A.T.E. - Systems And Advanced Technologies Engineering Srl
• Inventors: Brighenti, Attilio
• Granted Patent: US 8,060,273 B2
• Time in Patent Office: Days
• US Class Current: 701/29
• CPC Class Codes: G01M 13/022 Power-transmitting couplings or clutches

Abstract

A method for diagnosing a component of a vehicle, which presents the steps of: identifying a series of classes, each of which is associated to a given mode of operation of the component; identifying at least one control signal of the component; providing a mathematical model of the component designed to classify an evolution in time of the control signal, assigning a respective class to the evolution in time of the control signal itself; detecting a number of evolutions in time of the control signal; classifying each evolution in time of the control signal by means of the mathematical model of the component; and determining a diagnosis of the component on the basis of the classification of the evolutions in time of the control signal.

Claims

1. ) A method for diagnosing a component of a vehicle and comprising the steps of:

identifying a series of classes, each of which is associated to a given mode of operation of the component;

identifying at least one control signal of the component;

providing a mathematical model of the component designed to classify an evolution in time of the control signal, assigning a respective class to the evolution in time of the control signal itself;

detecting a number of evolutions in time of the control signal;

classifying each evolution in time of the control signal by means of the mathematical model of the component; and

determining a diagnosis of the component on the basis of the classification of the evolutions in time of the control signal.

2. ) The diagnostic method according to claim 1, wherein the step of classifying each evolution in time of the control signal by means of the mathematical model of the component envisages the further steps of:

assigning to the evolution in time of the control signal an index of membership in each class; and

identifying the evolution in time of the control signal as belonging to the class with respect to which it has the highest index of membership.

3. ) The diagnostic method according to claim 2, wherein the mathematical model of the component uses for each class a respective neural network, which supplies the index of membership in the class itself of an evolution in time of the control signal.

4. ) The diagnostic method according to claim 2 and comprising the further step of identifying the evolution in time of the control signal as not classified if it is not possible to identify a class with respect to which the evolution in time of the control signal has the highest index of membership.

5. ) The diagnostic method according to claim 4, wherein declared as not identifiable is a class with respect to which the evolution in time of the control signal presents the highest index of membership if the difference between the two highest indices of membership is not significant.

6. ) The diagnostic method according to claim 1 and comprising the further steps of:

obtaining a set of samples, each of which is made up of an evolution in time of the control signal obtained experimentally and a respective class of membership attributed with another method dictated by experience; and

using the set of samples for defining the mathematical model of the component.

7. ) The diagnostic method according to claim 6 and comprising the further steps of:

determining by means of the mathematical model of the component an index of validity of each evolution in time of the control signal, said index of validity indicating the degree of membership of the evolution in time of the control signal to the set of samples; and

assessing the reliability of the classification of each evolution in time of the control signal as a function of the respective index of validity.

8. ) The diagnostic method according to claim 7, wherein the mathematical model of the component uses a probabilistic algorithm for determining the index of validity.

9. ) The diagnostic method according to claim 1, wherein:

the component is a servo mechanical gear change (12) comprising at least one engagement actuator (16) that determines engagement of the gears; and

the control signal of the component is the position of the actuator (16) for engagement of the gear.

10. ) The diagnostic method according to claim 9, wherein at least the following three classes are envisaged:

normal engagement, false meshing, i.e, engagement in two goes, and engagement with clashing of the gears.

11. ) The diagnostic method according to claim 9, wherein at least the following four classes are envisaged:

engagement with synchronization pause, regular engagement, false meshing, i.e, engagement in two goes, and engagement with clashing of the gears.

12. ) The diagnostic method according to claim 9, wherein the mathematical model of the gear change (12) envisages at input the evolution in time of the position of the engagement actuator (16), the velocity of rotation of a crankshaft (6), and the gear engaged and/or to be engaged.

13. ) The diagnostic method according to claim 1, wherein the step of determining a diagnosis of the component on the basis of the classification of the evolutions in time of the control signal envisages the further steps of:

considering a set of temporally consecutive classifications of evolutions in time of the control signal;

determining the frequency percentage of each set of classifications; and

determining a diagnosis of the component on the basis of the frequency percentages.

14. ) The diagnostic method according to claim 13, wherein the diagnosis of the component is determined on the basis of the values of the frequency percentages and/or on the basis of the trends of the frequency percentages.

15. ) The diagnostic method according to claim 13, wherein a malfunctioning of the component is diagnosed if at least one class associated to a faulty behaviour has an excessively high frequency and/or has a tendency to increase progressively over time.

16. ) A method for diagnosing a component of a vehicle and comprising the steps of:

identifying a number of input signals of the component and a number of output signals of the component;

providing a mathematical model of the component in normal operation, which, as a function of the values of the input signals, is able to supply the values of the output signals;

measuring the actual values of the input signals;

measuring the actual values of the output signals;

determining the estimated values of the output signals using the mathematical model of the component and the actual values of the input signals;

comparing the estimated values of the output signals with the respective actual values of the output signals; and

determining a diagnosis of the component on the basis of the comparison between the estimated values of the output signals and the actual values of the output signals.

17. ) The diagnostic method according to claim 16, wherein the diagnosis of the component is determined on the basis of the deviation between the estimated values of the output signals and the actual values of the output signals and/or on the basis of the trend of the deviation between the estimated values of the output signals and the actual values of the output signals.

18. ) The diagnostic method according to claim 16, wherein one and the same signal can be simultaneously both an input signal and an output signal.

19. ) The diagnostic method according to claim 16, wherein the mathematical model of the component uses a neural network for simulating the output signals as a function of the input signals.

20. ) The diagnostic method according to claim 16 and comprising the further steps of:

obtaining a set of samples, each of which is made up of values measured experimentally of the input signals and of the output signals; and

using the set of samples for defining the mathematical model of the component.

21. ) The diagnostic method according to claim 20 and comprising the further steps of:

determining by means of the mathematical model of the component an index of validity of the estimated values of the output signals, said index of validity indicating the degree of membership of the actual values of the output signals to the set of samples; and

assessing the reliability of the diagnosis of the component as a function of the index of validity.

22. ) The diagnostic method according to claim 21, wherein the mathematical model of the component uses a probabilistic algorithm for determining the index of validity.

23. ) The diagnostic method according to claim 16, wherein the component is an alternator (24), which receives motion from an internal-combustion engine (5), generates the electrical energy used by an electrical load of the vehicle, and is electrically connected to a battery (25).

24. ) The diagnostic method according to claim 23, wherein the input signals comprise the velocity of rotation of a rotor of the alternator (24), the voltage of the battery (25), and the current supplied by the alternator (24).

25. ) The diagnostic method according to claim 24, wherein the input signals comprise the temperature of the alternator (24).

26. ) The diagnostic method according to claim 24, wherein the input signals comprise the temperature of an engine compartment in which the internal-combustion engine (5) is housed.

27. ) The diagnostic method according to claim 24, wherein the input signals comprise the temperature of a liquid for cooling the internal-combustion engine (5).

28. ) The diagnostic method according to claim 23, wherein the output signals comprise only the current supplied by the alternator (24).

29. ) The diagnostic method according to claim 16, wherein the component is a hydraulic circuit (19) comprising a tank (20) containing a driving fluid at ambient pressure, a pump (22) that draws from the tank (20), a hydraulic accumulator (21), which is supplied by the pump (22) and contains pressurized driving fluid, a number of hydraulic actuators (16, 17, 18) provided with chambers that can be filled with the pressurized fluid, and a number of solenoid valves (23) designed to connect the chambers of the hydraulic actuators (16, 17, 18) with the tank (20) or with the hydraulic accumulator (21).

30. ) The diagnostic method according to claim 29, wherein the hydraulic actuators drive a servo transmission (7) comprising a servo clutch (15) and a servo mechanical gear change (12).

31. ) The diagnostic method according to claim 29, wherein the input signals comprise the pressure (P_a) of the driving fluid within the hydraulic accumulator (21), the logic signal of activation of the pump (22), the logic signal of activation of the solenoid valves (23) and the temperature of the driving fluid.

32. ) The diagnostic method according to claim 29, wherein the output signals comprise the pressure (P_a) of the driving fluid within the hydraulic accumulator (21).

33. ) The diagnostic method according to claim 32, wherein the comparison between the estimated value of the pressure (P_a) of the driving fluid within the hydraulic accumulator (21) with the respective actual value of the pressure (P_a) of the driving fluid within the hydraulic accumulator (21) is carried out only at the end of a step of discharge of the hydraulic accumulator (21) and at the end of a step of charge of the hydraulic accumulator (21).

34. ) The diagnostic method according to claim 33 and comprising the further steps of:

calculating, at the end of a step of discharge of the hydraulic accumulator (21), a difference (Δ

Ps) of discharging pressure equal to the difference between the estimated value of the pressure (P_a) of the driving fluid within the hydraulic accumulator (21) and the respective actual value of the pressure (P_a) of the driving fluid within the hydraulic accumulator (21);

calculating, at the end of a step of charging of the hydraulic accumulator (21), a difference (Δ

Pc) of charging pressure equal to the difference between the estimated value of the pressure (P_a) of the driving fluid within the hydraulic accumulator (21) and the respective actual value of the pressure (P_a) of the driving fluid within the hydraulic accumulator (21); and

determining a diagnosis of the component on the basis of the difference (Δ

Ps) of discharging pressure and of the difference (Δ

Pc) of charging pressure.

35. ) The diagnostic method according to claim 29, wherein the output signals comprise a volumetric efficiency (η

_v) calculated for a number N of complete successive cycles and estimated by applying the following equation;

${\mathrm{\eta <br><br>}}_V=\frac{\underset{i}{\sum <br><br>}<br><br>\underset{j}{\sum <br><br>}<br><br>V_{\mathrm{ij}}+Q_{\mathrm{leakSTD}}·<br><br>T_{\mathrm{DUTIES}}}{\sum <br><br>T_{\mathrm{ONi}}·<br><br>Q_{\mathrm{PSTD}}}$

where

η

_v is the estimated volumetric efficiency;

Σ

_iΣ

_jΣ

_ij is the summation of all the “

j”

volumes of driving fluid consumed by the solenoid valves (23) during the N complete cycles “

i”

;

Q_leakSTD is the standard leakage flowrate of the hydraulic circuit (19);

T_DUTIES is the overall duration of the N complete cycles;

Σ

TON_i is the summation of the times of operation of the pump (22) during the N complete cycles;

Q_PSTD is the standard flowrate of the pump (22).

36. ) The diagnostic method according to claim 35, wherein the volumetric efficiency (η

_v) has an actual value that is equal to unity, and the estimated value of the volumetric efficiency (η

_v) is compared with unity for determining the diagnosis of the component.

37. ) A method for diagnosing a clutch of a vehicle, which mechanically connects a crankshaft (6) of an internal-combustion engine (5) to a primary shaft (13) of a gear change (12);

the diagnostic method comprising the steps of;

detecting the velocity (ω

_m) of rotation of the crankshaft (6);

detecting the velocity (ω

₁) of rotation of the primary shaft (13) of the gear change (12);

determining the difference (Δ

ω

) of velocity between the velocity (ω

_m) of rotation of the crankshaft (6) and the velocity (ω

₁) of rotation of the primary shaft (13) of the gear change (12) when the velocity (ω

_m) of rotation of the crankshaft (6) is higher than a first threshold value and the torque supplied by the internal-combustion engine (5) is higher than a second threshold value; and

determining a diagnosis of the clutch (15) on the basis of the difference (Δ

ω

) of velocity.

38. ) The diagnostic method according to claim 37, wherein the diagnosis of the clutch (15) is determined on the basis of the value of the difference (Δ

ω

) of velocity and/or on the basis of the trend of the differences (Δ

ω

) of velocity.

39. ) The diagnostic method according to claim 37, wherein the diagnosis of the clutch (15) is determined on the basis of the mean value of the difference (Δ

ω

) of velocity and/or on the basis of the trend of the average of the differences (Δ

ω

) of velocity.