WO2004110837A2

WO2004110837A2 - Thermal model for an electronic component

Info

Publication number: WO2004110837A2
Application number: PCT/US2004/017158
Authority: WO
Inventors: Wei Zhan
Original assignee: Kelsey-Hayes Company
Priority date: 2003-05-28
Filing date: 2004-05-28
Publication date: 2004-12-23
Also published as: WO2004110837B1; WO2004110837A3

Abstract

A method of protecting an electronic component, e.g. a FET controlling a pump motor, in a brake control system that uses a nonlinear relationship to determine an estimated operating temperature exceeding, or being equal to, a predetermined thereshold temperature (Tthres), the component is deactivated.

Description

THERMAL MODEL FOR AN

ELECTRONIC COMPONENT

Background of Invention

[001 ] This invention relates in general to electronically controlled vehicle brake systems and in particular to a thermal model for electronic components in an electronically controlled brake system.

[002] Electronically controlled brake systems are increasingly being included in motor vehicles. Such systems monitor vehicle parameters, such as the vehicle speed and individual wheel speeds, and selectively activate wheel brakes to control the vehicle dynamics. Such systems typically include several modes of operation. One operational mode includes an Anti-lock Brake System (ABS) in which excessive slippage of a wheel is detected during a braking cycle and the pressure being applied to the vehicle brakes is selectively adjusted to avoid a wheel lock-up condition. A second operational mode includes Traction Control (TC) which selectively applies pressure to the wheel brake associated with a slipping driven wheel during vehicle acceleration. Another operational mode includes Vehicle Stability Control (VSC) in which the vehicle wheel brakes are selectively applied to correct a potential spin-out of the vehicle. Vehicle stability control systems typically include accelerometers and yaw sensors to provide more vehicle motion data and also may control the engine in addition to the vehicle brakes.

[003] Referring now to the drawings, there is shown in FIG. 1, a typical electronically controlled brake circuit I of a dual-circuit vehicle brake system 10. A second brake circuit II, not shown, is structurally the same as the brake circuit I shown and functions in the same way. Lines carrying hydraulic brake fluid are shown as solid lines in Fig. 1, while electrical wiring is shown as dashed lines. [004] The vehicle brake system 10 includes a tandem master cylinder 12 with a brake fluid reservoir 14. A branching main brake line 16 leads to a pair of wheel brake cylinders 18 connected to the brake circuit I. A normally open isolation valve 20, with an integrated pressure limiting valve 22, is incorporated into the main brake line 16. Each wheel brake cylinder 18 is preceded by a normally open apply valve 24 which is open in its basic position.

[005] Each wheel brake cylinder 18 also is connected to a normally closed dump valve 26 from which a common return line 28 leads to the main brake line 16 and discharges between the isolation valve 20 and the apply valves 24. Incorporated into the return line 28 is a high pressure hydraulic pump 30, which is preceded by a reservoir 32 and followed by a damper 34. Typically, the hydraulic pump 30 includes a pair of reciprocating pistons driven by an electric motor. Each piston provides pressurized brake fluid to one of the brake circuits, I and IL

[006] Each pair of apply and dump valves 24 and 26 form a brake pressure modulating valve assembly for the corresponding wheel brake cylinder 18. The apply and dump valves 24 and 26 are selectively actuated to modulate the pressure applied to the associated wheel brake cylinder 18, in order to prevent or limit slip of the corresponding vehicle wheel upon braking (ABS), upon startup (TC), or selective actuation of one or more wheel brakes to reduce the vehicle cornering force in order to prevent loss of directional control of the vehicle (VSC). The brake pressure modulation is accomplished with pressurized brake fluid supplied by the high pressure hydraulic pump 30. The isolation valve 20 may be closed during the brake pressure modulation, in order to prevent feedback effects on the master cylinder 12. The apply and dump valves 24 and 26 may be combined into one 3/3-way valve, instead of the separate 2/2-way valves shown. [007] An intake line 36 leads from the master cylinder 12 to a suction side of the high pressure hydraulic pump 30. A normally closed intake valve 38 is disposed in the intake line 36. Through the intake line 36, when the intake valve 38 is open, the hydraulic pump 30 receives brake fluid directly from the master cylinder 12, which is needed for a rapid pressure buildup. A pressure sensor 41 is also connected to the master cylinder 12.

[008] All the valves 20, 24, 26, 38 are solenoid valves and are electrically connected to an Electronic Control Unit (ECU) 39 that encloses electronic components for controlling the brake system 10. Typically, a temperature sensor 40 is included in the ECU for monitoring the ambient temperature therein to provide protection from overheating for the electronic components. For clarity, only the electrical connection from the ECU 39 to the isolation valve 20 is shown in Fig. 1. The high pressure hydraulic pump also is electrically connected to the ECU 39, as are wheel speed sensors (not shown), other vehicle dynamic sensors (not shown) and a pressure sensor 41. A microprocessor (not shown) within the ECU 39 is programmed with a brake control system algorithm. The algorithm monitors the sensor data and is responsive thereto to selectively actuate the hydraulic pump 30 and the solenoid valves 20, 24, 26 and 38 to correct any potential vehicle dynamics problems.

[009] In order to supply an adequate quantity of brake fluid to the hydraulic pump 30 when the master cylinder 12 is not actuated, as during operation in the TC and VSC modes, the vehicle brake system 10 also includes a precharge pump 42. Typically, the precharge pump 42 may be disposed directly upon the brake fluid reservoir 14, and has a suction intake port that receives brake fluid from the brake fluid reservoir 14. Usually, the precharge pump 42 is a self-aspiring recirculating positive displacement pump, such as a gear pump and supplies a lower pressure than the high pressure hydraulic pump 30. As with the high pressure pump 3O₅ the precharge pump 42 is electrically connected to, and controlled by, the ECU 39.

[010] As shown in Fig. 1, the precharge pump 42 pumps brake fluid into the master cylinder 12 and the brake system 10 through a connection to the line labeled I. An orifice 44 restricts flow between the master cylinder 12 and the wheel brakes to allow the precharge pump 42 to build pressure within the brake system 10 when activated. When the precharge pump 42 is in operation, the high pressure hydraulic feed pump 30 can therefore be supplied, when needed, through the intake line 36 and the opened intake valve 38, with a sufficient quantity of brake fluid for a rapid pressure buildup in the wheel brake cylinders 18, even if the master cylinder 12 is not actuated.

[011] The brake system 10 illustrated in Fig. 1 is intended to be exemplary and the specific details of components and their arrangement may vary. Similarly, the location and connection of the precharge pump 42 also can vary from that shown in Fig. 1. However, the low pressure loading of brake circuit by the precharge pump 42, ensures that the wheel brakes of the vehicle are weakly applied before the brake control actually begins. Thus, a pressure build up by the high pressure pump during the initial stage of brake control system activation can become effective immediately, with the result that a very rapid and sensitive response to the brake control system becomes possible.

Summary of Invention

[012] This invention relates to a thermal model for a electronic component in an electronically controlled brake system.

[013] Both the high pressure hydraulic pump and the pre-charge pump, if present, in an electronically controlled vehicle brake system are driven by electric motors. Typically, each of the pump motors are energized by an associated motor Field Effect Transistor (FET) that acts as a switch between the vehicle voltage supply and the motor. The pump motor FETs are usually located in the ECU 39. Also, the ECU 39 may include other FETs that are used to supply power to other electrical components within the brake control system. Often the FETs are switched on and off to provide a pulsed voltage supply to the pump motor or other device to control the average current being supplied to the motor or device. As the FETs are switched on and off, their temperatures increase above the ambient temperature within the ECU 39. If the FETs become too hot, they may not function properly. Also, the FETs will raise the ambient temperature within the ECU and may thereby adversely affect the operation of the other electronic components that are located in the ECU. Therefore, it is desired that the FETs be disabled before they become too hot. Accordingly, in the past, a thermal model has been used to predict the temperature for each FET as a linear function of the time that the FET has been operating. The thermal models provided an estimated temperature for each FET and the microprocessor within the ECU 39 was responsive to the estimated temperature to disable the FET before it becomes excessively hot. However, such a model must be based upon the hottest operating conditions in order to not exceed the thermal constraint of the FET. Thus, prior art models do not take into account the ambient temperature of the surroundings of the FET. Under cold operating conditions, the prior art thermal model may prematurely disable the FET. Accordingly, it would be desirable to provide a better thermal model. Similarly, the temperatures of other components contained within the ECU, such as bipolar transistors, may increase above the ambient ECU temperature as they are operated. Accordingly, it also would be desirable to provide a better thermal model for the other electronic components.

The present invention contemplates a method for protecting an electronic component in a vehicle brake control system from overheating that includes the steps of providing a first increasing nonlinear relationship relating the temperature of the component to at least one operating parameter. The first nonlinear relationship is used while the component is operating to determine an estimated component temperature. The estimated operating temperature is then compared to a threshold temperature and the component is deactivated if the estimated operating temperature exceeds, or is equal to, the threshold temperature. The method further contemplates providing a second decreasing nonlinear relationship relating the temperature of the electronic component to at least one operating parameter and, upon the component ceasing to operate, using the second nonlinear relationship to determine an estimated component temperature. In the preferred embodiment, the electronic component is an electronic switch that provides power to a hydraulic fluid pump; however, the invention also can be practiced to provide thermal protection to any similar device.

[015] Various objects and advantages of this invention will become apparent to those skilled in the art from the following detailed description of the preferred embodiment, when read in light of the accompanying drawings.

Brief Description of Drawings

[016] Fig. 1 is schematic diagram of a brake control system that includes a pre- charge pump.

[017] Fig. 2 is an experimentally determined thermal graph that illustrates the increase of the temperature of a pre-charge pump motor FET during pump operation as a function of time.

[018] Fig. 3 an experimentally determined thermal graph that illustrates the decrease of the temperature of a pre-charge pump motor FET following pump operation as a function of time. [019] Fig. 4 is a graph that illustrates the estimated temperature of a pre-charge pump motor FET in accordance with the present invention.

[020] Fig. 5 illustrates the use of the graph shown in Fig. 2 in estimating a portion of the temperatures shown in Fig. 4.

[021] Fig. 6 illustrates the use of the graph shown in Fig. 3 in estimating another portion of the temperatures shown in Fig. 4.

[022] Fig. 7 illustrates the use of the graph shown in Fig. 2 in estimating another portion of the temperatures shown in Fig. 4.

[023] Fig. 8 is flow chart for an algorithm in accordance with the invention that utilizes the thermal graphs illustrated in Figs. 2 and 3.

[024] Fig. 9 is an alternate embodiment of the flow chart shown in Fig. 8.

Detailed Description

[025] Referring again to the drawings, there is illustrated in Fig. 2, a non-linear increasing curve showing the incremental change of temperature of an electronic component that is included in an ECU of a vehicle brake control system as a function of component operating time. In the preferred embodiment, the component is a pump motor FET that controls operation of either the high pressure hydraulic fluid pump or the precharge pump; however, the invention also may be applied to other driver components in the ECU. The change of temperature is shown along the vertical axis while time is shown along the horizontal axis. In the preferred embodiment, the curve is empirically derived from experimental data obtained by continuously operating the ECU in the laboratory and actually measuring the temperature of the electronic component. Alternately, the curve can be developed from analytical data or from other data supplied from the manufacturer. [026] Similarly, there is illustrated in Fig. 3, a non-linear decreasing curve showing the incremental change of temperature of the electronic component as the component cools off following operation as a function of non-operating time. Again, the change of temperature is shown along the vertical axis while time is shown along the horizontal axis. In the preferred embodiment, the curve also is empirically derived from experimental data obtained by continuously operating the ECU in the laboratory until the component reaches an elevated temperature and then deactivating the component. The temperature of the component is then measured as the component cools back to ambient temperature. Alternately, the curve can be developed from analytical data or from other data supplied from the manufacturer.

[027] The present invention contemplates using the empirical curves shown in

Figs. 2 and 3 to estimate the operating temperature, T, of the electronic component. The operating temperature T is then compared to a threshold temperature, T_threS5 and, if the operating temperature exceeds, or is equal to, the threshold temperature T_threS5 the component is deactivated.

[028] The operation of the invention is illustrated by the graph shown in Fig. 4, where the operating temperature T of a pump motor FET during several operating cycles of the pump is plotted as a function of time. In Fig. 4, the ambient temperature within the ECU is shown by the dashed line labeled 50, while the pump motor FET operating temperature is shown by the solid line labeled 52. As shown in Fig. 4, the ambient temperature varies slowly with time. Because the ECU 39 is typically located within the vehicle engine compartment, the ambient temperature tends to increase as the vehicle is operated. At time t_l5 the pump is actuated in response to a demand from the braking system ECU 39 and begins an pressure supply cycle. Prior to t_l5 it is assumed that the pump has not been used for an extended period of time and that the FET operating temperature T is the same as the ambient temperature T^_b within the ECU 39.

[029] At the beginning of the first pressure supply cycle, t_ls the ambient temperature, T^, within the ECU 39 is determined by the ECU microprocessor from the ECU temperature sensor 40. It is assumed that the pump motor FET is at the ambient temperature T^_b. Accordingly, the ambient temperature T^ at t_\ is equal to an initial temperature T₀ used to estimate the FET operating temperature T. Thus, the temperature for entering a first lookup table that stores the values of the increasing temperature curve shown in Fig. 2 is T₀ - T^_b, which is zero, and locates the pump actuation time ti upon the increasing temperature curve, as illustrated in Fig. 5. The ECU microprocessor then uses the stored first lookup table data to estimate the pump motor FET operating temperature T as a function of time. The temperature is estimated during each iteration of the control algorithm, which, in the preferred embodiment, is every 6 milliseconds. Should the operating estimated temperature T exceed the threshold temperature T_thres, the ECU microprocessor will disable the pump motor to prevent heat damage to the motor FET. The ECU microprocessor also will generate a signal to the vehicle operator, which in the preferred embodiment is the illumination of a warning light. The estimated operating temperatures are shown between X._λ and t₂ in Fig. 4 and replicate the corresponding portion of the non-linear increasing temperature curve, as shown in Fig. 5. The incremental change in estimated operating temperature at any point between t_\ and t₂ is given by the relationship:

T - T₀, as shown in Fig. 5.

[030] At time t₂, the pump is deactuated, ending the first pressure supply cycle.

The total increase in the motor FET estimated temperature is ΔTi and the current estimated temperature of pump motor FET at that time is the peak temperature T_p for the first cycle. As the motor FET begins to cool down, the difference ^* between the peak temperature T_p and the beginning, or initial, cycle temperature T₀ is used to enter a second lookup table that stores the values of the decreasing temperature curve shown in Fig. 3. Thus, the difference between the peak temperature T_p and the initial, cycle temperature T₀, T_p - T₀, locates the pump deactuation time t₂ upon the decreasing temperature curve, as illustrated in Fig. 6. The ECU microprocessor then uses the stored second lookup table data to estimate the pump motor FET operating temperature T as a function of time. Again, the temperature is estimated during each iteration of the control algorithm, which, in the preferred embodiment, is every 6 milliseconds. The resulting estimated operating temperatures are shown between X₂ and t₃ in Fig. 4 and replicate the corresponding portion of the non-linear decreasing temperature curve, as shown in Fig. 3. The incremental decrease in estimated operating temperature at any point between X₂ and t₃ is given by the relationship:

(T_p - T₀) - T, as shown in Fig. 6.

[031 ] At time X₃, the pump is again actuated, beginning another pressure supply cycle. The total decrease in the motor FET estimated temperature is ΔT_d at X₃ and the estimated motor FET temperature will have reached a minimum, or bottom, temperature T_b, after which the estimated motor FET temperature will begin to increase again. Accordingly, the ECU microprocessor reenters the first lookup table at T_b - T₀ to begin estimating the temperature increase. As the temperature increases, the incremental temperature change is given by the relationship:

T - (T_b- T₀), as shown in Fig. 7.

[032] As shown in Fig. 4, the motor FET temperature continues to increase until the pump is again deactivated at t₄. As also shown in Fig. 4, the FET operating temperature T does not reach the threshold temperature T_thres, so the pump is not deactivated in the example shown. The FET again begins to cool down and the ECU microprocessor continues to estimate T, but uses a new peak estimated temperature T_p', which is the estimated operating temperature at U- After a sufficiently long period of time, or reset time, has passed without operation of the pump, the operating temperature T approaches the ambient temperature T_amb within the ECU. At that point in time, which is shown as t₅ in Fig. 4, the operating temperature T is reset to the ambient temperature T_31nI3. In the preferred embodiment, the reset time is adjustable within a range of 1,000 to 65,535 iterations of the control algorithm and a value of 10,000 iterations, which is equivalent to a one minute reset time, is used.

[033] If the pump were reactuated before the operating temperature was reset to the ambient temperature, the ECU microprocessor would again use the first lookup table to determine the incremental temperature increase for each iteration. The first lookup table would be entered at T_b - T₀ where the initial cycle temperature T₀ remains the value set at ti.

[034] A flow chart for an algorithm for implementing the invention is shown in

Fig. 8. The algorithm is entered through block 60 and proceeds to functional block 62 where the temperature values are initialized with the initial operating temperature T_0, the estimated peak temperature T_p and the estimated bottom temperature T_b all set equal to the ambient ECU temperature T^. The algorithm then advances to functional block 64 where the temperature threshold T_thres is set. In functional block 64 the ambient ECU temperature is compared to a trimable low temperature value. In the preferred embodiment, the low temperature is within a range of -20⁰C. to -30⁰C; however, it will be appreciated that the invention may be practiced with other ranges for the low temperature. Again, in the preferred embodiment, if the ECU ambient temperature is below the selected low temperature, the temperature threshold T_thres is set within the range of 60-110⁰C; however, if the ECU ambient temperature is above the selected low temperature, the temperature threshold T_tiaes is set within the range of 90-150⁰C. Again, the invention also contemplates that the ranges for the temperature threshold T_thres can be other than described above. Once the temperature threshold T_thres has been set, the algorithm advances to decision block 66.

[035] In decision block 66, the algorithm determines whether or not the pump is running. If the pump is running, the algorithm transfers to functional block 68 where the first lookup table is entered based upon the motor FET estimated temperature at the beginning of the current pressure supply cycle. For the first pressure supply cycle, the estimated temperature will have been initialized at the ECU ambient temperature T_31J1I,. This will also be true if sufficient time has passed without pump operation to reset the operating temperature T to the ECU ambient temperature T^_b. Otherwise, the estimated temperature for entering the first lookup table will be the value of T_b - T₀ using the estimated bottom temperature T_b from the end of the previous cycle. The entry temperature is used to determine the starting time ti for increasing temperature curve shown in Fig. 2. The algorithm then advances to functional block 70.

[036] In functional block 70, the current time t₂ for the increasing curve is determined by adding the total elapsed time to the starting time ti. Then, in functional block 72, the first lookup table is used to obtain the current estimated operating temperature T that corresponds to the current time t₂. The algorithm then advances to decision block 74.

[037] In decision block 74, the current estimated operating temperature T is compared to the threshold temperature T_thres selected in functional block 64. If the current estimated operating temperature T is greater than, or equal to, the threshold temperature T_thres, the algorithm advances to functional block 76 where a thermal flag is set to disable the pump motor. In the preferred embodiment, a warning signal is also generated (not shown) to alert the vehicle operator that the pump has been disabled; however, this step is optional. The algorithm then returns to decision block 66. If the current estimated operating temperature T is less than the threshold temperature

in decision block 74, the algorithm returns directly to decision block 66.

[038] Returning to decision block 66, if the algorithm determines that the pump is not running, the algorithm transfers to decision block 80 where the duration of the time that the pump has been off is compared to the reset time. If the pump has been off for a period longer than the reset time, the algorithm transfers to functional block 62 where the values for the estimated initial temperature T₀, the estimated peak temperature T_p and the estimated bottom temperature T_b are all reset to the current ECU ambient temperature T^_t,. Use of the current ECU ambient temperature T^_b assures that any heating of the ECU due to operation of the vehicle is taken into account.

[039] If, in decision block 80, the duration of the time that the pump has been off is less than the reset time, the pump is in a cool down mode and the algorithm transfers to functional block 82 where the second lookup table is entered at the value of T_p - T₀ using the estimated peak temperature T_p at the end of the last pressure supply cycle. The entry temperature T_p - T₀ is used to determine the starting time t_s for the decreasing temperature curve shown in Fig. 3. The algorithm then advances to functional block 84.

[040] In functional block 84, the current time t_c for the decreasing curve is determined by adding the total elapsed time to the starting time t_s. Then, in functional block 86, the second lookup table is used to obtain the current estimated operating temperature T that corresponds to the current time t_c. The algorithm then advances to decision block 87.

[041] In decision block 87, the algorithm checks the status of the thermal flag.

If the thermal flag was not set in functional block 76, the algorithm transfers back to decision block 66 and begins a new iteration. If the thermal flag was set in functional block 76, the algorithm transfers to decision block 88.

[042] In decision block 88, the estimated operating temperature T is compared to a second threshold temperature T_thres₂ which is less than the threshold temperature T_thres utilized in decision block 74. The reduced threshold temperature T_threS2 introduces hysteresis into the algorithm to avoid hunting, or oscillations, about the threshold temperature T_thres. In the preferred embodiment, the reduced threshold temperature Tt_hres₂ is the first threshold temperature T_thre_s reduced by subtracting a hysteresis temperature T_hys that is selected for the particular brake system 10. If the motor FET has cooled sufficiently to preclude potential oscillations about the first threshold temperature T^_es, the algorithm advances to functional block 90 where the pump motor thermal flag is reset to enable the pump. The algorithm then returns to decision block 66. If the motor FET has not cooled sufficiently to preclude potential oscillations about the first threshold temperature T_threS5 the algorithm returns to directly to decision block 66 from decision block 88. The pump will remain disabled until the pump motor FET has cooled down sufficiently for the estimated operating temperature T to fall below the second threshold temperature T^_e^.

[043] While the preferred embodiment of the invention has described and illustrated above for a FET that controls a pump motor, it will be appreciated that the invention also may be utilized for other electronic components in the ECU that function as drivers. Thus, for example, it is contemplated that the invention also may be applied to protect a bipolar transistor or other electronic device from overheating during operation.

[044] The present invention also contemplates an alternate embodiment for predicting the temperature of FETs used to control pump motors when the motors are actuated. The alternate embodiment utilizes the fact that when the pump motor speed is being controlled by the microprocessor using Closed Loop Motor Speed Control or any other control method to maintain the motor speed at the desired level, the motor FET temperature is in general a function of the following factors: switching frequency, duty cycle (a percentage calculated as the motor on time divided by the total time), ECU ambient temperature, initial temperature, battery voltage, motor target speed, and motor load. The motor load is not measured, so some kind of estimation must be used. If the ECU ambient temperature, initial temperature, battery voltage, motor target speed are fixed, then the frequency and duty cycle change as the load changes. This allows use of the ambient temperature, initial temperature, battery voltage, motor target speed, frequency, and duty cycle, all of which are readily monitored, to predict the FET temperature.

[045] As before, when the vehicle ignition is initially turned on, the FET temperature is assumed to be the same as that of the temperature sensor inside the ECU, i.e. the ECU ambient temperature. Accordingly, during an ABS/TC/YSC event where the motor is pulsed on and off with certain target speed, the FET temperature increase can be estimated for an iteration k by the following relationship:

Δ_Temp_FET (k)= F(Temp_FET(k-l)-Temp_ambient₅ Voltge_battery, motor_target_speed, Switching_Freq, Duty_Cylce) (1)

[046] While it may be difficult to determine the above function, it is possible to approximate the function using test data. The simplest form of approximation is a linear approximation, as shown below:

Δ _Temp_FET(k) = CO + Cl*( Temp_FET(k-l) - Temρ_ambient) + C2*Voltage_batt + C3* motor_target_sρeed+C4*Switcbing_Freq+C5*DC (2): where, in the preferred embodiment, the coefficient Cl is negative. During operation of the motors, the switching frequency and duty cycle vary with time. Accordingly, the change in temperature will vary for each iteration k as the switching frequency and duty cycle vary. Additionally, while a battery voltage is indicated in the above equation, it will be realized that the equation also can use the voltage supplied by any power source, such as, for example, an alternator.

[047] It will be appreciated that the relationship shown above as equation (2) is the preferred embodiment. It will be appreciated that the invention also may be practiced with more or less terms included in equation (2) than shown above. Thus, the inventor contemplates that equation (2) may be modified to correspond to available data. For example, is not available, equation may be simply modified by setting the coefficient C2 to zero.

[048] In the preferred embodiment, the temperature increase, Δ _Temp_FET(k), shown in the formula above is evaluated every 200 ms; however, other time periods also may be used. The FET temperature is updated using the following relationship: Temp_FET(k) = Temp_FET(k-l) + Δ _Temp_FET(k) (3)

[049] While a first order polynomial is shown in equation (2) above, the invention also contemplates using higher order polynomials to approximate the function Δ_Temp_FET(k) for FET temperature increase prediction. The coefficients in the approximation can be estimated using test data using standard curve-fitting technique such as the Least Square Method. Note that all the variables in the right hand side of the equation are either monitored or can be easily calculated by the microprocessor.

[050] When the motor target speed is O₅ that is, the motor is shut off, the cool off curve shown in Fig. 6 is used for FET temperature estimation. As described above, the current FET temperature estimation and the ECU ambient are used to determine the initial point (t₀) on the cool off curve. The FET temperature estimation would then moves along this curve as long as the motor target speed remains zero.

[051 ] A flow chart for an algorithm for implementing the operation of the alternate embodiment described above is illustrated by the flow chart shown in Fig. 9, where blocks that are the same as the blocks shown in Fig. 8 have the same numerical identifiers. As can be seen in Fig. 9, the change occurs in the left column of the flow chart where the change in FET temperature, Δ_Temp_FET(k), is calculated in functional block 92 by formula (2) above. As also explained above, while a first order polynomial is shown in equation (2), the invention also may practiced using a higher order polynomials to approximate the temperature change. The estimated FET temperature is then calculated in functional block 94 using formula (3) from above where the change in temperature calculated in functional block 92 is added to the last estimated temperature. The operation of the rest of the algorithm is the same as shown by the flow chart in Fig. 8.

[052] While the preferred embodiment contemplates using the thermal model described above in a Vehicle Stability Control System, it will be appreciated that the thermal model also can be utilized in other systems when the load on the pump motor is approximately constant. Thus, the inventor also contemplates use of the thermal model in Traction Control and Anti-lock Brake Systems. The principle and mode of operation of this invention have been explained and illustrated in its preferred embodiment. However, it must be understood that this invention may be practiced otherwise than as specifically explained and illustrated without departing from its spirit or scope. Thus, while the preferred embodiment of the invention has been illustrated and described using empirically derived temperature-time curves, it will be appreciated that the invention also may be practiced using temperature-time curves obtained from different methods.

Claims

ClaimsWhat is claimed is:

1. A method for protecting an electronic component in a vehicle brake control system from overheating, the method comprising the steps of:

(a) providing a first increasing nonlinear relationship relating the temperature of a control system electronic component to at least one operating parameter;

(b) using the first nonlinear relationship while the electronic component is operating to determine an estimated component temperature;

(c) comparing the estimated temperature to a threshold temperature; and

(d) allowing the component to continue to operate only if the estimated temperature is less than the threshold temperature.

2. The method according to claim 1 wherein, during step (b), the nonlinear relationship is first entered from an initial temperature.

3. The method according to claim 2 wherein step (a) also includes providing a second decreasing nonlinear relationship relating the temperature of the electronic component to at least one operating parameter and, upon the component ceasing to operate, using the second nonlinear relationship to determine an estimated component temperature.

4. The method according to claim 3 wherein upon the component subsequently beginning to operate again, the current temperature obtained from the second nonlinear relationship is utilized to determine a starting point upon the first nonlinear relationship and then again using the first nonlinear relationship to determine an estimated component temperature.

5. The method according to claim 4 wherein, upon initial actuation of the control system, the initial temperature is the ambient temperature within a housing containing the electronic component.

6. The method according to claim 5 wherein, upon the electronic component remaining in an non-operating condition for a period of time greater than a predetermined reset time period, the entry temperature for the first nonlinear relationship is reset to the ambient temperature within the housing containing the electronic component.

7. The method according to claim 6 wherein said second non-linear relationship is a function of component operating time.

8. The method according to claim 7 wherein said first non-linear relationship is also a function of component operating time.

9. The method according to claim 7 wherein said first non-linear relationship is a function of a plurality of vehicle operating parameters.

10. The method according to claim 9 wherein said first non-linear relationship is represented by a first order polynomial.

11. The method according to claim 9 wherein the electronic component is used to control a pump motor and further wherein the first nonlinear relationship is represented by the following functional relationship:

Δ_Temρ_FET (k)= F(Temρ_FET(k-l)-Temρ_ambient, Voltge_supply, motor_target_speed, Switching_Freq, Duty_Cylce);

where; Δ_Temp_FET (k) is the change in component temperature for the time interval k;

Temp_FET(k-l) is the component temperature for the immediately preceding time interval;

Temp_ambient is the ambient temperature within the housing containing the electronic component

Voltge_supply is the voltage of power supply for the motor;

motor_target_speed is the desired motor of the motor;

Switching_Freq is the frequency of a pulse modulated voltage applied to the motor; and

Duty_Cylce is the percent on-time of the a pulse modulated voltage applied to the motor per cycle.

12. The method according to claim 11 wherein the first non-linear relationship is represented by the following first order polynomial:

Δ _Temρ_FET(k) = CO + Cl*( Temρ_FET(k-l) - Temρ_ambient) + C2*Voltage_batt + C3* motor_target_sρeed + C4*Switching_Freq + C5*DC;

with the coefficients determined empirically form experimental data.

13. The method according to claim 11 wherein the first non-linear relationship is represented by the following first order polynomial:

Δ _Temp_FET(k) = CO + Cl*( Temρ_FET(k-l) - Temρ_ambient) + C2*Voltage_batt + C3* motor_target_sρeed + C4*Switching_Freq + C5*DC;

with the coefficients determined analytically.

14. The method according to claim 11 wherein the pump is a high pressure hydraulic pump.

15. The method according to claim 14 wherein the pump is included in a vehicle stability control system.

16. The method according to claim 14 wherein the electronic component is a Field Effect Transistor.

17. The method according to claim 14 wherein the pump is a precharge pump.

18. The method according to claim 9 wherein the first non-linear relationship is represented by a polynomial that is greater than first order.

19. The method according to claim 8 wherein the electronic component is a bipolar transistor.