FR3010554A1 - METHOD FOR DETECTING AND PREVENTING FAILURE OF COMPONENT OF MOTOR VEHICLE - Google Patents

Abstract

The present invention relates to a method (30) for detecting failure of a component (20) of a motor vehicle, said component being connected to an electronic calculator (10), said method comprising a step (31) of searching for failure of said component during successive driving cycles and, when a failure has been detected, a step (32) of incrementing a failure confirmation counter. Further, when a failure is detected, said method (30) includes a step (33) of storing a failure context, said fault context including the value of each environment parameter of a predefined list associated with the component (20). Said method (30) also comprises a step (39) for resetting or decrementing the failure confirmation counter which is executed only if, during a driving cycle, no failure of the component has been detected then that the memorized failure context has been reproduced.

Description

The present invention belongs to the field of embedded electronics, and more particularly relates to a method for detecting and preventing a fault in a motor vehicle component connected to an electronic computer of said motor vehicle.

Nowadays, it is known to monitor the operation of certain components of a motor vehicle. In particular, it is known, in the case of the engine computer of a motor vehicle, to perform an on-board diagnostics (OBD in the English literature) of the components involved in the control of the engine of said 10 motor vehicle. The OBD diagnostic system integrated in the engine ECU monitors, in particular, all the components having an impact on the exhaust gases, in order to detect any loss of the ability to control the polluting emissions (002, NOx, etc.) in the atmosphere. The components monitored by the OBD diagnostic system are for example sensors (ambient temperature, water temperature, inlet air pressure, etc.), actuators (butterflies, valves, turbo, pumps, relays, etc.). If a component failure is detected, it is notified to the user of the vehicle by the activation of a standard indicator light ("Malfunction Indicator Lamp" or MIL in the English literature). Simultaneously, the fault is recorded in a non-volatile memory of the engine computer, which can be read back and erased by means of a diagnostic device via an interface which is also standardized. The format according to which faults are memorized ("Diagnostic Trouble Codes" or DTC in the English literature) is also standardized, to ensure better interoperability. The test procedures carried out by the OBD diagnostic system integrated in the engine computer are also standardized, and in particular provide for monitoring the operation of the various components during successive driving cycles. A driving cycle corresponds to a time interval during which the engine of the motor vehicle operates uninterrupted, which starts with a start of the engine and which ends with a stop of said engine. A failure of a component is notified (by the activation of the MIL indicator light and / or the storage in a non-volatile memory) only if said detected fault has been confirmed, that is to say if it has been detected during a predefined number (equal to three in Europe) of consecutive driving cycles.

For this purpose, the OBD diagnostic system includes, for each monitored component, a fault confirmation counter which is incremented with each new failure detection. If, before a failure is confirmed, a driving cycle 5 occurs without the same failure being detected, then the corresponding failure confirmation counter is decremented. A disadvantage of current OBD diagnostic systems is that they do not detect intermittent component failures, that is failures that do not necessarily occur at each driving cycle. Indeed, such intermittent failures are most often not confirmed, so that no failure notification is made. The present invention aims to overcome all or part of the limitations of the solutions of the prior art, including those described above, by providing a solution that can detect intermittent failures of components connected to an electronic computer. In addition, the present invention also aims to provide a solution that allows, in some embodiments and embodiments, to limit the occurrence of intermittent failures. For this purpose, and according to a first aspect, the present invention relates to a method for detecting failure of a component of a motor vehicle, said component being connected to an electronic computer of said motor vehicle, said method comprising a research step failure of said component during successive driving cycles of said motor vehicle and, when a failure has been detected, a step of incrementing a failure confirmation counter, said method further comprising a failure notification step when the fault confirmation counter reaches a predefined threshold value. In addition, when the failure confirmation counter is equal to a predefined initial value and a fault is detected, said method includes a step of storing a fault context, said fault context including the value of each parameter of environment of a predefined list associated with said component. Finally, said method comprises a step of resetting or decrementing the failure confirmation counter that is executed only if, during a driving cycle, no component failure has been detected while the parameter or parameters environment of the predefined list associated with said component have reproduced the stored fault context. Such arrangements make it possible to confirm certain intermittent failures detected during a driving cycle. Indeed, some component failures can occur only under certain conditions. It is in particular possible to identify, a priori, for each component to be monitored, a list of environmental parameters (ambient temperature, coolant temperature, intensity of an electric current flowing in the component, speed of the motor vehicle, speed motor, value of the voltage supplied by the battery, etc.) capable of influencing the behavior of said component. Because, when a failure of a component is detected, the values of the environment parameters for that component are stored in the form of a failure context, and because during each driving cycle, the failure confirmation counter is initialized or decremented only if said memorized failure context is reproduced, then the intermittent failures associated with the stored failure context can be confirmed. In particular modes of implementation, the fault detection method may further comprise one or more of the following characteristics, taken separately or in any technically possible combination.

In particular modes of implementation, the component is a sensor or an actuator of a motor vehicle engine. In particular modes of implementation, when a fault context has been memorized, said method comprises recurrent steps of: measuring, by one or more sensors connected to the electronic computer, of each environment parameter of the predefined list associated with the component, - determining whether the memorized failure context is reproduced by comparing the measured value (s) of the environment parameter (s) of said list with said memorized failure context. In particular modes of implementation, the failure notification step comprises the activation of visual and / or audible warning means of the motor vehicle, and / or the storage of the fault in a non-volatile memory of the computer. . In particular embodiments, said fault detection method includes a preliminary step of characterizing the component, during which the list associated with said component is constituted by the environment parameter or parameters identified as likely to be present. influence the operation of said component. According to a second aspect, the present invention relates to a method for preventing failure of a component of a motor vehicle, said component being connected to an electronic computer of said motor vehicle, comprising steps of: - failure detection of the component in implementing a fault detection method according to one of the embodiments of the invention, when a failure has been detected: control of at least one actuator of the motor vehicle adapted to influence the parameter or parameters environment of the predefined list associated with the component, said at least one actuator being controlled so as to limit the reproduction of the fault context stored by said environment parameter or parameters. According to a third aspect, the present invention relates to an electronic computer of a motor vehicle, connected or intended to be connected to several components of said motor vehicle, comprising: means configured to detect a component failure by implementing a detection method according to one of the embodiments of the invention, a non-volatile memory in which is memorized, for each component, a predefined list comprising at least one environment parameter capable of influencing the behavior of said component. In particular embodiments, the electronic computer may further comprise one or more of the following characteristics, taken separately or in any technically possible combination. In particular embodiments, the electronic computer further comprises means configured to prevent the appearance of a component failure by implementing a fault prevention method according to one of the implementation modes of the invention. 'invention. In particular embodiments, the components are sensors 25 and / or actuators of a motor vehicle engine. The invention will be better understood on reading the following description, given by way of non-limiting example, and with reference to the figures which represent: FIG. 1: a schematic representation of an electronic computer of a motor vehicle, FIG. 2: a diagram illustrating the main steps of a preferred embodiment of a fault detection method, FIGS. 3 and 4: time diagrams illustrating the operation of the fault detection method of FIG. 2 In these figures, identical references from one figure to another designate identical or similar elements. For the sake of clarity, the elements shown are not to scale unless otherwise stated.

The present invention relates firstly to a method 30 for detecting failure of one or more components 20 of a motor vehicle (not shown in the figures), connected to an electronic computer 10 of said motor vehicle. Figure 1 shows very schematically an electronic calculator 10 5 of a motor vehicle. In known manner, the electronic computer comprises a microprocessor 11 connected in particular to electronic memories 12, 13 by a data bus 14. In particular, the electronic computer comprises one or more nonvolatile electronic memories 12, in which are stored 10 information that can be modified (rewritable memory). The modifiable information may be computer program products in the form of program code instructions which, when executed by the microprocessor 11, enable the various tasks assigned to said electronic computer to be performed. The modifiable information may also be data used and / or written by the various computer program products, such as calibration data, monitoring data of the components connected to the electronic computer, etc. The electronic computer may also include one or more non-rewritable and non-volatile electronic memories (not shown in the figures), in which information which can not be modified is stored. The electronic computer also comprises one or more volatile electronic memories 13, in which information is copied temporarily during operation of the electronic computer. The electronic computer 10 is further connected to a plurality of components 20 via respective communication interfaces. The components 20 are for example sensors making measurements used by the electronic computer, actuators controlled by said electronic computer, etc. In the remainder of the description, non-limitative is placed in the case of fault detection of components 20 connected to a motor computer 10, that is to say an electronic computer which controls the operation of the motor. of the motor vehicle. The components to which the engine computer is connected, the operation of which must be monitored, are for example sensors (ambient temperature, water temperature, inlet air pressure, etc.), actuators (butterflies , valves, turbo, pumps, relays, etc.).

The engine computer 10 comprises, stored in a non-volatile electronic memory 12, a computer program product in the form of a set of program code instructions which, when executed by the microprocessor 11, implement all or part of the steps of a component failure detection method 20. In a variant, the motor computer comprises programmable logic circuits of the FPGA, PLD, etc. type, and / or specialized integrated circuits (ASICs). ), adapted to implement all or part of the steps of said component failure detection method 20. In other words, the engine computer comprises a set of 10 software configured means (specific computer program product ) and / or hardware (FPGA, PLD, ASIC, etc.) to implement the various steps of a fault detection method described hereinafter. FIG. 2 represents the main steps of a fault detection method, considering in a nonlimiting manner the case where the operation of a single component 20 is monitored. As illustrated in FIG. 2, the method of detecting failure of a component 20 firstly comprises a step 31 of troubleshooting component 20, which is preferably performed recurrently during each cycle. driving the motor vehicle. Recall that a driving cycle 20 corresponds to a time interval during which the engine of the motor vehicle operates uninterrupted, which starts with a start of the engine and which ends with a stop of said engine. Step 31 troubleshooting is considered known to those skilled in the art, including current OBD diagnostic systems. In addition, the troubleshooting step 31 may depend on the monitored component, i.e., the tests performed during said troubleshooting step may vary from a component to be monitored to a monitored component. other. Conventionally, it is checked during the troubleshooting step 31 if a predefined fault detection criterion is verified. For example, the detection criterion can be checked as soon as a predefined number Nd (Nd k 1) of tests, performed during successive executions of the troubleshooting step 31, have resulted in a result. considered abnormal. For this purpose, the engine computer may increment a failure detection counter each time a test on said component provides an abnormal result, said failure detection counter being initially zeroed at the beginning of the driving cycle. . If the failure detection counter reaches the Nd value before the end of a driving cycle, then a failure is considered detected during this driving cycle. If, on the other hand, the failure detection counter does not reach the Nd value before the end of a driving cycle, then no failure is considered detected during this driving cycle. When no failure is detected during the troubleshooting step 31 (reference 31a in FIG. 2), the failure detection method is preferably continued by performing the search step 31 again. component failure 20. When a failure is detected during the troubleshooting step 31 (reference 31b in Figure 2), the method of failure detection then comprises: - a step 32 of incrementing d a failure confirmation ccp counter associated with the monitored component; a step 33 for storing a failure context in the nonvolatile electronic memory 12 of the engine computer. The failure confirmation counter ccp is initialized to a predetermined initial value (equal to zero in the example illustrated in FIG. 2), and is stored in the non-volatile electronic memory 12. In the example illustrated in FIG. the 15 ccp failure confirmation counter is used to count the number of driving cycles for which a failure has been detected, and is incremented at most once per driving cycle. The failure context stored in memory step 33 includes the value of each environment parameter of a predefined list associated with the monitored component. The value of each environment parameter of said list is for example measured at the moment of detection of the fault by one or more sensors connected to the engine computer. Each environment parameter of the list associated with the monitored component 20 is a parameter identified as likely to influence the operation of said component 20. The environment parameter (s) 25 of the list are for example identified during a prior step. (Not shown in the figures) characterizing said component 20, and said list is for example stored in the non-volatile electronic memory 12 of the engine computer. The choice of suitable environmental parameters for a given component is considered to be within the abilities of those skilled in the art. The predefined list 30 associated with a component 20 comprises at least one environment parameter which may be for example chosen from the following environmental parameters: ambient temperature, coolant temperature, intensity of an electric current flowing in the component 20, speed of the motor vehicle, engine speed, value of the voltage supplied by the battery of the motor vehicle, etc. For example, in the case of a component 20 of the ambient temperature sensor type, the environment parameters to be memorized may be the speed of the motor vehicle and the temperature of the coolant. For a flywheel position sensor, the environment parameters to be memorized may be the engine speed, the speed of the motor vehicle and the ambient temperature. For a starter, the environment parameters to be memorized may be the continuous use time coupled to the ambient temperature. For a butterfly, the environment parameters to be memorized may be the requested angular amplitude and the current position of said butterfly. In the remainder of the description, one places oneself in a nonlimiting manner in the case where the list associated with the component 20 monitored comprises several environment parameters. The fault context is stored at least when the failure confirmation ccp counter is equal to its initial value when the failure is detected. Nothing, however, excludes, as illustrated in Figure 2, to memorize the fault context at each failure detection. If necessary, several failure contexts may be stored in the non-volatile electronic memory 12. However, in the case of an intermittent failure, the stored failure contexts will most often be close. Therefore, the failure context is preferably stored only when the failure confirmation ccp counter is equal to its initial value when the failure is detected. In the remainder of the description, one places oneself in a nonlimiting manner in the case where a single fault context is memorized for the component 20.

As shown in FIG. 2, the fault detection method further comprises, after a failure has been detected, a step 34 of determining whether the failure confirmation ccp counter has reached a threshold value Nc (Nc k 2) predefined, for example equal to three. If the ccp failure confirmation counter is less than the threshold value Nc (reference 34a in FIG. 2), then, in the example illustrated in FIG. 2, the execution of the steps of the failure detection method 30 is interrupted. for the current driving cycle, and will not resume until the next driving cycle. If, on the other hand, the failure confirmation counter CCp has reached the threshold value Nc (reference 34b in FIG. 2), then the fault detection method further comprises a step 35 of failure notification and the execution of the steps of FIG. For example, failure detection method 30 is interrupted for the component in question. It should be noted that the stopping of the engine and starting the engine are shown in Figure 2 to introduce the concept of driving cycle. After starting the engine, the execution of the failure detection method 30 preferably resumes in the troubleshooting step 31. However, we understand that the engine stop can intervene at any time. In Fig. 2, the engine stop is shown after the determination step 34 and the failure notification step only to indicate that it is not necessary after these steps and for the current driving cycle, to perform again step 31 troubleshooting for the component 20 considered. The step 35 of failure notification comprises, in particular modes of implementation, the activation by the engine computer of visual and / or audible warning means of the motor vehicle, and / or the memorization of the failure in the non-volatile electronic memory 12 of said engine computer 10. For example, in the case of an OBD diagnostic system, the step 35 of failure notification includes activating the MIL indicator light and storing the fault in the form of DTCs. As shown in FIG. 2, the failure detection method 30 also includes a step 39 of resetting the failure confirmation counter ccp. The reset step 39 is executed during a driving cycle only if no failure of the monitored component has been detected while the stored failure context has been reproduced by the one or more parameters of In preferred embodiments, the resetting of the failure confirmation ccp counter is accompanied by an erasure of the failure context stored in the nonvolatile electronic memory 12. In the preferred embodiment of the predefined list associated with the component 20. to verify whether the fault context stored in the nonvolatile electronic memory 12 has been reproduced, the method of failure detection comprises, in preferred embodiments: a measuring step 37, by one or more sensors connected to the engine computer 10, environment parameters of the predefined list associated with the component 20 monitored, - a step 38 of determining whether the stored fault context is reproduced by comparing the respective measured values of the environment parameters of the list with the stored failure context; the comparison of the values can be done exactly or taking into account a tolerance. In the nonlimiting example illustrated in FIG. 2, the failure detection method comprises a step 36 of determining whether the failure confirmation ccp counter is different from its initial value (in other words to determine whether a The fault context has been memorized during a previous driving cycle) and the measurement step 37 and the determination step 38 are executed only if the failure confirmation ccp counter is different from its initial value ( reference 36a in Figure 2). If, on the other hand, the failure confirmation counter ccp is equal to its initial value (reference 36b in FIG. 2), then the fault detection method preferably continues by executing step 31 of searching for a fault. component failure 20.

During the determination step 38, the proximity between the current values of the various environmental parameters of the list and the context of failure stored in the non-volatile electronic memory 12 is evaluated. This proximity can be evaluated in many different ways without the invention being modified in principle. According to a first example, several ranges of values can be predefined for each environment parameter and the memorized failure context comprises, for each environment parameter of the list, the range in which the value of said environment parameter was located when the fault has occurred. The stored failure context can then be considered to be reproduced if the current values of the environment parameters of the list are in the same ranges of values as those stored in said fault context. In another example, assuming that the list associated with the component 20 includes Np environment parameters, it is possible to consider that said Np environment parameters form a vector space of dimension Np. On this vector space, it is also possible to define a p-distance between two points (for example the Euclidean distance when p is equal to two). The memorized failure context can then be considered to be reproduced if the p-distance between, on the one hand, said memorized failure context and, on the other hand, the point defined by the current values of the Np environment parameters, is less than a predefined threshold value. According to another nonlimiting example, the stored failure context can be considered as reproduced if the difference between, on the one hand, the current value of each environment parameter and, on the other hand, the corresponding value stored in the context of the fault, is less than a predefined percentage (for example 10 (3/0) of said corresponding value stored in the fault context When, during the determination step 38, the stored fault context is considered to be reproduced ( reference 38a in Fig. 2), the reset step 39 is performed If, on the other hand, the stored failure context is not considered reproduced (reference 38b in Fig. 2), the reset step 39 In any case, the method of fault detection is preferably continued by performing again the step 31 of troubleshooting component 20. Figures 3 and 4 show diagrams my timing illustrating the operation of the failure detection method. In these figures, one places in a nonlimiting manner in the case where the component 20 monitored is the ambient temperature sensor. The list associated with the ambient temperature sensor includes the following environmental parameters: speed of the motor vehicle and temperature of the coolant. In addition, a threshold value Nc equal to three is considered. Parts a) and b) of FIGS. 3 and 4 show the respective time changes, during successive driving cycles DC1 to DC5, of the speed of the motor vehicle and the temperature of the coolant. Part c) of Figures 3 and 4 shows the various fault detections during driving cycles DC1 to DC5. On part c), a high state represents a failure detection while a low state represents an absence of failure detection. Part d) of FIGS. 3 and 4 represents the time evolution of a prior art fault confirmation counter aacp (current OBD diagnostic systems) initialized to zero. Part e) of FIGS. 3 and 4 represents the time evolution of the failure confirmation ccp counter according to the invention, initialized to zero. FIG. 3 illustrates the operation of the failure detection method 15 in the case of an intermittent failure of the ambient temperature sensor which reproduces deterministically when the following arbitrary conditions, called "failure conditions", are verified: speed of the motor vehicle less than a value Vp with the coolant temperature greater than a value Tp. The driving cycle DC1 starts at time TO. At the instant T1 of said DC1 driving cycle, the failure conditions are checked and a failure occurs and is detected. Following this failure detection, the failure confirmation aacp counter and the failure confirmation ccp counter are both incremented to one. In addition, the fault context is stored in the non-volatile electronic memory 12 of the engine computer.

The driving cycle DC2 starts at time T2. The failure conditions are not verified during said DC2 driving cycle, so that no failure occurs. The aacp failure confirmation counter is decremented to zero at the end of said DC2 driving cycle. On the other hand, the memorized failure context is not reproduced, so that the failure confirmation ccp counter is not reinitialized.

The driving cycle DC3 starts at time T3. At the instant T4 of said DC3 driving cycle, the failure conditions are checked and a failure occurs and is detected. Following this failure detection, the aacp failure confirmation counter is incremented to one and the failure confirmation ccp counter is incremented to two. The DC4 driving cycle starts at time T5. The failure conditions are not checked during said DC4 driving cycle, so that no failure occurs. The aacp failure confirmation counter is reset to zero at the end of said DC4 driving cycle. On the other hand, the memorized failure context is not reproduced, so that the failure confirmation ccp counter is not reinitialized. The DC5 driving cycle starts at time T6. At the instant T7 of said DC5 driving cycle, the failure conditions are checked and a failure occurs and is detected. Following this failure detection, the failure confirmation aacp counter is incremented to one and the failure confirmation ccp counter is again incremented and reaches the threshold value Nc. The failure notification step 35 is therefore executed. FIG. 4 illustrates the operation of the failure detection method 10 in the case of a random failure and isolated from the ambient temperature sensor which does not reproduce over time and therefore does not need to be notified. . The driving cycle DC1 starts at time TO. At the instant T1 of said DC1 driving cycle, the random and isolated failure occurs and is detected. Following this failure detection, the failure confirmation aacp counter and the failure confirmation ccp counter are both incremented to one. In addition, the fault context (vehicle speed equal to Vp and coolant temperature equal to Tp) is stored in the non-volatile electronic memory 12 of the engine computer. The driving cycle DC2 starts at time T2. No failure occurs during said DC2 driving cycle. The failure confirmation aacp counter is reset to zero at the end of said DC2 driving cycle. On the other hand, the memorized failure context is not reproduced, so that the failure confirmation ccp counter is not reinitialized. The DC3 driving cycle starts at time T3. No failure occurs during said DC3 driving cycle. In addition, the memorized failure context is not reproduced, so that the failure confirmation ccp counter is not reset. The DC4 driving cycle starts at time T4. No failure occurs during said DC4 driving cycle. In addition, the stored failure context is not reproduced, so that the failure confirmation ccp counter is not reset. The DC5 driving cycle starts at time T5. No failure occurs during said DC5 driving cycle. On the other hand, the memorized failure context is reproduced at time T6, so that the failure confirmation ccp counter is reset to zero. In particular embodiments, the engine computer may comprise a set of means configured in software (specific computer program product) and / or hardware (FPGA, PLD, ASIC, etc.) to implement a method method of preventing the failure of the component 20, the main steps of which are the following (not illustrated by figures): a step of detecting the failure of the component 20 by implementing a method of detecting failure according to one of the modes implementation of the invention, - when a failure has been detected: a motor vehicle actuators control step adapted to influence the environmental parameters of the predefined list associated with the component 20, said actuators being controlled by the engine computer 10 so as to limit the reproduction of the fault context stored by said environment parameters. By "limiting the reproduction of the stored fault context" is meant that the actuators are controlled so as to prevent the reproduction of said fault context or, if it is not possible to prevent the reproduction of said fault context, that said actuators are controlled so as to minimize the reproduction time of said failure context. For example, in the case of the ambient temperature sensor and the above-mentioned failure conditions, it is possible in particular to prevent the motor fans from being activated in a preventive manner so as to prevent the coolant from reaching the temperature Tp. It is also possible to clamp the engine of the vehicle to prevent the coolant from reaching the temperature Tp, etc. More generally, it should be noted that the embodiments and embodiments considered above have been described by way of non-limiting examples, and that other variants are therefore possible. In particular, the invention has been described by considering the failure detection for a single component 20. Nothing excludes, according to other examples, to implement the failure detection method to monitor the operation of several of the 25 components 20 connected to the engine computer, in which case the steps illustrated in FIG. 2 are executed for each of the components 20. In preferred embodiments, the failure detection method is implemented to monitor all the components. 20 connected to the engine computer 10. The non-volatile electronic memory 12 of the engine computer 10 stores, for each of the components 20 to 30 monitor, a predefined list of environment parameters and possibly a list of preventive actions to be implemented in case of failure detection. In addition, the invention has been described by considering the case of components 20 connected to a motor computer. Nothing excludes, according to other examples, to consider other electronic computers of a motor vehicle (interior computer 35, power steering computer, etc.). In addition, the invention has been described by considering that the failure confirmation ccp counter is reset when, during a driving cycle, no failure of the monitored component has been detected while the stored failure context has been detected. has been reproduced. In other examples, nothing precludes decrementing said failure confirmation ccp counter rather than resetting it, when no failure of the monitored component has been detected while the stored failure context has been reproduced. The above description clearly illustrates that by its different characteristics and advantages, the present invention achieves the objectives it has set for itself. In particular, the present invention makes it possible to detect deterministic intermittent failures. In addition, it should be noted that the present invention can be simply integrated with current OBD diagnostic systems, by adding the failure confirmation ccp counter according to the invention to the current failure confirmation aacp counter, or by replacing said counter aacp of confirmation of current failure by said counter ccp failure confirmation according to the invention.

Claims (9)

REVENDICATIONS1. Method (30) for detecting failure of a component (20) of a motor vehicle, said component being connected to an electronic calculator (10) of said motor vehicle, said method comprising a step (31) for troubleshooting said component during successive driving cycles of said motor vehicle and, when a failure has been detected, a step (32) of incrementing a failure confirmation counter, said method (30) further comprising a step (35) of failure notification when the failure confirmation counter reaches a predefined threshold value, characterized in that: - when the failure confirmation counter is equal to a predefined initial value and a failure is detected, said method (30) comprises a step (33) for storing a failure context, said fault context comprising the value of each environment parameter of a predefined list associated with the component, - said method (30) comprises a step (39) for resetting or decrementing the failure confirmation counter which is executed only if, during a driving cycle, no failure of the component has been detected while the or the environment parameters of the predefined list associated with said component have reproduced the stored fault context. 2. Method (30) according to claim 1, characterized in that the component (20) is a sensor or an actuator of a motor vehicle engine. 3. Method (30) according to any one of the preceding claims, characterized in that, when a fault context has been stored, said method comprises recurrent steps of: - (37) measurement, by one or more connected sensors to the electronic calculator (10), of each environment parameter of the predefined list associated with the component (20), - (38) determining whether the stored fault context is reproduced by comparison of the measured value (s) of the at least one parameter of environment of said list to said stored failure context. 4. Method (30) according to any one of the preceding claims, characterized in that the step (35) of failure notification comprises the activation of visual and / or audible warning means of the motor vehicle, and / or storing the fault in a non-volatile memory (12) of the electronic computer (10). 5. Method (30) according to any one of the preceding claims, characterized in that it comprises a preliminary step of characterizing the component (20), during which the list associated with said component is constituted by the parameter or parameters environment identified as likely to influence the operation of said component. 6. A method of preventing failure of a component (20) of a motor vehicle, said component being connected to an electronic computer (10) of said motor vehicle, characterized in that said method comprises steps of: - failure detection component (20) by implementing a method (30) according to one of the preceding claims, - when a failure has been detected: control of at least one actuator of the motor vehicle adapted to influence the parameter or parameters of environment of the predefined list associated with the component (20), said at least one actuator being controlled so as to limit the reproduction of the fault context stored by the at least one environment parameter. Motor vehicle electronic calculator (10), connected or intended to be connected to several components (20) of said motor vehicle, characterized in that it comprises means configured to detect a component failure by implementing a method ( 30) according to any one of claims 1 to 5, said electronic calculator (10) comprising a non-volatile memory (12) in which is stored, for each component (20), a predefined list comprising at least one environment parameter likely to influence the behavior of said component. 8. Calculator (10) according to claim 7, characterized in that it further comprises means configured to prevent the appearance of a component failure by implementing a method according to claim 6. 9. Calculator (10) according to any one of claims 7 to 8, characterized in that the components (20) are sensors and / or actuators of a motor vehicle engine.