CN104481715A - Fault detection method of high-pressure fuel pump - Google Patents

Abstract

The present invention provides a high-pressure fuel pump fault detection method, including the following steps: step 1, collecting the instantaneous fuel pressure in the fuel accumulator, filtering the instantaneous fuel pressure, and retaining the fluctuations corresponding to the fuel supply frequency and the fuel injection frequency Component; step 2, in each high-pressure fuel pump fuel supply cycle, calculate the fuel pressure increment according to the difference between the maximum value and the minimum value of the instantaneous fuel pressure fluctuation component; step 3, under the preset working conditions, The failure of the high-pressure fuel pump is judged by counting the probability that the fuel pressure increment exceeds the threshold within the set period. The method can timely detect the failure of the high-pressure fuel pump, the reduction of the fuel supply capacity and aging, and send out early warning information to inform maintenance personnel to adjust the high-pressure fuel pump, so as to prevent further deterioration of the high-pressure fuel pump from causing emission and safety problems, and reduce maintenance costs.

Description

High Pressure Fuel Pump Fault Detection Method

technical field

The invention relates to the field of internal combustion engines, in particular to a fault detection method for a high-pressure fuel pump.

Background technique

Chinese patent CN103047125A proposes a method and device for detecting oil supply failure of a high-pressure plunger pump. This method first calculates the rail pressure values corresponding to the N crankshaft signal teeth in each oil supply cycle, and calculates the difference between the maximum and minimum values corresponding to the N rail pressure values, which is the rail pressure in each oil supply cycle Rise value, secondly calculate the average value of all plunger rail pressure rise values, and finally calculate the ratio of the rail pressure rise value of each plunger to the average value, if the ratio is within the range, the oil pump is not faulty, otherwise, there is a fault. This patent has the following deficiencies: it does not consider the influence of the advance angle on the rise of the rail pressure; it does not filter the rail pressure corresponding to N crankshaft teeth; the patent only calculates the maximum and minimum values of the rail pressure during the oil supply cycle The difference between the difference and the oil supply angle corresponding to the difference is not considered, so it is impossible to accurately judge the failure of the plunger’s oil supply capacity; The rail pressure rise value is compared with its historical data, so the aging failure of the plunger cannot be accurately judged.

Chinese patent CN102828876A proposes a method and device for monitoring unbalanced fuel supply faults. This method calculates the theoretical oil supply period of the high-pressure pump according to the engine speed and the number of engine cylinders, obtains the peak value of each rail pressure and its corresponding time point, and calculates the time interval between two adjacent peaks as the actual oil supply period. If it is within the preset value range of several integral multiples of the theoretical fuel supply period, it is judged that the high-pressure fuel supply is unbalanced. This method has the following deficiencies: the influence of the advance angle on the rail pressure rise is not considered; the specific method for obtaining the peak value of the rail pressure is not explained; the patent can only detect high pressure when there is a large difference in the working capacity of each cylinder of the high-pressure pump The oil supply is unbalanced, and when the working capacity of each cylinder of the high-pressure pump changes to the same degree, the method described in this patent cannot be detected.

The detection methods in the prior art mainly have the following problems: ①. The characteristic quantity that reflects the working capacity of the high-pressure fuel pump cannot be accurately selected, and it is impossible to accurately judge the failure of the high-pressure fuel pump such as the reduction of the fuel supply capacity; ②. The characteristic quantity of the pump working capacity is compared with its historical data, so the aging fault of the high-pressure fuel pump cannot be accurately judged; ③, only when there is a large difference in the working capacity of each cylinder of the high-pressure pump, can the unbalanced state of high-pressure fuel supply be detected. It is impossible to detect the same degree of change in the working capacity of each cylinder of the high-pressure pump.

Contents of the invention

The object of the present invention is to provide a method for fault detection of a high-pressure fuel pump, specifically a method for detecting faults such as failure of a high-pressure fuel pump, reduced pumping capacity, and aging. The technical scheme adopted in the present invention is:

A high-pressure fuel pump failure detection method, comprising the steps of:

Step 1, collecting the instantaneous fuel pressure in the fuel pressure accumulator, filtering the instantaneous fuel pressure, and retaining the fluctuation components corresponding to the fuel supply frequency and the fuel injection frequency;

Step 2: In each fuel supply cycle of the high-pressure fuel pump, calculate the fuel pressure increment according to the difference between the maximum value and the minimum value of the instantaneous fuel pressure fluctuation component;

Step 3, under the pre-set operating conditions, calculate the probability that the fuel pressure increment exceeds the threshold within a set period to determine the failure of the high-pressure fuel pump.

Further, in step 1, the instantaneous fuel pressure in the fuel pressure accumulator can be collected at a fixed time period or at a fixed angle period, and the sampling period of the fuel pressure must satisfy that the instantaneous fuel pressure collected with this sampling period can reflect a high-pressure fuel pump. The change characteristics of the fuel pressure during the fuel supply cycle.

Furthermore, if the instantaneous fuel pressure is sampled at a fixed time period, the instantaneous fuel pressure signal is processed by a time domain filter, and if the instantaneous fuel pressure is sampled at a fixed angle cycle, the instantaneous fuel pressure signal is processed by an angle domain filter.

Furthermore, the applicable time-domain filter or angle-domain filter is a bandpass filter, which includes two passbands, and the centers of the two passbands are the fuel supply frequency and the fuel injection frequency respectively.

Further, the calculation in step 2 is specifically: calculating the maximum value and minimum value of the instantaneous fuel pressure fluctuation component and the corresponding engine angle or time, and comparing the difference between the maximum value and the minimum value with the time experienced from the minimum value to the maximum value The ratio of engine angle or time as fuel pressure increment.

Further, Step 3 specifically includes: calculating the average value of the fuel pressure increment in N (N is greater than or equal to 1) engine working cycles, multiplying the average value by the high-pressure fuel pump failure threshold coefficient to obtain the high-pressure fuel pump failure threshold, if the high-pressure fuel pump If the probability that the fuel pressure increment corresponding to a pump cylinder is less than the failure threshold exceeds the set probability, it is judged that the high-pressure fuel pump of the cylinder is invalid; the average value is multiplied by the high-pressure fuel pump capacity reduction threshold coefficient to obtain the high-pressure fuel pump fuel supply capacity reduction threshold , if the probability that the fuel pressure increment corresponding to a certain cylinder of the high-pressure fuel pump is less than the capacity reduction threshold exceeds the set probability, it is judged that the fuel supply capacity of the high-pressure fuel pump of the cylinder is reduced. Among them, the failure threshold coefficient and capacity reduction threshold coefficient of the high-pressure fuel pump are determined according to the fuel pressure in the fuel pressure accumulator and the fuel injection quantity of the engine.

Further, step 3 also includes: determining the aging threshold of the high-pressure fuel pump according to the fuel pressure in the fuel pressure accumulator and the fuel injection quantity of the engine, if the probability that the fuel pressure increment corresponding to a certain cylinder of the high-pressure fuel pump is less than the aging threshold of the high-pressure fuel pump exceeds If the probability is set, it is judged that the high-pressure fuel pump of the cylinder has an aging fault.

Furthermore, the aging threshold of the high-pressure fuel pump is determined by manual calibration or automatic learning; the manual calibration process is: according to the fuel pressure and fuel injection volume, a two-dimensional table is made, and then the corresponding aging threshold under different working conditions is filled in. The current fuel pressure and fuel injection quantity can be checked in the above table; the automatic learning process is: according to the current fuel pressure and fuel injection quantity, a two-dimensional fuel pressure increment table is made, and the new high-pressure fuel pump learns in different working conditions at the beginning of use. and fill in the above table; and make an aging coefficient table according to the running time of the internal combustion engine, and fill in the corresponding aging coefficient in this table; when in use, check the above self-study according to the current internal combustion engine operating conditions The corresponding fuel pressure increment can be obtained from the fuel pressure increment table; then check the aging coefficient table according to the current cumulative running time of the engine to obtain the corresponding aging coefficient; finally, multiply the fuel pressure increment obtained by looking up the table by the aging coefficient to obtain the The aging threshold of the high pressure pump.

Further, when performing the above fault judgment, the failure fault of the high-pressure fuel pump and the reduction fault of the fuel supply capacity are first judged, and then the aging fault is judged.

The differences between the high-pressure fuel pump fault detection method described in the present invention and existing solutions include: 1. The instantaneous fuel pressure can be collected at a fixed time period or a fixed angle period, and the instantaneous fuel pressure can be processed through a band-pass filter. The center of the passband is the fuel supply frequency and the fuel injection frequency. ② The ratio of the difference between the maximum value and the minimum value of the instantaneous fuel pressure in a fuel supply cycle to the elapsed time or angle is used as the rail pressure increment to reflect the fuel supply capacity of the high-pressure fuel pump. ③By setting the high-pressure fuel pump failure threshold, fuel supply capacity reduction threshold and aging threshold to judge the high-pressure fuel pump failure, fuel supply capacity reduction and aging faults.

The method can timely detect the failure of the high-pressure fuel pump, the reduction of fuel supply capacity and aging, and send out early warning information to inform the maintenance personnel to adjust the high-pressure fuel pump, so as to prevent further deterioration of the high-pressure fuel pump and cause emission and safety problems, and reduce maintenance costs.

Description of drawings

Fig. 1 is a schematic structural diagram of the high-pressure common rail fuel system of the present invention.

Fig. 2A is a diagram of the original signal of instantaneous fuel pressure in the present invention.

Fig. 2B is a diagram of the instantaneous fuel pressure signal after the original signal is filtered according to the present invention.

FIG. 3A is a schematic diagram of the high-pressure fuel pump failure threshold coefficient of the present invention.

FIG. 3B is a schematic diagram of the capacity reduction threshold coefficient of the high-pressure fuel pump in the present invention.

Fig. 4A is a diagram of the instantaneous fuel pressure signal when the one-cylinder high-pressure pump of the present invention fails.

4B is a schematic diagram of fuel pressure increment and failure judgment threshold when the one-cylinder high-pressure pump fails in the present invention.

FIG. 5A is a schematic diagram of the aging threshold of the high-pressure fuel pump in the present invention.

Fig. 5B is a schematic diagram of the aging coefficient of the high-pressure fuel pump of the present invention.

Fig. 6 is a flow chart of judging implementation conditions of high-pressure pump fault monitoring in the present invention.

Fig. 7 is a flow chart of the implementation of high-pressure pump fault detection in the present invention.

Detailed ways

The engine of this embodiment is an example of a diesel internal combustion engine.

The principle and flow of the high-pressure fuel pump fault detection method of the internal combustion engine high-pressure fuel pump fault detection method of the present invention will be described in detail below in conjunction with the accompanying drawings by taking the high-pressure common rail fuel injection system as an example.

Figure 1 is a schematic diagram of the high-pressure common rail fuel system. In the figure, the fuel is sucked from the fuel tank 1 with the primary filter to the fine fuel filter 2, and a part of the fuel is pressurized in the plunger chamber of the high-pressure fuel pump 3 (hereinafter referred to as the high-pressure pump) to form high-pressure fuel, and the fuel is discharged from the outlet valve port of the fuel pump. It flows through the high-pressure oil pipe and collects into the common rail pipe 5 to provide a stable and continuous high-pressure fuel source for the high-pressure injection of the injector 7, and the excess part flows back to the fuel tank 1 from the overflow valve on the oil pump together with the return oil of the injector 7; The high-pressure fuel flows from the common rail pipe 5 through the high-pressure fuel pipe to the fuel injector 7 of each cylinder; the fuel injector 7 injects the fuel into each engine according to the characteristic injection characteristics according to the given time and given width of the pulse output by the electronic control unit ECU8. in the combustion chamber of the cylinder. A fuel pressure sensor 6 is installed at one end of the common rail pipe 5 as a rail pressure sensor to monitor the fuel pressure in the common rail pipe in real time. When the fuel pressure exceeds the maximum allowable value, the pressure relief valve 4 opens, and the fuel pressure in the common rail pipe drops rapidly to a safe range to ensure the safety of the entire system. The electronic control unit 8 of the common rail system collects the state parameters of the diesel engine and the common rail system detected by each sensor in real time, sends out accurate current pulse signals through the built-in control strategy and stored data, and makes the corresponding common rail pump solenoid valve and fuel injector Solenoid valves etc. generate electromagnetic force to drive the corresponding actuators to act, so that the fuel supply, rail pressure, fuel injection angle and fuel injection volume can be adjusted according to the demand. The sensors 9 used in the common rail fuel injection system include: speed sensor, rail pressure sensor, coolant temperature sensor, fuel temperature sensor, crankshaft angle sensor (or camshaft angle sensor), accelerator pedal sensor, etc. Also equipped with: vehicle speed sensor, air flow sensor, atmospheric pressure sensor, boost pressure sensor, atmospheric temperature sensor and other sensors. The actuator driving signal 10 of the electronic control unit 8 includes: the driving signal of the solenoid valve of the fuel injector and the solenoid valve of the high-pressure fuel pump.

The instantaneous fuel pressure in the common rail pipe 5 as a pressure accumulator can be collected through the rail pressure sensor (fuel pressure sensor 6 ).

Fig. 2A is an original signal diagram of instantaneous fuel pressure collected at a fixed angle of every 3CA under the condition of internal combustion engine speed of 1000rpm and target fuel pressure of 850bar. CA is the unit of crank angle, 1CA is 1 degree in 360 degrees. The acquisition method of the instantaneous fuel pressure signal is not limited to the periodic sampling at a fixed angle in this embodiment, and may also be sampled at a fixed time period. The differential interference signals generated during the operation of the vehicle and the internal combustion engine, as well as the interference signals generated during the signal acquisition process of the sensor, wiring harness and electronic control unit, make the high-frequency glitch signal appear in the original signal of instantaneous fuel pressure as shown in Figure 2A.

Fig. 2B is a diagram of the instantaneous fuel pressure signal after filtering the original signal. For the original signal of instantaneous fuel pressure sampled at a fixed time period, it is filtered by a time domain filter; for the original signal of instantaneous fuel pressure sampled at a fixed angle period, it is filtered by an angle domain filter. The filter is a bandpass filter with two passbands. The centers of the two passbands are the fuel supply frequency (that is, the fuel supply frequency) and the fuel injection frequency (that is, the fuel injection frequency). Taking a 6-cylinder internal combustion engine and a 2-cylinder in-line high-pressure fuel pump as examples, for the angle domain filter, the reference frequency is the operating frequency of the internal combustion engine, that is, the frequency corresponding to one cycle of the internal combustion engine is 1, and the fuel supply frequency and fuel injection frequency are respectively 2 and 6, set two passbands [2-△, 2+△] and [6-△, 6+△] in the angle domain filter, △ can be determined according to the actual situation. The passband of the time domain filter and the passband of the angle domain filter are set in the same way.

Calculate the fuel pressure increment based on the filtered instantaneous fuel pressure, calculate the minimum and maximum values of the instantaneous fuel pressure and their corresponding angles or time during the fuel supply cycle of the high-pressure fuel pump for each cylinder, and calculate the maximum and minimum values The ratio of the difference to the engine angle or time elapsed from minimum to maximum is the fuel pressure increment. When calculating the fuel pressure increment, the fuel injection quantity, injection advance angle, fuel pressure value and internal combustion engine speed all have an impact on the instantaneous fuel pressure increment. The larger the injection volume, the greater the fuel pressure increment; the earlier the injection advance angle is before the top dead center of fuel supply, the smaller the fuel pressure increment is, and the injection advance angle has no effect on the fuel pressure increment after the fuel supply top dead center; The fuel pressure increment is directly proportional to the fuel pressure, the higher the fuel pressure, the greater the fuel pressure increment; the fuel pressure increment is inversely proportional to the internal combustion engine speed, the higher the speed, the smaller the fuel pressure increment.

Fig. 3A is a schematic diagram of the high pressure pump failure threshold coefficient. Calculate the average value of the fuel pressure increment in N (N greater than or equal to 1) engine working cycles, and multiply the average value of the fuel pressure increment by the high-pressure fuel pump failure threshold coefficient to obtain the high-pressure fuel pump failure threshold. If a cylinder of the high-pressure fuel pump If the probability that the corresponding fuel pressure increment is smaller than the failure threshold exceeds the set probability, it is judged that the high-pressure fuel pump of the cylinder is invalid. The failure threshold coefficient of the high-pressure pump is determined according to the fuel pressure and fuel injection quantity, as shown in Fig. 3A. Fig. 3B is a schematic diagram of the reduction coefficient of the oil pumping capacity of the high-pressure pump. Multiply the average value of the fuel pressure increment by the high-pressure fuel pump capacity reduction threshold coefficient to obtain the high-pressure fuel pump fuel supply capacity reduction threshold. If the probability that the fuel pressure increment corresponding to a cylinder of the high-pressure fuel pump is less than the capacity reduction threshold exceeds the set probability, Then it is judged that the fuel supply capacity of the high-pressure fuel pump of the cylinder is reduced. The reduction factor of the pumping capacity of the high-pressure pump is determined according to the fuel pressure and fuel injection quantity. The high-pressure fuel pump capacity reduction threshold coefficient is determined according to the fuel pressure and fuel injection quantity, as shown in Figure 3B.

FIG. 4A is a diagram of instantaneous fuel pressure signals when one of the two-cylinder high-pressure fuel pumps fails. When the high-pressure pump of one cylinder fails, the fuel pressure increment obviously decreases or even appears negative in one pumping period, while the fuel pressure increment obviously increases during the pumping period of the other cylinder’s high-pressure pump. feature, it can quickly and accurately judge which cylinder high-pressure pump fails. Fig. 4B is a schematic diagram of the fuel pressure increment and the failure judgment threshold when the high-pressure pump of one cylinder fails. When a cylinder high-pressure pump fails, the corresponding fuel pressure increment is less than the failure judgment threshold of the high-pressure fuel pump. In a certain period, the probability of the fuel pressure increment being less than the high-pressure fuel pump failure judgment threshold can be used to determine the high-pressure fuel pump failure. Fault. For example, an actual judgment example is that if the fuel pressure increment is less than the high-pressure fuel pump failure judgment threshold for more than 60 times within 100 times, then it is judged that the high-pressure fuel pump fails.

If there is no failure of the high-pressure fuel pump of a certain cylinder, but its pumping capacity is severely reduced, the pumping capacity of the high-pressure fuel pump can be judged by counting the probability that the fuel pressure increment is less than the threshold value of the high-pressure fuel pump’s pumping capacity reduction within a certain period. Ability to reduce failure. If the probability that the fuel pressure increment corresponding to a certain cylinder of the high-pressure fuel pump is smaller than the capacity reduction threshold exceeds the set probability, it is judged that the fuel supply capacity of the high-pressure fuel pump of the cylinder is reduced.

Fig. 5A is a schematic diagram of the aging threshold of the high-pressure pump. In a specific embodiment of the present invention, as the use time increases, due to mechanical wear and other reasons, the high-pressure fuel pump will age, resulting in a decrease in the overall fuel supply capacity. For this fault, the present invention judges by calibrating the aging threshold and aging characteristic coefficient of the high-pressure fuel pump. The specific method is: determine the aging threshold of the high-pressure fuel pump through manual calibration or automatic learning; the manual calibration process is: make a two-dimensional table according to the fuel pressure and fuel injection quantity, and then fill in the corresponding aging threshold under different working conditions. The current fuel pressure and fuel injection quantity can be checked in the above table; the automatic learning process is: according to the current fuel pressure and fuel injection quantity, a two-dimensional fuel pressure increment table is made, and the new high-pressure fuel pump learns in different working conditions at the beginning of use. According to the fuel pressure increment below, fill in the above table; make an aging coefficient table according to the running time of the internal combustion engine, and fill in the corresponding aging coefficient in this table; when using, check the above self-study according to the current internal combustion The corresponding fuel pressure increment can be obtained from the fuel pressure increment table; then check the aging coefficient table according to the current cumulative running time of the engine to obtain the corresponding aging coefficient; finally, multiply the fuel pressure increment obtained from the table lookup by the aging coefficient to obtain the High pressure pump aging threshold. Fig. 5B is a schematic diagram of the aging coefficient of the high-pressure pump. The aging coefficient corresponding to the new high-pressure fuel pump is 1, and the aging coefficient decreases as the service time increases. The change characteristics of the aging coefficient are related to the design and manufacture characteristics of the high-pressure fuel pump.

Fig. 6 is a flow chart of judging implementation conditions of high-pressure pump fault monitoring. At first in step 11, it is judged whether condition A is satisfied, if condition A is satisfied, then step 12 is carried out, if not satisfied, then step 19 is carried out, and this condition judgment ends. Condition A may be that the engine speed is within a certain range. In step 12, it is judged whether condition B is satisfied, if condition B is satisfied, proceed to step 13, if not, proceed to step 19, and end this condition judgment. Condition B may be that the fuel injection quantity is within a certain range. In step 13, it is judged whether the condition C is satisfied, if the condition is satisfied, then proceed to step 14, if not, then proceed to step 19, and end this condition judgment. Condition C may be that the injection advance angle is within a certain range. In step 14, it is judged whether the condition D is satisfied, if the condition is satisfied, then proceed to step 15, if not, then proceed to step 19, and end this condition judgment. Condition D may be that the fuel pressure is within a certain range. In step 15, it is judged whether the condition E is satisfied, if the condition is satisfied, then proceed to step 16, if not, then proceed to step 19, and end this condition judgment. The condition E may be that the fuel pressure variation is within a certain range. In step 16 a counter is set which increments by one. In step 17, it is judged whether the counter has reached a given threshold, if not, then return to step 11, and start a new condition judgment, if it reaches a given threshold, then proceed to step 18. In step 18, the high-pressure fuel pump failure detection enable flag is set.

Fig. 7 is a flow chart of the implementation of high-pressure pump fault detection. In step 20, it is judged whether the high-pressure fuel pump fault detection condition is satisfied, if not, then end this detection, if yes, then go to step 21. In step 21, the fuel pressure increase is calculated according to the method described above. Next, in step 22 , the failure threshold of the high-pressure fuel pump is calculated. Next in step 23, it is judged whether the high-pressure pump fails, if yes, execute step 28, if not, execute step 24. In step 24, calculate the threshold value of the low oil pumping capacity of the high pressure pump, and then judge whether the high pressure pump oil pumping capacity has decreased in step 25, if yes, go to step 28, if not, go to step 26. In step 26, the aging threshold of the high-pressure pump is calculated, and then in step 27, it is judged whether the high-pressure pump is aging. In step 28, process the fault judgment result of the high-pressure pump, such as storing fault codes and the like.

The principles and implementation methods of the present invention have been described by using specific examples herein. The descriptions of the above implementation methods are only used to help understand the method and core idea of the present invention. There will be changes in the specific implementation and application scope of the method. In summary, the content of this specification should not be construed as limiting the present invention.

Claims (9)

1. A high-pressure fuel pump fault detection method, is characterized in that, comprises the steps: Step 1, collecting the instantaneous fuel pressure in the fuel pressure accumulator, filtering the instantaneous fuel pressure, and retaining the fluctuation components corresponding to the fuel supply frequency and the fuel injection frequency; Step 2: In each fuel supply cycle of the high-pressure fuel pump, calculate the fuel pressure increment according to the difference between the maximum value and the minimum value of the instantaneous fuel pressure fluctuation component; Step 3, under the pre-set operating conditions, calculate the probability that the fuel pressure increment exceeds the threshold within a set period to determine the failure of the high-pressure fuel pump. 2. The high-pressure fuel pump fault detection method as claimed in claim 1, characterized in that: In the first step, the instantaneous fuel pressure in the fuel pressure accumulator is collected by a fixed time period or a fixed angle period. 3. The high-pressure fuel pump fault detection method as claimed in claim 2, characterized in that: If the instantaneous fuel pressure is sampled at a fixed time period, the instantaneous fuel pressure signal is processed by a time domain filter; if the instantaneous fuel pressure is sampled at a fixed angle cycle, the instantaneous fuel pressure signal is processed by an angle domain filter. 4. The high-pressure fuel pump fault detection method as claimed in claim 3, characterized in that: The applicable time-domain filter or angle-domain filter is a bandpass filter comprising two passbands centered at the fuel supply frequency and the fuel injection frequency respectively. 5. The high-pressure fuel pump fault detection method as claimed in claim 4, wherein the calculation in step 2 is specifically: Calculate the maximum value and minimum value of the instantaneous fuel pressure fluctuation component and its corresponding engine angle or time, and use the ratio of the difference between the maximum value and the minimum value to the engine angle or time experienced from the minimum value to the maximum value as the fuel pressure increment . 6. The high-pressure fuel pump fault detection method as claimed in claim 1, 2, 3 or 4, wherein step 3 specifically comprises: Calculate the average value of the fuel pressure increment in N engine working cycles, and multiply the average value by the failure threshold coefficient of the high-pressure fuel pump to obtain the failure threshold of the high-pressure fuel pump. If the fuel pressure increment corresponding to a certain cylinder of the high-pressure fuel pump is less than the failure threshold If the probability exceeds the set probability, it is judged that the high-pressure fuel pump of the cylinder is invalid; the average value is multiplied by the high-pressure fuel pump capacity reduction threshold coefficient to obtain the high-pressure fuel pump fuel supply capacity reduction threshold. If the fuel pressure increase corresponding to a certain cylinder of the high-pressure fuel pump If the probability of being less than the capacity reduction threshold exceeds the set probability, it is judged that the fuel supply capacity of the high-pressure fuel pump of the cylinder is reduced; N is greater than or equal to 1. 7. The high-pressure fuel pump fault detection method as claimed in claim 6, wherein step 3 further comprises: The aging threshold of the high-pressure fuel pump is determined according to the fuel pressure in the fuel accumulator and the fuel injection quantity of the engine. If the probability that the fuel pressure increment corresponding to a certain cylinder of the high-pressure fuel pump is less than the aging threshold of the high-pressure fuel pump exceeds the set probability, then the cylinder is judged The high pressure fuel pump has an aging failure. 8. The high-pressure fuel pump fault detection method as claimed in claim 7, characterized in that: The aging threshold of the high-pressure fuel pump is determined by manual calibration or automatic learning; the manual calibration process is: according to the fuel pressure and fuel injection volume, a two-dimensional table is made, and then the corresponding aging threshold under different working conditions is filled in. The fuel injection quantity can be checked in the above table; the automatic learning process is: according to the current fuel pressure and fuel injection quantity, a two-dimensional fuel pressure increment table is made, and the new high-pressure fuel pump learns the fuel pressure under different working conditions when it starts to use Increment and fill in the above table; and make an aging coefficient table according to the running time of the internal combustion engine, and fill in the corresponding aging coefficient in this table; when using, check the fuel pressure increase obtained from the above self-study according to the current internal combustion engine operating conditions Then check the aging coefficient table according to the accumulated running time of the current engine to obtain the corresponding aging coefficient; finally, multiply the fuel pressure increment obtained by looking up the table by the aging coefficient to obtain the aging of the high-pressure pump under the current working condition threshold. 9. The high-pressure fuel pump fault detection method as claimed in claim 7, characterized in that: When performing the above fault judgment, first judge the failure fault of the high-pressure fuel pump and the fault of reduced fuel supply capacity, and then judge the aging fault.