Field of the invention
The present disclosure relates to a method of detecting a degree of filter clogging in an internal combustion engine.
Background of the invention
Filtered Fuel is a primary need for functionality & durability of fuel injection components. Imperfect Filtration leads to malfunction of components which further affects engine performance. The filter in due course of time becomes clogged due to the trapping of suspended parti eies as they pass through a porous medium. This trapping progressively impedes and eventually stops the flow of the fuel. Conventional methods of detecting filter clogging in an internal combustion engine use sensors inside the internal combustion engine to measure a differential pressure across the filter. However this method is expensive and not robust.
Patent Application US6672147B1, Method for detecting clogging in a fuel filter in an internal combustion engine supply circuit concerns a method for detecting clogging in a fuel filter, between a fuel pressure regulator imposing a pressure towards the internal combustion engine downstream of the filter and an electric motor pump compressing fuel coming from the tank towards the regulator through the filter which consists in: determining the fuel pressure at the pump output and by considering it as the fuel pressure at the filter intake; determining the fuel pressure at the filter outlet as being the pressure imposed by the regulator; determining the pressure drop of the filter from the difference between the

filter input and output pressure levels, and by comparing at least a value based on the pressure drop with at least a reference value to deduce therefrom information concerning the clogging condition of the filter.
Brief description of the accompanying drawings
[0004] Fig. 1 is a flow-chart for a method of detecting a degree of
filter clogging in an internal combustion engine.
Detailed description of the drawings
[0005] Figure 1 is a flow chart for a method of detecting a degree of
filter clogging in an internal combustion engine using an Electronic control unit (ECU) adapted to detect a degree of clogging in a fuel filter of an internal combustion engine. In step (101), the ECU receives a measured values of a set of parameters. In step (102) the ECU stores a matrix comprising a correlation factor between each of the set of parameters and value of differential pressure. In step (103) the ECU calculates a predicted value of differential pressure based on the measured values of the set of parameters and the matrix. In step (104) the ECU linearizes the predicted value of differential pressure with a dynamic value of actual differential pressure. The linearized value of predicted differential pressure gives the degree of filter clogging. In step (105) the ECU indicates the value of degree of filter clogging to the vehicle user.
[0006] The measured value of the set of parameters are the
instantaneous values of the set of parameters. The set of parameters measured by the ECU include but are not limited to battery voltage,

coolant temperature, torque, engine speed, torque demand state, main injection energizing time, pilot injection energizing time, rail pressure, total system quantity requirement, metering unit actual current etc.
[0007] In step (102) the ECU stores a matrix comprising a
correlation factor between each of the set of parameters and value of differential pressure. A matrix comprising a correlation factor between each of the set of parameters and value of differential pressure is derived after experimentation with various values of each of the parameters and the corresponding change in value of actual differential pressure.

Table 1.
[0008] Table 1 illustrates a correlation between some important
parameters in the set of parameters and differential pressure. The

correlation factor lies between -1 to 1. The factor -1 corresponds to high correlation and inverse proportionality, where the factor 1 corresponds to high correlation and direct proportionality. For example battery voltage is correlated to differential pressure and directly proportional to differential pressure by a factor of 0.67. Similarly Engine torque is correlated to differential pressure and inversely proportional to differential pressure by a factor of 0.73.
[0009] In step 103 the correlation between all the above parameters
and the differential pressure is used to calculate a predicted value of differential pressure. The instantaneous value of the parameters listed in Table 1 are measured and received by the ECU. They are then superimposed with the corresponding correlation factor to calculate a predicted value of differential pressure.
[0010] In step (104) the ECU linearizes the predicted value of
differential pressure with a dynamic value of actual differential pressure. The dynamic value of actual differential pressure is derived from a self-learning algorithm using the actual value of differential pressure measured at various instances. A model is built in ECU using equations from a self-learning algorithm using the pressure values and engine parameters recorded for different instances. The dynamic value of the actual differential pressure is predicted using this model with the corresponding correlated engine parameter values. The various instances correspond to various values of the boundary conditions. These boundary conditions include but it is not limited to engine running state, engine speed idle,

accelerator pedal not pressed and the like. In an embodiment the self-learning algorithm is a decision tree algorithm which is fed with values of actual differential pressure and engine parameters measured at above mentioned various boundary conditions to give a predicted dynamic value of actual differential pressure. The self-learning algorithm need not be limited to decision tree alone, but can be extended to other regression algorithms.
[0011] The engine parameter values are recorded for a specific time
and the mean of the spread of predicted delta pressure values are fit into a curve. This curve is formed using the already measured differential pressure values measured for different clogging conditions. The predicted value from the curve indicates the degree of filter clogging.
[0012] In step 105 the ECU indicates the degree of filter clogging to
the vehicle user. In an embodiment the ECU indicates the value of filter clogging by displaying the value on a display unit provided on the dashboard of the vehicle. In another embodiment the ECU indicates the value of filter clogging by an audio notification to the user of the vehicle.
[0013] This idea is to develop a method of detecting a degree of
clogging in a fuel filter based on analysis & prognosis of various data parameters of a Common Rail system. The degree of filter clogging is indicative of health and efficiency of the filter. This is an accurate and more precise method of determining filter clogging as it takes into account various parameters indicative of dynamic health of filter. This method

eliminates the need of special filter clogging sensors to be installed inside the internal combustion engine, hence offering a cost-effective and more accurate solution for the problem of checking the degree of filter clogging.
[0014] It must be understood that the embodiments explained in the
above detailed description are only illustrative and do not limit the scope of this invention. Any modification to the method of detecting a degree of filter clogging in an internal combustion engine are envisaged and form a part of this invention. The scope of this invention is limited only by the claims.

We Claim:
1. A method (100) of detecting a degree of clogging in a fuel filter of an internal combustion engine using an electronic control unit (ECU), said method comprising the following steps: receiving (101) measured values of a set of parameters, characterized in that method:
Storing (102) inside the ECU a matrix comprising a correlation factor between each of the set of parameters and value of differential pressure;
Calculating (103) a predicted value of differential pressure based on the measured values of the set of parameters and the matrix;
Linearizing (104) the predicted value of differential pressure with a dynamic value of actual differential pressure, where the linearized value of predicted differential pressure gives the degree of filter clogging;
indicating (105) the value of degree of filter clogging to the vehicle user.
2. The method as claimed in 1, where the dynamic value of actual differential pressure is derived from a self-learning algorithm using the actual value of differential pressure measured at various instances.

3. The method as claimed in 1, where the indication of the value of degree of filter clogging can be either an audio indication or a visual indication.
4. An Electronic Control Unit (ECU) adapted to detect a degree of clogging in a fuel filter of an internal combustion engine, where the said ECU is capable of:
Receiving (101) measured values of a set of parameters;
storing (102) a matrix comprising a correlation factor between each of the set of parameters and value of differential pressure;
calculating (103) a predicted value of differential pressure based on the measured values of the set of parameters and the matrix;
linearizing (104) the predicted value of differential pressure with a dynamic value of actual differential pressure, where the linearized value of predicted differential pressure gives the degree of filter clogging;
indicating (105) the value of degree of filter clogging to the vehicle user.
5. The ECU as claimed in Claim 3, where the dynamic value of actual
differential pressure is derived from a self-learning algorithm using

the actual value of differential pressure measured at various instances. 6. The ECU as claimed in Claim 3, where the indication of the value of degree of filter clogging can be either an audio indication or a visual indication.