Description:Complete Specification:
The following specification describes and ascertains the nature of this invention and the manner in which it is to be performed

Field of the invention
[0001] The present disclosure relates to DPF (diesel particulate filter) prognostics. In particular the present invention discloses a method and system to monitor the working of a Diesel Particulate filter (DPF).

Background of the invention
[0002] Diesel engines produce particulate particles during combustion of the fuel/air mix due to incomplete combustion. This is known as Diesel particulate matter. Diesel Particulate matter results from the incomplete combustion of diesel fuel and produces soot (black carbon) particles. Soot particles from diesel engines worsen the particulate matter pollution in the air and are harmful to health. Hence new legislation norms such as the Bharat Stage VI have made mandatory Diesel Particulate Filter (DPF) to meet the particulate matter emission targets. DPF accumulates/collects the soot particles. At a certain interval soot particles are fully collected inside the DPF which will create a back pressure to engine and not allow the flow of exhaust. So, it is required to clean the DPF when the soot threshold met, in a process called regeneration.

[0003] Collected carbon particulates are removed from the filter, continuously or periodically, through thermal regeneration. This regeneration of DPF is accomplished by programming engine to run (when the filter is full) in a manner that it elevates the exhaust temperature, in conjunction with an extra fuel injected in the exhaust stream. During DPF regeneration, upstream of DPF temperature is required around 600degC for efficient regeneration. To raise these temperature late post fuel injections are injected, which is oxidized by a catalyst that will help to raise the temperature.

[0004] However, a regeneration is not always perfect and often inefficacious in burning the soot completely. This is attributed to various reasons such inadequate temperature for a required duration, inadequate post fuel injection and the like. DPF being a very critical component, it is important to maintain and monitor its healthiness. A poorly performing DPF will impact power, fuel economy, throttle response. There is need for a system in the vehicle that identifies root cause of the undesired performance thereby giving proper recommendations for inspection to the technicians/experts and creating advance alert.

[0005] Patent application “US2015167517AA” discloses a “Method for detecting abnormally frequent diesel particulate filter regeneration, engine and exhaust aftertreatment system, and warning system”. The application provides a method for detecting abnormally frequent diesel particulate filter (DPF) regeneration. The method includes measuring a pressure drop across the DPF and using the measured pressure drop to calculate a pressure drop based soot load estimate, calculating soot output from an engine model and using the calculated soot output to calculate an emissions based soot load estimate, comparing the pressure drop based soot load estimate with the emissions based soot load estimate, and providing a warning if a difference between the pressure drop based soot load estimate and the emissions based soot load estimate exceeds a predetermined.

Brief description of the accompanying drawings
[0006] An embodiment of the invention is described with reference to the following accompanying drawings:
[0007] Figure 1 depicts a system (100) to monitor the working of a Diesel Particulate filter (DPF (106)) in an exhaust gas treatment (EGT) system of a vehicle;
[0008] Figure 2 illustrates method steps (200) to monitor the working of a Diesel Particulate filter (DPF (106)) in an exhaust gas treatment (EGT) system;
[0009] Figure 3 (3a,3b,3c,3d) is a graphical representation of characteristics for a set of parameters during a regeneration event.

Detailed description of the drawings

[0010] Figure 1 depicts a system (100) to monitor the working of a Diesel Particulate filter (DPF (106)) in an exhaust gas treatment (EGT) system of a vehicle. The EGT system comprises an oxidation catalyst (104), a diesel particulate filter (DPF (106)), and at least an electronic control unit (ECU (103)) along with other relevant components and sensors known to a person skilled in the art that. Only the components that have any bearing on the invention are described herein. The DPF (106) is located downstream of an oxidation catalyst (104).

[0011] The DPF (106) is a filter having a mesh substrate or cartridge designed to remove diesel particulate matter or soot from the exhaust gas of a diesel engine. Collected carbon particulates in the mesh known as soot. They are removed from the filter, continuously or periodically, through thermal regeneration. This regeneration of DPF (106) is accomplished by programming engine to run (when the filter is full) in a manner that it elevates the exhaust temperature and burns off the soot accumulated in the DPF (106). The elevated temperature is provided by inducing exothermic reactions in the oxidation catalyst (104). To measure this temperature two temperature sensors, upstream (102) and downstream (105) of the oxidation catalyst (104) are provided. Exothermic reactions occur in the oxidation catalyst (104) due to the injection of hydrocarbons (fuel) during the process of regeneration.

[0012] Other sensors like a lambda sensor are provided in the exhaust gas pipe (101) to measure O2 quantity in the air mass coming from the exhaust manifold. A delta pressure sensor (107) installed across the DPF (106) measure pressure difference across the DPF (106). This pressure difference indicates the quantity of accumulation of soot. Modelled soot mass or Mass of soot loaded into the DPF (106) (referred as Msot) is calculated based on the correlation of delta pressures across DPF (106) for known volume flow rate of air, with corrections applied with respect to exhaust temperature, altitude pressure etc. The ECU (103) is in communication with the plurality of sensors and other ancillary components of the fuel injection system that influence the working of EGT. For example, the ECU (103) calculates a post injection fuel quantity (PoI1) based on the input parameters such as injector energizing time and rail pressure from a fuel injector and a rail pressure sensor respectively.

[0013] The ECU (103) is a logic circuitry implemented as any or a combination of: one or more microchips or integrated circuits interconnected using a parent board, hardwired logic, and executed by a microprocessor, firmware, an application specific integrated circuit (ASIC), and/or a field programmable gate array (FPGA). The ECU (103) is required to maintain a raised temperature (through post fuel injection) for a pre-determined amount of time and till a sufficient soot load is decreased in the DPF (106). The temperature is maintained downstream of the oxidation catalyst (104) by injection of fuel in sufficient quantity and for a suitable duration in the exhaust gas pipe (101).

[0014] The important non-limiting feature of the ECU (103) is the function it performs. The ECU (103) is configured to transmit values of a set of parameters from the ECU (103) to a processor (108). The set of parameters comprise one or more from a group of modelled soot mass (Msot), post-injection quantity (PoI) , O2 quantity, temperature upstream (T4) of oxidation catalyst (104), temperature downstream (T5) of the oxidation catalyst (104).

[0015] The ECU (103) is in communication with a processor (108). The processor (108) can either be a logic circuitry or a software program that respond to and processes logical instructions to get a meaningful result. The processor (108) can reside within the vehicle or remotely across a cloud or server. In an exemplary embodiment of the present invention the ECU (103) communication with the processor (108) that is remotely located via a telematics unit in the vehicle.

[0016] The processor (108) is characterized by its function. The processor (108) is configured to: extract a characteristic each for the transmitted values of the set of parameters ; Compare each of the extracted characteristics with pre-defined ideal set of characteristics for each of the set of parameters stored in the processor (108); record anomalies in the transmitted values for each of the parameters based on comparison; monitor the working of the DPF (106) based on an analysis of the recorded anomalies.

[0017] The processor (108) extracts a mean for the post-injection quantity and O2 quantity of the post- during a regeneration event. The processor (108) computes a gradient of the modelled soot mass during the regeneration event. The processor (108) computes a mean and standard deviation of the values upstream of oxidation catalyst (104) and temperature downstream of the oxidation catalyst (104) is during a regeneration event. An AI module within the processor (108) evaluates a degree of deviation of the extracted characteristic from the pre-defined ideal characteristics for each of the parameters.

[0018] The processor (108) executes the AI module. AI module with reference to this disclosure can be explained as a reference or an inference set of data, which is use different forms of correlation matrices. Using these models and the data from these models, correlations can be established between different types of data to arrive at some logical understanding of the data. A person skilled in the art would be aware of the different types of AI models such as linear regression, naïve bayes classifier, support vector machine, neural networks and the like. The AI module may be implemented as a set of software instructions, combination of software and hardware or any combination of the same.

[0019] It should be understood at the outset that, although exemplary embodiments are illustrated in the figures and described below, the present disclosure should in no way be limited to the exemplary implementations and techniques illustrated in the drawings and described below.

[0020] Figure 2 illustrates method steps to monitor the working of a Diesel Particulate filter (DPF (106)) in an exhaust gas treatment (EGT) system. The relevant components of the EGT system along with an ECU (103) and processor (108) have been elucidated in accordance with figure 1.

[0021] Method step 201 comprises transmitting values of a set of parameters from the ECU (103) to a processor (108). The set of parameters comprise one or more from a group of modelled soot mass (Msot), post-injection fuel quantity (PoI), O2 quantity, temperature upstream of oxidation catalyst (104) (T4), temperature downstream of the oxidation catalyst (104) (T5). Modelled soot mass or Mass of soot loaded into the DPF (106) (referred as Msot) is calculated based on the correlation of delta pressures measured across the DPF (106) for known volume flow rate of air, with corrections applied with respect to exhaust temperature, altitude pressure etc. The ECU (103) calculates the post injection fuel quantity (PoI) based on the input parameters such as injector energizing time and rail pressure from a fuel injection and a rail pressure sensor respectively. O2 quantity is measured by the lambda sensor. T4 and T5 values again are measured by the respective temperature sensors. The values of these parameters are transmitted to the processor (108) by the ECU (103). For processor (108)s that are remotely located transmission from the ECU (103) happens via a vehicle telematics unit. The transmission of values of these parameters can happen continuously or only during the regeneration event.

[0022] Method step 202 comprises extracting a characteristic each for the transmitted values of the set of parameters. The extracted characteristic for the post-injection quantity and O2 quantity is a mean of the post-injection quantity and O2 quantity during a regeneration event. The extracted characteristic for the value of modelled soot mass comprises a gradient of the modelled soot mass during the regeneration event. The extracted characteristic for the value of upstream of oxidation catalyst (104) and temperature downstream of the oxidation catalyst (104) is mean and standard deviation of the values during a regeneration event.

[0023] Figure 3 is a graphical representation of characteristics for a set of parameters during a regeneration event. Figure 3a illustrates characteristics of Msot during the regeneration event. Msot (soot mass ) exhibits less negative gradient for a bad regeneration and more negative gradient for a good regeneration. Figure 3b illustrates characteristics of O2 quantity during the regeneration event. O2 quantity has drastic drops for a bad regeneration and less drops for a good regeneration. Figure 3c illustrates characteristics of post fuel injection (PoI) during the regeneration event. PoI is low for a bad regeneration and high for good regeneration. Figure 3d illustrates characteristics of T5 temperature (temperature downstream of the oxidation catalyst (104)) during the regeneration event. T5 temperature has a high rise for a longer duration with less frequent drop in good regeneration where low rise for low duration with more frequent drops in bad regeneration

[0024] Method step 203 comprises comparing each of the extracted characteristics with pre-defined ideal set of characteristics for each of the set of parameters. Ideal characteristics of each of the parameters for a good regeneration are used to train the AI model and stored in the processor (108). Method step 204 comprises recording anomalies in the transmitted values for each of the parameters based on comparison. The AI model is trained on ideal regeneration data and is sensitive towards any deviation in the set of parameters provided as an input and identifies them as anomalous.

[0025] Method step 205 comprises monitoring the working of the DPF (106) based on an analysis of the recorded anomalies. Analysis of the recorded anomalies comprises evaluating a degree of deviation of the extracted characteristic from the pre-defined ideal characteristics for each of the parameters by means of the AI module in the processor (108). In an exemplary embodiment of the present invention Isolation forest algorithm is used .Isolation forest at each node picks a random feature and a random threshold value to divide the dataset into two parts. The AI model obtains the distance of the characteristics of the input parameters from the average of the characteristics of the input parameters in the training dataset. It then assesses the contribution of each of the parameter to the total deviation. A list of root causes is then identified based on the deviation from the good regeneration data.

[0026] A person skilled in the art will appreciate that while these method steps describes only a series of steps to accomplish the objectives, these methodologies may be implemented with adaptation and modification to the system disclosed. This idea to develop a data analytics based prognostics for DPF (106) monitors critical parameters related to DPF (106) and creates an advance alert for the end user if performance reduction is observed.

[0027] 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 and adaptation to a method and system to monitor the working of a Diesel Particulate filter (DPF (106)) are envisaged and form a part of this invention. The scope of this invention is limited only by the claims.
, Claims:We Claim:
1. A method (200) to monitor the working of a Diesel Particulate filter (DPF (106)) in an exhaust gas treatment (EGT) system of vehicle, the EGT system comprising the DPF (106) located downstream of an oxidation catalyst (104) and at least a plurality of sensors configured to communicate with an Electronic control unit, the method steps comprising: transmitting (201) values of a set of parameters from the ECU (103) to a processor (108) ; the method steps characterized by:
extracting (202) a characteristic each for the transmitted values of the set of parameters ;
comparing (203) each of the extracted characteristics with pre-defined ideal set of characteristics for each of the set of parameters;
recording (204) anomalies in the transmitted values for each of the parameters based on comparison;
monitoring (205) the working of the DPF (106) based on an analysis of the recorded anomalies.

2. The method (200) to monitor the working of a Diesel Particulate filter (DPF (106)) as claimed in claim 1, wherein the set of parameters comprise one or more from a group of modelled soot mass, post-injection quantity, O2 quantity, temperature upstream of oxidation catalyst (104), temperature downstream of the oxidation catalyst (104).

3. The method (200) to monitor the working of a Diesel Particulate filter (DPF (106)) as claimed in claim 1, wherein the extracted characteristic for the post-injection quantity and the O2 quantity is a mean of the post-injection quantity and a mean of the O2 quantity during a regeneration event.

4. The method (200) to monitor the working of a Diesel Particulate filter (DPF (106)) as claimed in claim 1, wherein the extracted characteristic for the value of modelled soot mass comprises a gradient of the modelled soot mass during the regeneration event.

5. The method (200) to monitor the working of a Diesel Particulate filter (DPF (106)) as claimed in claim 1, wherein the extracted characteristic for the value of temperature (T4) upstream of oxidation catalyst (104) and temperature (T5) downstream of the oxidation catalyst (104) is mean and standard deviation of the values during the regeneration event.

6. The method (200) to monitor the working of a Diesel Particulate filter (DPF (106)) as claimed in claim 1, wherein analysis of the recorded anomalies comprises evaluating a degree of deviation of the extracted characteristic from the pre-defined ideal characteristics for each of the parameters by means of an AI module in the processor (108).

7. A system (100) to monitor the working of a Diesel Particulate filter (DPF (106)), in an exhaust gas treatment (EGT) system of vehicle, the EGT system comprising the DPF (106) located downstream of an oxidation catalyst (104) and at least a plurality of sensors configured to communicate with an Electronic control unit (ECU (103)), the ECU (103) configured to transmit values of a set of parameters from the ECU (103) to a processor (108): characterized in that system:
the processor (108) configured to:
extract a characteristic each for the transmitted values of the set of parameters ;
compare each of the extracted characteristics with pre-defined ideal set of characteristics for each of the set of parameters stored in the processor (108);
record anomalies in the transmitted values for each of the parameters based on comparison;
monitor the working of the DPF (106) based on an analysis of the recorded anomalies.

8. The system (100) to monitor the working of a Diesel Particulate filter (DPF (106)) as claimed in claim 7, wherein the set of parameters comprise one or more from a group of modelled soot mass, post-injection quantity, O2 quantity, temperature upstream of oxidation catalyst (104), temperature downstream of the oxidation catalyst (104).

9. The system (100) to monitor the working of a Diesel Particulate filter (DPF (106)) as claimed in claim 7, wherein the processor (108) extracts a mean for the post-injection quantity and the O2 quantity during a regeneration event.

10. The system (100) to monitor the working of a Diesel Particulate filter (DPF (106)) as claimed in claim 7, wherein the processor (108) computes a gradient of the modelled soot mass during the regeneration event.

11. The system (100) to monitor the working of a Diesel Particulate filter (DPF (106)) as claimed in claim 7, wherein the processor (108) computes a mean and standard deviation of the values of temperature (T4) upstream of oxidation catalyst (104) and temperature (T5) downstream of the oxidation catalyst (104) during the regeneration event.

12. The system (100) to monitor the working of a Diesel Particulate filter (DPF (106)) as claimed in claim 7, wherein processor (108) comprises an AI module that evaluates a degree of deviation of the extracted characteristic from the pre-defined ideal characteristics for each of the parameters