TECHNICAL FIELD

[0001] The present subject matter relates to spark ignition engines. More particularly, detection of occurrence of an event in a spark ignition engine.

BACKGROUND

[0002] Internal Combustion (IC) engines are significantly contributing to pollution and global warming by emitting various exhaust gases into the environment. An IC engine produces power by burning of a fossil fuel that emits harmful gases, such as, CO, HC, NOx and hydrocarbons. The emission of gases has been and is deteriorating the environmental condition and thus there is an imminent need for automobile manufacturers to try and reduce the extent of emissions from an IC engine powertrain to previously unimagined levels.

[0003] In addition to the control cum reduction of emission of gases, the Automobile industry is moving rapidly towards implementing “On board diagnostic (OBD)” on vehicles for intimating user about the status of vehicle conditions. OBD system is subdivided in multiple categories based on the regime in different countries, e.g. OBD phase I & OBD phase II. Typically, OBD II focuses on three main aspects - Engine misfire detection, Catalytic convertor monitoring, and Lambda sensor monitoring. However, this may vary from one jurisdiction to other depending on the definition by local bodies.

[0004] To improve engine combustion efficiency & reduce emissions, it is essential to detect and monitor events, such as, misfire, knocking, etc., in an IC engine. Misfire occurs when the injected air fuel mixture does not bum at all or partially bums. Misfiring of the IC engine affects the quality of combustion and degrades the performance of a catalyst convertor, thereby leading to increase in emissions as well as drop in durability of the system, which is undesirable. BRIEF DESCRIPTION OF DRAWINGS

[0005] The detailed description is described with reference to the accompanying figures. The same numbers are used throughout the drawings to reference like features and components.

[0006] Fig. 1 exemplarily illustrates an ignition system with an ion current measurement circuit of an engine;

[0007] Fig. 2 exemplarily illustrates an event detection system electrically coupled to the ignition system exemplarily illustrated in Fig. 1;

[0008] Fig. 3 exemplarily illustrates a graphical representation of variation in the ion current signal, measured by the ion current measurement circuit, with respect to time;

[0009] Figs. 4-5 exemplarily illustrate graphical representations of the ion current signal in misfire condition and No-misfire condition, in frequency domain; [00010] Fig. 6 exemplarily illustrates a schematic diagram showing the determination of occurrence of the misfire from the ion current signal, based on the number of frequency bands determined by the frequency band determination unit;

[00011] Figs. 7-9 exemplarily illustrate graphical representation of the variation in the digital ion current signal in different frequency bands, during misfire condition and no-misfire condition;

[00012] Fig. 10 exemplarily illustrates a flowchart showing a method comprising steps for determining occurrence of a misfire in an internal combustion engine; [00013] Fig. 11 exemplarily illustrates a flow chart comprising steps of determining the occurrence of misfire in the engine; and

[00014] Fig. 12 exemplarily illustrates a flow chart comprising steps of determining the occurrence of misfire in the engine.

DETAILED DESCRIPTION OF THE INVENTION

[00015] Misfire events in an Internal Combustion (IC) engine can be categorized as partial or complete, based on the amount of combustion occurring during a particular engine cycle. In most engines, identification of misfire is performed by monitoring angular acceleration of engine crank-shaft. Any misfire results in momentary change in the angular acceleration of the crankshaft which is sensed to determine occurrence of a misfire event. However, single cylinder engines with lower capacity (e.g. capacity less than 200 cubic centimeter) provide challenges to detect misfire owing to low mechanical inertia of the IC engine using, when the same approach of angular acceleration is used. The problem of misfire identification for the single cylinder IC engines turns out to be additionally challenging due to presence of various load disturbances on the power train especially when the engine is employed in a vehicle.

[00016] Several other techniques are designed for the detection of engine misfire for the single cylinder engines. Such techniques include analysis of instantaneous crankshaft speed, analysis of in-cylinder pressure, analysis of instantaneous crankshaft torque, etc. Evaluation of the crankshaft speed for detection of misfire faces a lot of challenges due to low mechanical inertial and load disturbances in engines. These aspects adversely impact the reliability with which the engine misfire is detected. Alternate solutions, to address the problem of detection of misfire event in the engine, utilize ion current generated in a sparking event in the engine. When air-fuel mixture ignites inside the IC engine cylinder, air particles get ionized. By applying a suitable high-voltage on a spark plug, it is possible to measure the ion current as the amount of ion current reflects the level of ionization of the air-fuel mixture. Thus, the flow of ion current depends on the combustion event. The ion current signal can be captured with the help of an ion current measurement circuit. The captured signal needs to be processed to detect a misfire. Multiple techniques can be applied to differentiate the misfire from normal combustion with the help of the ion current signal.

[00017] An integral of the ion current signal is a common technique to detect misfire. However, the integral of ion current signal shows error for partial misfire as the value of integral coincides for misfire and No-misfire condition. There are various reasons which can result in engine misfires. Electrical failure in the ignition coil circuit is one of them. No flow of ion current in the ignition coil helps to detect a complete misfire. However, in cases where a feeble spark is generated across the spark plug resulting in a partial misfire, the detection using the ion current signal demands for filtering and processing the ion current signal to extract the information.

[00018] In application of the IC engine on vehicles, following are the disadvantages of occurrence of a misfire condition: Due to the occurrence of misfire in a combustion event, fuel will be wasted as there is no spark for fuel to bum. This will degrade the performance of the vehicle due to reduction in mileage. Misfire detection will intimate driver about the misfire events in the vehicle and by investigating and rectifying the causes of misfire, the user can improve the performance of vehicle. Detection and reduction of misfire will have an impact on the durability of vehicle. Further, misfire has direct impact on power/pick up of vehicle as there is loss of combustion event. The user may feel a sudden jerk in driving due to misfire. These may cause discomfort to the user of the vehicle. Also, due to misfire, the unbumt fuel in the exhaust affects the life of the catalyst in the converter and has direct impact on emissions. Therefore, there exists a need for detecting events, such as, no-misfires, complete misfires, and partial misfires effectively based on ion current sensing in engines for smooth experience, durability, and adherence to emission norms by the engine.

[00019] The present subject matter discloses a method of configuring magnitude and shape of ion current signal generated during a sparking event in an engine for information about a misfire (complete or partial) event in the engine. Thus, the ion current measurement is indicative of the combustion event as well as the quality of combustion in the engine. The measured ion current signal is configured to display dynamics in no-misfire condition compared to that of a misfire condition. Frequency spectrum of the ion current signal is used to extract the essential information and to differentiate between the no-misfire condition and the misfire condition in the engine.

[00020] In an embodiment, an event detection system of an engine is disclosed that comprises an ion current measurement circuit, a predetermined number of frequency band limiter modules, a predetermined number of frequency domain conversion units, and an ion signal analyzer. The ion current measurement circuit measures an ion current signal in an ignition coil generated during a spark event in a spark plug. The predetermined number of frequency band limiter modules, communicatively coupled to the ion current measurement circuit, generate a predetermined number of band fdtered ion current signals. The predetermined number of frequency domain conversion units, communicatively coupled to the predetermined number of frequency band limiter modules, converts the predetermined number of band fdtered ion current signals to a predetermined number of digital ion current signals in frequency domain. The ion signal analyzer, communicatively coupled to the predetermined number of frequency domain conversion units, analyzes and compares an amplitude of each of the predetermined number of digital ion current signals with an amplitude threshold of each of the predetermined number of digital ion current signals to determine occurrence of an event in the engine.

[00021] In an embodiment, the event detection system further comprises a frequency band determination unit for determining a predetermined number of frequency bands in the measured ion current signal based on variation in voltage levels of the measured ion current signal. Each of the predetermined number of frequency bands corresponds to each of the predetermined number of frequency band limiter modules. Each of the predetermined number of frequency band limiter modules corresponds to each of the predetermined number of band fdtered ion current signals. Each of the predetermined number of band fdtered ion current signals corresponds to each of the predetermined number of digital ion current signals. In an embodiment, the predetermined number of frequency bands is three. Each of the predetermined number of frequency band limiter modules is a bandpass fdter with a corresponding low cut-off frequency and a corresponding high cut-off frequency.

[00022] Each of the predetermined frequency domain conversion units perform Fourier Transformation of each of the predetermined number of band fdtered ion current signals in time domain to obtain each of the predetermined number of digital ion current signals in frequency domain. The Fourier Transformation is an FFT algorithm applied on each of the predetermined number of band fdtered ion current signals.

[00023] In an embodiment, the event detection system further comprises a notification unit for generating and notifying a user of the engine on occurrence of the event, based on a comparative analysis of the amplitude of each of the predetermined number of digital ion current signals by the ion signal analyzer against a prestored amplitude threshold. In an embodiment, the event detection system further comprises an amplitude database server for storing amplitude thresholds corresponding to frequency bands in the measured ion current signal. [00024] In another embodiment, a method for determining occurrence of an event in an internal combustion engine is disclosed. The method is implemented by the event detection system disclosed above. The method comprises the steps of: measuring voltage levels of an ion current signal received from an ignition coil of the IC engine, during a sparking event in a spark plug, by the ion current measurement circuit. Further, the step of filtering the received ion current signal, by the predetermined number of frequency band limiter modules, to generate a predetermined number of band filtered ion current signals is disclosed. Further, the method comprises the steps of converting the predetermined number of band filtered ion current signals to a predetermined number of digital ion current signals in frequency domain by the predetermined number of frequency domain conversion units; and analysing an amplitude of each of the predetermined number of digital ion current signals with an amplitude threshold corresponding to a frequency band of each of the predetermined number of digital ion current signals, by the ion signal analyzer, to consequentially determine occurrence of the event in the engine.

[00025] The method further comprises determining a predetermined number of frequency bands in the measured ion current signal based on variation in voltage levels of the measured ion current signal and storing the amplitude thresholds in an amplitude threshold server (205) by a frequency band determination unit of the event detection system. The conversion of the predetermined number of band filtered ion current signals in time domain to frequency domain comprises performing Fourier Transformation of each of the predetermined number of band filtered ion current signals. The Fourier Transformation is a Fast Fourier Transform (FFT) algorithm applied on each of the predetermined number of band filtered ion current signals.

[00026] In an embodiment, analyzing an amplitude of each of the predetermined number of digital ion current signals by the ion signal analyzer comprises: determining corresponding frequency band of each of the predetermined number of digital ion current signals, performing a lookup for the amplitude threshold, corresponding to the determined frequency band of each of the predetermined number of digital ion current signals, in an amplitude database server, comparing the amplitude of each of the predetermined number of digital ion current signals with the corresponding amplitude threshold, and determining the occurrence of the event in the engine based on the comparison. The method further comprises generating and notifying a user of the engine on the occurrence of the event by a notification unit of the event detection system, based on the comparative analysis by the ion signal analyzer. The method ffurther comprises the step of storing event fault information history in an electronic control unit for engine diagnostic analysis. The method utilises FFT and bandpass filters for identification of events, such as, misfires as well as partial misfires in the engines.

[00027] Fig. 1 exemplarily illustrates an ignition system 100 with an ion current measurement circuit 107 of an engine. The ignition system 100 consists of an ignition coil with a primary side 101 and secondary side 103, a spark plug 106, and a control circuit 105, for example, an electronic control unit (ECU) with an electrical switching device 104 to produce a high voltage spike required to generate a spark in a cylinder of the engine. The primary coil 101 is connected between a battery 102 and the electrical switching device 104. When the electrical switching device 104 is in a closed state, the primary side 101 of the ignition coil stores energy. As soon as the control circuit 105 changes the state of the electrical switching device 104 to open, a voltage e.g. 400V is generated in the primary side 101 of the ignition coil due to the sudden interruption of current flow in the primary side 101. A secondary high voltage of around 20-25 kV (depending upon the turn ratio of primary & secondary coils) is generated at the spark plug 106, which results into voltage breakdown and thus allows the flow of ion current across the spark plug 106. The ion current measurement circuit 107 is connected at the secondary side 103 to provide a biasing voltage which in turn generates the flow of ion current. The ion current measurement circuit 107 consists of a capacitor which is charged during trigger of the spark. Once the spark is triggered, the charge held by the capacitor generates a potential difference across the spark electrodes, which results in flow of the ion current. The ion current measurement circuit 107 is connected to a terminal 103B of the secondary side 103 of the ignition coil to detect misfire. The captured ion current signal needs to be filtered and processed to extract the essential information pertaining to combustion event in the engine.

[00028] Fig. 2 exemplarily illustrates an event detection system 200 electrically coupled to the ignition system 100 exemplarily illustrated in Fig. 1. The event detection system 200 comprises the ion current measurement circuit 107 connected to the secondary side 103 of the ignition coil, N frequency band limiter modules 203, N frequency domain conversion units 204, and an ion signal analyzer 206. The number N is a predetermined number varying between 1 to Infinity. The event detection system 200 comprises a frequency band determination unit 202 that is electrically coupled with the ion current measurement circuit 107.

[00029] The ion current measurement circuit 107 measures the ion current signal flowing through the secondary side of the ignition coil. The ion current measurement circuit 107 measures the voltage level of the ion current signal. The variation in the voltage levels of the ion current signal during a misfire event and a no-misfire event is shown in Fig. 3. Based on the variation in the voltage levels of the ion current signal with respect to time, the frequency band determination unit 202 determines number of frequency bands N, low cut-off frequency, and high cut-off frequency of each of the N frequency bands. When the ion current signal exemplarily illustrated in Fig. 3 is converted to frequency domain as exemplarily illustrated in Fig. 4, the frequency band determination unit 202 determines a low cut-off frequency and a high cut-off frequency for the entire ion current signal in frequency domain, where there is a predetermined amount of variation in amplitudes of the ion current signal during both misfire condition and no-misfire condition. Based on the variation in the amplitudes of the ion current signal in frequency domain signal, the frequency band determination unit identifies N number of frequency bands, each with a low cut-off frequency and a high cut-off frequency as disclosed in the description of Figs. 3-4.

[00030] The number of frequency band limiter modules is N which is equal to the number of frequency bands (N) determined by the frequency band determination unit 202. The N frequency band limiter modules 203 filter the ion current signal and generate N number of band filtered ion current signals. Each of the frequency band limiter modules 203 is, for example, a bandpass filter operating in a frequency band. Each frequency band is defined with a low cut-off frequency and a high cut-off frequency. Each of the frequency band limiter modules filters the ion current signal between the low cut-off frequency and the high cut-off frequency of that frequency band. Each frequency band limiter module outputs a band filtered ion current signal. Each frequency band limiter module utilizes voltage at the secondary side of the ignition coil, voltage at the primary side of the ignition coil, and engine speed as a reference, for filtering the ion current signal. Predetermined engine speeds are used to trigger the start and end of the detection of the misfire by the event detection system. Between the engine speeds, the ion current signal filtering and processing is carried out by the event detection system. The detection of the misfire is performed in real-time continuously, since no data is stored for later processing.

[00031] Each of the N frequency domain conversion units 204 acts on a band filtered ion current signal out of the N number of band filtered signals. The N frequency domain conversion units 204 convert the N band filtered ion current signals in time domain to frequency domain, since the frequency spectrum of the ion current signal provides information on the combustion event. The N frequency domain conversion units 204 generate N digital ion current signals. Each digital ion current signal comprises multiple frequency components with corresponding amplitudes as will be disclosed in the detailed description of Figs. 7-9. The frequency domain conversion unit 204 performs FFT on the band fdtered ion current signal and obtains a digital ion current signal which is sent to the ion signal analyzer 206. The ion signal analyzer 206 analyzes the amplitudes of the frequency components constituting the FFT spectrum of the band fdtered digital ion current signal to detect occurrence of misfire. The ion signal analyzer 206 analyzes the amplitudes of each of the frequency components of each digital ion current signal with an amplitude threshold corresponding to the digital ion current signals. The amplitude thresholds are stored in an amplitude database server 205. The amplitude thresholds correspond to the frequency bands of the ion current signal received by the ion current measurement circuit 107. Simultaneous to the determination of number of frequency bands by the frequency band determination unit 202, the amplitude database server 205 is populated with the amplitude thresholds corresponding to the frequency bands obtained based on experimental studies or theoretical computations which correspond to misfire of the engine. [00032] The ion signal analyzer 206 determines the corresponding frequency band of each of the predetermined number of digital ion current signals and performs a lookup for the corresponding amplitude threshold AT in the amplitude database in the amplitude database server 205. Further, the ion signal analyzer 206 compares the amplitude A of the frequency components of each of the predetermined number of digital ion current signals (DICS) with the corresponding amplitude threshold and determines the occurrence of the misfire in the engine 201 based on the comparison. In an embodiment, the ion analyzer 206 may set a misfire flag and a no-misfire flag based on the analysis.

[00033] The event detection system 200 further comprises a notification unit 207 for generating and notifying a user of the engine 201 on occurrence of the misfire, based on the analysis by the ion signal analyzer 206. Based on the misfire flag and the no-misfire flag, the notification unit 207 sends a notification to a user device of the user. The user device may be a smart phone, a desktop, a laptop, an instrument cluster of the vehicle, etc. The notification may be a voice alert, a text notification communicating an error code and the related information on the identified fault to the user. As per an alternate embodiment, the alert or misfire fault information history may be stored in the electronic control unit (ECU) 105 or any storage unit. The notification or stored history may later on be used for diagnosis and rectification of the fault by a service engineer.

[00034] Fig. 3 exemplarily illustrates a graphical representation of variation in the ion current signal, measured by the ion current measurement circuit 107, with respect to time. As can be seen, the voltage levels of the ion current signal exhibits dynamic voltage variation of signal pattern (amplitude and frequency) at the initiation of the misfire condition as compared to the no-misfire condition. The variation in the voltage levels are analyzed by other components of the event detection system 200 discussed in detailed description of Fig. 2 to determine occurrence of the misfire. The frequency band determination unit 202 performs Fourier transformation on the ion current signal illustrated in Fig. 3 without using bandpass filters to identify the number of frequency bands, the high cut-off frequency and the low cut-off frequency of each frequency band.

[00035] Figs. 4-5 exemplarily illustrate graphical representations of the ion current signal in misfire condition and no-misfire condition, in frequency domain. As exemplarily illustrated in Fig. 4, the ion current signal in frequency domain is obtained on performing Fourier transformation of the ion current signal exemplarily illustrated in Fig. 3 by the frequency band determination unit 202. The graphical representation shows variation in the amplitudes of the frequency components of the ion current signal in the range f2 Hz to f8 Hz, during misfire condition and no-misfire condition. The frequencies vary from fO Hz, fl Hz, f2 Hz, ...to.., fl 3 Hz, fl4 Hz. f2 Hz is 2 times of fl Hz, f3 is 3 times of fl Hz,...., fl4 is 14 times of fl Hz. The amplitude of the frequency components varies from A0, Al, A2, ..., to A9, where A2 is 2 times of Al, A3 is 3 times of Al,...., A9 is 9 times of Al.

[00036] From this graphical representation, it is evident that the ion current signal comprises noise due to vibration and engine dynamics and to remove the induced noise, filters are to be used. Using a single frequency band limiter module with high cut-off frequency f8 Hz and low cut-off frequency f2 Hz, the graphical representation of the ion current signal in frequency domain, illustrated in Fig. 5, is obtained. As exemplarily illustrated in Fig. 5, between the frequencies the f2 Hz and f8 Hz, the variation in the amplitudes of the frequency components is different in different frequency bands. Based on the variation in the amplitudes of the frequency components, the frequency band determination unit 202 determines N number of frequency bands, and the high cut-off frequency and low cut-off frequency for each frequency band. Exemplarily, from the graphical representation of Fig. 5, the frequency band determination unit 202 determines three frequency bands and the high cut-off frequency and low cut-off frequency for each of them. The bandpass fdter 1 has a low cut-off frequency of fl Hz and high cut-off frequency of f3 Hz. The bandpass filter 2 has a low cut-off frequency of f3 Hz and a high cut-off frequency of £5 Hz. The bandpass filter 3 has a low cut-off frequency of f5 Hz and a high cut-off frequency of f7 Hz. Since N=3 i.e. three frequency bands are identified, the number of frequency band limiter modules 203 are also three and the frequency domain conversion units 204 are also three as exemplarily illustrated in Fig. 6.

[00037] Fig. 6 exemplarily illustrates a schematic diagram showing the determination of occurrence of the misfire from the ion current signal, based on the number of frequency bands determined by the frequency band determination unit 202. As disclosed in above example, there are three frequency band limiter modules 203a, 203b, and 203c, each with a high cut-off frequency and a low cut-off frequency. The frequency band limiter modules are bandpass filters. The bandpass filter 1 203a has a low cut-off frequency of fl Hz and high cut-off frequency of £3 Hz. The bandpass filter 2 203b has a low cut-off frequency of £3 Hz and a high cut-off frequency of £5 Hz. The bandpass filter 3 203c has a low cut-off frequency of £5 Hz and a high cut-off frequency of f7 Hz. The frequency band determination unit 202 selects the low cut-off frequency and the high cut-off frequency such that the ion current signal shows significant variation during misfire and No-misfire condition between these two frequencies. Each bandpass filter 203a, 203b, and 203c generates a band filtered ion current signal. The band filtered ion current signal in time domain is converted to frequency domain using Fourier transformation. Fast Fourier Transform (FFT) is then applied to the band filtered signal after each bandpass filter block. The three FFT blocks 204a, 204b, and 204c perform FFT and generate three digital ion current signals. From the frequency spectrum of the digital ion current signal, the ion signal analyzer 206 can detect the misfire in each frequency band by comparing with an appropriate predetermined threshold from experimental studies, as will be disclosed further. In an embodiment, the frequency band limiter modules 203, the frequency domain conversion units 204, and the ion signal analyzer 206 are embodied as software modules in the Engine/Electronic Control Unit 105.

[00038] Figs. 7-9 exemplarily illustrate graphical representation of the variation in the digital ion current signal in different frequency bands, during misfire condition and No-misfire condition. It is observed from the frequency spectra in Figs. 7-9 that FFT of each band filtered ion current signal shows variation in misfire and No-misfire condition The digital ion current signal exemplarily illustrated in Fig. 7 is generated from a combination of the bandpass filter 1 203a and the FFT block 1 204a and lies between a low cut-off frequency of fl Hz and high cut-off frequency of f3 Hz. It can be observed that the amplitude of the frequency components of the digital ion current signal in Fig. 7 is above A9 at approximately fl Hz and the amplitude of the frequency components is above A4 from fl Hz to f3 Hz for No-misfire condition. However, the amplitude of the frequency components is below A7 at approximately fl Hz and the amplitude of the frequency components is below A2 beyond fl Hz to f3 Hz for misfire condition. Based on the amplitudes of the frequency components in the frequency band fl Hz to f3 Hz, the frequency band determination unit 202 determines amplitude thresholds and stores the amplitude thresholds corresponding to the frequencies in the amplitude database server 205 for further use by the ion signal analyzer 206.

[00039] The digital ion current signal exemplarily illustrated in Fig. 8 is generated from a combination of the bandpass filter 2 203b and the FFT block 2 204b and lies between a low cut-off frequency of f3 Hz and a high cut-off frequency of f5 Hz. It can be observed that the peak amplitude of the frequency components is greater than A7 between f4 Hz and f6 Hz for No-misfire condition, whereas the peak amplitude of the frequency components is varying from A6 to A8 between f4 Hz to f6 Hz for misfire condition. Sometimes two digital ion current signals for the misfire condition and the No-misfire condition are coinciding at same frequency and the amplitude band is overlapping from A6 to A8. However, there are multiple peaks in No-misfire condition with amplitude greater than A8 but there is only single peak in misfire condition. Based on the amplitudes of the frequency components in the frequency band f3 Hz to f5 Hz, the frequency band determination unit 202 determines amplitude thresholds and stores the amplitude thresholds corresponding to the frequencies in the amplitude database server 205 for further use by the ion signal analyzer 206.

[00040] The digital ion current signal exemplarily illustrated in Fig. 9 is generated from a combination of the bandpass filter 3 203c and the FFT block 3 204c and lies between a low cut-off frequency of £5 Hz and high cut-off frequency of f7 Hz. It can be observed that the amplitude of the frequency components is above A4 for all frequencies varying from f5 Hz to f7 Hz for No-misfire condition, whereas the amplitude is below A4 for all frequencies varying from f5 Hz to f7 Hz for misfire condition. Based on the amplitudes of the frequency components in the frequency band f5 Hz to f7 Hz, the frequency band determination unit 202 determines amplitude thresholds and stores the amplitude thresholds corresponding to the frequencies in the amplitude database unit 205 for further use by the ion signal analyzer 206.

[00041] The frequency band determination unit 202 determines the amplitude thresholds for No-misfire condition for the digital ion current signal in Fig. 9 to be above A9 at approximately fl Hz and above A4 from fl Hz to £3 Hz. The amplitude thresholds for the misfire condition are below A7 at approximately fl Hz and below A2 beyond fl Hz to £3 Hz. The frequency band determination unit 202 determines the amplitude thresholds for No-misfire condition to be multiple frequency components with amplitude above A7 in the frequency band £3 Hz to f5 Hz. The amplitude threshold for the misfire condition is no multiple frequency components with amplitude above A7 in the frequency band £3 Hz to f5 Hz. The frequency band determination unit 202 determines the amplitude threshold for No-misfire condition to be above A4 for all frequencies varying from f5 Hz to f7 Hz. The amplitude threshold for the misfire condition is to be below A4 for all frequencies varying from f5 Hz to f7 Hz in the frequency band f5 Hz to f7 Hz. These amplitude thresholds corresponding to the frequencies are stored in the amplitude database server 205.

[00042] When an ion current signal is received from the ion current measurement circuit 107, the 3 frequency band limiter modules 203a, 203b, 203c generate 3 band filtered ion current signals and the 3 frequency conversion units 204a, 204b, 204c generate the 3 digital ion current signals. Each digital ion current signal belonging to a frequency band is examined by the ion signal analyzer 206 to determine occurrence of misfire. For a digital ion current signal from a specific frequency band, the ion analyzer 206 compares the amplitudes of the frequency components in the digital ion current signal with the amplitude thresholds corresponding to the frequencies pre-stored in the amplitude database of the amplitude database server 205. On comparing with the amplitudes of the frequency components with the amplitude thresholds, the ion signal analyzer 206 may detect the occurrence of the misfire and activates the misfire flag as exemplarily illustrated in Fig. 6. Based on the misfire flag, the notification unit 207 generates a notification to notify the occurrence of the misfire to the user of the engine 201.

[00043] Fig. 10 exemplarily illustrates a flowchart showing a method comprising steps for determining occurrence of a misfire in an internal combustion engine 201. The method is implemented by the event detection system 200 exemplarily illustrated in Fig. 2. At step 1001, the ion current measurement circuit 107 measures voltage levels of an ion current signal received from an ignition coil 101 and 103 of the IC engine 201, during a sparking event in a spark plug 106. At step

1002, the frequency band limiter modules 203 filter the received ion current signal to generate a predetermined number of band filtered ion current signals. At step

1003, the frequency domain conversion units 204 convert the predetermined number of band filtered ion current signals in time domain to a predetermined number of digital ion current signals in frequency domain. At step 1004, the ion signal analyzer 206 analyzes amplitude of each of the predetermined number of digital ion current signals with an amplitude threshold corresponding to a frequency band of each of the predetermined number of digital ion current signals stored in the amplitude database server 205. Based on the analysis, the ion signal analyzer 206 generates a misfire flag that indicates occurrence of the misfire in the engine 201.

[00044] Fig. 11 exemplarily illustrates a flow chart comprising steps of determining the occurrence of misfire in the engine 201. At step 1101, the ion current measurement circuit 107 measures and captures or stores ion current signal in every engine cycle. At step 1102, the ion current signal is filtered with a cut-off frequency to remove noise components in higher frequencies such as fl2-fl6 Hz in Fig. 3. At step 1103, the bandpass filter 1 203a with a frequency band of fl to f3 Hz is used. At step 1104, the bandpass filter 2 203b with a frequency band of f3 to f5 Hz is used. At step 1105, the bandpass filter 3 203c with a frequency band of f5 to f7 Hz is used. The ion current signal is passed through the bandpass filter 1 203a, the bandpass filter 2 203b, and the bandpass filter 3 203c. For each band filtered ion current signal, Digital Fourier Transform (DFT) is computed using FFT algorithm to convert the time domain band filtered ion current signal to frequency domain digital ion current signal at the steps 1106, 1107, and 1108 respectively. The ion signal analyzer 206 performs comparison check for threshold condition of each digital ion current signal. The amplitude thresholds are determined from the Figs. 7-9 as disclosed in the detailed description above. At step 1109, the digital ion current signal in the frequency band fl Hz to f3 Hz is checked if the peak amplitude of the frequency components of the digital ion current signal is above A9 at fl Hz and above A4 from fl Hz to f3 Hz. If yes, the occurrence of misfire is confirmed to have been detected by the ion signal analyzer 206 and a malfunction indicator lamp (MIL) indication is generated by the notification unit 207 at step 1113. If No, the ion signal analyzer 206 confirms that No-misfire is detected at step 1112. [00045] At step 1110, the digital ion current signal in the frequency band £5 Hz to f7 Hz is checked if the amplitude of the frequency components of the digital ion current signal is above A3 from f5 Hz to f7 Hz. If yes, occurrence of the misfire is confirmed to have been detected by the ion signal analyzer 206 and a MIL indication is generated by the notification unit 207 at step 1113. If answer to amplitude comparison is No, the ion signal analyzer 206 confirms that No-misfire is detected at step 1112. At step 1111, the digital ion current signal in the frequency band £3 Hz to f5 Hz is checked if there are multiple peaks with amplitude greater than A7 from £3 Hz to f5 Hz. If answer to amplitude comparison is yes, the misfire is detected by the ion signal analyzer 206 and a MIL indication is generated by the notification unit 207 at step 1113. If answer to amplitude comparison is No, the ion signal analyzer 206 detects that No-misfire is detected at step 1112.

[00046] In this method, the misfire is detected if any of the band filtered ion current signal satisfies the threshold condition. However, to improve accuracy of the method of determining occurrence of a misfire, the misfire may be detected if and only if the threshold condition is satisfied in all the three frequency bands (N), as exemplarily illustrated in Fig. 12.

[00047] Fig. 12 exemplarily illustrates a flow chart comprising steps of determining the occurrence of misfire in the engine 201. At step 1201, the ion current measurement circuit 107 ensures and captures ion current signal in every engine cycle. At step 1202, the ion current signal is filtered with a cut-off frequency to remove noise components in higher frequencies such as fl2-fl6 Hz in Fig. 3. At step 1203, the bandpass filter 1 203a with a frequency band of fl to £3 Hz is used. At step 1204, the bandpass filter 2 203b with a frequency band of £3 to f5 Hz is used. At step 1205, the bandpass filter 3 203c with a frequency band of f5 to f7 Hz is used. The ion current signal is passed through the bandpass filter 1 203a, the bandpass filter 2 203b, and the bandpass filter 3 203c. For each band filtered ion current signal, DFT is computed using FFT algorithm to convert the time domain band filtered ion current signal to frequency domain digital ion current signal at the steps 1206, 1207, and 1208 respectively. The ion signal analyzer 206 performs check for threshold condition of each digital ion current signal. The amplitude thresholds are determined from the Figs. 7-9 and are disclosed in the detailed description above. At step 1209, the digital ion current signal in the frequency band fl Hz to f3 Hz is checked and compared if the peak amplitude of the frequency components of the digital ion current signal is above A9 at fl Hz and above A4 from fl Hz to f3 Hz. If yes, the threshold condition for the frequency band f3 Hz to £5 Hz in step 1211 is examined by the ion signal analyzer 206. If condition in step 1209 is No, the ion signal analyzer 206 determines and infers that No-misfire is detected at step 1212.

[00048] At step 1211, the digital ion current signal in the frequency band £3 Hz to f5 Hz is checked and compared if there are multiple peaks with amplitude greater than A7 from £3 Hz to f5 Hz. If condition in step 1211 is yes, the threshold condition for the frequency band £5 Hz to f7 Hz in step 1210 is examined by the ion signal analyzer 206. If condition in step 1211 is No, the ion signal analyzer 206 detects that No-misfire is detected at step 1212. Thus, if the condition in step 1209 is yes and the condition in step 1211 is no, the ion signal analyzer 206 determines and infers that No-misfire is detected at step 1212. At step 1210, the digital ion current signal in the frequency band f5 Hz to f7 Hz is checked and compared if the amplitude of the frequency components of the digital ion current signal is above A3 from f5 Hz to f7 Hz. If condition in step 1210 is yes, the misfire is detected by the ion signal analyzer 206 and a MIL indication is generated by the notification unit 207 at step 1213. If condition in step 1210 is No, the ion signal analyzer 206 detects that No-misfire is detected at step 1212. Thus, if the condition in step 1211 is yes and the condition in step 1210 is no, the ion signal analyzer 206 determines and infers that No-misfire is detected at step 1212

[00049] The prevent invention provides the following technological advancements in the field of on-board diagnostics of IC engines as follows: The present invention discloses a new method and circuit to detect a misfire and a partial misfire in a spark ignition engine by analyzing ion current signal in frequency domain with the help of digital signal processing techniques, storing the amplitude threshold data in the hardware for analysis cum comparison and subsequently inferring the misfire condition. Essential information is extracted in frequency domain by comparing ion current signal during ‘Misfire and No-misfire condition’ and appropriate threshold conditions are preset to differentiate misfire from normal combustion. The current subject matter enables overcoming the drawbacks of the known art and ensures the misfire is accurately detected thereby enabling reliable control and reduction of emissions for an engine especially of a small capacity.

[00050] Also, the sparking event is a very fast phenomenon with a maximum time of few milliseconds. Ion current signal flows through the circuit rapidly, the moment spark ends. Hence, in order to collect large set of data pertaining to the sparking event in a short duration, the frequency domain conversion units with very high sampling rates are utilised. The frequency band limiter modules are chosen based on the number of frequency bands identified by the frequency band determination unit. The frequency band limiter modules filter out signal within defined band so that the digital ion currents signal of respective frequency bands can be compared accurately. Also, the frequency bands are intelligently configured such that the digital ion current signal shows significant statistical variations in both misfire and No-misfire condition. The engine speed and the voltage on the primary side of the ignition coil are both used as reference to initiate the misfire detection by the frequency band limiter modules, the frequency domain conversion units, and the ion signal analyzer. These two signals are used to improve the reliability of the event detection system. If incase voltage of the primary side of the ignition coil is not detected due to some fault, engine speed will act as a trigger for the event detection system in real time.

[00051] Misfire detection based on ion current measurement will intimate user of the engine about the misfire events in the spark ignition engine of a vehicle in real time and by investigating and rectifying the causes of misfire, the performance of the vehicle is improved, thereby improving reliability, durability, mileage of the vehicle, and comfort offered to the user of the vehicle. Also, timely detection of misfire in the engine and rectification of the fault reduces catalyst and lambda performance degradation in the vehicle.

[00052] In addition to detecting the occurrence of the misfire, the event detection system of the present invention can perform knock detection, determine spark plug timing errors and perform spark duration measurement, perform analysis of quality of combustion in the engine, and aid in spark plug maintenance of the engine, using ion current sensing.

[00053] Improvements and modifications may be incorporated herein without deviating from the scope of the invention.

LIST OF REFERENCE NUMERALS

100- Ignition system

101-primary side of ignition coil

102- battery positive terminal 103- secondary side of ignition coil

104-control unit

105 -switching device

106- spark plug

107- ion current measurement circuit 200-Event detection system

201 -Engine

202-Frequency band determination unit

203-Frequency band limiter modules 203a- Bandpass fdter 1 203b- Bandpass fdter 2

203c- Bandpass fdter 3

204-Frequency domain conversion units 204a- FFT block 1

204b- FFT block 2 204c- FFT block 3

205 -Amplitude database server

206-Ion signal analyzer

207-Notification unit