Event detection system

Abstract

An event detection system (200) for determining occurrence of an event in an internal combustion engine (201) is disclosed, comprising: an ion current measurement circuit (107), a predetermined number of band limiter modules (203), a predetermined number of frequency domain conversion units (204), and an ion signal analyzer (206). The band limiter modules (203) generate band filtered ion current signals from received ion current signals generated during a spark event in a spark plug (106). The frequency domain conversion units (204) convert the band filtered ion current signals to a frequency domain to obtain digital ion current signals. The ion analyzer analyzes an amplitude of each of the digital ion current signals with an amplitude threshold of each of the digital ion current signals to determine occurrence of the event in the engine. Misfire detection ensures that the vehicle complies with OBD II regulatory specifications implemented by regulatory agencies.

Description

Technical Field

[0001] This topic pertains to spark-ignition engines. More specifically, it relates to detecting events occurring in spark-ignition engines. Background Technology

[0002] Internal combustion engines (ICs) are a major contributor to pollution and global warming by emitting various exhaust gases into the environment. IC engines generate power by burning fossil fuels that emit harmful gases such as CO, HC, NOx, and hydrocarbons. These emissions have been and continue to worsen environmental conditions, making it imperative for automakers to attempt to reduce the emissions from IC engine powertrains to previously unimaginable levels.

[0003] Beyond controlling and reducing emissions, the automotive industry is rapidly shifting towards implementing On-Board Diagnostics (OBD) to inform users of their vehicle's status. OBD systems are classified into several types according to different national regulations, such as OBD Stage I and OBD Stage II. Typically, OBD II focuses on three aspects—engine misfire detection, catalytic converter monitoring, and Lambda sensor monitoring. However, this can vary across different jurisdictions depending on the definition provided by local authorities.

[0004] To improve engine combustion efficiency and reduce emissions, detecting and monitoring events such as misfires and knocking in IC engines is crucial. Misfires occur when the injected air-fuel mixture fails to burn completely or only partially. Misfires in IC engines affect combustion quality and degrade catalytic converter performance, leading to unwanted emissions increases and reduced system durability. Attached Figure Description

[0005] The accompanying drawings provide a detailed description. The same reference numerals are used in all the drawings to denote similar features and components.

[0006] Figure 1 An ignition system with an ion current measurement circuit for the engine is shown as an example;

[0007] Figure 2 An example is shown of an electrical connection to Figure 1 An event detection system for an ignition system is illustrated in the example.

[0008] Figure 3 A graphical representation of the change of the ion current signal relative to time, as measured by the ion current measurement circuit, is shown as an example.

[0009] Figures 4 to 5 An exemplary graphical representation of the ion current signal in the frequency domain, under both the misfire and non-misfire conditions, is shown.

[0010] Figure 6 An exemplary diagram illustrates how misfire is determined based on the number of frequency bands determined by the frequency band determination unit and the ion current signal.

[0011] Figures 7 to 9 A graphical representation of the variation of the digital ion current signal in different frequency bands is shown as an example during fire-out and fire-free conditions;

[0012] Figure 10 An exemplary flowchart is shown, including steps for determining that a misfire has occurred in an internal combustion engine;

[0013] Figure 11 An exemplary flowchart is shown, including the steps of determining that a misfire has occurred in the engine; and

[0014] Figure 12 An exemplary flowchart is shown, including the steps of determining that a misfire has occurred in the engine. Detailed Implementation

[0015] Misfire events in internal combustion engines (ICs) can be classified as partial or complete based on the amount of combustion occurring during a specific engine cycle. In most engines, misfire identification is performed by monitoring the angular acceleration of the engine crankshaft. Any misfire results in a transient change in the crankshaft's angular acceleration, and detecting this transient change determines the occurrence of a misfire event. However, when using the same angular acceleration method, detecting misfires is challenging for single-cylinder engines with lower capacities (e.g., less than 200 cubic centimeters) due to the low mechanical inertia of the IC engine used. The problem of misfire identification in single-cylinder IC engines becomes even more challenging due to various load disturbances in the powertrain (especially when the engine is used in a vehicle).

[0016] Several other techniques for detecting misfires in single-cylinder engines were also designed. These techniques include instantaneous crankshaft speed analysis, in-cylinder pressure analysis, and instantaneous crankshaft torque analysis. Due to the low mechanical inertia and load disturbances in the engine, the crankshaft speed used for misfire detection faces many challenges. These aspects adversely affect the reliability of engine misfire detection. An alternative solution to the problem of detecting misfire events in engines is to utilize the ion current generated during a spark event in the engine. When the air-fuel mixture ignites in the cylinder of an IC engine, air particles ionize. By applying an appropriate high voltage to the spark plug, the ion current can be measured because the magnitude of the ion current reflects the degree of ionization of the air-fuel mixture. Therefore, the ion current flow 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 misfires. With the help of the ion current signal, various techniques can be used to distinguish between misfires and normal combustion.

[0017] Integrating the ion current signal is a standard technique for detecting misfires. However, integrating the ion current signal cannot accurately indicate partial misfires because the integral value is the same in both misfire and no-misfire states. Various factors can cause engine misfires. Electrical faults in the ignition coil circuit are one of them. The absence of ion current flow in the ignition coil helps detect complete misfires. However, in cases where a weak spark at the spark plug causes a partial misfire, detection using the ion current signal requires filtering and processing to extract information.

[0018] In the application of IC engines in automobiles, the disadvantages of misfire are as follows: fuel is wasted due to the lack of a spark for combustion. This reduces vehicle performance due to decreased mileage. Misfire detection allows the driver to understand misfire events in the vehicle, and by investigating and correcting the cause, the user can improve vehicle performance. Detection and reduction of misfires will impact vehicle durability. Furthermore, misfires directly affect vehicle power / acceleration due to the loss of combustion. Users may experience sudden jolting while driving due to misfires, which can cause discomfort. Similarly, unburned fuel in the exhaust gas due to misfires affects the lifespan of the catalyst in the converter and directly impacts emissions. Therefore, it is necessary to effectively detect events such as no-misfire, complete misfire, and partial misfire based on ion current sensing in the engine to achieve a smooth engine experience, durability, and compliance with emission standards.

[0019] This subject matter discloses a method for determining the magnitude and shape of an ion current signal generated during an ignition event in an engine to obtain information about a misfire (complete or partial) event in the engine. Therefore, the ion current measurement indicates the combustion event and combustion quality in the engine. The measured ion current signal is configured to display the dynamics in a misfire state compared to a misfire state. The spectrum of the ion current signal is used to extract basic information and distinguish between a misfire state and a misfire state in the engine.

[0020] In one embodiment, an engine event detection system is disclosed, including an ion current measurement circuit, a predetermined number of band limiter modules, a predetermined number of frequency domain conversion units, and an ion signal analyzer. The ion current measurement circuit measures the ion current signal in the ignition coil generated during a spark event in the spark plug. The predetermined number of band limiter modules, communicatively connected to the ion current measurement circuit, generate a predetermined number of band-filtered ion current signals. The predetermined number of frequency domain conversion units, communicatively connected to the predetermined number of band limiter modules, convert the predetermined number of band-filtered ion current signals into a predetermined number of digital ion current signals in the frequency domain. The ion signal analyzer, communicatively connected to the predetermined number of frequency domain conversion units, analyzes the amplitude of each of the predetermined number of digital ion current signals using an amplitude threshold of each of the predetermined number of digital ion current signals and compares the amplitude of each of the predetermined number of digital ion current signals with the amplitude threshold of each of the predetermined number of digital ion current signals to determine the occurrence of an event in the engine.

[0021] In one embodiment, the event detection system further includes a frequency band determination unit for determining a predetermined number of frequency bands in the measured ion current signal based on changes in the voltage level of the measured ion current signal. Each of the predetermined number of frequency bands corresponds to each of a predetermined number of frequency band limiter modules. Each of the predetermined number of frequency band limiter modules corresponds to each of a predetermined number of band-filtered ion current signals. Each of the predetermined number of band-filtered ion current signals corresponds to each of a predetermined number of digital ion current signals. In one embodiment, the predetermined number of frequency bands is three. Each of the predetermined number of frequency band limiter modules is a bandpass filter having a corresponding low cutoff frequency and a corresponding high cutoff frequency.

[0022] Each of the predetermined frequency domain transformation units performs a Fourier transform on each of a predetermined number of band-filtered ion current signals in the time domain to obtain a predetermined number of digital ion current signals in the frequency domain. The Fourier transform is an FFT algorithm applied to each of the predetermined number of band-filtered ion current signals.

[0023] In one embodiment, the event detection system further includes a notification unit for generating and notifying the engine user of the occurrence of an event based on a comparative analysis of the amplitude of each of a predetermined number of digital ion current signals obtained by an ion signal analyzer relative to a pre-stored amplitude threshold. In one embodiment, the event detection system further includes an amplitude database server for storing amplitude thresholds corresponding to frequency bands in the measured ion current signals.

[0024] In another embodiment, a method for determining the occurrence of an event in an internal combustion engine is disclosed. This method is implemented by the event detection system disclosed above. The method includes the following steps: during a spark event in the spark plug, measuring the voltage level of an ion current signal received from the ignition coil of an IC engine by an ion current measurement circuit. Furthermore, the method discloses a step of filtering the received ion current signal by a predetermined number of band limiter modules to generate a predetermined number of band-filtered ion current signals. Additionally, the method includes a step of converting the predetermined number of band-filtered ion current signals into a predetermined number of digital ion current signals by a predetermined number of frequency domain conversion units; and an ion signal analyzer analyzing the amplitude of each of the predetermined number of digital ion signals using amplitude thresholds corresponding to the frequency bands of each of the predetermined number of digital ion current signals to necessarily determine the occurrence of an event in the engine.

[0025] The method further includes determining a predetermined number of frequency bands in the measured ion current signal by a frequency band determination unit of the event detection system based on changes in the voltage level of the measured ion current signal, and storing amplitude thresholds in an amplitude threshold server (205). The conversion of the predetermined number of band-filtered ion current signals in the time domain to the frequency domain includes performing a Fourier transform on each of the predetermined number of band-filtered ion currents. The Fourier transform is a Fast Fourier Transform (FFT) algorithm applied to each of the predetermined number of band-filtered ion current signals.

[0026] In one embodiment, analyzing the amplitude of each of a predetermined number of digital ion current signals by an ion signal analyzer includes: determining a corresponding frequency band for each of the predetermined number of digital ion current signals; searching an amplitude database server for an amplitude threshold corresponding to the determined frequency band for each of the predetermined number of digital ion signals; comparing the amplitude of each of the predetermined number of digital ion currents with the corresponding amplitude threshold; and determining the occurrence of an event in the engine based on the comparison analysis by the ion signal analyzer. The method also includes generating and notifying the engine user of the occurrence of the event by a notification unit of the event detection system based on the comparison analysis by the ion signal analyzer. The method further includes the step of storing event fault information history in the electronic control unit for engine diagnostic analysis. The method utilizes FFT and bandpass filters to identify events such as misfires and partial misfires in the engine.

[0027] Figure 1 An ignition system 100 with an ion current measurement circuit 107 for an engine is illustrated exemplarily. The ignition system 100 comprises an ignition coil with a primary side 101 and a secondary side 103, a spark plug 106, and a control circuit 105, such as an electronic control unit (ECU) with an electrical switching device 104, to generate the high-voltage peak required to produce a spark in the engine cylinder. The primary side coil 101 is connected between a battery 102 and the electrical switching device 104. When the electrical switching device 104 is closed, the primary side 101 of the ignition coil stores energy. Once the control circuit 105 changes the state of the electrical switching device 104 to open, a voltage, for example, 400V, is generated in the primary side 101 of the ignition coil due to the sudden termination of current in the primary side 101. A secondary high voltage of approximately 20-25kV (depending on the turns ratio of the primary and secondary coils) is generated at the spark plug 106, causing voltage breakdown and allowing an ion current to flow through the spark plug 106. An ion current measurement circuit 107 is connected to the secondary side 103 to provide a bias voltage, thereby generating an ion current flow. The ion current measurement circuit 107 consists of a capacitor that charges during spark ignition. Once the spark is ignited, the charge held by the capacitor creates a potential difference across the spark electrodes, resulting in an ion current flow. The ion current measurement circuit 107 is connected to 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 basic information related to combustion events in the engine.

[0028] Figure 2 An exemplary illustration shows an electrical connection to, as in Figure 1An event detection system 200 of an ignition system 100 is illustrated exemplarily. The event detection system 200 includes an ion current measurement circuit 107 connected to the secondary side 103 of the ignition coil, N 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 and infinity. The event detection system 200 includes a band determination unit 202 electrically connected to the ion current measurement circuit 107.

[0029] 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 also measures the voltage level of the ion current signal. Figure 3 The diagram shows the voltage level changes of the ion current signal during fire incidents and the absence of fire incidents. Based on the voltage level changes of the ion current signal relative to time, the frequency band determination unit 202 determines the number of frequency bands N, the low cutoff frequency, and the high cutoff frequency of each of the N frequency bands. When in Figure 3 The ion current signal illustrated in the example is converted to... Figure 4 In the frequency domain, as exemplarily shown, the frequency band determination unit 202 determines the low cutoff frequency and high cutoff frequency of the entire ion current signal in the frequency domain, wherein the amplitude of the ion current signal varies by a predetermined amount under both fire-fighting and fire-free conditions. Based on the amplitude variation of the ion current signal in the frequency domain signal, the frequency band determination unit identifies N frequency bands, such as... Figures 3 to 4 Each of the publicly disclosed ones has a low cutoff frequency and a high cutoff frequency.

[0030] The number of band limiter modules is N, which is equal to the number of bands (N) determined by the band determination unit 202. N band limiter modules 203 filter the ion current signal and generate N band-filtered ion current signals. Each of the band limiter modules 203 is, for example, a bandpass filter operating within a band. Each band is defined by a low cutoff frequency and a high cutoff frequency. Each of the band limiter modules filters the ion current signal between the low and high cutoff frequencies of that band. Each band limiter module outputs a band-filtered ion current signal. Each band limiter module filters the ion current signal based on the voltage on the secondary side of the ignition coil, the voltage on the primary side of the ignition coil, and the engine speed. A predetermined engine speed is used to trigger the start and end of misfire detection by the event detection system. Between engine speeds, the event detection system performs ion current signal filtering and processing. Since no data is stored for subsequent processing, misfire detection is performed continuously in real time.

[0031] Each of the N frequency-domain converter units 204 operates on a filtered ion current signal from among N filtered signals. Since the spectrum of the ion current signal provides information about the combustion event, the N frequency-domain converter units 204 convert the N filtered ion current signals from the time domain to the frequency domain. The N frequency-domain converter units 204 generate N digital ion current signals. (As in...) Figures 7 to 9 As detailed in the disclosure, each digital ion current signal includes multiple frequency components with corresponding amplitudes. A frequency domain converter unit 204 performs an FFT on the filtered ion current signal and acquires the digital ion current signal sent to an ion signal analyzer 206. The ion signal analyzer 206 analyzes the amplitudes of the frequency components constituting the FFT spectrum of the filtered digital ion current signal to detect misfire. The ion signal analyzer 206 uses amplitude thresholds corresponding to the digital ion current signals to analyze the amplitude of each frequency component of each digital ion current. The amplitude thresholds are stored in an amplitude database server 205. The amplitude thresholds correspond to the frequency bands of the ion current signals received by the ion current measurement circuit 107. While the number of frequency bands is determined by the frequency band determination unit 202, the amplitude database server 205 is populated with amplitude thresholds corresponding to the frequency bands obtained based on experimental studies or theoretical calculations corresponding to engine misfires.

[0032] Ion signal analyzer 206 determines the corresponding frequency band for each of the digital ion current signals and performs a lookup of the corresponding amplitude threshold AT in the amplitude database of amplitude database server 205. Furthermore, ion signal analyzer 206 compares the amplitude A of the frequency component of each of a predetermined number of digital ion current signals (DICS) with the corresponding amplitude threshold and determines, based on this comparison, that a misfire has occurred in engine 201. In one embodiment, ion signal analyzer 206 can set a misfire flag and a no-misfire flag based on the analysis.

[0033] The event detection system 200 also includes a notification unit 207, which is used to generate and notify the user of the engine 201 of a misfire based on the analysis generated by the ion signal analyzer 206. Based on the misfire indicator and no-misfire indicator, the notification unit 207 sends a notification to the user's user device. The user device can be a smartphone, desktop computer, laptop, car dashboard, etc. The notification can be a voice alarm, a text notification conveying error codes and relevant information about the identified fault to the user. According to an alternative embodiment, the alarm or misfire fault information history can be stored in the electronic control unit (ECU) 105 or any storage unit. This notification or stored history can be used later by service engineers for fault diagnosis and correction.

[0034] Figure 3An exemplary graphical representation of the change in the ion current signal relative to time, measured by the ion current measurement circuit 107, is shown. It can be seen that, compared to the non-flame-out state, at the start of the flame-out state, the voltage level of the ion current signal exhibits a dynamic voltage change in the signal pattern (amplitude and frequency). Figure 2 The event detection system 200, described in detail, analyzes changes in voltage levels to determine if a fire has occurred. The frequency band determination unit 202 determines the frequency band without using a bandpass filter. Figure 3 The ion current signal shown is subjected to a Fourier transform to identify the number of frequency bands, the high cutoff frequency and the low cutoff frequency of each band.

[0035] Figures 4 to 5 The graphical representation of the ion current signal in the frequency domain is illustrated exemplarily under both misfire and non-misfire conditions. (As shown in...) Figure 4 As exemplarily shown, the ion current signal in the frequency domain is determined by the frequency band determination unit 202, such as in... Figure 3 The ion current signal illustrated is obtained after undergoing a Fourier transform. The figure shows the frequency components of the ion current signal varying from f2 Hz to f8 Hz during both the fire-fighting and fire-free states. The frequency varies from f0 Hz, f1 Hz, f2 Hz to f13 Hz, f14 Hz. f2 Hz is twice f1 Hz, f3 is three times f1 Hz, and so on, with f14 being 14 times f1 Hz. The amplitude of the frequency components varies from A0, A1, A2 to A9, where A2 is twice A1, A3 is three times A1, and so on, with A9 being nine times A1.

[0036] As can be clearly seen from the graphical representation, the ion current signal includes noise generated by vibration and engine dynamics. To remove this noise, a filter is required. Using a single-frequency limiter module with a high cutoff frequency of f8 Hz and a low cutoff frequency of f2 Hz, the following results were obtained: Figure 5 The image shows a graphical representation of the ion current signal in the frequency domain. (As shown in...) Figure 5 As exemplarily shown, the amplitude variation of the frequency component differs across different frequency bands between frequencies f2 Hz and f8 Hz. Based on the amplitude variation of the frequency component, the frequency band determination unit 202 determines N frequency bands and the high cutoff frequency and low cutoff frequency of each band. For example, from... Figure 5 In the graphical representation, the frequency band determination unit 202 determines three frequency bands and the high and low cutoff frequencies in each band. Bandpass filter 1 has a low cutoff frequency of f1 Hz and a high cutoff frequency of f3 Hz. Bandpass filter 2 has a low cutoff frequency of f3 Hz and a high cutoff frequency of f5 Hz. Bandpass filter 3 has a low cutoff frequency of f5 Hz and a high cutoff frequency of f7 Hz. Since N = 3, that is, three frequency bands are identified, so as in Figure 6 As exemplarily shown, there are also three frequency band limiter modules 203 and three frequency domain conversion units 204.

[0037] Figure 6 An exemplary diagram illustrates the determination of misfire occurrence from an ion current signal based on the number of frequency bands determined by the frequency band determination unit 202. As disclosed in the example above, there are three frequency band limiter modules 203a, 203b, and 203c, each with a high cutoff frequency and a low cutoff frequency. The frequency band limiter modules are bandpass filters. Bandpass filter 1203a has a low cutoff frequency of f1 Hz and a high cutoff frequency of f3 Hz. Bandpass filter 2203b has a low cutoff frequency of f3 Hz and a high cutoff frequency of f5 Hz. Bandpass filter 3203c has a low cutoff frequency of f5 Hz and a high cutoff frequency of f7 Hz. The frequency band determination unit 202 selects the low and high cutoff frequencies such that the ion current signal exhibits a significant change between misfire and non-misfire states 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 the time domain is converted to the frequency domain using a Fourier transform. Then, after each bandpass filter block, a Fast Fourier Transform (FFT) is applied to the bandpass-filtered signal. Three FFT blocks 204a, 204b, and 204c perform the FFT and generate three digital ion current signals. From the spectrum of the digital ion current signals, the ion signal analyzer 206 can detect misfires in each frequency band by comparing them with appropriate predetermined thresholds from experimental studies, as will be further disclosed. In one embodiment, the band limiter module 203, the frequency domain conversion unit 204, and the ion signal analyzer 206 are embodied as software modules within the engine / electronic control unit 105.

[0038] Figures 7 to 9 A graphical representation of the variations in digital ion current signals in different frequency bands during misfire and non-misfire conditions is illustrated, as exemplarily. Figures 7 to 9 The spectrum clearly shows that the FFT of each filtered ion current signal changes under both off-fire and no-off-fire conditions. Figure 7 The digital ion current signal exemplarily shown is generated by a combination of a bandpass filter 1203a and an FFT block 1204a, and lies between a low cutoff frequency of f1 Hz and a high cutoff frequency of f3 Hz. It can be observed that, for a no-flashover state, in Figure 7The amplitude of the frequency component of the digital ion current signal is higher than A9 at approximately f1 Hz, and higher than A4 between f1 Hz and f3 Hz. However, in the case of an off-fire, the amplitude of the frequency component is lower than A7 at approximately f1 Hz, and lower than A2 between f1 Hz and f3 Hz. Based on the amplitude of the frequency component in the frequency band from f1 Hz to f3 Hz, the frequency band determination unit 202 determines the amplitude threshold and stores the amplitude threshold corresponding to the frequency in the amplitude database server 205 for further use by the ion signal analyzer 206.

[0039] Figure 8 The digital ion current signal, exemplarily shown, is generated from a combination of a bandpass filter 2203b and an FFT block 2204b, and lies between a low cutoff frequency of f3 Hz and a high cutoff frequency of f5 Hz. It can be observed that, for the no-flashover state, the peak amplitude of the frequency component is greater than A7 between f4 Hz and f6 Hz, while for the flashover state, the peak amplitude of the frequency component varies from A6 to A8 between f4 Hz and f6 Hz. Sometimes, the two digital ion current signals for the flashover and no-flashover states coincide at the same frequency, and the amplitude bands overlap from A6 to A8. However, in the no-flashover state, there are multiple peaks with amplitudes greater than A8, but in the flashover state, there is only one peak. Based on the amplitude of the frequency component in the frequency band from f3 Hz to f5 Hz, the frequency band determination unit 202 determines an amplitude threshold and stores the amplitude threshold corresponding to the frequency in the amplitude database server 205 for further use by the ion signal analyzer 206.

[0040] Figure 9 The digital ion current signal, exemplarily shown, is generated from a combination of a bandpass filter 3203c and an FFT block 3204c, and lies between a low cutoff frequency of f5 Hz and a high cutoff frequency of f7 Hz. It can be observed that, for the no-flashover state, the amplitude of the frequency components is above A4 for all frequencies varying from f5 Hz to f7 Hz, while for the flashover state, the amplitude is below A4 for all frequencies varying from f5 Hz to f7 Hz. Based on the amplitude of the frequency components in the band f5 Hz to f7 Hz, the band determination unit 202 determines an amplitude threshold and stores the corresponding amplitude threshold in the amplitude database server 205 for further use by the ion signal analyzer 206.

[0041] Frequency band determination unit 202 determines for in Figure 9The amplitude threshold of the digital ion current signal in the no-flashover state is higher than A9 at approximately f1 Hz and higher than A4 in the range of f1 Hz to f3 Hz. The amplitude threshold for the flashover state is lower than A7 at approximately f1 Hz and lower than A2 outside the range of f1 Hz to f3 Hz. The frequency band determination unit 202 determines the amplitude threshold for the no-flashover state as multiple frequency components in the frequency band f3 Hz to f5 Hz with amplitudes higher than A7. In the frequency band f3 Hz to f5 Hz, the amplitude threshold for the flashover state is multiple frequency components without amplitudes exceeding A7. The frequency band determination unit 202 determines that for all frequencies varying from f5 Hz to f7 Hz, the amplitude threshold for the no-flashover state is higher than A4. For all frequencies varying from f5 Hz to f7 Hz in the frequency band f5 Hz to f7 Hz, the amplitude threshold for the flashover state is lower than A4. These amplitude thresholds correspond to frequencies stored in the amplitude database server 205.

[0042] When an ion current signal is received from the ion current measurement circuit 107, three band limiter modules 203a, 203b, and 203c generate three band-filtered ion current signals, and three frequency conversion units 204a, 204b, and 204c generate three digital ion current signals. The ion signal analyzer 206 examines each digital ion current signal belonging to a specific band to determine if an off-center fire has occurred. For a digital ion signal from a particular band, the ion analyzer 206 compares the amplitude of the frequency component in the digital ion current with an amplitude threshold corresponding to a frequency pre-stored in the amplitude database server 205. Figure 6 As exemplarily shown, the ion signal analyzer 206 can detect the occurrence of a misfire and activate a misfire flag by comparing the amplitude of the frequency component with an amplitude threshold. Based on the misfire flag, the notification unit 207 generates a notification to inform the user of the engine 201 of the occurrence of a misfire.

[0043] Figure 10 An exemplary flowchart is shown, illustrating a method including steps for determining a misfire in an internal combustion engine 201. (As in...) Figure 2The method is exemplarily illustrated by an event detection system 200. In step 1001, during a spark event in spark plug 106, ion current measurement circuit 107 measures the voltage level of the ion current signals received from ignition coils 101 and 103 of IC engine 201. In step 1002, band limiter module 203 filters the received ion current signals to generate a predetermined number of band-filtered ion current signals. In step 1003, frequency domain conversion unit 204 converts the predetermined number of band-filtered ion current signals in the time domain to a predetermined number of digital ion current signals in the frequency domain. In step 1004, ion signal analyzer 206 analyzes the amplitude of each of the predetermined number of digital ion current signals using an amplitude threshold corresponding to the frequency band of each of the predetermined number of digital ion current signals stored in amplitude database server 205. Based on the above analysis, ion signal analyzer 206 generates a misfire indicator, indicating that a misfire has occurred in engine 201.

[0044] Figure 11 An exemplary flowchart is shown, including steps for determining a misfire in engine 201. In step 1101, ion current measurement circuit 107 measures and captures or stores the ion current signal in each engine cycle. In step 1102, the ion current signal is filtered through a cutoff frequency to remove higher frequencies, such as... Figure 3 The noise component in the f12-f16 Hz frequency band is analyzed. In step 1103, a bandpass filter 1 203a with a frequency band of f1 to f3 Hz is used. In step 1104, a bandpass filter 2203b with a frequency band of f3 to f5 Hz is used. In step 1105, a bandpass filter 3203c with a frequency band of f5 to f7 Hz is used. The ion current signal passes through bandpass filters 1 203a, 2203b, and 3203c. For each band-filtered ion current signal, a digital Fourier transform (DFT) is calculated using an FFT algorithm in steps 1106, 1107, and 1108 to convert the time-domain band-filtered ion current signal to a frequency-domain digital ion current signal. The ion signal analyzer 206 performs a comparison check on the threshold conditions of each digital ion current signal. As disclosed in the detailed description above, the amplitude threshold is determined by... Figures 7 to 9 Confirmed. In step 1109, if the peak amplitude of the frequency component of the digital ion current signal is higher than A9 at f1 Hz and higher than A4 in the range of f1 Hz to f3 Hz, then the digital ion current signal in the frequency band f1 Hz to f3 Hz is checked. If yes, in step 1113, it is confirmed that a misfire has been detected by the ion signal analyzer 206, and a fault indicator light (MIL) is generated by the notification unit 207. If no, in step 1112, the ion signal analyzer 206 confirms that no misfire has been detected.

[0045] In step 1110, if the amplitude of the frequency component of the digital ion current signal is higher than A3 in the range of f5 Hz to f7 Hz, the digital ion current signal in the frequency band f5 Hz to f7 Hz is checked. If yes, in step 1113, the ion signal analyzer 206 confirms that an off-fire has been detected, and the notification unit 207 generates a MIL indication. If the answer to the amplitude comparison is "no", in step 1112, the ion signal analyzer 206 confirms that no off-fire has been detected. In step 1111, if there are multiple peaks with amplitudes greater than A7 in the range of f3 Hz to f5 Hz, the digital ion current signal in the frequency band f3 Hz to f5 Hz is checked. If the answer to the amplitude comparison is "yes", in step 1113, the ion signal analyzer 206 detects an off-fire, and the notification unit 207 generates a MIL indication. If the answer to the amplitude comparison is "no", in step 1112, the ion signal analyzer 206 detects no off-fire.

[0046] In this method, an ignition is detected if any filtered ion current signal meets a threshold condition. However, to improve the accuracy of the method for determining the occurrence of an ignition, such as... Figure 12 As exemplarily shown, a fire can be detected if and only if the threshold conditions for all three frequency bands (N) are met.

[0047] Figure 12 An exemplary flowchart is shown, including steps for determining a misfire in engine 201. In step 1201, ion current measurement circuit 107 ensures and captures the ion current signal in each engine cycle. In step 1202, the ion current signal is filtered through a cutoff frequency to remove higher frequencies, such as those in... Figure 3 The noise component in the f12-f16 Hz frequency band is analyzed. In step 1203, a bandpass filter 1 203a with a frequency band of f1 to f3 Hz is used. In step 1204, a bandpass filter 2203b with a frequency band of f3 to f5 Hz is used. In step 1205, a bandpass filter 3203c with a frequency band of f5 to f7 Hz is used. The ion current signal passes through bandpass filters 1 203a, 2203b, and 3203c. For each band-filtered ion current signal, a digital Fourier transform (DFT) is calculated using an FFT algorithm in steps 1206, 1207, and 1208 to convert the time-domain band-filtered ion current signal to a frequency-domain digital ion current signal. The ion signal analyzer 206 performs a check of threshold conditions for each ion current signal. As disclosed in the detailed description above, the amplitude threshold is determined by… Figures 7 to 9Confirmed. In step 1209, if the peak amplitude of the frequency component of the digital ion current signal is higher than A9 at f1 Hz and higher than A4 in the range of f1 Hz to f3 Hz, then the digital ion current signal within the frequency band f1 Hz to f3 Hz is checked and compared. If yes, then the threshold condition for the frequency band f3 Hz to f5 Hz in step 1211 is checked by the ion signal analyzer 206. If the condition in step 1209 is "no", then in step 1212, the ion signal analyzer 206 determines and infers that no misfire was detected.

[0048] In step 1211, if multiple peaks with amplitudes greater than A7 exist from f3 Hz to f5 Hz, the digital ion current signal in the frequency band f3 Hz to f5 Hz is checked and compared. If the condition in step 1211 is "yes", the threshold condition for the frequency band f5 Hz to f7 Hz in step 1210 is checked by the ion signal analyzer 206. If the condition in step 1211 is "no", the ion signal analyzer 206 detects no misfire in step 1212. Therefore, 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 in step 121. In step 1210, if the amplitude of the frequency component of the digital ion current signal is higher than A3 between f5 Hz and f7 Hz, the digital ion current signal in the frequency band f5 Hz to f7 Hz is checked and compared. If the condition in step 1210 is "yes", in step 1213, the ion signal analyzer 206 detects a misfire, and the notification unit 207 generates a MIL indication. If the condition in step 1210 is "no", the ion signal analyzer 206 detects in step 1212 that no misfire was detected. Therefore, 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 was detected in step 1212.

[0049] This invention provides the following technical improvements in the field of on-board diagnostics for IC engines: It offers a novel method and circuit that, through digital signal processing technology, analyzes ion current signals in the frequency domain, stores amplitude threshold data in hardware for analysis and comparison, and then infers the misfire state to detect misfires and partial misfires in spark-ignition engines. By comparing ion current signals during "misfire and no-misfire states," basic information is extracted in the frequency domain, and appropriate threshold conditions are preset to distinguish between misfires and normal combustion. This invention overcomes the shortcomings of known technologies, ensuring accurate detection of misfires, thereby enabling reliable control and reduction of emissions, especially in small-capacity engines.

[0050] Furthermore, spark events are very rapid phenomena, lasting up to a few milliseconds. The ion current signal flows through the circuit quickly, and the spark ends instantaneously. Therefore, to collect a large amount of data related to spark events in a short time, a frequency domain conversion unit with a very high sampling rate is used. The band limiter module is selected based on the number of bands identified by the band determination unit. The band limiter module filters out signals within the defined band, thus allowing for accurate comparison of the digital ion current signals in each band. Simultaneously, intelligent band configuration ensures that the digital ion current signal exhibits significant statistical changes under both misfire and no-misfire conditions. Engine speed and the voltage on the primary side of the ignition coil are used as references to initiate misfire detection by the band limiter module, the frequency domain conversion unit, and the ion signal analyzer. These two signals are used to improve the reliability of the event detection system. If the voltage on the primary side of the ignition coil is not detected due to some fault, the engine speed is used as a real-time trigger for the event detection system.

[0051] Misfire detection based on ion current measurement allows engine users to understand misfire events in the vehicle's spark-ignition engine in real time, and improves vehicle performance by investigating and correcting the causes of misfires, thereby enhancing reliability, durability, vehicle mileage, and the comfort provided to vehicle users. Furthermore, timely detection and correction of misfires in the engine reduces the degradation of catalyst and lambda performance in the vehicle.

[0052] In addition to detecting misfires, the event detection system of the present invention can perform knock detection, determine spark plug timing errors and perform spark duration measurement, perform combustion quality analysis in the engine, and use ion current sensing to assist in spark plug maintenance of the engine.

[0053] Improvements and modifications may be incorporated herein without departing from the scope of the invention.

[0054] Explanation of reference numerals in the attached figures

[0055] 100 Ignition System

[0056] 101 Primary side of ignition coil

[0057] 102 Battery positive terminal

[0058] 103 Secondary side of ignition coil

[0059] 104 Control Unit

[0060] 105 Switchgear

[0061] 106 Spark Plug

[0062] 107 Ion Current Measurement Circuit

[0063] 200 Event Detection System

[0064] 201 engine

[0065] 202 Frequency Band Determination Unit

[0066] 203 Band Limiter Module

[0067] 203a Bandpass Filter 1

[0068] 203b Bandpass Filter 2

[0069] 203c bandpass filter 3

[0070] 204 frequency domain conversion units

[0071] 204a FFT Block 1

[0072] 204b FFT block 2

[0073] 204c FFT block 3

[0074] 205 Amplitude Database Server

[0075] 206 Ion Signal Analyzer

[0076] 207 Notification Unit

Claims

1. A misfire event detection system (200) for an engine (201), the misfire event detection system (200) comprising: An ion current measurement circuit is used to measure the ion current signal in the ignition coils (101 and 103) generated during a spark event in the spark plug (106); A predetermined number (N) of bandwidth limiter modules (203) are communicatively connected to an ion current measurement circuit (107) for generating a predetermined number (N) of filtered ion current signals. and A predetermined number (N) of frequency domain conversion units (204) are communicatively connected to the predetermined number of frequency band limiter modules (203) for converting the predetermined number of band-filtered ion current signals into a predetermined number (N) of digital ion current signals in the frequency domain. and An ion signal analyzer (206) is communicatively connected to the predetermined number (N) frequency domain conversion units (204) and configured to analyze the amplitude of each of the predetermined number of digital ion current signals using an amplitude threshold of each of the predetermined number of digital ion current signals and compare the amplitude of each of the predetermined number of digital ion current signals with the amplitude threshold of each of the predetermined number of digital ion current signals to infer the occurrence of a misfire event in the engine (201).

2. The misfire event detection system (200) of claim 1, further comprising: A frequency band determination unit (202) is used to determine a predetermined number of frequency bands (N) in the measured ion current signal based on the change in voltage level of the measured ion current signal, wherein each of the predetermined number of frequency bands (N) corresponds to each of the predetermined number of frequency band limiter modules (203).

3. The fire incident detection system (200) according to claim 2, wherein, Each of the predetermined number of band limiter modules (203) corresponds to each of the predetermined number of band-filtered ion current signals.

4. The fire incident detection system (200) according to claim 3, wherein, Each of the predetermined number of filtered ion current signals corresponds to each of the predetermined number of digital ion current signals.

5. The fire incident detection system (200) according to claim 2, wherein, The predetermined number of frequency bands (N) is three frequency bands.

6. The fire incident detection system (200) according to claim 1, wherein, Each of the predetermined number (N) frequency domain conversion units (204) performs a Fourier transform on each of the predetermined number of filtered ion current signals in the time domain to obtain each of the predetermined number of digital ion current signals in the frequency domain.

7. The fire incident detection system (200) according to claim 6, wherein, The Fourier transform is an FFT algorithm applied to each of the predetermined number of filtered ion current signals.

8. The fire incident detection system (200) according to claim 1 further includes: The notification unit (207) is used to generate and notify the user of the engine (201) of the occurrence of the event based on the comparative analysis of the amplitude of each of the predetermined number of digital ion current signals by the ion signal analyzer (206) relative to a pre-stored amplitude threshold.

9. The fire incident detection system (200) according to claim 1 further includes: An amplitude database server (205) stores amplitude thresholds corresponding to the frequency bands in the measured ion current signal.

10. The fire incident detection system (200) according to claim 1, wherein, Each of the predetermined number of band limiter modules (203) is a bandpass filter having a corresponding low cutoff frequency and a corresponding high cutoff frequency.

11. A method for determining the occurrence of a misfire event in an engine (201), the method being implemented by a misfire event detection system (200), the misfire event detection system (200) comprising: Ion current measurement circuit (107); A predetermined number of bandwidth limiter modules (203) are communicatively connected to the ion current measurement circuit (107). A predetermined number of frequency domain conversion units (204) are communicatively connected to the predetermined number of frequency band limiter modules (203); and An ion signal analyzer (206) is communicatively connected to the predetermined number of frequency domain conversion units (204). The method includes the following steps: During a spark event in the spark plug (106), the voltage level of the ion current signal received from the ignition coils (101 and 103) of the engine (201) is measured by the ion current measuring circuit (107); The received ion current signal is filtered by the predetermined number of band limiter modules (203) to generate a predetermined number of band-filtered ion current signals; The predetermined number of filtered ion current signals are converted into a predetermined number of digital ion current signals in the frequency domain by the predetermined number of frequency domain conversion units (204); as well as The ion signal analyzer (206) analyzes the amplitude of each of the predetermined number of digital ion current signals using the amplitude threshold of the frequency band corresponding to each of the predetermined number of digital ion current signals and compares the amplitude of each of the predetermined number of digital ion current signals with the amplitude threshold of the frequency band corresponding to each of the predetermined number of digital ion current signals to determine the occurrence of the misfire event in the engine (201).

12. The method of claim 11, further comprising: A predetermined number of frequency bands in the measured ion current signal are determined based on the change in voltage level of the ion current signal measured by the frequency band determination unit (202) of the fire event detection system (200), and the amplitude threshold is stored in the amplitude threshold server (205).

13. The method according to claim 12, wherein, Each of the predetermined number of frequency bands corresponds to each of the predetermined number of frequency band limiter modules (203).

14. The method according to claim 13, wherein, Each of the predetermined number of band limiter modules (203) corresponds to each of the predetermined number of band-filtered ion current signals.

15. The method according to claim 14, wherein, Each of the predetermined number of filtered ion current signals corresponds to each of the predetermined number of digital ion current signals.

16. The method according to claim 12, wherein, The predetermined number of frequency bands (N) is three frequency bands.

17. The method according to claim 11, wherein, Converting the predetermined number of band-filtered ion current signals in the time domain to the frequency domain includes performing a Fourier transform on each of the predetermined number of band-filtered ion current signals.

18. The method according to claim 17, wherein, The Fourier transform is an FFT algorithm applied to each of the predetermined number of filtered ion current signals.

19. The method according to claim 11, wherein, The analysis of the amplitude of each of the predetermined number of digital ion current signals by the ion signal analyzer (206) includes: Determine the corresponding frequency band for each of the predetermined number of digital ion current signals; A lookup of the amplitude threshold corresponding to the determined frequency band of each of the predetermined number of digital ion current signals is performed in the amplitude database server (205); The amplitude of each of the predetermined number of digital ion current signals is compared with a corresponding amplitude threshold; and Based on the comparison, it is determined that the misfire event in the engine (201) has occurred.

20. The method of claim 11, further comprising: Based on the comparative analysis by the ion signal analyzer (206), the notification unit (207) of the misfire event detection system (200) generates and notifies the user of the engine (201) of the occurrence of the misfire event.

21. The method of claim 11, further comprising: The history of misfire incident fault information is stored in the electronic control unit (ECU) 105 for engine diagnostic analysis.

Images